HIST-AID: Leveraging Historical Patient Reports for Enhanced Multi-Modal Automatic Diagnosis

Haoxu Huang, Cem Deniz, Kyunghyun Cho, Sumit Chopra, Divyam Madaan
Proceedings of the 4th Machine Learning for Health Symposium, PMLR 259:502-523, 2025.

Abstract

Chest X-ray imaging is a widely accessible and non-invasive diagnostic tool for detecting thoracic abnormalities. While numerous AI models assist radiologists in interpreting these images, most overlook patients’ historical data. To bridge this gap, we introduce Temporal MIMIC dataset, which integrates five years of patient history, including radiographic scans and reports from MIMIC-CXR and MIMIC-IV, encompassing 12,221 patients and thirteen pathologies. Building on this, we present HIST-AID, a framework that enhances automatic diagnostic accuracy using historical reports. HIST-AID emulates the radiologist’s comprehensive approach, leveraging historical data to improve diagnostic accuracy. Our experiments demonstrate significant improvements, with AUROC increasing by 6.56% and AUPRC by 9.51% compared to models that rely solely on radiographic scans. These gains were consistently observed across diverse demo- graphic groups, including variations in gender, age, and racial categories. We show that while recent data boost performance, older data may reduce accuracy due to changes in patient conditions. Our work paves the potential of incor- porating historical data for more reliable automatic diagnosis, providing critical support for clinical decision-making.

Cite this Paper


BibTeX
@InProceedings{pmlr-v259-huang25a, title = {HIST-AID: Leveraging Historical Patient Reports for Enhanced Multi-Modal Automatic Diagnosis}, author = {Huang, Haoxu and Deniz, Cem and Cho, Kyunghyun and Chopra, Sumit and Madaan, Divyam}, booktitle = {Proceedings of the 4th Machine Learning for Health Symposium}, pages = {502--523}, year = {2025}, editor = {Hegselmann, Stefan and Zhou, Helen and Healey, Elizabeth and Chang, Trenton and Ellington, Caleb and Mhasawade, Vishwali and Tonekaboni, Sana and Argaw, Peniel and Zhang, Haoran}, volume = {259}, series = {Proceedings of Machine Learning Research}, month = {15--16 Dec}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v259/main/assets/huang25a/huang25a.pdf}, url = {https://proceedings.mlr.press/v259/huang25a.html}, abstract = {Chest X-ray imaging is a widely accessible and non-invasive diagnostic tool for detecting thoracic abnormalities. While numerous AI models assist radiologists in interpreting these images, most overlook patients’ historical data. To bridge this gap, we introduce Temporal MIMIC dataset, which integrates five years of patient history, including radiographic scans and reports from MIMIC-CXR and MIMIC-IV, encompassing 12,221 patients and thirteen pathologies. Building on this, we present HIST-AID, a framework that enhances automatic diagnostic accuracy using historical reports. HIST-AID emulates the radiologist’s comprehensive approach, leveraging historical data to improve diagnostic accuracy. Our experiments demonstrate significant improvements, with AUROC increasing by 6.56% and AUPRC by 9.51% compared to models that rely solely on radiographic scans. These gains were consistently observed across diverse demo- graphic groups, including variations in gender, age, and racial categories. We show that while recent data boost performance, older data may reduce accuracy due to changes in patient conditions. Our work paves the potential of incor- porating historical data for more reliable automatic diagnosis, providing critical support for clinical decision-making.} }
Endnote
%0 Conference Paper %T HIST-AID: Leveraging Historical Patient Reports for Enhanced Multi-Modal Automatic Diagnosis %A Haoxu Huang %A Cem Deniz %A Kyunghyun Cho %A Sumit Chopra %A Divyam Madaan %B Proceedings of the 4th Machine Learning for Health Symposium %C Proceedings of Machine Learning Research %D 2025 %E Stefan Hegselmann %E Helen Zhou %E Elizabeth Healey %E Trenton Chang %E Caleb Ellington %E Vishwali Mhasawade %E Sana Tonekaboni %E Peniel Argaw %E Haoran Zhang %F pmlr-v259-huang25a %I PMLR %P 502--523 %U https://proceedings.mlr.press/v259/huang25a.html %V 259 %X Chest X-ray imaging is a widely accessible and non-invasive diagnostic tool for detecting thoracic abnormalities. While numerous AI models assist radiologists in interpreting these images, most overlook patients’ historical data. To bridge this gap, we introduce Temporal MIMIC dataset, which integrates five years of patient history, including radiographic scans and reports from MIMIC-CXR and MIMIC-IV, encompassing 12,221 patients and thirteen pathologies. Building on this, we present HIST-AID, a framework that enhances automatic diagnostic accuracy using historical reports. HIST-AID emulates the radiologist’s comprehensive approach, leveraging historical data to improve diagnostic accuracy. Our experiments demonstrate significant improvements, with AUROC increasing by 6.56% and AUPRC by 9.51% compared to models that rely solely on radiographic scans. These gains were consistently observed across diverse demo- graphic groups, including variations in gender, age, and racial categories. We show that while recent data boost performance, older data may reduce accuracy due to changes in patient conditions. Our work paves the potential of incor- porating historical data for more reliable automatic diagnosis, providing critical support for clinical decision-making.
APA
Huang, H., Deniz, C., Cho, K., Chopra, S. & Madaan, D.. (2025). HIST-AID: Leveraging Historical Patient Reports for Enhanced Multi-Modal Automatic Diagnosis. Proceedings of the 4th Machine Learning for Health Symposium, in Proceedings of Machine Learning Research 259:502-523 Available from https://proceedings.mlr.press/v259/huang25a.html.

Related Material