v149: Proceedings of Machine Learning for Healthcare 2021

2026-06-10 · Source: Proceedings of Machine Learning Research · Field: Health & Wellbeing — Medical Devices & Health Technology, Health & Medical Research, Clinical Care & Medical Practice · Depth: Expert, medium

Summary

Volume 149 of the "Proceedings of the 6th Machine Learning for Healthcare Conference," held virtually on August 6-7, 2021, presents a wide array of machine learning applications tailored for clinical environments. Key research areas include model selection for offline reinforcement learning in healthcare settings and knowledge graph-based question answering with Electronic Health Records (EHRs). Significant contributions also cover uncertainty-aware time-to-event prediction, deep learning for medical imaging analysis (e.g., CheXbreak for chest X-rays, ECG diagnosis, fMRI), and advancements in clinical natural language processing for medical code prediction and dialogue summarization using models like GPT-3. The conference further highlights innovations in federated classifier selection for clinical collaborations, intraoperative adverse event detection, and risk score learning for conditions such as COVID-19 and Alzheimer's Disease, alongside new datasets like MIMIC-SBDH.

Key takeaway

The 6th Machine Learning for Healthcare Conference presents diverse advancements in applying ML to critical clinical challenges. Papers introduce novel techniques such as deep generative models for EHR synthesis (EVA, POPCORN), uncertainty-aware survival analysis (Deep Kernel AFT, Deep Cox Mixtures), and robust diagnostic tools for medical imaging (CheXbreak, MedAug). These innovations offer practical solutions for improving diagnostics, personalizing patient predictions, and enhancing data utility for healthcare professionals and researchers.

Topics

• Clinical Machine Learning
• Electronic Health Records
• Medical Imaging Analysis
• Survival Analysis
• Offline Reinforcement Learning

Best for: AI Scientist, Research Scientist, Machine Learning Engineer

Related on AIssential

• v182: Proceedings of MLHC 2022 — The 7th Machine Learning for Healthcare Conference (MLHC 2022), held in Durham, NC, on August 5-6, 2022, showcased diverse advancements in applying machine learning to critical medical challenges.
• v219: Proceedings of Machine Learning for Healthcare 2023 — Volume 219 presents the proceedings of the 8th Machine Learning for Healthcare Conference, held on August 11-12, 2023, in New York, USA, featuring a diverse collection of research applying machine lea…
• UniRG: Scaling medical imaging report generation with multimodal reinforcement learning — Reinforcement learning with clinically meaningful rewards significantly improves medical vision-language model reliability and generality.
• A Reproducible Log-Driven AutoML Framework for Interpretable Pipeline Optimization in Healthcare Risk Prediction — Reproducible, log-driven AutoML reveals healthcare risk prediction pipeline component importance and search space redundancy.
• Comparing decentralized machine learning and AI clinical models to local and centralized alternatives: a systematic review — Decentralized learning offers clinically acceptable performance in healthcare, especially for privacy-constrained contexts, outperforming local learning.
• From Hazard Functions to Language Space: Cox-Supervised Distillation of Survival Risk into a Large Language Model — Large language models can internalize and predict survival risk from structured clinical data via text-based distillation.

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by Proceedings of Machine Learning Research.