The field of clinical predictive modeling and decision support is rapidly advancing, driven by the increasing availability of electronic health records (EHRs) and advances in machine learning and artificial intelligence. Recent developments have focused on improving the accuracy and interpretability of predictive models, as well as enhancing their ability to support clinical decision-making. Notably, the use of large language models and multimodal learning approaches has shown promise in improving the performance of predictive models. Furthermore, there is a growing emphasis on developing transparent and explainable models that can provide insights into the factors driving predictions, which is critical for building trust and ensuring safe deployment in clinical settings. Noteworthy papers in this area include Toward Scalable Early Cancer Detection, which demonstrated the potential of EHR-based predictive models to identify high-risk individuals, and CURENet, which introduced a multimodal model for efficient chronic disease prediction. Additionally, the development of benchmarks such as VitalBench and SurvBench is expected to facilitate the evaluation and comparison of predictive models, driving further innovation in the field.
Advances in Clinical Predictive Modeling and Decision Support
Sources
Toward Scalable Early Cancer Detection: Evaluating EHR-Based Predictive Models Against Traditional Screening Criteria
CURENet: Combining Unified Representations for Efficient Chronic Disease Prediction
Early GVHD Prediction in Liver Transplantation via Multi-Modal Deep Learning on Imbalanced EHR Data
ClinStructor: AI-Powered Structuring of Unstructured Clinical Texts
Additive Large Language Models for Semi-Structured Text
SurvBench: A Standardised Preprocessing Pipeline for Multi-Modal Electronic Health Record Survival Analysis
Interpretable Fine-Gray Deep Survival Model for Competing Risks: Predicting Post-Discharge Foot Complications for Diabetic Patients in Ontario
A Smart-Glasses for Emergency Medical Services via Multimodal Multitask Learning
Artificial Intelligence-Enabled Spirometry for Early Detection of Right Heart Failure
Towards Multimodal Representation Learning in Paediatric Kidney Disease
Generalist Foundation Models Are Not Clinical Enough for Hospital Operations
VitalBench: A Rigorous Multi-Center Benchmark for Long-Term Vital Sign Prediction in Intraoperative Care
Real-Time Mobile Video Analytics for Pre-arrival Emergency Medical Services
Hybrid Modeling of Photoplethysmography for Non-invasive Monitoring of Cardiovascular Parameters
A Method for Characterizing Disease Progression from Acute Kidney Injury to Chronic Kidney Disease
Machine Learning Models for Predicting Smoking-Related Health Decline and Disease Risk
FRIENDS GUI: A graphical user interface for data collection and visualization of vaping behavior from a passive vaping monitor
Cost-Aware Prediction (CAP): An LLM-Enhanced Machine Learning Pipeline and Decision Support System for Heart Failure Mortality Prediction
Transparent Early ICU Mortality Prediction with Clinical Transformer and Per-Case Modality Attribution