EUVIMEDEUVIMED entdecken
Crossref

Explainable AI for Anaemia Prediction: Enhancing Clinical Transparency through SHAP and LIME Interpretability

Mounika K, M Himasree, U Dinesh, T Dinesh, M Shivamani

International Journal of Leading Research Publication · 20.03.2026 · 7 (3)

DOI: 10.70528/ijlrp.v7.i3.2008 ↗

Kurz gefasst

Fachartikel in der Zeitschrift International Journal of Leading Research Publication, online erschienen am 20.03.2026. Verfasst von Mounika K und 4 weiteren Autor:innen.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Anaemia is a condition characterized by a deficiency of healthy red blood cells or haemoglobin, resulting in a reduced capacity of the blood to carry oxygen to the body's tissues. It is a global health concern requiring accurate, interpretable diagnostic tools to improve patient outcomes. Existing prediction systems predominantly use statistical techniques and black-box AI models, which suffer from limited transparency and practical applicability, restricting their adoption in clinical settings. These models fail to provide the actionable insights necessary for informed clinical decisions, often relying on small, non-generalizable datasets and lacking real-world integration. Addressing these limitations, the proposed system advances the field by deploying a transparent and explainable AI (XAI) methodology for anaemia prediction, utilizing SHapley Additive exPlanations (SHAP) and Local Interpretable Model-agnostic Explanations (LIME) to offer clear, interpretable feature contributions behind each diagnosis. This approach facilitates understanding for clinicians, enhances model credibility, and bridges the gap between predictive accuracy and clinical interpretability. Future directions include extending the application of the model to wider populations, integrating continuous health data from wearable devices, and seamless adoption within Electronic Health Records (EHR) systems. These enhancements promise superior reliability, data-driven insights, and improved patient trust, ultimately enabling timely interventions and more effective anaemia management across varied healthcare environments.

Zitieren

APA

K, M., Himasree, M., Dinesh, U., Dinesh, T., & Shivamani, M. (2026). Explainable AI for Anaemia Prediction: Enhancing Clinical Transparency through SHAP and LIME Interpretability. International Journal of Leading Research Publication, 7(3). https://doi.org/10.70528/ijlrp.v7.i3.2008

Vancouver

K M, Himasree M, Dinesh U, Dinesh T, Shivamani M. Explainable AI for Anaemia Prediction: Enhancing Clinical Transparency through SHAP and LIME Interpretability. International Journal of Leading Research Publication. 2026;7(3). doi:10.70528/ijlrp.v7.i3.2008

BibTeX
@article{K2026,
  author = {K, Mounika and Himasree, M and Dinesh, U and Dinesh, T and Shivamani, M},
  title = {Explainable AI for Anaemia Prediction: Enhancing Clinical Transparency through SHAP and LIME Interpretability},
  journal = {International Journal of Leading Research Publication},
  year = {2026},
  volume = {7},
  number = {3},
  doi = {10.70528/ijlrp.v7.i3.2008}
}
Kostenlos starten →