The paper introduces a bounded updating framework for ICU prediction models that freezes the physiological encoder while allowing updates only to the treatment pathway and fusion head. Experiments on 84,792 MIMIC-IV ICU stays across four temporal shifts show that selective adaptation yields more stable explanations—higher rank correlation, better top‑5 feature agreement, and improved retrieval stability—compared to full model adaptation. Predictive performance varies by outcome, but the results demonstrate that explanation stability is governed by the structural boundaries of allowed adaptation rather than merely by freezing components.
By Fatema Ferdous Tamanna, K. M. Merajul Arefin, Md. Abdul Masud
arXiv:2607. 19020v1 Announce Type: cross Abstract: Background: Clinical decision support systems degrade silently as treatment protocols evolve, yet standard adaptation methods treat models as monolithic blocks, unable to distinguish stable patient physiology from shifting institutional practice.
By Fatema Ferdous Tamanna, K. M. Merajul Arefin, Md. Abdul Masud
The paper investigates how to evaluate bounded model updates for ICU prediction models using the MIMIC‑IV database. After auditing the data for recording artifacts and cleaning it, the authors compare a bounded update that retrains only treatment‑context and fusion parameters with full fine‑tuning across two model families and two targets, finding that bounded updates are non‑inferior or superior in most cases but depend on target, model family, and learning‑rate tuning. The study also shows that audits should precede update evaluation and that the update boundary should be chosen per target.
By Fatema Ferdous Tamanna, K. M. Merajul Arefin, Md. Abdul Masud
arXiv:2608. 12805v1 Announce Type: new Abstract: Access to clinical data is essential for developing reliable healthcare machine learning systems, but direct use of electronic health records is constrained by privacy regulation, institutional review, data-use agreements, and the risk of re-identification.
By Akanta Das, Al Amin Farhad, Mrinmoy Sarkar Anto, David Rehkopf, Ayin Vala, Tanmoy Sarkar Pias
arXiv:2607. 26752v1 Announce Type: new Abstract: Medical world models aim to learn a latent state of patient or organ physiology and a transition function that forecasts how that state evolves under interventions, supporting downstream tasks from imaging-based diagnosis to digital-twin treatment planning.
By Behraj Khan, Shabir Ahmad, Syed Ahmad Chan Bukhari, Tahir Qasim Syed
arXiv:2607. 20453v1 Announce Type: cross Abstract: Large language models show promise for clinical prediction, but zero-shot performance on specialized tasks is limited by incomplete domain knowledge, especially for smaller locally deployable models.
By Jessica Sena, Shesadree Priyadarshani, Miguel Contreras, Bharat Gandhi, Scott Siegel, Subhash Nerella, Parisa Rashidi
arXiv:2605.20292v2 Announce Type: replace
Abstract: Numerical time-series models effectively process irregular electronic health record (EHR) trajectories, but do not expose which temporal patterns s...
By Kwanhyung Lee, Juhwan Choi, Jongheon Kim, Joohyung Lee, Hyeongwon Jang, Jeonguk Lee, Jisoo Jung, Eunho Yang
The paper introduces a formal framework and benchmark for time‑series world models (TSWMs) that separates state, actions, and exogenous inputs, and defines a new metric called mechanism consistency to evaluate whether model predictions move in the expected direction when actions change. Experiments on eight public datasets show that using a frozen latent prediction space and gated output fusion improves prediction accuracy, while prediction error and mechanism consistency often diverge, with the best‑performing models sometimes failing to exhibit consistent directional responses. Adding a directional supervision loss significantly boosts mechanism consistency without affecting mean‑absolute error, providing a practical recipe for building more reliable TSWMs.
By Haochen Zhang, Jiaheng Guo, Zhen Xu, Zachary Plotkin, Nicholas Konz, Zhen Tan, Tianlong Chen
INTERVenE introduces Transformer models that use a knowledge‑based temporal abstraction (KBTA) token stream of named clinical concepts instead of raw measurements, enabling per‑token attributions to resolve directly to clinical concepts. Two variants are offered: an auto‑regressive decoder that generates future abstraction trajectories with step‑wise risk readouts, and a bidirectional encoder that jointly predicts risk and time‑to‑event in a single pass. On 57,078 MIMIC‑IV admissions, the encoder variant outperforms neural baselines with a support‑weighted AUPRC of 0.672 and AUROC of 0.901, while the decoder provides complementary token‑level risk trajectories.
By Shahar Oded, Yuval Shahar
arXiv:2610.02957v1 Announce Type: new
Abstract: World models learn state evolution from trajectories, making access to temporal context a central training requirement. Federated learning can use dist...
By Yipan Wei, Zhaokun Yan, Ziming Hong, Jiaqi Wu, Lixu Wang
arXiv:2610.33592v1 Announce Type: new
Abstract: Augmented large language model (LLM) systems have reported high benchmark accuracies in rare disease diagnosis, yet their real-world clinical utility r...
By Jian Yang, Peilin Liu, Xiaohan Li, Songda Li, Yu Cao, Yuqing Feng, Haomin Li
arXiv:2607. 00127v1 Announce Type: new Abstract: Survival analysis models time-to-event data, but in clinical settings training data are costly and scarce: events accrue over years of follow-up, cohorts are small, and privacy regulations restrict sharing across institutions.
By Niccol\`o Maria Rizzi, Eugenio Lomurno, Alberto Archetti, Matteo Matteucci