Medical Information Mart for Intensive
Medical Information Mart for Intensive Care (MIMIC) datasets, particularly MIMIC-III and MIMIC-IV, are driving significant advancements in applying machine learning to various clinical tasks. Current research focuses on improving the accuracy and efficiency of diagnostic tools, including mortality prediction, disease classification (both cardiac and non-cardiac), and report generation using models like XGBoost and large language models (LLMs), often incorporating multimodal data (e.g., images, time series, and clinical notes). This work holds substantial promise for improving clinical decision-making, resource allocation, and ultimately, patient outcomes by leveraging the rich information contained within electronic health records.
Papers
LTCXNet: Advancing Chest X-Ray Analysis with Solutions for Long-Tailed Multi-Label Classification and Fairness Challenges
Chin-Wei Huang, Mu-Yi Shen, Kuan-Chang Shih, Shih-Chih Lin, Chi-Yu Chen, Po-Chih Kuo
HIST-AID: Leveraging Historical Patient Reports for Enhanced Multi-Modal Automatic Diagnosis
Haoxu Huang, Cem M. Deniz, Kyunghyun Cho, Sumit Chopra, Divyam Madaan