MIMIC-III

Overview edit

Variants: MIMIC-III

Associated Benchmarks

This dataset is used in 7 benchmarks:

• Medical Code Prediction - Metrics: Micro-F1, Macro-F1, Micro-AUC, Macro-AUC, Precision@5, Precision@8, Precision@15, mAP
• Length-of-Stay prediction - Metrics: Accuracy (LOS>3 Days), Accuracy (LOS>7 Days)
• Multi-Label Classification Of Biomedical Texts - Metrics: Micro F1
• Multi-Label Text Classification - Metrics: AUC, Macro F1, Macro Precision, Macro Recall, Micro Precision, Micro Recall, Micro-F1, P@5, Precision, Recall
• Mortality Prediction - Metrics: F1 score, Precision, Recall, Accuracy
• Blood pressure estimation - Metrics: MAE for SBP [mmHg], MAE for DBP [mmHg], Mean Squared Error, MAE
• Multivariate Time Series Forecasting - Metrics: MSE, NegLL

Recent Benchmark Submissions

Research Papers

Recent papers with results on this dataset:

• A General Knowledge Injection Framework for ICD Coding (2025) -
• An Unsupervised Approach to Achieve Supervised-Level Explainability in Healthcare Records (2024) -
• Functional Latent Dynamics for Irregularly Sampled Time Series Forecasting (2024) -
• Forecasting Irregularly Sampled Time Series using Graphs (2023) -
• Knowledge Injected Prompt Based Fine-tuning for Multi-label Few-shot ICD Coding (2022) -
• Unsupervised Pre-Training on Patient Population Graphs for Patient-Level Predictions (2022) -
• Code Synonyms Do Matter: Multiple Synonyms Matching Network for Automatic ICD Coding (2022) -
• Neural Flows: Efficient Alternative to Neural ODEs (2021) -
• Read, Attend, and Code: Pushing the Limits of Medical Codes Prediction from Clinical Notes by Machines (2021) -
• MeDAL: Medical Abbreviation Disambiguation Dataset for Natural Language Understanding Pretraining (2020) -
• Explainable Automated Coding of Clinical Notes using Hierarchical Label-wise Attention Networks and Label Embedding Initialisation (2020) -
• Predicting Multiple ICD-10 Codes from Brazilian-Portuguese Clinical Notes (2020) -
• A Label Attention Model for ICD Coding from Clinical Text (2020) -
• ICD Coding from Clinical Text Using Multi-Filter Residual Convolutional Neural Network (2019) -
• MIMIC-Extract: A Data Extraction, Preprocessing, and Representation Pipeline for MIMIC-III (2019) -
• GRU-ODE-Bayes: Continuous modeling of sporadically-observed time series (2019) -
• Early hospital mortality prediction using vital signals (2018) -
• Explainable Prediction of Medical Codes from Clinical Text (2018) -
• An Empirical Evaluation of Deep Learning for ICD-9 Code Assignment using MIMIC-III Clinical Notes (2018) -
• Long-term Blood Pressure Prediction with Deep Recurrent Neural Networks (2017) -