Topic
healthcare ai
EEG Foundation Models Show Promise for Burst-Suppression Detection in ICU Without Patient-Specific Calibration
A new study on arXiv evaluates three EEG foundation models—REVE-base, LUNA-large, and LuMamba-Tiny—for automatic burst-suppression detection in ICU patients, finding REVE-base achieves the highest event-based F1-score (0.868) and reduces burst-per-minute error by 52.1% compared to a task-specific EEGNet baseline.
SL-S4Wave: Self-Supervised Learning Framework Improves ECG and EEG Analysis with State Space Models
Researchers propose SL-S4Wave, a self-supervised learning framework combining contrastive learning with structured state space models (S4) to analyze long-sequence physiological waveforms. The model outperforms state-of-the-art baselines in arrhythmia detection and EEG tasks, demonstrates strong label efficiency, and generalizes to unseen arrhythmia types.
AI Model Predicts Five-Year Heart Failure Risk from 24-Hour ECG Data
Researchers at Technion and Leumit Health Services developed DeepHHF, a deep learning model that analyzes 24-hour single-lead ECG recordings to predict heart failure risk within five years. The model achieved an AUC of 0.80, outperforming shorter segments and clinical scores, and identified high-risk individuals with a twofold chance of hospitalization or death.
Before the Labels: How Dataset Construction Biases Suicidality Detection in Clinical Text
A new paper from arXiv argues that clinical NLP datasets built from electronic health records encode specific operationalizations of suicidality, shaped by governance constraints, ICD-based cohort selection, and annotation practices. The authors demonstrate that identical labels can subsume heterogeneous clinical framings, raising concerns for AI-driven healthcare decisions.
Agentic RAG Pipeline Achieves 96.5% Clinician Acceptance in Clinical Information Extraction
Standard retrieval-augmented generation fails on clinical data due to missing metadata and cross-document dependencies. Researchers at University Medicine Essen deployed ACIE, an on-premise agentic RAG pipeline, that reasons over complete patient contexts and grounds answers in source passages. In an independent study with 7,326 clinician judgments, extractions were accepted 96.5% of the time, with per-type acceptance ranging from 80% to 99%.
AI Pace-Mapping System Uses Continual Learning to Cut Pacing Sites by 67% for Ventricular Tachycardia
Researchers introduced cAPM, a continuous AI-assisted pace-mapping system for ventricular tachycardia ablation. Using a surrogate neural network, active learning, and continual learning, cAPM reduced required pacing sites from 13.7 to 4.5 while improving localization probability from 38% to 81% within 5 mm accuracy, compared to the state-of-the-art active-learning method.
Deterministic Integrity Gates Verify LLM-Assisted Clinical Manuscripts Without False Positives
A new architecture from arXiv introduces deterministic integrity gates for verifying LLM-assisted clinical manuscripts. The MedSci Skills toolkit uses 43 skills with a 21-detector deterministic tier, catching all 27 injected defects with zero false positives, compared to an LLM reviewer's 11 detections.
Federated Medical Image Segmentation under Real-World Label Noise: A Benchmark Suite for Noisy Label Learning Method Selection
Federated learning enables collaborative medical image segmentation without centralizing sensitive data, but real-world label noise hampers deployment. A new benchmark suite combines diverse real-world noisy datasets, client-noise scenarios, and targeted evaluation to support systematic assessment of federated noisy label learning methods, addressing the gap left by synthetic noise studies.
CAP Achieves 87.6% Improvement in Respiratory Rate Prediction via Patient-Level PPG Learning
Researchers introduce Clinical Anchored Pretraining (CAP) for PPG signals, which anchors representations to patient-level clinical semantics. CAP outperforms baselines on four tasks, with a remarkable 87.6% relative improvement in respiratory rate prediction and average 26.7% gain across tasks.
Medical Image Segmentation Survey: U-Net, Transformers, SAM and Clinical Translation Challenges
A new arXiv survey systematically reviews medical image segmentation methods based on U-Net, Transformer, and SAM architectures. It covers public datasets, evaluation metrics, and key challenges, aiming to guide future research and clinical adoption. The authors have made all related resources publicly available on GitHub.
VIBEMed Introduces Self-Evolving Multi-Agent Framework for Clinical Decision Support
VIBEMed is a multi-agent framework with a built-in self-evolution mechanism and architecture-level safety sandbox for clinical decision support. It integrates three specialized agents: Clinical Diagnostic Agent (CDA), Therapeutic Execution Agent (TEA), and Clinical Evolution Manager Agent (CEMA). Experimental results show superior performance in complex clinical cases requiring integrated decision-making and longitudinal planning, particularly in oncology.