Barttender: An approachable & interpretable way to compare medical imaging and non-imaging data

Ayush Singla, Shakson Isaac, Chirag J Patel
Proceedings of the 4th Machine Learning for Health Symposium, PMLR 259:976-990, 2025.

Abstract

Imaging-based deep learning has transformed healthcare research, yet its clinical adoption remains limited due to challenges in comparing imaging models with traditional non-imaging and tabular data. To bridge this gap, we introduce Barttender, an interpretable framework that uses deep learning for the direct comparison of the utility of imaging versus non-imaging tabular data for tasks like disease prediction. Barttender converts non-imaging tabular features, such as scalar data from electronic health records, into grayscale bars, facilitating an interpretable and scalable deep learning based modeling of both data modalities. Our framework allows researchers to evaluate differences in utility through performance measures, as well as local (sample-level) and global (population-level) explanations. We introduce a novel measure to define global feature importances for image-based deep learning models, which we call gIoU. Experiments on the CheXpert and MIMIC datasets with chest X-rays and scalar data from electronic health records show that Barttender performs comparably to traditional methods and offers enhanced explainability using deep learning models.

Cite this Paper


BibTeX
@InProceedings{pmlr-v259-singla25a, title = {Barttender: An approachable & interpretable way to compare medical imaging and non-imaging data}, author = {Singla, Ayush and Isaac, Shakson and Patel, Chirag J}, booktitle = {Proceedings of the 4th Machine Learning for Health Symposium}, pages = {976--990}, 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/singla25a/singla25a.pdf}, url = {https://proceedings.mlr.press/v259/singla25a.html}, abstract = {Imaging-based deep learning has transformed healthcare research, yet its clinical adoption remains limited due to challenges in comparing imaging models with traditional non-imaging and tabular data. To bridge this gap, we introduce Barttender, an interpretable framework that uses deep learning for the direct comparison of the utility of imaging versus non-imaging tabular data for tasks like disease prediction. Barttender converts non-imaging tabular features, such as scalar data from electronic health records, into grayscale bars, facilitating an interpretable and scalable deep learning based modeling of both data modalities. Our framework allows researchers to evaluate differences in utility through performance measures, as well as local (sample-level) and global (population-level) explanations. We introduce a novel measure to define global feature importances for image-based deep learning models, which we call gIoU. Experiments on the CheXpert and MIMIC datasets with chest X-rays and scalar data from electronic health records show that Barttender performs comparably to traditional methods and offers enhanced explainability using deep learning models.} }
Endnote
%0 Conference Paper %T Barttender: An approachable & interpretable way to compare medical imaging and non-imaging data %A Ayush Singla %A Shakson Isaac %A Chirag J Patel %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-singla25a %I PMLR %P 976--990 %U https://proceedings.mlr.press/v259/singla25a.html %V 259 %X Imaging-based deep learning has transformed healthcare research, yet its clinical adoption remains limited due to challenges in comparing imaging models with traditional non-imaging and tabular data. To bridge this gap, we introduce Barttender, an interpretable framework that uses deep learning for the direct comparison of the utility of imaging versus non-imaging tabular data for tasks like disease prediction. Barttender converts non-imaging tabular features, such as scalar data from electronic health records, into grayscale bars, facilitating an interpretable and scalable deep learning based modeling of both data modalities. Our framework allows researchers to evaluate differences in utility through performance measures, as well as local (sample-level) and global (population-level) explanations. We introduce a novel measure to define global feature importances for image-based deep learning models, which we call gIoU. Experiments on the CheXpert and MIMIC datasets with chest X-rays and scalar data from electronic health records show that Barttender performs comparably to traditional methods and offers enhanced explainability using deep learning models.
APA
Singla, A., Isaac, S. & Patel, C.J.. (2025). Barttender: An approachable & interpretable way to compare medical imaging and non-imaging data. Proceedings of the 4th Machine Learning for Health Symposium, in Proceedings of Machine Learning Research 259:976-990 Available from https://proceedings.mlr.press/v259/singla25a.html.

Related Material