Papers for

healthcare ai engineers

Papers whose findings have a practical use for this group, as judged from the abstract. Open a paper to read what it means in practice.

Study treats AI agents like people to improve their safety and reliability

Do AI Agents need Preventive Care? An Epidemiologic Cohort Design for Agentic AI

Abstract: As AI agents increasingly perform tasks autonomously across sectors, including healthcare and biomedical research, understanding how their behaviors develop and influence outcomes over time becomes increasingly important. Existing evaluations largely focus on model capabilities and performance on predefined tasks. Here, we propose an epidemiologic cohort design for studying agentic AI, drawing on principles established in human population studies. The framework treats individual agents as members of a cohort, their underlying model as genetic makeup, accumulated work as their time scale, and naturally occurring behaviors as exposures that may influence subsequent outcomes. By following agents longitudinally, this approach enables investigation of behavioral determinants of adverse events, estimation of their potential preventable burden under causal assumptions, and characterization of behavioral changes over an agent's work life. We illustrate the design through two prospective observational cohorts examining curiosity and probing as a safety question, and data checking and analytical accuracy as a reliability question. Together, the framework and illustrative studies provide a generalizable blueprint for designing epidemiologic cohorts of AI agents.

Mon 9 NovArtificial IntelligenceMultiagent Systems
The gist
AI agents are computer programs that work on their own to do tasks. This paper suggests studying these agents like doctors study groups of people over time to understand their behavior and outcomes. The researchers propose treating each agent’s work history and behaviors as if they were a person's health and lifestyle, to find what causes problems or makes them more reliable. They show this idea by looking at how curiosity or careful checking affects AI agent safety and accuracy. This method could help design better and safer AI systems by watching their actions continuously.
Open → 2611.13333v1

Neural clustering improves personalized treatment rules with many options

Neural Supervised Clustering for Learning Individualized Treatment Rules with Many Treatments

Abstract: Learning an individualized treatment rule (ITR) with many treatments requires estimating treatment-specific effects from limited observations per arm. Treatments with similar conditional responses can share information, but linear contrast models may miss nonlinear heterogeneity. We propose nonlinear supervised clustering fusion (NSCF), a neural Q-learning framework with a shared nonlinear covariate representation and low-dimensional treatment embeddings. Pairwise fusion groups treatments with similar learned contrasts, yielding an ITR and an interpretable partition. An alternating algorithm combines neural training, embedding fusion, validation-based partition selection, and unpenalized grouped refitting. NSCF lowers regret relative to a strong neural reference in nonlinear exact-sharing simulations; perturbations characterize the limits of hard sharing. In a melanoma xenograft application, NSCF improves average policy value over the neural reference. These results support nonlinear representation learning with supervised treatment fusion when treatment effects exhibit shared structure.

Tue 13 OctMachine Learning
The gist
When deciding the best treatment for someone from many options, there is often too little data for each choice to know what works best. The authors propose a neural method that groups similar treatments based on how they work and learns from shared patterns, even if effects are complex and nonlinear. This method both improves treatment choice and produces understandable groupings of treatments. They tested it with simulations and a cancer treatment example, finding it performs better than some strong existing methods.
Open → 2610.19104v1

EviGen improves clinical outcome predictions with verifiable evidence

EviGen: Predictive Evidence Scaffolding for Verifiable Clinical Rationale Generation

Abstract: Longitudinal electronic health records (EHRs) capture years of patient history across notes, codes, labs, and procedures, and contain evidence needed to reason about likely clinical outcomes. However, comprehensive clinician review of these records is impractical, and LLM-based processing is costly and often unreliable, missing some relevant observations while hallucinating others. We therefore propose EviGen, a three-layer framework for verifiable clinical rationale generation that addresses these challenges. The first layer is a patient-conditioned retriever that uses learnable queries to find evidence predictive of, not just textually relevant to, a clinical outcome and ranks it by prediction attribution scores. The second layer is an LLM generator that consumes this ranked evidence as a scaffold to produce a clinical rationale grounded in the retrieved spans. The third layer is a process-supervised verifier that checks the generated rationale at the reasoning-step level, flagging unreliable claims. Across three medical prediction datasets, EviGen improves prediction performance and rationale faithfulness over full-context LLM and RAG baselines, and is preferred by clinical reviewers in a usability evaluation.

Wed 16 SeptComputation and Language
The gist
Long-term patient records hold lots of valuable clues for predicting medical outcomes, but reading all this information is too time-consuming for doctors. The authors built EviGen to automatically find the most predictive pieces of evidence from patient histories and then generate clear explanations for clinical predictions. This system also double-checks its own reasoning to avoid mistakes. When tested, EviGen gave better predictions and explanations than other AI methods and was preferred by medical reviewers.
Open → 2609.18852v1

Question-guided token pruning reduces privacy risks in vision language models

Don't Send What You Don't Need: Question-Guided Token Pruning as a Privacy Defense for Vision-Language Models

Abstract: Visual Question Answering (VQA) with Vision-Language Models (VLMs) is increasingly used in privacy-sensitive and bandwidth-constrained settings. Federated Learning (FL), Split Learning (SL), and U-Shaped Split Learning (USL) keep raw data local, but transmitting all visual tokens across a model partition remains costly and can expose private information. We propose QPriv-VL, a question-guided, privacy-aware token-pruning framework for FL, SL, and USL that prunes visual tokens before transmission based on task utility and privacy sensitivity. Its core component is a lightweight Dynamic Threshold Predictor (DTP) that jointly estimates a sample-specific pruning ratio and a token-level retention mask in one forward pass. DTP combines question relevance, computed from cross-modal similarity between visual patches and the pooled question embedding, with a sensitivity signal derived from frozen DINOv2 features. This allows the model to suppress potentially sensitive regions while preserving patches useful for answering the question, without requiring sensitivity labels. We evaluate QPriv-VL on GQA, OK-VQA, VQAv2, SLAKE, VQA-RAD, and PathVQA against four privacy attack families: FSHA, FORA, iDLG, and attribute-inference membership inference attacks. DTP matches or outperforms fixed-ratio pruning while using substantially fewer transmitted tokens. On VQA-RAD, it reduces membership-inference attack success from 0.99 to 0.76-0.79, lowers FSHA and FORA reconstruction PSNR relative to fixed-ratio pruning, and preserves competitive VQA accuracy using about 40% of the original visual-token budget. A sensitivity exclusion ratio of 1.20 +/- 0.18 indicates preferential removal of privacy-sensitive patches, while explainability analysis shows that retention adapts to question semantics rather than generic visual saliency.

Mon 14 SeptComputer Vision and Pattern RecognitionArtificial IntelligenceCryptography and Security
The gist
When AI systems answer questions about images, sometimes sending all image details risks exposing private information. The authors created a way to smartly pick only the important parts of an image related to the question, while hiding sensitive parts. Their method uses a tiny helper that decides which image pieces to keep or drop based on how private or useful they are. This reduces how much data is shared and lowers chances of privacy attacks, all without hurting the AI’s ability to answer questions well.
Open → 2609.15671v1