index.xml

<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>MLO Lab</title><link>https://mlo-lab.github.io/</link><atom:link href="https://mlo-lab.github.io/index.xml" rel="self" type="application/rss+xml"/><description>MLO Lab</description><generator>Wowchemy (https://wowchemy.com)</generator><language>en-us</language><copyright>© 2026 MLO Lab</copyright><lastBuildDate>Tue, 24 Feb 2026 09:51:10 +0100</lastBuildDate><image><url>https://mlo-lab.github.io/media/logo_huedbb8239ffd36c55e33765ff7c80fb90_78702_300x300_fit_lanczos_2.png</url><title>MLO Lab</title><link>https://mlo-lab.github.io/</link></image><item><title>Florian presents at the 6th Rhein-Main Cancer Retreat</title><link>https://mlo-lab.github.io/post/confernce-rmcr-2026/</link><pubDate>Tue, 24 Feb 2026 09:51:10 +0100</pubDate><guid>https://mlo-lab.github.io/post/confernce-rmcr-2026/</guid><description>&lt;p>Florian represented our lab at the 6th Rhein-Main Cancer Retreat in Glashütten (23–24 February), where he presented our latest advances in AI-based modelling of multi-omics data.His talk highlighted how integrative machine learning approaches can transform complex molecular datasets into clinically actionable insights. By combining heterogeneous omics layers, our models aim to improve patient stratification, enhance model interpretability, and support the translation of computational discoveries into clinical practice.&lt;/p></description></item><item><title>Guest Talk | Ziqi Kang – Quantitative Spatial Analysis of the Tumor Microenvironment</title><link>https://mlo-lab.github.io/post/guest-talk-kang/</link><pubDate>Thu, 12 Feb 2026 10:13:40 +0100</pubDate><guid>https://mlo-lab.github.io/post/guest-talk-kang/</guid><description>&lt;p>On the 12th of February, we welcomed Ziqi Kang (PhD student, Färkkilä Lab, University of Helsinki) for an online guest talk on quantitative spatial analysis of the tumor microenvironment in high-grade serous ovarian cancer.
Ziqi presented SPACEstat, a method tackling a key bottleneck in spatial biology: generating robust, comparable spatial measurements across slides and patient cohorts from multiplexed spatial proteomics data (e.g., tCyCIF). The framework quantifies spatial organization in biologically interpretable ways — tumor–stroma interface architecture, immune infiltration patterns, cellular neighborhoods, and cell–cell interaction networks — enabling single-cell–level comparisons across samples.
The talk concluded with ongoing directions extending spatial analyses across multiple biological scales to better connect cellular interactions with tissue-level architecture.&lt;/p>
&lt;p>Thank you, Ziqi, for the presentation and the lively discussion!&lt;/p></description></item><item><title>Group Retreat 2026</title><link>https://mlo-lab.github.io/post/retreat-2026/</link><pubDate>Thu, 05 Feb 2026 08:47:12 +0100</pubDate><guid>https://mlo-lab.github.io/post/retreat-2026/</guid><description>&lt;p>For our 2026 group retreat, the entire team gathered at Bildungshaus Kloster Schöntal 03–05 February. Set in the beautiful monastery surroundings, the retreat gave us dedicated time to discuss organizational topics, exchange ideas on science, and deepen our understanding of our wide range of research areas.&lt;/p>
&lt;p>Beyond the sessions, the retreat was also about connection, with lively conversations and a friendly (and competitive) round of table tennis in the evenings. Overall, it was a wonderful opportunity to come together as a team, strengthen collaboration, and return with renewed focus and a shared perspective.&lt;/p></description></item><item><title>Insights into FLAME: Visit by Dr. Marius Herr</title><link>https://mlo-lab.github.io/post/flame-project-herr/</link><pubDate>Thu, 04 Dec 2025 13:58:55 +0100</pubDate><guid>https://mlo-lab.github.io/post/flame-project-herr/</guid><description>&lt;p>We hosted Dr. Marius Herr from the University of Tübingen, who presented FLAME — &lt;a href="https://docs.privateaim.net" target="_blank" rel="noopener">Federated Learning and Analyses in Medicine&lt;/a>, a platform developed within the privateAIM initiative. He outlined how patient data can be analyzed in a privacy-preserving manner and demonstrated how our methods could be applied to clinical data through FLAME.
The subsequent discussions provided practical insights into the requirements for clinically ready applications of such systems.&lt;/p>
&lt;p>We warmly thank Dr. Herr for his visit and the constructive exchange. We look forward to staying in contact and exploring potential collaborations as FLAME moves toward national implementation.&lt;/p></description></item><item><title>Welcome to our newest team member, Azza</title><link>https://mlo-lab.github.io/post/azza-welcome-to-the-team/</link><pubDate>Mon, 17 Nov 2025 10:38:11 +0100</pubDate><guid>https://mlo-lab.github.io/post/azza-welcome-to-the-team/</guid><description>&lt;p>Azza joins our ML sub-team as a doctoral researcher working on uncertainty quantification and estimation for trustworthy AI. She brings experience in applied machine learning and language model engineering and will support our ongoing research on reliable ML methods in oncology.&lt;/p></description></item><item><title>Our paper “Fine-Grained Uncertainty Decomposition in Large Language Models: A Spectral Approach” is now available at AAAI</title><link>https://mlo-lab.github.io/post/walha-uncertainty-decomposition/</link><pubDate>Sun, 16 Nov 2025 14:18:17 +0100</pubDate><guid>https://mlo-lab.github.io/post/walha-uncertainty-decomposition/</guid><description>&lt;p>This work presents Spectral Uncertainty, a new way to decompose uncertainty in large language models. Using the Von Neumann entropy, the method distinguishes aleatoric from epistemic uncertainty and incorporates detailed semantic structure in model outputs. Across multiple benchmarks, it outperforms current approaches in estimating uncertainty.&lt;/p></description></item><item><title>Florian named among the world’s most cited researchers — huge congrats to the whole MLO Lab team!</title><link>https://mlo-lab.github.io/post/buettner-most-cited-2025/</link><pubDate>Fri, 14 Nov 2025 14:41:04 +0100</pubDate><guid>https://mlo-lab.github.io/post/buettner-most-cited-2025/</guid><description>&lt;p>A proud moment: Florian has been listed as a Highly Cited Researcher 2025, placing him among the top 1% of scientists worldwide. This recognition reflects the shared work, ideas, and energy that move our lab forward every day. Proud to have this team! &lt;a href="https://aktuelles.uni-frankfurt.de/en/english/seven-goethe-university-researchers-among-the-most-cited-scientists-in-the-world/" target="_blank" rel="noopener">Official Announcement&lt;/a>&lt;/p></description></item><item><title>Welcome to our newest team member, Rashika</title><link>https://mlo-lab.github.io/post/rashika-welcome-to-the-team/</link><pubDate>Mon, 10 Nov 2025 14:43:30 +0100</pubDate><guid>https://mlo-lab.github.io/post/rashika-welcome-to-the-team/</guid><description>&lt;p>Rashika joins the MLO Lab as a doctoral researcher working on probabilistic machine learning for multi-omic data integration and foundation-model approaches for single-cell representation learning. She will contribute to our research on machine learning methods for cancer data analysis.&lt;/p></description></item><item><title>Welcome to the team, Hendrik</title><link>https://mlo-lab.github.io/post/hendrik-welcome-to-the-team/</link><pubDate>Sat, 18 Oct 2025 09:29:04 +0100</pubDate><guid>https://mlo-lab.github.io/post/hendrik-welcome-to-the-team/</guid><description>&lt;p>Hendrik joins our ML sub-team, working on machine learning methods that combine large-scale EHR data with other data types such as genomics. He focuses on causality and uncertainty estimation and will support our ongoing methodological work.&lt;/p></description></item><item><title>Welcome to our newest team member, Rashika</title><link>https://mlo-lab.github.io/post/leo-welcome-to-the-team/</link><pubDate>Thu, 02 Oct 2025 09:40:09 +0100</pubDate><guid>https://mlo-lab.github.io/post/leo-welcome-to-the-team/</guid><description>&lt;p>Leo joins our group as a shared doctoral researcher working on the integration and interpretation of spatial transcriptomics and spatial proteomics data. Their research focuses on spatial multi-omics analyses to study immunotherapy effects in glioblastoma.&lt;/p></description></item><item><title>Welcome to the team, Yusuf</title><link>https://mlo-lab.github.io/post/yusuf-welcome-to-the-team/</link><pubDate>Wed, 01 Oct 2025 09:36:24 +0100</pubDate><guid>https://mlo-lab.github.io/post/yusuf-welcome-to-the-team/</guid><description>&lt;p>Yussuf joins our bioinformatics sub-team as a doctoral researcher working on interpretable probabilistic machine learning models for multimodal data integration. He applies these methods to projects on novel mRNA technologies for colorectal and pancreatic cancer and will support our ongoing research efforts.&lt;/p></description></item><item><title>Quantitative Imaging in Oncology</title><link>https://mlo-lab.github.io/project/medical-imaging/</link><pubDate>Wed, 24 Sep 2025 15:55:09 +0200</pubDate><guid>https://mlo-lab.github.io/project/medical-imaging/</guid><description>&lt;p>Quantitative imaging in oncology combines advanced microscopy, image analysis and machine learning to study cancer in unprecedented detail. By extracting rich spatial and molecular information from tissues, we aim to better understand how tumors grow, respond to treatments and interact with their surroundings. In our lab, we have developed colocatome frameworks to map and quantify in situ cellular organization, revealing how microenvironments regulate cell behavior. In parallel, we extract reproducible radiomic features such as texture, intensity and shape, together with quantitative MRI metrics like relaxation times, and apply interpretable machine-learning models to link these imaging biomarkers to clinical outcomes. This enables precise tumor localization and non-invasive monitoring of disease progression.&lt;/p>
&lt;h3 id="quantitative-imaging-and-spatial-analysis">Quantitative Imaging and Spatial Analysis&lt;/h3>
&lt;p>Our work focuses on developing computational frameworks to analyze high-resolution multiplex microscopy images of tissues such as bone marrow, generated by &lt;a href="https://www.kokkaliarislab.com/" target="_blank" rel="noopener">Quantitative Spatial Cancer Biology - Kokkaliaris lab&lt;/a>. We specialize in extracting spatial and morphological features from complex, multi-modal image data.&lt;/p>
&lt;p>We have developed a framework that enables the integration and analysis of multiple biological replicates with complementary information in a shared spatial reference space. Building on this, we established a pipeline to extract and quantify spatial remodeling of the cellular neighborhood during the aging process. These tools allow us to investigate how hematopoietic stem cells, blood vessels, megakaryocytes, adipocytes, and stromal components are spatially organized and how these patterns evolve with age or in response to treatment.&lt;/p>
&lt;p>&lt;img src="bone_marrow.avif" alt="Bone Marrow Tissue">&lt;/p>
&lt;h3 id="radiomics-and-quantitative-mri">Radiomics and Quantitative MRI&lt;/h3>
&lt;p>Our imaging research also includes radiological data analysis, with a focus on transitioning from qualitative interpretation to quantitative, reproducible metrics. In radiomics, we extract features such as texture, intensity, and shape from defined regions of interest within MRI scans, and use machine learning models to link these features to clinical outcomes. We emphasize interpretability and robustness, testing models on controlled environments to avoid confounding due to real-patient variability and ethical exposure limitations.&lt;/p>
&lt;p>In parallel, we employ quantitative MRI (qMRI) techniques to move beyond traditional contrast-based imaging. By measuring intrinsic physical properties of tissues, e.g. relaxation times, we obtain microstructural insights into tissue composition, particularly in brain imaging. Combining qMRI with machine learning enables precise localization of tumors and assessment of disease progression, further enhancing the clinical utility of radiological data.&lt;/p></description></item><item><title>Our paper “Learning interpretable representations of single-cell multi-omics data with multi-output Gaussian processes” has been published in Nucleic Acids Research.</title><link>https://mlo-lab.github.io/post/moslehi-multioutput-gp/</link><pubDate>Tue, 12 Aug 2025 16:11:09 +0100</pubDate><guid>https://mlo-lab.github.io/post/moslehi-multioutput-gp/</guid><description>&lt;p>We present a unified framework that combines expressive neural embeddings with interpretable multi-output Gaussian processes for single-cell genomics. Joint representations of cells and genes reveal meaningful links between cell clusters and their marker genes via an interpretable gene-relevance map. &lt;a href="https://academic.oup.com/nar/article/53/14/gkaf630/8210588" target="_blank" rel="noopener">Published in Nucleic Acids Research&lt;/a>.&lt;/p></description></item><item><title>An autonomous agent for auditing and improving the reliability of clinical AI models — now published.</title><link>https://mlo-lab.github.io/post/kuhn-autonomous-agent-clinical-ai/</link><pubDate>Tue, 08 Jul 2025 14:41:12 +0100</pubDate><guid>https://mlo-lab.github.io/post/kuhn-autonomous-agent-clinical-ai/</guid><description>&lt;p>We introduce ModelAuditor, a self-reflective agent that simulates clinically relevant distribution shifts and produces interpretable reports on likely failure modes. Across multiple medical imaging domains, it recovers up to 25% of performance lost under shift while providing actionable deployment insights.&lt;/p></description></item><item><title>Learning interpretable representations of single-cell multi-omics data with multi-output Gaussian processes</title><link>https://mlo-lab.github.io/publication/10-1093-nargkaf-630/</link><pubDate>Tue, 01 Jul 2025 00:00:00 +0000</pubDate><guid>https://mlo-lab.github.io/publication/10-1093-nargkaf-630/</guid><description/></item><item><title>Trustworthy Machine Learning in Biomedical Research</title><link>https://mlo-lab.github.io/project/trustworthy-ml/</link><pubDate>Tue, 10 Jun 2025 22:46:01 +0200</pubDate><guid>https://mlo-lab.github.io/project/trustworthy-ml/</guid><description>&lt;p>As machine learning becomes increasingly central to biomedical discovery and clinical decision-making, ensuring the reliability, fairness, and interpretability of these models is critical. In our lab, we are committed to developing and applying machine learning methods that are not only accurate but also &lt;strong>trustworthy&lt;/strong>, meaning they are robust to noise, generalizable across datasets, transparent in their decision-making, and aligned with ethical and clinical standards.&lt;/p>
&lt;p>Our work spans multiple aspects of trustworthy ML, including uncertainty quantification, model calibration, interpretability, fairness in predictive models, and robustness to distributional shifts. These components are especially important in healthcare, where decisions influenced by models can have direct consequences for patients.&lt;/p>
&lt;p>In the context of multi-omics data, single-cell analysis, and quantitative imaging, we embed trustworthiness principles throughout the model development pipeline, from data preprocessing and integration to prediction and interpretation. This ensures that our computational outputs can be confidently used to guide biological insight and translational applications.&lt;/p>
&lt;h3 id="model-calibration-under-distribution-shift">Model Calibration Under Distribution Shift&lt;/h3>
&lt;p>Current-generation neural networks exhibit systematic underconfidence rather than the overconfidence reported in earlier models, and demonstrate improved calibration robustness under distribution shift. However, post-hoc calibration methods become less effective or even detrimental under severe shifts. Our analysis across ImageNet and biomedical datasets reveals that calibration insights from web-scraped benchmarks have limited transferability to specialized domains, where convolutional architectures consistently outperform transformers regardless of model generation. This work challenges established calibration paradigms and emphasizes the need for domain-specific architectural evaluation beyond standard benchmarks.
&lt;img src="distribution_shift_post-hoc.png" alt="Distribution Shift Calibration">&lt;/p>
&lt;p>[&lt;a href="https://arxiv.org/abs/2506.09593" target="_blank" rel="noopener">pdf&lt;/a>, &lt;a href="https://github.com/MLO-lab/ModelTransformer" target="_blank" rel="noopener">repo&lt;/a>]&lt;/p>
&lt;h3 id="uncertainty-quantification-for-classification-and-applications">Uncertainty Quantification for Classification and Applications&lt;/h3>
&lt;p>Reliably estimating the uncertainty of a prediction throughout the model lifecycle is crucial in many safety-critical applications. Since ML-based decision models are increasingly deployed in dynamic environments, understanding when and why a model might fail becomes as important as achieving accurate predictive performance. In our group, we focus on developing theoretically sounded methods for uncertainty quantification that remain robust across different applications, enabling more trustworthy and transparent AI systems.
&lt;img src="BI_plot.png" alt="BI">&lt;/p>
&lt;h4 id="uncertainty-estimates-of-predictions-via-a-general-bias-variance-decomposition-aistats-2023">Uncertainty Estimates of Predictions via a General Bias-Variance Decomposition (AISTATS 2023)&lt;/h4>
&lt;p>Proper scoring rules (e.g., Brier score or negative log-likelihood) are commonly used as loss functions in machine learning, as they are designed to assign optimal predictions to the target distribution. However, it remains unclear how to decompose these scores in a way that a component capturing the model’s predictive uncertainty arises. To address this, we derive a general bias-variance decomposition for proper scoring rules, where the Bregman Information (BI) naturally emerges as the variance term. This new theoretical insight has practical implications for classification tasks: since the decomposition applies to the cross-entropy loss, it allows us to quantify predictive uncertainty directly in the logit space (the standard output of neural networks) without requiring a normalisation step. Extensive empirical results demonstrate the effectiveness and robustness of this method, particularly in out-of-distribution settings. [&lt;a href="https://proceedings.mlr.press/v206/gruber23a/gruber23a.pdf" target="_blank" rel="noopener">pdf&lt;/a>, &lt;a href="https://github.com/MLO-lab/Uncertainty_Estimates_via_BVD" target="_blank" rel="noopener">repo&lt;/a>]&lt;/p>
&lt;h4 id="how-to-leverage-predictive-uncertainty-estimates-for-reducing-catastrophic-forgetting-in-online-continual-learning-tmlr-2025">How to Leverage Predictive Uncertainty Estimates for Reducing Catastrophic Forgetting in Online Continual Learning (TMLR 2025)&lt;/h4>
&lt;p>In many real-world scenarios, we want models to continuously learn new information without forgetting what they already know. In memory-based online continual learning, a key challenge is managing a limited memory buffer to mitigate catastrophic forgetting (CF) — but what is the best strategy for selecting samples to store in the memory? Under an uncertainty lens, we investigate what characteristics make samples effective in alleviating CF. Starting from the examination of the properties and behaviours of popular uncertainty estimates, we identify that they mostly capture the irreducible aleatoric uncertainty and hypothesise that a better strategy should focus on the epistemic uncertainty instead. To this end, we propose using Bregman Information – derived from our general bias-variance decomposition of strictly proper scores – as an effective estimator of epistemic uncertainty, leading to improved memory population strategy and reduced forgetting. [&lt;a href="https://openreview.net/pdf?id=dczXe0S1oL" target="_blank" rel="noopener">pdf&lt;/a>, &lt;a href="https://github.com/MLO-lab/uncertainty_estimates_for_CF" target="_blank" rel="noopener">repo&lt;/a>]&lt;/p>
&lt;h4 id="federated-continual-learning-goes-online-uncertainty-aware-memory-management-for-vision-tasks-and-beyond--iclr-2025">Federated Continual Learning Goes Online: Uncertainty-Aware Memory Management for Vision Tasks and Beyond (ICLR 2025)&lt;/h4>
&lt;p>Federated Continual Learning (FCL) is a powerful paradigm that combines the privacy-preserving benefits of Federated Learning (FL) with the ability to learn sequentially over time, as in Continual Learning (CL). However, catastrophic forgetting still remains a major challenge. Most existing FCL methods rely on generative models, assuming an offline setting where all task data are available beforehand. But in real-world applications, data often arrives sequentially in small chunks — a challenge that remains largely unaddressed. To address this, we introduce a novel framework for online federated continual learning. To address scenarios where storing the full dataset locally is impractical, we propose an effective memory-based baseline that integrates uncertainty-aware updates — based on Bregman Information — with random replay to reduce catastrophic forgetting. Unlike generative approaches, our uncertainty-based solution is simple to implement and adaptable across different data modalities. [&lt;a href="https://openreview.net/pdf?id=f65RuQgVlp" target="_blank" rel="noopener">pdf&lt;/a>, &lt;a href="https://github.com/MLO-lab/online-FCL" target="_blank" rel="noopener">repo&lt;/a>]&lt;/p>
&lt;h3 id="uncertainty-quantification-for-generative-ai">Uncertainty Quantification for Generative AI&lt;/h3>
&lt;p>Generative AI models in general and large language models in particular have emerged as a disruptive technology that has been rapidly democratized. Their use in critical domains such as medicine, scientific research, and politics has raised serious concerns about reliability. Consequently, robust estimation of their uncertainty is essential to build trust and to prevent the potentially severe consequences of failures. We develop methods to quantify and calibrate their uncertainty across different modalities, while accounting for the specific characteristics of each modality.
&lt;img src="UQ_LLM_plot.png" alt="UQ_LLM">&lt;/p>
&lt;h4 id="a-bias-variance-covariance-decomposition-of-kernel-scores-for-generative-models-icml-2024">A Bias-Variance-Covariance Decomposition of Kernel Scores for Generative Models (ICML 2024)&lt;/h4>
&lt;p>This paper tackles a core gap in generative AI: there’s no unified, theory-grounded way to assess generalization and uncertainty across modalities or closed-source models. We introduce the first bias–variance–covariance decomposition for kernel scores, yielding kernel-based measures that can be estimated directly from generated samples, without access to the underlying model. Because kernels work from samples alone and are computed based on vector representations of these samples, the framework applies uniformly to images, audio, and language. In experiments, the approach explains generalization behavior (including mode collapse patterns) and delivers stronger uncertainty signals, even for closed-source LLMs.&lt;/p>
&lt;p>[&lt;a href="https://proceedings.mlr.press/v235/gruber24a.html" target="_blank" rel="noopener">pdf&lt;/a>, &lt;a href="https://github.com/MLO-lab/BVCD_generative_models" target="_blank" rel="noopener">repo&lt;/a>]&lt;/p></description></item><item><title>Computational Molecular Medicine</title><link>https://mlo-lab.github.io/project/comp-medicine/</link><pubDate>Tue, 10 Jun 2025 22:36:09 +0200</pubDate><guid>https://mlo-lab.github.io/project/comp-medicine/</guid><description>&lt;p>Single-cell multi-omics data offer powerful opportunities to study disease at unprecedented resolution, but they also present significant challenges. The data are often sparse, noisy, and extremely high-dimensional, with technical differences between batches or donors that can obscure true biological signals. Our lab combines advanced computational methods with high-dimensional biological data to exludes technical artifacts and uncover mechanisms of disease progresssion and therapy response, with a particular emphasis on cancer and metabolic disorder.&lt;/p>
&lt;h3 id="single-cell-multi-omics-for-clinical-cohorts">Single-Cell Multi-Omics for Clinical Cohorts&lt;/h3>
&lt;p>We specialize in the analysis of clinical single-cell multi-omics datasets, particularly from cancer and metabolic disease cohorts. Using state-of-the-art machine learning techniques, we analyze single-cell RNA sequencing and chromatin accessibility data to characterize cellular heterogeneity and regulatory dynamics. Our pipeline includes robust dimensionality reduction, clustering, and batch correction methods, allowing us to identify distinct cell populations and states across individuals. Through probabilistic graphical modeling and motif enrichment analysis, we reconstruct gene regulatory networks that govern disease-specific transcriptional programs. These approaches allow us to overcome the sparsity and noise inherent to single-cell data and extract biologically meaningful patterns that inform prognosis and therapeutic strategy.&lt;/p>
&lt;h3 id="multi-omics-for-mouse-models-of-cancer-progression">Multi-Omics for Mouse Models of Cancer Progression&lt;/h3>
&lt;p>To explore tumor development and treatment effects in vivo, we analyze multi-omics data generated from mouse models, including xenografts and genetically induced cancers. These datasets are complex, encompassing multiple axes of variation such as treatment regimens, time points, and tumor subtypes. To disentangle these factors, we develop tailored probabilistic latent variable models (LVMs) that reveal how sources of variability interact and which molecular features are relevant to human disease.
Recent projects:&lt;/p>
&lt;ul>
&lt;li>
&lt;p>Multi-Omics combined with &lt;strong>lineage tracing&lt;/strong> technology now allow us to quantify the clonal connectivity between different cell populations and infer the temporal dynamics of cell populations. Using mechanistic modeling, we can uncover the directionality of differentiation trajectories and the dynamical properties of the clones [&lt;a href="https://www.biorxiv.org/content/10.1101/2025.09.10.674954v1.full.pdf" target="_blank" rel="noopener">pdf&lt;/a>].&lt;/p>
&lt;/li>
&lt;li>
&lt;p>Analyzing cancerous mouse models provides valuable insights for developing &lt;strong>personalized oncology&lt;/strong> approaches. We will integrate patient data at an early stage in this process using a forward and reverse translation technique. This method ensures that the results are clinically relevant and enables us to identify patients who are eligible for a new treatment. For example, we are working on a TRR project that investigates how ubiquitination impacts DNA damage repair in AML in order to identify a new anticancer target [&lt;a href="https://ubiqancer.de/project/a13/" target="_blank" rel="noopener">project&lt;/a>].&lt;/p>
&lt;/li>
&lt;/ul></description></item><item><title>Interpretable Integration of Multi-Omics Data</title><link>https://mlo-lab.github.io/project/multi-omics/</link><pubDate>Tue, 10 Jun 2025 22:11:00 +0200</pubDate><guid>https://mlo-lab.github.io/project/multi-omics/</guid><description>&lt;p>Understanding the complexity of cancer requires methods that can integrate equally complex biological data. In our lab, we are committed to developing probabilistic models that bring together multiple molecular layers, including genomics, epigenomics, transcriptomics, proteomics and metabolomics, to provide a holistic view of each patient. These models uncover hidden structure by capturing both shared and modality-specific variation, allowing us to reduce noise and reveal biologically meaningful patterns. By modeling system-level responses to perturbations such as drug treatments or environmental changes, we aim to generate representations that are not only statistically robust but also interpretable, enabling new biological insights that can be directly validated and translated into clinical understanding.&lt;/p>
&lt;h3 id="muvi">MuVI&lt;/h3>
&lt;p>MuVI is a general-purpose probabilistic latent variable model for multi-omics integration that incorporates prior biological knowledge into its structure. It uses pathway annotations, gene sets, or cell-type signatures to guide the discovery of latent factors that explain variation across different data types. Even when this prior knowledge is noisy or incomplete, MuVI is able to learn biologically relevant dimensions, enabling scientists to interpret the sources of variation in the data more clearly and to relate them to known mechanisms.
&lt;img src="muvi.png" alt="MuVI">&lt;/p>
&lt;h3 id="music">MUSIC&lt;/h3>
&lt;p>MUSIC (MUltiview baySIan tensor deComposition) extends probabilistic modeling to high-dimensional array data, such as time-series or condition-specific measurements. It jointly decomposes collections of heterogeneous tensors, e.g. patient × gene × time or patient × protein × condition, into shared and modality-specific components. With structured sparsity priors and efficient variational inference, MUSIC scales to large datasets, handles missing data, and yields interpretable embeddings. We have applied it to cancer drug-response studies and single-cell leukemia data, where it revealed meaningful molecular signatures associated with disease pathways.&lt;/p>
&lt;h3 id="momo-gp">MOMO-GP&lt;/h3>
&lt;p>MOMO-GP (Multi-Omic Multi-output Gaussian Processes) addresses the challenge of learning interpretable representations from single-cell multi-omics data, which are typically high-dimensional, sparse, and nonlinear. Unlike traditional methods that trade off interpretability for modeling power, MOMO-GP combines neural networks with Gaussian Processes to achieve both. It learns separate latent embeddings for cells and features, as well as shared and modality-specific components in the multi-view setting. By modeling gene relevance explicitly, MOMO-GP connects cell clusters to marker genes, making the learned structure readily interpretable in biological terms.&lt;/p>
&lt;h3 id="joana">JOANA&lt;/h3>
&lt;p>JOANA is a probabilistic model for pathway enrichment analysis (PEA) that overcomes limitations of classical approaches like Over-Representation Analysis (ORA) and Functional Class Scoring (FCS). While methods such as GSEA work with continuous scores, they typically operate on a single omics layer and can yield overly broad sets of enriched pathways. JOANA improves on this by modeling enrichment scores across multiple omics layers using mixtures of beta distributions within a Bayesian framework. This allows it to estimate the probability of pathway enrichment both within and across modalities, yielding higher precision and more biologically relevant results.&lt;/p>
&lt;h3 id="mofaflex">MOFAFLEX&lt;/h3>
&lt;p>MOFAFLEX is our upcoming framework for flexible and interpretable multi-omics integration. Designed to generalize the principles behind models like MuVI and MUSIC, MOFAFLEX supports heterogeneous data types, modular priors, and scalable inference. Its architecture allows for tailored modeling of real-world datasets, balancing interpretability with modeling flexibility. MOFAFLEX is currently under active development and will provide a unified foundation for future applications in cancer biology and beyond.
&lt;img src="mofaflex.png" alt="MOFA-FLEX">&lt;/p></description></item><item><title>Application-driven validation of posteriors in inverse problems, published in Medical Image Analysis.</title><link>https://mlo-lab.github.io/post/buettner-posterior-inverse-problems/</link><pubDate>Tue, 01 Apr 2025 15:10:20 +0100</pubDate><guid>https://mlo-lab.github.io/post/buettner-posterior-inverse-problems/</guid><description>&lt;p>We present the first systematic framework for application-driven validation of posterior-based methods in inverse problems. Adapting concepts from object detection enables mode-centric validation with interpretable, application-focused metrics, demonstrated on multiple medical imaging use cases. Published in Medical Image Analysis.&lt;/p></description></item><item><title>Our paper on bidirectional human-AI visual alignment is out at the ICLR 2025 Workshop!</title><link>https://mlo-lab.github.io/post/buettner-lvlm-aided-visiual-alignment/</link><pubDate>Thu, 06 Mar 2025 15:34:56 +0100</pubDate><guid>https://mlo-lab.github.io/post/buettner-lvlm-aided-visiual-alignment/</guid><description>&lt;p>We introduce LVLM-Aided Visual Alignment (LVLM-VA), which aligns small vision models with human domain knowledge using large vision-language models. A bidirectional interface translates model behavior into natural language and expert instructions into image-level critiques, improving performance while reducing fine-grained feedback needs. Published at the ICLR 2025 Workshop on Bidirectional Human-AI Alignment.&lt;/p></description></item><item><title>Forget forgetting — our TMLR paper shows how uncertainty helps models keep their memory straight!</title><link>https://mlo-lab.github.io/post/serra-predictive-uncertainty-catastrophic-forgetting/</link><pubDate>Tue, 04 Mar 2025 15:32:51 +0100</pubDate><guid>https://mlo-lab.github.io/post/serra-predictive-uncertainty-catastrophic-forgetting/</guid><description>&lt;p>We analyze how predictive uncertainty can guide memory management to mitigate catastrophic forgetting and introduce a generalized-variance–based uncertainty measure. Uncertainty-aware sampling improves retention across tasks. Published in the Journal of Transactions on Machine Learning Research.&lt;/p></description></item><item><title>Incremental Uncertainty-aware Performance Monitoring with Active Labeling Intervention</title><link>https://mlo-lab.github.io/publication/pmlr-v-258-koebler-25-a/</link><pubDate>Sat, 01 Mar 2025 00:00:00 +0000</pubDate><guid>https://mlo-lab.github.io/publication/pmlr-v-258-koebler-25-a/</guid><description/></item><item><title>New at AISTATS: IUPM — a label-free method for reliable model monitoring under drift.</title><link>https://mlo-lab.github.io/post/buettner-iupm-intervention/</link><pubDate>Wed, 22 Jan 2025 15:47:44 +0100</pubDate><guid>https://mlo-lab.github.io/post/buettner-iupm-intervention/</guid><description>&lt;p>We propose IUPM, a label-free method for tracking performance under gradual distribution shifts using optimal transport. IUPM quantifies uncertainty in its estimates and guides targeted labeling to restore reliability, outperforming existing baselines across scenarios. Published in the Proceedings of the 28th International Conference on Artificial Intelligence and Statistics (AISTATS).&lt;/p></description></item><item><title>Our ICLR paper proves that not everything needs to be forgotten — tackling catastrophic forgetting head-on!</title><link>https://mlo-lab.github.io/post/serra-continous-learning-memory-managment/</link><pubDate>Wed, 22 Jan 2025 15:24:57 +0100</pubDate><guid>https://mlo-lab.github.io/post/serra-continous-learning-memory-managment/</guid><description>&lt;p>We propose an uncertainty-aware memory-based approach for online Federated Continual Learning. Using a Bregman Information estimator to guide selective replay, the method reduces catastrophic forgetting across modalities while preserving privacy and communication efficiency. Presented at the Thirteenth International Conference on Learning Representations (ICLR).&lt;/p></description></item><item><title>Unsupervised and efficient — our latest work exposes and mitigates shortcut learning!</title><link>https://mlo-lab.github.io/post/kuhn-unsupervised-shortcut-transformers/</link><pubDate>Wed, 01 Jan 2025 15:16:16 +0100</pubDate><guid>https://mlo-lab.github.io/post/kuhn-unsupervised-shortcut-transformers/</guid><description>&lt;p>We introduce an unsupervised framework to detect and mitigate shortcut learning in transformers. The method improves both worst-group and average accuracy while reducing annotation effort, offering interpretable insights for experts and running efficiently on consumer hardware.&lt;/p></description></item><item><title>An autonomous agent for auditing and improving the reliability of clinical AI models</title><link>https://mlo-lab.github.io/publication/kuhn-2025-autonomousagentauditingimproving/</link><pubDate>Wed, 01 Jan 2025 00:00:00 +0000</pubDate><guid>https://mlo-lab.github.io/publication/kuhn-2025-autonomousagentauditingimproving/</guid><description/></item><item><title>Application-driven validation of posteriors in inverse problems</title><link>https://mlo-lab.github.io/publication/adler-2025-application/</link><pubDate>Wed, 01 Jan 2025 00:00:00 +0000</pubDate><guid>https://mlo-lab.github.io/publication/adler-2025-application/</guid><description/></item><item><title>Decoding heart failure subtypes with neural networks via differential explanation analysis</title><link>https://mlo-lab.github.io/publication/ruz-2025-decoding/</link><pubDate>Wed, 01 Jan 2025 00:00:00 +0000</pubDate><guid>https://mlo-lab.github.io/publication/ruz-2025-decoding/</guid><description/></item><item><title>Efficient Unsupervised Shortcut Learning Detection and Mitigation in Transformers</title><link>https://mlo-lab.github.io/publication/kuhn-2025-efficientunsupervisedshortcutlearning/</link><pubDate>Wed, 01 Jan 2025 00:00:00 +0000</pubDate><guid>https://mlo-lab.github.io/publication/kuhn-2025-efficientunsupervisedshortcutlearning/</guid><description/></item><item><title>Federated Continual Learning Goes Online: Uncertainty-Aware Memory Management for Vision Tasks and Beyond</title><link>https://mlo-lab.github.io/publication/serra-2025-federated/</link><pubDate>Wed, 01 Jan 2025 00:00:00 +0000</pubDate><guid>https://mlo-lab.github.io/publication/serra-2025-federated/</guid><description/></item><item><title>Fine-Grained Uncertainty Decomposition in Large Language Models: A Spectral Approach</title><link>https://mlo-lab.github.io/publication/walha-2025-fine/</link><pubDate>Wed, 01 Jan 2025 00:00:00 +0000</pubDate><guid>https://mlo-lab.github.io/publication/walha-2025-fine/</guid><description/></item><item><title>How to Leverage Predictive Uncertainty Estimates for Reducing Catastrophic Forgetting in Online Continual Learning</title><link>https://mlo-lab.github.io/publication/serra-2025-how/</link><pubDate>Wed, 01 Jan 2025 00:00:00 +0000</pubDate><guid>https://mlo-lab.github.io/publication/serra-2025-how/</guid><description/></item><item><title>Improving Perturbation-based Explanations by Understanding the Role of Uncertainty Calibration</title><link>https://mlo-lab.github.io/publication/decker-2025-improving/</link><pubDate>Wed, 01 Jan 2025 00:00:00 +0000</pubDate><guid>https://mlo-lab.github.io/publication/decker-2025-improving/</guid><description/></item><item><title>Towards LVLM-Aided Alignment of Task-Specific Vision Models</title><link>https://mlo-lab.github.io/publication/koebler-2025-towards/</link><pubDate>Wed, 01 Jan 2025 00:00:00 +0000</pubDate><guid>https://mlo-lab.github.io/publication/koebler-2025-towards/</guid><description/></item><item><title>Our latest work on deep learning for metabolomics just appeared in Scientific Reports!</title><link>https://mlo-lab.github.io/post/buettner-metabolic-changes-graph-embeddings/</link><pubDate>Thu, 28 Nov 2024 13:26:13 +0100</pubDate><guid>https://mlo-lab.github.io/post/buettner-metabolic-changes-graph-embeddings/</guid><description>&lt;p>We introduce GEMNA, a deep learning framework for mass spectrometry–based metabolomics that uses graph and edge embeddings with anomaly detection. GEMNA outperforms traditional tools in untargeted studies, producing clearer clusters and improved biological insights. Published in &lt;a href="https://www.nature.com/articles/s41598-024-80955-5" target="_blank" rel="noopener">Scientific Reports&lt;/a>&lt;/p></description></item><item><title>Rounding out our research hat trick with new insights into interpretable image synthesis!</title><link>https://mlo-lab.github.io/post/gruber-diagnostic_image_generators/</link><pubDate>Mon, 02 Sep 2024 11:27:52 +0100</pubDate><guid>https://mlo-lab.github.io/post/gruber-diagnostic_image_generators/</guid><description>&lt;p>Our latest work presents a new approach to disentangle image generation performance by decomposing cosine similarity into cluster-level contributions using central kernel alignment. This allows us to quantify how different pixel regions contribute to overall image quality, enabling more fine-grained evaluation and improved explainability of generative models across real-world use cases. Published at the Interpretable AI: Past, Present and Future Workshop at NeurIPS 2024.&lt;/p></description></item><item><title>Our framework for explanatory model monitoring was featured at KDD</title><link>https://mlo-lab.github.io/post/buettner-feature-shift-performance/</link><pubDate>Sat, 24 Aug 2024 14:57:12 +0100</pubDate><guid>https://mlo-lab.github.io/post/buettner-feature-shift-performance/</guid><description>&lt;p>We introduce Explanatory Performance Estimation (XPE), a method that explains model behavior under feature shifts by linking performance changes to interpretable input characteristics using Optimal Transport and Shapley Values. This enables explanatory model monitoring across image, audio, and tabular data. Published in the Proceedings of the 30th ACM SIGKDD Conference on Knowledge Discovery and Data Mining.&lt;/p></description></item><item><title>Our latest paper at HCI explores how human gaze can make AI more interpretable and reliable!</title><link>https://mlo-lab.github.io/post/buettner-expert-eyes-algning-industrial-ai/</link><pubDate>Mon, 29 Jul 2024 14:21:54 +0100</pubDate><guid>https://mlo-lab.github.io/post/buettner-expert-eyes-algning-industrial-ai/</guid><description>&lt;p>We show how integrating human gaze information aligns human and machine attention, improving model robustness and explainability. Demonstrated on real-world visual quality inspection, the approach highlights the value of explicit human knowledge in training trustworthy AI. Published at the &lt;a href="https://link.springer.com/book/10.1007/978-3-031-60611-3" target="_blank" rel="noopener">International Conference on Human-Computer Interaction&lt;/a>.&lt;/p></description></item><item><title>New at ICML: improving the stability of feature attributions through optimal combinations!</title><link>https://mlo-lab.github.io/post/buettner-provably-better-explanations/</link><pubDate>Sun, 07 Jul 2024 14:14:20 +0100</pubDate><guid>https://mlo-lab.github.io/post/buettner-provably-better-explanations/</guid><description>&lt;p>We improve the quality of feature attributions by optimally combining multiple explanation methods. Our convex-combination strategy enhances robustness and faithfulness, consistently outperforming individual methods and baselines across architectures. Published at the International Conference on Machine Learning.&lt;/p></description></item><item><title>Advancing global understanding of evaluation metrics — now published in Nature Methods!</title><link>https://mlo-lab.github.io/post/buettner-metric-related-pitfalls/</link><pubDate>Mon, 12 Feb 2024 14:24:23 +0100</pubDate><guid>https://mlo-lab.github.io/post/buettner-metric-related-pitfalls/</guid><description>&lt;p>This work compiles a domain-agnostic taxonomy of pitfalls in validation metrics, based on a multistage Delphi process and community feedback. It offers practical guidance to improve evaluation practices across application domains. Published in &lt;a href="https://www.nature.com/articles/s41592-023-02150-0" target="_blank" rel="noopener">Nature Methods&lt;/a>.&lt;/p></description></item><item><title>Metrics Reloaded — now published in Nature Methods.</title><link>https://mlo-lab.github.io/post/buettner-metrics-reloaded-image-validation/</link><pubDate>Mon, 12 Feb 2024 14:10:16 +0100</pubDate><guid>https://mlo-lab.github.io/post/buettner-metrics-reloaded-image-validation/</guid><description>&lt;p>We present Metrics Reloaded, a comprehensive framework for problem-aware selection of evaluation metrics in biomedical image analysis. Developed through a large international Delphi process, it introduces the concept of a problem fingerprint to guide researchers toward meaningful and domain-relevant validation. Published in &lt;a href="https://www.nature.com/articles/s41592-023-02151-z" target="_blank" rel="noopener">Nature Methods&lt;/a>.&lt;/p></description></item><item><title>A Bias-Variance-Covariance Decomposition of Kernel Scores for Generative Models</title><link>https://mlo-lab.github.io/publication/gruber-2024-bias/</link><pubDate>Mon, 01 Jan 2024 00:00:00 +0000</pubDate><guid>https://mlo-lab.github.io/publication/gruber-2024-bias/</guid><description/></item><item><title>Consistent and Asymptotically Unbiased Estimation of Proper Calibration Errors</title><link>https://mlo-lab.github.io/publication/popordanoska-2024-consistent/</link><pubDate>Mon, 01 Jan 2024 00:00:00 +0000</pubDate><guid>https://mlo-lab.github.io/publication/popordanoska-2024-consistent/</guid><description/></item><item><title>Disentangling Mean Embeddings for Better Diagnostics of Image Generators</title><link>https://mlo-lab.github.io/publication/gruber-2024-disentangling/</link><pubDate>Mon, 01 Jan 2024 00:00:00 +0000</pubDate><guid>https://mlo-lab.github.io/publication/gruber-2024-disentangling/</guid><description/></item><item><title>Explanatory Model Monitoring to Understand the Effects of Feature Shifts on Performance</title><link>https://mlo-lab.github.io/publication/decker-2024-explanatory/</link><pubDate>Mon, 01 Jan 2024 00:00:00 +0000</pubDate><guid>https://mlo-lab.github.io/publication/decker-2024-explanatory/</guid><description/></item><item><title>Exploratory analysis of metabolic changes using mass spectrometry data and graph embeddings</title><link>https://mlo-lab.github.io/publication/alvarez-2024-exploratory/</link><pubDate>Mon, 01 Jan 2024 00:00:00 +0000</pubDate><guid>https://mlo-lab.github.io/publication/alvarez-2024-exploratory/</guid><description/></item><item><title>Metrics reloaded: recommendations for image analysis validation</title><link>https://mlo-lab.github.io/publication/maier-2024-metrics/</link><pubDate>Mon, 01 Jan 2024 00:00:00 +0000</pubDate><guid>https://mlo-lab.github.io/publication/maier-2024-metrics/</guid><description/></item><item><title>Provably Better Explanations with Optimized Aggregation of Feature Attributions</title><link>https://mlo-lab.github.io/publication/decker-2024-provably/</link><pubDate>Mon, 01 Jan 2024 00:00:00 +0000</pubDate><guid>https://mlo-lab.github.io/publication/decker-2024-provably/</guid><description/></item><item><title>Through the Eyes of the Expert: Aligning Human and Machine Attention for Industrial AI</title><link>https://mlo-lab.github.io/publication/koebler-2024-through/</link><pubDate>Mon, 01 Jan 2024 00:00:00 +0000</pubDate><guid>https://mlo-lab.github.io/publication/koebler-2024-through/</guid><description/></item><item><title>Understanding metric-related pitfalls in image analysis validation</title><link>https://mlo-lab.github.io/publication/reinke-2024-understanding/</link><pubDate>Mon, 01 Jan 2024 00:00:00 +0000</pubDate><guid>https://mlo-lab.github.io/publication/reinke-2024-understanding/</guid><description/></item><item><title>Bridging calibration and refinement — our latest work at AISTATS!</title><link>https://mlo-lab.github.io/post/gruber-unbiased-calibration-errors/</link><pubDate>Thu, 14 Dec 2023 11:14:12 +0100</pubDate><guid>https://mlo-lab.github.io/post/gruber-unbiased-calibration-errors/</guid><description>&lt;p>We present a general method for consistent and asymptotically unbiased estimation of proper calibration errors and refinement terms. Introducing the Kullback–Leibler calibration error, we reveal its connection to f-divergences and information monotonicity in neural networks. Published at the International Conference on Artificial Intelligence and Statistics.&lt;/p></description></item><item><title>Our kernel uncertainty framework is in at ICML!</title><link>https://mlo-lab.github.io/post/gruber-kernel-scores/</link><pubDate>Mon, 09 Oct 2023 10:34:16 +0100</pubDate><guid>https://mlo-lab.github.io/post/gruber-kernel-scores/</guid><description>&lt;p>Our latest work introduces the first bias–variance–covariance decomposition for kernel scores, providing a unified framework for uncertainty estimation in generative models. We show how kernel-based entropy and variance capture uncertainty across image, audio, and language generation — even in closed-source models. Published at the International Conference on Machine Learning.&lt;/p></description></item><item><title>Paper accepted at ICML CompBio 2023!</title><link>https://mlo-lab.github.io/post/qoku-cellij-icml-23/</link><pubDate>Tue, 25 Jul 2023 22:16:23 +0200</pubDate><guid>https://mlo-lab.github.io/post/qoku-cellij-icml-23/</guid><description>&lt;p>&lt;strong>Paper accepted! Our latest work on developing a versatile framework for rapid prototyping and training of a wide range of factor analysis models for multi-omics data got accepted at this year&amp;rsquo;s ICML workshop on computational biology!&lt;/strong>&lt;/p></description></item><item><title>ERC Consolidator Grant for Florian!</title><link>https://mlo-lab.github.io/post/buettner-erc-grant-23/</link><pubDate>Sun, 22 Jan 2023 16:16:26 +0200</pubDate><guid>https://mlo-lab.github.io/post/buettner-erc-grant-23/</guid><description>&lt;p>&lt;strong>With 2 million Euros in funding from the European Research Council, we will be developing AI models to support doctors in the diagnosis and treatment of cancer. &lt;a href="https://www.dkfz.de/de/presse/pressemitteilungen/2023/dkfz-pm-23-07-ERC-Consolidator-Grant-fuer-DKTK-Forscher-Florian-Buettner.php" target="_blank" rel="noopener">Read more&lt;/a>!&lt;/strong>&lt;/p></description></item><item><title>Paper accepted at AISTATS 2023!</title><link>https://mlo-lab.github.io/post/gruber-uncertainty-aistats-23/</link><pubDate>Fri, 20 Jan 2023 16:16:03 +0200</pubDate><guid>https://mlo-lab.github.io/post/gruber-uncertainty-aistats-23/</guid><description>&lt;p>&lt;strong>Paper accepted! Our latest work on a general bias-variance decomposition for proper scores got accepted at this year&amp;rsquo;s AISTATS conference!&lt;/strong>&lt;/p></description></item><item><title>Paper accepted at AISTATS 2023!</title><link>https://mlo-lab.github.io/post/qoku-encoding-aistats-23/</link><pubDate>Fri, 20 Jan 2023 16:15:47 +0200</pubDate><guid>https://mlo-lab.github.io/post/qoku-encoding-aistats-23/</guid><description>&lt;p>&lt;strong>Paper accepted! Our latest work on multi-view latent variable models with structured sparsity got accepted at this year&amp;rsquo;s AISTATS conference!&lt;/strong>&lt;/p></description></item><item><title>Paper accepted at AAAI 2023!</title><link>https://mlo-lab.github.io/post/hekler-test-aaai-23/</link><pubDate>Sat, 14 Jan 2023 16:15:05 +0200</pubDate><guid>https://mlo-lab.github.io/post/hekler-test-aaai-23/</guid><description>&lt;p>&lt;strong>Paper accepted! Our latest work on quantifying uncertainty under real-world conditions got accepted at this year&amp;rsquo;s AAAI conference on artificial intelligence!&lt;/strong>&lt;/p></description></item><item><title>Encoding domain knowledge in multi-view latent variable models: A bayesian approach with structured sparsity</title><link>https://mlo-lab.github.io/publication/qoku-2023-encoding/</link><pubDate>Sun, 01 Jan 2023 00:00:00 +0000</pubDate><guid>https://mlo-lab.github.io/publication/qoku-2023-encoding/</guid><description/></item><item><title>Multimodal analysis methods in predictive biomedicine</title><link>https://mlo-lab.github.io/publication/qoku-2023-multimodal/</link><pubDate>Sun, 01 Jan 2023 00:00:00 +0000</pubDate><guid>https://mlo-lab.github.io/publication/qoku-2023-multimodal/</guid><description/></item><item><title>Test Time Augmentation Meets Post-hoc Calibration: Uncertainty Quantification under Real-World Conditions</title><link>https://mlo-lab.github.io/publication/hekler-2023-test/</link><pubDate>Sun, 01 Jan 2023 00:00:00 +0000</pubDate><guid>https://mlo-lab.github.io/publication/hekler-2023-test/</guid><description/></item><item><title>Uncertainty Estimates of Predictions via a General Bias-Variance Decomposition</title><link>https://mlo-lab.github.io/publication/gruber-2023-uncertainty/</link><pubDate>Sun, 01 Jan 2023 00:00:00 +0000</pubDate><guid>https://mlo-lab.github.io/publication/gruber-2023-uncertainty/</guid><description/></item><item><title>Paper accepted at ECCV 2022!</title><link>https://mlo-lab.github.io/post/tomani-eccv-22/</link><pubDate>Fri, 08 Jul 2022 17:13:49 +0200</pubDate><guid>https://mlo-lab.github.io/post/tomani-eccv-22/</guid><description>&lt;p>&lt;strong>Paper accepted! Our latest work on boosting the expressive power in post-hoc uncertainty calibration got accepted at this year&amp;rsquo;s ECCV!&lt;/strong>&lt;/p></description></item><item><title>Paper accepted at Cancer Cell 2022!</title><link>https://mlo-lab.github.io/post/wolf-aml-cancercell-22/</link><pubDate>Fri, 08 Jul 2022 16:33:37 +0200</pubDate><guid>https://mlo-lab.github.io/post/wolf-aml-cancercell-22/</guid><description>&lt;p>&lt;strong>Paper accepted! Our latest work on characterizing proteogenomic subtypes of AML got accepted at this year&amp;rsquo;s Cancer Cell!&lt;/strong>&lt;/p></description></item><item><title>Better uncertainty calibration via proper scores for classification and beyond</title><link>https://mlo-lab.github.io/publication/gruber-2022-better/</link><pubDate>Sat, 01 Jan 2022 00:00:00 +0000</pubDate><guid>https://mlo-lab.github.io/publication/gruber-2022-better/</guid><description/></item><item><title>Inflammatory fibroblasts mediate resistance to neoadjuvant therapy in rectal cancer</title><link>https://mlo-lab.github.io/publication/nicolas-2022-inflammatory/</link><pubDate>Sat, 01 Jan 2022 00:00:00 +0000</pubDate><guid>https://mlo-lab.github.io/publication/nicolas-2022-inflammatory/</guid><description/></item><item><title>Single cell analyses identify a highly regenerative and homogenous human CD34+ hematopoietic stem cell population</title><link>https://mlo-lab.github.io/publication/anjos-2022-single/</link><pubDate>Sat, 01 Jan 2022 00:00:00 +0000</pubDate><guid>https://mlo-lab.github.io/publication/anjos-2022-single/</guid><description/></item><item><title>The proteogenomic subtypes of acute myeloid leukemia</title><link>https://mlo-lab.github.io/publication/jayavelu-2022-proteogenomic/</link><pubDate>Sat, 01 Jan 2022 00:00:00 +0000</pubDate><guid>https://mlo-lab.github.io/publication/jayavelu-2022-proteogenomic/</guid><description/></item><item><title>Paper accepted at KDD 2021!</title><link>https://mlo-lab.github.io/post/spexlvm-kdd/</link><pubDate>Tue, 13 Jul 2021 16:30:04 +0200</pubDate><guid>https://mlo-lab.github.io/post/spexlvm-kdd/</guid><description>&lt;p>Latent variable models are powerful statistical tools that can uncover relevant variation between patients or cells, by inferring unobserved hidden states from observable high-dimensional data. A major shortcoming of current methods, however, is their inability to learn sparse and interpretable hidden states. Additionally, in settings where partial knowledge on the latent structure of the data is readily available, a statistically sound integration of prior information into current methods is challenging. To address these issues, we propose spex-LVM, a factorial latent variable model with &lt;strong>s&lt;/strong>parse &lt;strong>p&lt;/strong>riors to encourage the inference of &lt;strong>ex&lt;/strong>plainable factors driven by domain-relevant information. spex-LVM utilizes existing knowledge of curated biomedical pathways to automatically assign annotated attributes to latent factors, yielding interpretable results tailored to the corresponding domain of interest. Evaluations on simulated and real single-cell RNA-seq datasets demonstrate that our model robustly identifies relevant structure in an inherently explainable manner, distinguishes technical noise from sources of biomedical variation, and provides data-driven adaptations of existing pathway annotations.&lt;/p></description></item><item><title>Paper accepted at UAI!</title><link>https://mlo-lab.github.io/post/paper-uai-yang/</link><pubDate>Thu, 20 May 2021 16:59:59 +0200</pubDate><guid>https://mlo-lab.github.io/post/paper-uai-yang/</guid><description>&lt;p>&lt;strong>Paper accepted! Our latest work on multi-output Gaussian Process Latent Variable models got accepted at this year&amp;rsquo;s UAI!&lt;/strong>&lt;/p></description></item><item><title>Workshop on XAI and Trustworthiness in Healthcare at KDD 2021!</title><link>https://mlo-lab.github.io/post/workshop/</link><pubDate>Tue, 13 Apr 2021 00:00:00 +0000</pubDate><guid>https://mlo-lab.github.io/post/workshop/</guid><description>&lt;p>&lt;strong>We organize a workshop at this year&amp;rsquo;s KDD&lt;/strong>&lt;/p>
&lt;p>Check out the workshop website &lt;a href="https://dshealthkdd.github.io/dshealth-2021/" target="_blank">here&lt;/a>.
Paper submission deadline is 10th May 2021.&lt;/p>
&lt;h1 id="overview">Overview&lt;/h1>
&lt;p>Healthcare is, traditionally, a knowledge-driven enterprise with an enormous amount of data - both structured and unstructured. These data can impact positively on the development of data-driven health care including precision medicine and precision public health. In recent years, large scale medical/clinical datasets, such as “omics” data and radiology reports are increasingly available. We have also witnessed an increasing number of successful AI/ML applications using such datasets to address problems such as drug repurposing and preliminary screening of radiology reports. To facilitate the adoption of such AI/ML in practice, we have simultaneously witnessed an increasing adoption/innovation of using explainability methods to analyze/present AI for Health. In this deep learning era, What is the current status of AI/ML applications in healthcare? What are the standard methods of explaining such AI models for healthcare? What are the roles of causality in AI/ML practices? What are the state-of-the-art developments in causal AI in health and health care domains? What are the limitations and how are the different facets of trust and explanations (see figure 1 below) being addressed in practice? Can knowledge-backed AI lead to more robust and interpretable models? How do data scientists and physicians apply this knowledge in collaboration and via human-centered AI approaches to further the field and improve healthcare? How are regulatory requirements for transparency and trustworthiness of models and data privacy being defined and how can they be fulfilled? After witnessing so many great achievements from deep learning lately, we propose to invite world-leading experts from both data science and healthcare to discuss and debate the path forward for practical applications of AI/ML in healthcare, including demos, early work, and critiques on the current state and the path forward for explainability and trustworthiness in AI. More specifically, we plan to attract high-quality original research from emerging areas with significant implications in healthcare and invite open discussions on controversial yet crucial topics regarding healthcare transformation&lt;/p>
&lt;h1 id="key-dates">Key dates&lt;/h1>
&lt;ul>
&lt;li>Paper Submission opens: Apr 15, 2021&lt;/li>
&lt;li>Paper Submission deadline: May 10, 2021&lt;/li>
&lt;li>Acceptance Notice: Jun 10, 2021&lt;/li>
&lt;li>Workshop Date: Aug 14, 2021&lt;/li>
&lt;/ul>
&lt;p>All deadlines correspond to 11:59 PM Hawaii Standard Time ( HST).&lt;/p></description></item><item><title>Hierarchical Domain Invariant Variational Auto-Encoding with weak domain supervision</title><link>https://mlo-lab.github.io/publication/sun-2021-hierarchical/</link><pubDate>Fri, 01 Jan 2021 00:00:00 +0000</pubDate><guid>https://mlo-lab.github.io/publication/sun-2021-hierarchical/</guid><description/></item><item><title>Multi-output Gaussian Processes for uncertainty-aware recommender systems</title><link>https://mlo-lab.github.io/publication/yang-2021-multi/</link><pubDate>Fri, 01 Jan 2021 00:00:00 +0000</pubDate><guid>https://mlo-lab.github.io/publication/yang-2021-multi/</guid><description/></item><item><title>Parameterized Temperature Scaling for Boosting the Expressive Power in Post-Hoc Uncertainty Calibration</title><link>https://mlo-lab.github.io/publication/tomani-2021-parameterized/</link><pubDate>Fri, 01 Jan 2021 00:00:00 +0000</pubDate><guid>https://mlo-lab.github.io/publication/tomani-2021-parameterized/</guid><description/></item><item><title>Towards trustworthy predictions from deep neural networks with fast adversarial calibration</title><link>https://mlo-lab.github.io/publication/tomani-2021-towards/</link><pubDate>Fri, 01 Jan 2021 00:00:00 +0000</pubDate><guid>https://mlo-lab.github.io/publication/tomani-2021-towards/</guid><description/></item><item><title>Post-hoc Uncertainty Calibration for Domain Drift Scenarios</title><link>https://mlo-lab.github.io/publication/tomani-post-hoc-2020/</link><pubDate>Wed, 01 Jan 2020 00:00:00 +0000</pubDate><guid>https://mlo-lab.github.io/publication/tomani-post-hoc-2020/</guid><description/></item><item><title>Single-cell RNA-sequencing of differentiating iPS cells reveals dynamic genetic effects on gene expression</title><link>https://mlo-lab.github.io/publication/cuomo-single-cell-2020/</link><pubDate>Wed, 01 Jan 2020 00:00:00 +0000</pubDate><guid>https://mlo-lab.github.io/publication/cuomo-single-cell-2020/</guid><description/></item><item><title>TIMELY: Improving Labeling Consistency in Medical Imaging for Cell Type Classification</title><link>https://mlo-lab.github.io/publication/liu-timely-2020/</link><pubDate>Wed, 01 Jan 2020 00:00:00 +0000</pubDate><guid>https://mlo-lab.github.io/publication/liu-timely-2020/</guid><description/></item><item><title>Ano-rectal wall dose-surface maps localize the dosimetric benefit of hydrogel rectum spacers in prostate cancer radiotherapy</title><link>https://mlo-lab.github.io/publication/vanneste-ano-rectal-2019/</link><pubDate>Tue, 01 Jan 2019 00:00:00 +0000</pubDate><guid>https://mlo-lab.github.io/publication/vanneste-ano-rectal-2019/</guid><description/></item><item><title>Document informed neural autoregressive topic models with distributional prior</title><link>https://mlo-lab.github.io/publication/gupta-document-2019/</link><pubDate>Tue, 01 Jan 2019 00:00:00 +0000</pubDate><guid>https://mlo-lab.github.io/publication/gupta-document-2019/</guid><description/></item><item><title>Metabolic regulation of pluripotency and germ cell fate through α-ketoglutarate</title><link>https://mlo-lab.github.io/publication/tischler-metabolic-2019/</link><pubDate>Tue, 01 Jan 2019 00:00:00 +0000</pubDate><guid>https://mlo-lab.github.io/publication/tischler-metabolic-2019/</guid><description/></item><item><title>Single cell multi-omics profiling reveals a hierarchical epigenetic landscape during mammalian germ layer specification</title><link>https://mlo-lab.github.io/publication/argelaguet-single-2019/</link><pubDate>Tue, 01 Jan 2019 00:00:00 +0000</pubDate><guid>https://mlo-lab.github.io/publication/argelaguet-single-2019/</guid><description/></item><item><title>Impressum</title><link>https://mlo-lab.github.io/terms/</link><pubDate>Thu, 28 Jun 2018 00:00:00 +0100</pubDate><guid>https://mlo-lab.github.io/terms/</guid><description>&lt;h2 id="anbieter-der-internetpräsenz">Anbieter der Internetpräsenz&lt;/h2>
&lt;p>Prof. Dr. Florian Buettner&lt;br>
Deutsches Konsortium für Translationale Krebsforschung (DKTK)&lt;br>
Deutsches Krebsforschungszentrum&lt;br>
Goethe University Frankfurt&lt;/p>
&lt;p>Theodor-Stern-Kai 7&lt;br>
60596 Frankfurt am Main&lt;/p>
&lt;h3 id="kontakt">Kontakt&lt;/h3>
&lt;p>Telefon: +49 696 30186212&lt;br>
E-Mail: florian.buettner|dkfz-heidelberg.de&lt;/p>
&lt;h2 id="streitschlichtung">Streitschlichtung&lt;/h2>
&lt;p>Die Europäische Kommission stellt eine Plattform zur Online-Streitbeilegung (OS) bereit: &lt;a href="https://ec.europa.eu/consumers/odr">https://ec.europa.eu/consumers/odr&lt;/a>.
Unsere E-Mail-Adresse finden Sie oben im Impressum.&lt;/p>
&lt;p>Wir sind nicht bereit oder verpflichtet, an Streitbeilegungsverfahren vor einer Verbraucherschlichtungsstelle teilzunehmen.&lt;/p>
&lt;h3 id="haftung-für-inhalte">Haftung für Inhalte&lt;/h3>
&lt;p>Als Diensteanbieter sind wir gemäß § 7 Abs.1 TMG für eigene Inhalte auf diesen Seiten nach den allgemeinen Gesetzen verantwortlich. Nach §§ 8 bis 10 TMG sind wir als Diensteanbieter jedoch nicht verpflichtet, übermittelte oder gespeicherte fremde Informationen zu überwachen oder nach Umständen zu forschen, die auf eine rechtswidrige Tätigkeit hinweisen.&lt;/p>
&lt;p>Verpflichtungen zur Entfernung oder Sperrung der Nutzung von Informationen nach den allgemeinen Gesetzen bleiben hiervon unberührt. Eine diesbezügliche Haftung ist jedoch erst ab dem Zeitpunkt der Kenntnis einer konkreten Rechtsverletzung möglich. Bei Bekanntwerden von entsprechenden Rechtsverletzungen werden wir diese Inhalte umgehend entfernen.&lt;/p>
&lt;h3 id="haftung-für-links">Haftung für Links&lt;/h3>
&lt;p>Unser Angebot enthält Links zu externen Websites Dritter, auf deren Inhalte wir keinen Einfluss haben. Deshalb können wir für diese fremden Inhalte auch keine Gewähr übernehmen. Für die Inhalte der verlinkten Seiten ist stets der jeweilige Anbieter oder Betreiber der Seiten verantwortlich. Die verlinkten Seiten wurden zum Zeitpunkt der Verlinkung auf mögliche Rechtsverstöße überprüft. Rechtswidrige Inhalte waren zum Zeitpunkt der Verlinkung nicht erkennbar.&lt;/p>
&lt;p>Eine permanente inhaltliche Kontrolle der verlinkten Seiten ist jedoch ohne konkrete Anhaltspunkte einer Rechtsverletzung nicht zumutbar. Bei Bekanntwerden von Rechtsverletzungen werden wir derartige Links umgehend entfernen.&lt;/p>
&lt;h3 id="urheberrecht">Urheberrecht&lt;/h3>
&lt;p>Die durch die Seitenbetreiber erstellten Inhalte und Werke auf diesen Seiten unterliegen dem deutschen Urheberrecht. Die Vervielfältigung, Bearbeitung, Verbreitung und jede Art der Verwertung außerhalb der Grenzen des Urheberrechtes bedürfen der schriftlichen Zustimmung des jeweiligen Autors bzw. Erstellers. Downloads und Kopien dieser Seite sind nur für den privaten, nicht kommerziellen Gebrauch gestattet.&lt;/p>
&lt;p>Soweit die Inhalte auf dieser Seite nicht vom Betreiber erstellt wurden, werden die Urheberrechte Dritter beachtet. Insbesondere werden Inhalte Dritter als solche gekennzeichnet. Sollten Sie trotzdem auf eine Urheberrechtsverletzung aufmerksam werden, bitten wir um einen entsprechenden Hinweis. Bei Bekanntwerden von Rechtsverletzungen werden wir derartige Inhalte umgehend entfernen.&lt;/p></description></item><item><title>Design and selection of machine learning methods using radiomics and dosiomics for normal tissue complication probability modeling of xerostomia</title><link>https://mlo-lab.github.io/publication/gabrys-design-2018/</link><pubDate>Mon, 01 Jan 2018 00:00:00 +0000</pubDate><guid>https://mlo-lab.github.io/publication/gabrys-design-2018/</guid><description/></item><item><title>Document informed neural autoregressive topic models</title><link>https://mlo-lab.github.io/publication/gupta-document-2018/</link><pubDate>Mon, 01 Jan 2018 00:00:00 +0000</pubDate><guid>https://mlo-lab.github.io/publication/gupta-document-2018/</guid><description/></item><item><title>Multi-Omics Factor Analysis—a framework for unsupervised integration of multi-omics data sets</title><link>https://mlo-lab.github.io/publication/argelaguet-multi-omics-2018/</link><pubDate>Mon, 01 Jan 2018 00:00:00 +0000</pubDate><guid>https://mlo-lab.github.io/publication/argelaguet-multi-omics-2018/</guid><description/></item><item><title>texttovec: Deep contextualized neural autoregressive models of language with distributed compositional prior</title><link>https://mlo-lab.github.io/publication/gupta-texttovec-2018-1/</link><pubDate>Mon, 01 Jan 2018 00:00:00 +0000</pubDate><guid>https://mlo-lab.github.io/publication/gupta-texttovec-2018-1/</guid><description/></item><item><title>Texttovec: Deep contextualized neural autoregressive topic models of language with distributed compositional prior</title><link>https://mlo-lab.github.io/publication/gupta-texttovec-2018/</link><pubDate>Mon, 01 Jan 2018 00:00:00 +0000</pubDate><guid>https://mlo-lab.github.io/publication/gupta-texttovec-2018/</guid><description/></item><item><title>A divergent population of autoantigen-responsive CD4+ T cells in infants prior to β cell autoimmunity</title><link>https://mlo-lab.github.io/publication/heninger-divergent-2017/</link><pubDate>Sun, 01 Jan 2017 00:00:00 +0000</pubDate><guid>https://mlo-lab.github.io/publication/heninger-divergent-2017/</guid><description/></item><item><title>cgcorrect: a method to correct for confounding cell–cell variation due to cell growth in single-cell transcriptomics</title><link>https://mlo-lab.github.io/publication/blasi-cgcorrect-2017/</link><pubDate>Sun, 01 Jan 2017 00:00:00 +0000</pubDate><guid>https://mlo-lab.github.io/publication/blasi-cgcorrect-2017/</guid><description/></item><item><title>f-scLVM: scalable and versatile factor analysis for single-cell RNA-seq</title><link>https://mlo-lab.github.io/publication/buettner-f-sclvm-2017/</link><pubDate>Sun, 01 Jan 2017 00:00:00 +0000</pubDate><guid>https://mlo-lab.github.io/publication/buettner-f-sclvm-2017/</guid><description/></item><item><title>Non-targeted metabolomic approach reveals two distinct types of metabolic responses to telomerase dysfunction in S. cerevisiae</title><link>https://mlo-lab.github.io/publication/buettner-non-targeted-2017/</link><pubDate>Sun, 01 Jan 2017 00:00:00 +0000</pubDate><guid>https://mlo-lab.github.io/publication/buettner-non-targeted-2017/</guid><description/></item><item><title>Parotid gland mean dose as a xerostomia predictor in low-dose domains</title><link>https://mlo-lab.github.io/publication/gabrys-parotid-2017/</link><pubDate>Sun, 01 Jan 2017 00:00:00 +0000</pubDate><guid>https://mlo-lab.github.io/publication/gabrys-parotid-2017/</guid><description/></item><item><title>Prospective identification of hematopoietic lineage choice by deep learning</title><link>https://mlo-lab.github.io/publication/buggenthin-prospective-2017/</link><pubDate>Sun, 01 Jan 2017 00:00:00 +0000</pubDate><guid>https://mlo-lab.github.io/publication/buggenthin-prospective-2017/</guid><description/></item><item><title>Vitamin A-retinoic acid signaling regulates hematopoietic stem cell dormancy</title><link>https://mlo-lab.github.io/publication/cabezas-wallscheid-vitamin-2017/</link><pubDate>Sun, 01 Jan 2017 00:00:00 +0000</pubDate><guid>https://mlo-lab.github.io/publication/cabezas-wallscheid-vitamin-2017/</guid><description/></item><item><title>destiny: diffusion maps for large-scale single-cell data in R</title><link>https://mlo-lab.github.io/publication/angerer-destiny-2016/</link><pubDate>Fri, 01 Jan 2016 00:00:00 +0000</pubDate><guid>https://mlo-lab.github.io/publication/angerer-destiny-2016/</guid><description/></item><item><title>Unbiased prediction and feature selection in high-dimensional survival regression</title><link>https://mlo-lab.github.io/publication/laimighofer-unbiased-2016/</link><pubDate>Fri, 01 Jan 2016 00:00:00 +0000</pubDate><guid>https://mlo-lab.github.io/publication/laimighofer-unbiased-2016/</guid><description/></item><item><title>Combined single-cell functional and gene expression analysis resolves heterogeneity within stem cell populations</title><link>https://mlo-lab.github.io/publication/wilson-combined-2015/</link><pubDate>Thu, 01 Jan 2015 00:00:00 +0000</pubDate><guid>https://mlo-lab.github.io/publication/wilson-combined-2015/</guid><description/></item><item><title>Computational analysis of cell-to-cell heterogeneity in single-cell RNA-sequencing data reveals hidden subpopulations of cells</title><link>https://mlo-lab.github.io/publication/buettner-computational-2015/</link><pubDate>Thu, 01 Jan 2015 00:00:00 +0000</pubDate><guid>https://mlo-lab.github.io/publication/buettner-computational-2015/</guid><description/></item><item><title>Computational assignment of cell-cycle stage from single-cell transcriptome data</title><link>https://mlo-lab.github.io/publication/scialdone-computational-2015/</link><pubDate>Thu, 01 Jan 2015 00:00:00 +0000</pubDate><guid>https://mlo-lab.github.io/publication/scialdone-computational-2015/</guid><description/></item><item><title>Decoding the regulatory network of early blood development from single-cell gene expression measurements</title><link>https://mlo-lab.github.io/publication/moignard-decoding-2015/</link><pubDate>Thu, 01 Jan 2015 00:00:00 +0000</pubDate><guid>https://mlo-lab.github.io/publication/moignard-decoding-2015/</guid><description/></item><item><title>Diffusion maps for high-dimensional single-cell analysis of differentiation data</title><link>https://mlo-lab.github.io/publication/haghverdi-diffusion-2015/</link><pubDate>Thu, 01 Jan 2015 00:00:00 +0000</pubDate><guid>https://mlo-lab.github.io/publication/haghverdi-diffusion-2015/</guid><description/></item><item><title>Effects of high-dose oral insulin on immune responses in children at high risk for type 1 diabetes: the Pre-POINT randomized clinical trial</title><link>https://mlo-lab.github.io/publication/bonifacio-effects-2015/</link><pubDate>Thu, 01 Jan 2015 00:00:00 +0000</pubDate><guid>https://mlo-lab.github.io/publication/bonifacio-effects-2015/</guid><description/></item><item><title>RAMONA: a Web application for gene set analysis on multilevel omics data</title><link>https://mlo-lab.github.io/publication/sass-ramona-2015/</link><pubDate>Thu, 01 Jan 2015 00:00:00 +0000</pubDate><guid>https://mlo-lab.github.io/publication/sass-ramona-2015/</guid><description/></item><item><title>Single-cell transcriptomic reconstruction reveals cell cycle and multi-lineage differentiation defects in Bcl11a-deficient hematopoietic stem cells</title><link>https://mlo-lab.github.io/publication/tsang-single-cell-2015/</link><pubDate>Thu, 01 Jan 2015 00:00:00 +0000</pubDate><guid>https://mlo-lab.github.io/publication/tsang-single-cell-2015/</guid><description/></item><item><title>Feature ranking of type 1 diabetes susceptibility genes improves prediction of type 1 diabetes</title><link>https://mlo-lab.github.io/publication/winkler-feature-2014/</link><pubDate>Wed, 01 Jan 2014 00:00:00 +0000</pubDate><guid>https://mlo-lab.github.io/publication/winkler-feature-2014/</guid><description/></item><item><title>Molecular phenotypic profiling of a Saccharomyces cerevisiae strain at the single-cell level</title><link>https://mlo-lab.github.io/publication/schmidt-molecular-2014/</link><pubDate>Wed, 01 Jan 2014 00:00:00 +0000</pubDate><guid>https://mlo-lab.github.io/publication/schmidt-molecular-2014/</guid><description/></item><item><title>Probabilistic PCA of censored data: accounting for uncertainties in the visualization of high-throughput single-cell qPCR data</title><link>https://mlo-lab.github.io/publication/buettner-probabilistic-2014/</link><pubDate>Wed, 01 Jan 2014 00:00:00 +0000</pubDate><guid>https://mlo-lab.github.io/publication/buettner-probabilistic-2014/</guid><description/></item><item><title>Two non-parametric methods for derivation of constraints from radiotherapy dose–histogram data</title><link>https://mlo-lab.github.io/publication/ebert-two-2014/</link><pubDate>Wed, 01 Jan 2014 00:00:00 +0000</pubDate><guid>https://mlo-lab.github.io/publication/ebert-two-2014/</guid><description/></item><item><title>A modular framework for gene set analysis integrating multilevel omics data</title><link>https://mlo-lab.github.io/publication/sass-modular-2013/</link><pubDate>Tue, 01 Jan 2013 00:00:00 +0000</pubDate><guid>https://mlo-lab.github.io/publication/sass-modular-2013/</guid><description/></item><item><title>Characterization of transcriptional networks in blood stem and progenitor cells using high-throughput single-cell gene expression analysis</title><link>https://mlo-lab.github.io/publication/moignard-characterization-2013/</link><pubDate>Tue, 01 Jan 2013 00:00:00 +0000</pubDate><guid>https://mlo-lab.github.io/publication/moignard-characterization-2013/</guid><description/></item><item><title>A novel approach for resolving differences in single-cell gene expression patterns from zygote to blastocyst</title><link>https://mlo-lab.github.io/publication/buettner-novel-2012/</link><pubDate>Sun, 01 Jan 2012 00:00:00 +0000</pubDate><guid>https://mlo-lab.github.io/publication/buettner-novel-2012/</guid><description/></item><item><title>Novel approaches to improve the therapeutic index of head and neck radiotherapy: an analysis of data from the PARSPORT randomised phase III trial</title><link>https://mlo-lab.github.io/publication/buettner-novel-2012-1/</link><pubDate>Sun, 01 Jan 2012 00:00:00 +0000</pubDate><guid>https://mlo-lab.github.io/publication/buettner-novel-2012-1/</guid><description/></item><item><title>The dose–response of the anal sphincter region–an analysis of data from the MRC RT01 trial</title><link>https://mlo-lab.github.io/publication/buettner-doseresponse-2012/</link><pubDate>Sun, 01 Jan 2012 00:00:00 +0000</pubDate><guid>https://mlo-lab.github.io/publication/buettner-doseresponse-2012/</guid><description/></item><item><title>Modeling late rectal toxicities based on a parameterized representation of the 3D dose distribution</title><link>https://mlo-lab.github.io/publication/buettner-modeling-2011/</link><pubDate>Sat, 01 Jan 2011 00:00:00 +0000</pubDate><guid>https://mlo-lab.github.io/publication/buettner-modeling-2011/</guid><description/></item><item><title>Using Bayesian logistic regression to evaluate a new type of dosimetric constraint for prostate radiotherapy treatment planning</title><link>https://mlo-lab.github.io/publication/buettner-using-2010/</link><pubDate>Fri, 01 Jan 2010 00:00:00 +0000</pubDate><guid>https://mlo-lab.github.io/publication/buettner-using-2010/</guid><description/></item><item><title>Assessing correlations between the spatial distribution of the dose to the rectal wall and late rectal toxicity after prostate radiotherapy: an analysis of data from the MRC RT01 trial (ISRCTN 47772397)</title><link>https://mlo-lab.github.io/publication/buettner-assessing-2009/</link><pubDate>Thu, 01 Jan 2009 00:00:00 +0000</pubDate><guid>https://mlo-lab.github.io/publication/buettner-assessing-2009/</guid><description/></item><item><title>Using dose-surface maps to predict radiation-induced rectal bleeding: a neural network approach</title><link>https://mlo-lab.github.io/publication/buettner-using-2009/</link><pubDate>Thu, 01 Jan 2009 00:00:00 +0000</pubDate><guid>https://mlo-lab.github.io/publication/buettner-using-2009/</guid><description/></item><item><title>Dosimetric impact of motion in free-breathing and gated lung radiotherapy: A 4D Monte Carlo study of intrafraction and interfraction effects</title><link>https://mlo-lab.github.io/publication/seco-dosimetric-2008/</link><pubDate>Tue, 01 Jan 2008 00:00:00 +0000</pubDate><guid>https://mlo-lab.github.io/publication/seco-dosimetric-2008/</guid><description/></item><item><title>Optical trapping dynamics for cell identification</title><link>https://mlo-lab.github.io/publication/volpe-optical-2006/</link><pubDate>Sun, 01 Jan 2006 00:00:00 +0000</pubDate><guid>https://mlo-lab.github.io/publication/volpe-optical-2006/</guid><description/></item><item><title/><link>https://mlo-lab.github.io/admin/config.yml</link><pubDate>Mon, 01 Jan 0001 00:00:00 +0000</pubDate><guid>https://mlo-lab.github.io/admin/config.yml</guid><description/></item></channel></rss>