In cross-silo Federated Learning (FL), client selection is critical to ensure high model performance, yet it remains challenging due to data quality decompensation, budget constraints, and incentive compatibility. As training progresses, these factors exacerbate client heterogeneity and degrade global performance. Most existing approaches treat these challenges in isolation, making jointly optimizing multiple factors difficult. To address this, we propose Shapley-Bid Reputation Optimized Federated Learning (SBRO-FL), a unified framework integrating dynamic bidding, reputation modeling, and cost-aware selection. Clients submit bids based on their perceived data quality, and their contributions are evaluated using Shapley values to quantify their marginal impact on the global model. A reputation system, inspired by prospect theory, captures historical performance while penalizing inconsistency. The client selection problem is formulated as a 0-1 integer program that maximizes reputation-weighted utility under budget constraints. Experiments on FashionMNIST, EMNIST, CIFAR-10, and SVHN datasets show that SBRO-FL improves accuracy, convergence speed, and robustness, even in adversarial and low-bid interference scenarios. Our results highlight the importance of balancing data reliability, incentive compatibility, and cost efficiency to enable scalable and trustworthy FL deployments.
The proliferation of Large Language Models (LLMs) and Intelligent Virtual Agents acting as psychotherapists presents significant opportunities for expanding mental healthcare access. However, their deployment has also been linked to serious adverse outcomes, including user harm and suicide, facilitated by a lack of standardized evaluation methodologies capable of capturing the nuanced risks of therapeutic interaction. Current evaluation techniques lack the sensitivity to detect subtle changes in patient cognition and behavior during therapy sessions that may lead to subsequent decompensation. We introduce a novel risk taxonomy specifically designed for the systematic evaluation of conversational AI psychotherapists. Developed through an iterative process including review of the psychotherapy risk literature, qualitative interviews with clinical and legal experts, and alignment with established clinical criteria (e.g., DSM-5) and existing assessment tools (e.g., NEQ, UE-ATR), the taxonomy aims to provide a structured approach to identifying and assessing user/patient harms. We provide a high-level overview of this taxonomy, detailing its grounding, and discuss potential use cases. We discuss two use cases in detail: monitoring cognitive model-based risk factors during a counseling conversation to detect unsafe deviations, in both human-AI counseling sessions and in automated benchmarking of AI psychotherapists with simulated patients. The proposed taxonomy offers a foundational step towards establishing safer and more responsible innovation in the domain of AI-driven mental health support.
We present an end-to-end deep learning framework for automated liver cirrhosis stage estimation from multi-sequence MRI. Cirrhosis is the severe scarring (fibrosis) of the liver and a common endpoint of various chronic liver diseases. Early diagnosis is vital to prevent complications such as decompensation and cancer, which significantly decreases life expectancy. However, diagnosing cirrhosis in its early stages is challenging, and patients often present with life-threatening complications. Our approach integrates multi-scale feature learning with sequence-specific attention mechanisms to capture subtle tissue variations across cirrhosis progression stages. Using CirrMRI600+, a large-scale publicly available dataset of 628 high-resolution MRI scans from 339 patients, we demonstrate state-of-the-art performance in three-stage cirrhosis classification. Our best model achieves 72.8% accuracy on T1W and 63.8% on T2W sequences, significantly outperforming traditional radiomics-based approaches. Through extensive ablation studies, we show that our architecture effectively learns stage-specific imaging biomarkers. We establish new benchmarks for automated cirrhosis staging and provide insights for developing clinically applicable deep learning systems. The source code will be available at https://github.com/JunZengz/CirrhosisStage.
We propose the Multimodal Clinical Benchmark for Emergency Care (MC-BEC), a comprehensive benchmark for evaluating foundation models in Emergency Medicine using a dataset of 100K+ continuously monitored Emergency Department visits from 2020-2022. MC-BEC focuses on clinically relevant prediction tasks at timescales from minutes to days, including predicting patient decompensation, disposition, and emergency department (ED) revisit, and includes a standardized evaluation framework with train-test splits and evaluation metrics. The multimodal dataset includes a wide range of detailed clinical data, including triage information, prior diagnoses and medications, continuously measured vital signs, electrocardiogram and photoplethysmograph waveforms, orders placed and medications administered throughout the visit, free-text reports of imaging studies, and information on ED diagnosis, disposition, and subsequent revisits. We provide performance baselines for each prediction task to enable the evaluation of multimodal, multitask models. We believe that MC-BEC will encourage researchers to develop more effective, generalizable, and accessible foundation models for multimodal clinical data.
Models that can predict the occurrence of events ahead of time with low false-alarm rates are critical to the acceptance of decision support systems in the medical community. This challenging task is typically treated as a simple binary classification, ignoring temporal dependencies between samples, whereas we propose to exploit this structure. We first introduce a common theoretical framework unifying dynamic survival analysis and early event prediction. Following an analysis of objectives from both fields, we propose Temporal Label Smoothing (TLS), a simpler, yet best-performing method that preserves prediction monotonicity over time. By focusing the objective on areas with a stronger predictive signal, TLS improves performance over all baselines on two large-scale benchmark tasks. Gains are particularly notable along clinically relevant measures, such as event recall at low false-alarm rates. TLS reduces the number of missed events by up to a factor of two over previously used approaches in early event prediction.
Objective: Clinical notes contain information not present elsewhere, including drug response and symptoms, all of which are highly important when predicting key outcomes in acute care patients. We propose the automatic annotation of phenotypes from clinical notes as a method to capture essential information, which is complementary to typically used vital signs and laboratory test results, to predict outcomes in the Intensive Care Unit (ICU). Methods: We develop a novel phenotype annotation model to annotate phenotypic features of patients which are then used as input features of predictive models to predict ICU patient outcomes. We demonstrate and validate our approach conducting experiments on three ICU prediction tasks including in-hospital mortality, physiological decompensation and length of stay for over 24,000 patients by using MIMIC-III dataset. Results: The predictive models incorporating phenotypic information achieve 0.845 (AUC-ROC) to predict in-hospital mortality, 0.839 (AUC-ROC) for physiological decompensation and 0.430 (Kappa) for length of stay, all of which consistently outperform the baseline models leveraging only vital signs and laboratory test results. Moreover, we conduct a thorough interpretability study, showing that phenotypes provide valuable insights at the patient and cohort levels. Conclusion: The proposed approach demonstrates phenotypic information complements traditionally used vital signs and laboratory test results, improving significantly forecast of outcomes in the ICU.
Objective: Worldwide, heart failure (HF) is a major cause of morbidity and mortality and one of the leading causes of hospitalization. Early detection of HF symptoms and pro-active management may reduce adverse events. Approach: Twenty-eight participants were monitored using a smartphone app after discharge from hospitals, and each clinical event during the enrollment (N=110 clinical events) was recorded. Motion, social, location, and clinical survey data collected via the smartphone-based monitoring system were used to develop and validate an algorithm for predicting or classifying HF decompensation events (hospitalizations or clinic visit) versus clinic monitoring visits in which they were determined to be compensated or stable. Models based on single modality as well as early and late fusion approaches combining patient-reported outcomes and passive smartphone data were evaluated. Results: The highest AUCPr for classifying decompensation with a late fusion approach was 0.80 using leave one subject out cross-validation. Significance: Passively collected data from smartphones, especially when combined with weekly patient-reported outcomes, may reflect behavioral and physiological changes due to HF and thus could enable prediction of HF decompensation.
The Large Scale Visual Recognition Challenge based on the well-known Imagenet dataset catalyzed an intense flurry of progress in computer vision. Benchmark tasks have propelled other sub-fields of machine learning forward at an equally impressive pace, but in healthcare it has primarily been image processing tasks, such as in dermatology and radiology, that have experienced similar benchmark-driven progress. In the present study, we performed a comprehensive review of benchmarks in medical machine learning for structured data, identifying one based on the Medical Information Mart for Intensive Care (MIMIC-III) that allows the first direct comparison of predictive performance and thus the evaluation of progress on four clinical prediction tasks: mortality, length of stay, phenotyping, and patient decompensation. We find that little meaningful progress has been made over a 3 year period on these tasks, despite significant community engagement. Through our meta-analysis, we find that the performance of deep recurrent models is only superior to logistic regression on certain tasks. We conclude with a synthesis of these results, possible explanations, and a list of desirable qualities for future benchmarks in medical machine learning.
Respiratory complications due to coronavirus disease COVID-19 have claimed tens of thousands of lives in 2020. Many cases of COVID-19 escalate from Severe Acute Respiratory Syndrome (SARS-CoV-2) to viral pneumonia to acute respiratory distress syndrome (ARDS) to death. Extracorporeal membranous oxygenation (ECMO) is a life-sustaining oxygenation and ventilation therapy that may be used for patients with severe ARDS when mechanical ventilation is insufficient to sustain life. While early planning and surgical cannulation for ECMO can increase survival, clinicians report the lack of a risk score hinders these efforts. In this work, we leverage machine learning techniques to develop the PEER score, used to highlight critically ill patients with viral or unspecified pneumonia at high risk of mortality or decompensation in a subpopulation eligible for ECMO. The PEER score is validated on two large, publicly available critical care databases and predicts mortality at least as well as other existing risk scores. Stratifying our cohorts into low-risk and high-risk groups, we find that the high-risk group also has a higher proportion of decompensation indicators such as vasopressor and ventilator use. Finally, the PEER score is provided in the form of a nomogram for direct calculation of patient risk, and can be used to highlight at-risk patients among critical care patients eligible for ECMO.
Effective triage is critical to mitigating the effect of increased volume by accurately determining patient acuity, need for resources, and establishing effective acuity-based patient prioritization. The purpose of this retrospective study was to determine whether historical EHR data can be extracted and synthesized with clinical natural language processing (C-NLP) and the latest ML algorithms (KATE) to produce highly accurate ESI predictive models. An ML model (KATE) for the triage process was developed using 166,175 patient encounters from two participating hospitals. The model was then tested against a gold set that was derived from a random sample of triage encounters at the study sites and correct acuity assignments were recorded by study clinicians using the Emergency Severity Index (ESI) standard as a guide. At the two study sites, KATE predicted accurate ESI acuity assignments 75.9% of the time, compared to nurses (59.8%) and average individual study clinicians (75.3%). KATE accuracy was 26.9% higher than the average nurse accuracy (p-value < 0.0001). On the boundary between ESI 2 and ESI 3 acuity assignments, which relates to the risk of decompensation, KATE was 93.2% higher with 80% accuracy, compared to triage nurses with 41.4% accuracy (p-value < 0.0001). KATE provides a triage acuity assignment substantially more accurate than the triage nurses in this study sample. KATE operates independently of contextual factors, unaffected by the external pressures that can cause under triage and may mitigate the racial and social biases that can negatively affect the accuracy of triage assignment. Future research should focus on the impact of KATE providing feedback to triage nurses in real time, KATEs impact on mortality and morbidity, ED throughput, resource optimization, and nursing outcomes.
Progress of machine learning in critical care has been difficult to track, in part due to absence of public benchmarks. Other fields of research (such as computer vision and natural language processing) have established various competitions and public benchmarks. Recent availability of large clinical datasets has enabled the possibility of establishing public benchmarks. Taking advantage of this opportunity, we propose a public benchmark suite to address four areas of critical care, namely mortality prediction, estimation of length of stay, patient phenotyping and risk of decompensation. We define each task and compare the performance of both clinical models as well as baseline and deep learning models using eICU critical care dataset of around 73,000 patients. This is the first public benchmark on a multi-centre critical care dataset, comparing the performance of clinical gold standard with our predictive model. We also investigate the impact of numerical variables as well as handling of categorical variables on each of the defined tasks. The source code, detailing our methods and experiments is publicly available such that anyone can replicate our results and build upon our work.
Monitoring patients in ICU is a challenging and high-cost task. Hence, predicting the condition of patients during their ICU stay can help provide better acute care and plan the hospital's resources. There has been continuous progress in machine learning research for ICU management, and most of this work has focused on using time series signals recorded by ICU instruments. In our work, we show that adding clinical notes as another modality improves the performance of the model for three benchmark tasks: in-hospital mortality prediction, modeling decompensation, and length of stay forecasting that play an important role in ICU management. While the time-series data is measured at regular intervals, doctor notes are charted at irregular times, making it challenging to model them together. We propose a method to model them jointly, achieving considerable improvement across benchmark tasks over baseline time-series model. Our implementation can be found at \url{https://github.com/kaggarwal/ClinicalNotesICU}.
Patient subtyping based on temporal observations can lead to significantly nuanced subtyping that acknowledges the dynamic characteristics of diseases. Existing methods for subtyping trajectories treat the evolution of clinical observations as a homogeneous process or employ data available at regular intervals. In reality, diseases may have transient underlying states and a state-dependent observation pattern. In our paper, we present an approach to subtype irregular patient data while acknowledging the underlying progression of disease states. Our approach consists of two components: a probabilistic model to determine the likelihood of a patient's observation trajectory and a mixture model to measure similarity between asynchronous patient trajectories. We demonstrate our model by discovering subtypes of progression to hemodynamic instability (requiring cardiovascular intervention) in a patient cohort from a multi-institution ICU dataset. We find three primary patterns: two of which show classic signs of decompensation (rising heart rate with dropping blood pressure), with one of these showing a faster course of decompensation than the other. The third pattern has transient period of low heart rate and blood pressure. We also show that our model results in a 13% reduction in average cross-entropy error compared to a model with no state progression when forecasting vital signs.
With the improvement of medical data capturing, vast amount of continuous patient monitoring data, e.g., electrocardiogram (ECG), real-time vital signs and medications, become available for clinical decision support at intensive care units (ICUs). However, it becomes increasingly challenging to model such data, due to high density of the monitoring data, heterogeneous data types and the requirement for interpretable models. Integration of these high-density monitoring data with the discrete clinical events (including diagnosis, medications, labs) is challenging but potentially rewarding since richness and granularity in such multimodal data increase the possibilities for accurate detection of complex problems and predicting outcomes (e.g., length of stay and mortality). We propose Recurrent Attentive and Intensive Model (RAIM) for jointly analyzing continuous monitoring data and discrete clinical events. RAIM introduces an efficient attention mechanism for continuous monitoring data (e.g., ECG), which is guided by discrete clinical events (e.g, medication usage). We apply RAIM in predicting physiological decompensation and length of stay in those critically ill patients at ICU. With evaluations on MIMIC- III Waveform Database Matched Subset, we obtain an AUC-ROC score of 90.18% for predicting decompensation and an accuracy of 86.82% for forecasting length of stay with our final model, which outperforms our six baseline models.