Integration of Explainable Artificial Intelligence (XAI)

The gap

Evidence profile

Sourced from the future work and future-work section of the source papers, classified as general, spanning 3 journals.

Supporting evidence — 3 representative gaps

• Parkinson's Disease Prediction Using ML (2026) · Zenodo (CERN European Organization for Nuclear Research) · doi

  and for Lack of Model Interpretability: A limitation of the proposed system is the limited interpretability of the trained model. Future work can address this by incorporating Explainable AI (XAI) techniques, such as feature attribution methods, to provide insight into how input features influence model predictions. through Operating on Simulated Data: The current implementation relies on simulated patient input provided files. Future development may involve integrating real-time data sources, such as wearable sensors or Internet of Things (IoT) devices, to better capture real-world variability. static CSV VI FUTURE ENHANCEMENTS Further improvements are required to extend the proposed system beyond its current implementation and evaluate its suitability for real-world clinical use. Expanding the Dataset: Future work will involve using larger and more diverse datasets to improve model robustness and generalizability. Improving Model Interpretability: Future work may incorporate Explainable AI (XAI) techniques to improve understanding of how input features influence model predictions. Real-World Evaluation: Future work may involve evaluating the system using real-world clinical data and deployment scenarios, including data obtained from wearable or IoT-based devices. symptoms.

  generalfuture workevidence 5/5
  Keywords: model future real world interpretability system input involve proposed explainable techniques features influence predictions simulated
• Machine Learning Approach for Calculating Healthcare Protection Premiums (2026) · International Journal for Research in Applied Science and Engineering Technology · doi

  Integration of Explainable Artificial Intelligence (XAI) techniques to improve transparency and interpretability of predictions. Deployment of the prediction model as a cloud-based web application for real-time premium estimation. Incorporation of Electronic Health Records (EHRs) to improve prediction accuracy.

  generalfuture-work sectionevidence 5/5
  Keywords: integration explainable artificial intelligence xai techniques improve transparency
• Recent Advances in AI for Automated ICD Coding: A Systematic Literature Review (2026) · Journal of Medical Systems · doi

  To address the challenges in automated ICD coding, future research should focus on using diverse datasets and preprocessing techniques. The development of more advanced machine learning and deep learning approaches is necessary to improve model performance. Future studies should prioritize the use of external validation and model interpretability to improve clinical trust.

  generalfuture-work sectionevidence 5/5
  Keywords: address challenges automated icd coding future research focus

Questions about this gap