AI-Augmented Clinical Diagnostics: Integrating Deep Learning with Electronic Health Records and Imaging Data
DOI:
https://doi.org/10.15662/IJEETR.2024.0602007Keywords:
Deep learning, Electronic health records, Medical imaging, Multimodal fusion, Clinical decision support, AI-augmented diagnosticsAbstract
AI-enhanced clinical diagnostics will make the healthcare revolution by integrating deep learning with the rich contextual information stored in electronic health records (EHRs) and high-resolution imaging data. Fusion Multimodal fusions of pixel-level results of radiology or pathology images with structured and unstructured EHR data have been demonstrated to have higher diagnostics accuracy, risk stratification, and disease subtyping compared to single-modality models. The current literature has shown that combining convolutional neural networks (CNNs) with imaging and deep models of EHR, like recurrent or transformer, provides clinically relevant predictions of diseases like Alzheimer, breast cancer, and cardiometabolic diseases. Nevertheless, their implementation in practice in clinical settings has not yet been widespread since there is difficulty in terms of data quality, interoperability, algorithmic bias, interpretability, and EHR integration.
This paper suggests a single AI-enhanced clinical diagnostics system that incorporates deep learning pipelines of imaging and EHR data into an end-to-end and clinically consistent architecture. These frameworks include (1) ingestion of data, preprocessing, representation learning, multi-modal fusion (early, late, and hybrid), and explainability layers (inherently embedded in a clinical decision support and reporting interface). It is created to provide potential support to validation, transparent model behavior, and EHR seam, to enable clinician trust and adoption. We also describe a strategy of evaluation that integrates the conventional machine learning measures with the workflow and user-focused measures. The framework suggested will be beneficial in helping the researchers and healthcare organizations operationalize AI-enhanced, multimodal diagnostic systems that should advance beyond proof-of-concept models and be reliable, equitable, and scalable clinical applications.
References
1. Jiang F, Jiang Y, Zhi H, et al. Artificial intelligence in healthcare: past, present and future. Stroke and Vascular Neurology. 2017;2(4). doi: 10.1136/svn-2017-000101.
2. He J, Baxter SL, Xu J, Xu J, Zhou X, Zhang K. The practical implementation of artificial intelligence technologies in medicine. Nature Medicine. 2019;25(1):30–36. doi: 10.1038/s41591-018-0307-0.
3. Davenport T, Kalakota R. The potential for artificial intelligence in healthcare. Future Healthcare Journal. 2019;6(2):94. doi: 10.7861/futurehosp.6-2-94.
4. Bi WL, Hosny A, Schabath MB, et al. Artificial intelligence in cancer imaging: clinical challenges and applications. CA: A Cancer Journal for Clinicians. 2019;69(2):127–157. doi: 10.3322/caac.21552.
5. Johnson KW, Torres Soto J, Glicksberg BS, et al. Artificial intelligence in cardiology. Journal of the American College of Cardiology. 2018;71(23):2668–2679. doi: 10.1016/j.jacc.2018.03.521.
6. Rajalakshmi R, Subashini R, Anjana RM, Mohan V. Automated diabetic retinopathy detection in smartphone-based fundus photography using artificial intelligence. Eye. 2018;32(6):1138–1144. doi: 10.1038/s41433-018-0064-9.
7. Ye J, Yao L, Shen J, Janarthanam R, Luo Y. Predicting mortality in critically ill patients with diabetes using machine learning and clinical notes. BMC Medical Informatics and Decision Making. 2020;20(11):1–7. doi: 10.1186/s12911-020-01318-4.
8. Casey JA, Schwartz BS, Stewart WF, Adler NE. Using electronic health records for population health research: a review of methods and applications. Annual Review of Public Health. 2016;37. doi: 10.1146/annurev-publhealth-032315-021353.
9. Cohen DJ, Keller SR, Hayes GR, Dorr DA, Ash JS, Sittig DF. Integrating patient-generated health data into clinical care settings or clinical decision-making: lessons learned from Project HealthDesign. JMIR Human Factors. 2016;3(2):e5919. doi: 10.2196/humanfactors.5919.
10. Mandel JC, Kreda DA, Mandl KD, Kohane IS, Ramoni RB. SMART on FHIR: a standards-based, interoperable apps platform for electronic health records. Journal of the American Medical Informatics Association. 2016;23(5):899–908. doi: 10.1093/jamia/ocv189.
11. Braunstein ML. Healthcare in the age of interoperability: the promise of fast healthcare interoperability resources. IEEE Pulse. 2018;9(6):24–27. doi: 10.1109/MPUL.2018.2869317.
12. Plastiras P, O’Sullivan D. Exchanging personal health data with electronic health records: a standardized information model for patient generated health data and observations of daily living. International Journal of Medical Informatics. 2018;120:116–125. doi: 10.1016/j.ijmedinf.2018.10.006
