XAI for Better Exploitation of Text in Medical Decision Support
Open accessAbstract
In electronic health records, text data is considered a valuable resource as it complements a medical history and may contain information that cannot be easily included in tables.But why does the inclusion of clinical texts as additional input into multimodal models, not always significantly improve the performance of medical decision-support systems?Explainable AI (XAI) might provide the answer.We examine which information in text and structured data influences the performance of models in the context of multimodal decision support for biomedical tasks.Using data from an intensive care unit and targeting a mortality prediction task, we compare information that has been considered relevant by XAI methods to the opinion of a physician.
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Integrity signals
Registry
Cites
Statistics
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
Attention mechanisms such as gated attention, cross-attention, or transformer encoders are increasingly adopted in multimodal AI.
“attention mechanisms such as gated attention [40,41], cross-attention [42–45], or transformer encoders [20,36,46–49] are increasingly adopted.”
From Multimodal Artificial Intelligence for Precision Critical Care: A Scoping Review (2026) · cites this paper as [40]
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Cites
The publisher hasn’t deposited this paper’s reference list.
Research works printed but not linked: 9
- Edward Choi, Mohammad Taha Bahadori, Andy Schuetz, Walter F. Stewart, and Jimeng Sun. 2016. Doctor ai: Predicting clinical events via recurrent neural networks. In Proceedings of the 1st Machine Learning for Healthcare Conference, volume 56 of Proceedings of Machine Learning Research, pages 301-318, Northeastern University, Boston, MA, USA. PMLR.
- Kexin Huang, Jaan Altosaar, and Rajesh Ranganath. 2019. Clinicalbert: Modeling clinical notes and predicting hospital readmission. arXiv preprint arXiv:1904.05342. · arXiv
- Alistair E W Johnson, Tom J Pollard, Lu Shen, Li- Wei H Lehman, Mengling Feng, Mohammad Ghas- semi, Benjamin Moody, Peter Szolovits, Leo An- thony Celi, and Roger G Mark. 2016. Mimic-iii, a freely accessible critical care database. Scientific data, 3(1):1-9.
- Weimin Lyu, Xinyu Dong, Rachel Wong, Songzhu Zheng, Kayley Abell-Hart, Fusheng Wang, and Chao Chen. 2022. A multimodal transformer: Fusing clini- cal notes with structured ehr data for interpretable in- hospital mortality prediction. AMIA ... Annual Sym- posium proceedings. AMIA Symposium, 2022:719- 728.
- Mitchell Naylor, Christi French, Samantha Terker, and Uday Kamath. 2021. Quantifying explainability in NLP and analyzing algorithms for performance- explainability tradeoff. Interpretable ML in Health- care workshop at ICML 2021.
- Zhi Qiao, X. Wu, Shen Ge, and Wei Fan. 2019. Mnn: Multimodal attentional neural networks for diagno- sis prediction. In International Joint Conference on Artificial Intelligence.
- Roland Roller, Aljoscha Burchardt, Nils Feldhus, Laura Seiffe, Klemens Budde, Simon Ronicke, and Bilgin Osmanodja. 2022a. An annotated corpus of textual explanations for clinical decision support. In Pro- ceedings of the Thirteenth Language Resources and Evaluation Conference, pages 2317-2326, Marseille, France. European Language Resources Association.
- Mukund Sundararajan, Ankur Taly, and Qiqi Yan. 2017. Axiomatic attribution for deep networks. In Interna- tional conference on machine learning, pages 3319- 3328. PMLR.
- Betty van Aken, Jens-Michalis Papaioannou, Manuel Mayrdorfer, Klemens Budde, Felix A. Gers, and Alexander Löser. 2021. Clinical outcome prediction from admission notes using self-supervised knowl- edge integration. In Proceedings of the 16th Con- ference of the European Chapter of the Association for Computational Linguistics: Main Volume, EACL 2021, Online, April 19 -23, 2021, pages 881-893. Association for Computational Linguistics.
Cited by
Sorted by citations
Ahn, J., Kang, B. G., Chang, M. Y., & Yoon, S. (2025) · Clinical and Experimental Otorhinolaryngology · doi:10.21053/ceo.2025-00121 · Cited by 10 · see citing works
XAI-based Data Visualization in Multimodal Medical DataPreprint
Sharma, S., Singh, M., Mcdaid, L. J., & Bhattacharyya, S. (2025) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.1101/2025.07.11.664302 · Cited by 6 · see citing works
Kandala, R., Moharir, A. K., & Nayak, D. A. (2025) · IEEE ... International Conference on Data Mining workshops · doi:10.1109/icdmw69685.2025.00156 · Cited by 5 · see citing works
Multimodal Artificial Intelligence for Precision Critical Care: A Scoping Review
Yang, M., Shi, N., Chen, H., Po, E. Y. P., Ng, P. Y., Lai, P. C. K., Wu, Y.-C., & Fong, D. Y. T. (2026) · Health Data Science · doi:10.34133/hds.0356 · Cited by 3 · see citing works
“While simple concatenation of deeply extracted features remains the most straightforward approach, attention mechanisms such as gated attention [ 40 , 41 ], cross-attention [ 42 – 45 ], or trans-former encoders [ 20 , 36 , 46 – 49 ] are increasingly adopted.”
Naik, M. D., Deepika, G., Jawalkar, A. P., Rajesh, M., Munigala, M., Srinivas, B., & Kumari, R. K. (2026) · AIP conference proceedings · doi:10.1063/5.0342272
- Authors
- Ravichandran, A. M., Grune, J., Feldhus, N., Burchardt, A., Roller, R., & Möller, S.
- Affiliations
- German Research Centre for Artificial Intelligence (DE)
- Venue
- Proceedings of the 23rd Workshop on Biomedical Natural Language Processing
- Published
- 2024
- Pages
- 506-513
- OpenAlex
- W4402671001
- Cited by
- 5
- References
- 0
- Citations by year
- 2026: 2 · 2025: 3
- FWCI
- 1.47
- Type
- conference paper
- Language
- en
- Open access
- hybrid
- doi.org
- Licence
- cc-by
- Locations
- Workshop on Biomedical Language Processing (BioNLP) (publishedVersion, cc-by)
- Funders
- Bundesministerium für Bildung und Forschung (01GP2202C)
- Topics
- Physical Sciences › Computer Science › Artificial Intelligence › Machine Learning in Healthcare
- Physical Sciences › Computer Science › Artificial Intelligence › Explainable Artificial Intelligence (XAI)
- Health Sciences › Medicine › Health Informatics › Artificial Intelligence in Healthcare and Education
- Keywords
- explainable AI, clinical text, electronic health records, multimodal models, clinical decision support, mortality prediction, intensive care unit, physician attitudes, structured data, clinical information extraction
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
