arXiv Machine Learning September 1, 2026 By Nadejda Drenska, Matthew Lemoine, Gowri Priya Sunkara, Yu Wang, Sri Lakshmi Sravani Devarakonda, Steven B. Heymsfield

Target-Aware State-Adaptive $p$-Dirichlet Graph Neural Regression for Non-Invasive Body-Composition Estimation

Read the original on arXiv Machine Learning →

The Flow has not summarised this story yet — read it at arXiv Machine Learning.

arXiv Machine Learning

Aug 18

Deep learning-based computed tomography (CT) derived body composition classifier for colorectal cancer patients

arXiv:2608. 15712v1 Announce Type: cross Abstract: Background: Accurate body composition analysis using Computed Tomography (CT) scans is essential for assessing skeletal muscle area (SMA) and skeletal muscle density (SMD), key markers of nutritional status in cancer patients.

By Eve Harling (James Watt School of Engineering, College of Science & Engineering, University of Glasgow, Glasgow, UK), Chattarin Pumtako (Academic Unit of Surgery, School of Medicine, College of Medical Veterinary & Life Sciences, University of Glasgow, Glasgow, UK), Bernd Porr (James Watt School of Engineering, College of Science & Engineering, University of Glasgow, Glasgow, UK), Donald C McMillan (Academic Unit of Surgery, School of Medicine, College of Medical Veterinary & Life Sciences, University of Glasgow, Glasgow, UK), Ross D Dolan (Academic Unit of Surgery, School of Medicine, College of Medical Veterinary & Life Sciences, University of Glasgow, Glasgow, UK)

computer-vision

arXiv Machine Learning

Aug 12

GARLIC: Graph Attention-based Relational Learning of Multivariate Time Series in Intensive Care

arXiv:2608. 10969v1 Announce Type: new Abstract: Healthcare data, such as Intensive Care Unit (ICU) records, comprise heterogeneous multivariate time series sampled at irregular intervals with pervasive missingness.

By Ruirui Wang, Yanke Li, Manuel G\"unther, Diego Paez-Granados

benchmarks

arXiv Machine Learning

Jul 21

Differentiable latent structure discovery for interpretable forecasting in clinical time series

arXiv:2604. 27967v2 Announce Type: replace Abstract: Background: We introduce StructGP, a continuous-time multi-task Gaussian process that couples process convolutions with differentiable structure learning to uncover a sparse, ordered directed acyclic graph (DAG) of inter-variable dependencies while preserving principled uncertainty.

By Ivan Lerner, Jean Feydy, Alexandre Kalimouttou, Anita Burgun, Francis Bach

efficiency

arXiv Machine Learning

Jun 29

Autoencoder Architectures for Athlete Performance Scoring from Wearable Telemetry

arXiv:2606. 28145v1 Announce Type: new Abstract: Wearable devices produce large, high dimensional training logs for everyday runners, and interpretation rather than data collection is now the limiting step.

By Mateusz Kubita, Jan Zubalewicz, Krzysztof Siwek

safety

arXiv Machine Learning

Sep 22

GLR-MM: Graph-Based Global-Local Reconstruction for Robust Multimodal Chest X-ray and EHR Representation Learning under Missing Modalities

arXiv:2609.23876v1 Announce Type: new Abstract: Clinical multimodal models must often predict before all chest X-ray (CXR) and electronic health record (EHR) inputs are available. Existing approaches...

By Surbhi Sharma, Nikhil Manali, Devesh Maheshwari

rag multimodal

arXiv AI

Sep 1

Predicting Future Organ Dysfunction in ICU Patients Using Temporal Convolutional Networks on MIMIC-IV Data

arXiv:2608.29301v1 Announce Type: new Abstract: Predicting future organ dysfunction in Intensive Care Unit (ICU) patients is critical for early clinical intervention, yet existing machine learning ap...

By Razan Albouq, Asra Aslam

safety

One email a morning, machine-written

Related stories

Deep learning-based computed tomography (CT) derived body composition classifier for colorectal cancer patients

GARLIC: Graph Attention-based Relational Learning of Multivariate Time Series in Intensive Care

Differentiable latent structure discovery for interpretable forecasting in clinical time series

Autoencoder Architectures for Athlete Performance Scoring from Wearable Telemetry

GLR-MM: Graph-Based Global-Local Reconstruction for Robust Multimodal Chest X-ray and EHR Representation Learning under Missing Modalities

Predicting Future Organ Dysfunction in ICU Patients Using Temporal Convolutional Networks on MIMIC-IV Data