EUVIMEDEUVIMED entdecken
Crossref

Explainable Artificial Intelligence with Deep Representation Learning Model for Polycystic Ovary Syndrome Detection Using Clinical Features

Veena Rani, Chandrasekar Venkatachalam, Karthick Raghunath K M.

Journal of Innovative Image Processing · 11.07.2026 · 8 (3) · 916-937

DOI: 10.36548/jiip.2026.3.009 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Journal of Innovative Image Processing, online erschienen am 11.07.2026. Verfasst von Veena Rani und 2 weiteren Autor:innen. Die Arbeit nennt 25 Literaturangaben.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Polycystic Ovary Syndrome (PCOS) is a prevalent endocrine condition affecting females and requires timely diagnosis for appropriate management and to mitigate long-term adverse problems. It is a complex condition identified by medical features such as hormonal imbalance, the existence of numerous ovarian follicles, irregular menstrual cycles, and anovulation. Over the past few years, Artificial Intelligence and deep learning approaches have demonstrated positive outcomes in medical applications by allowing effective analysis of intricate medical data. This study introduces an Explainable Artificial Intelligence-based Deep Representation Learning Model for Polycystic Ovary Syndrome Detection (XAIDRL-PCOSD). The proposed approach aims to support early and reliable detection of PCOS by leveraging real-world patient data. The model applies a hybrid feature selection strategy combining mutual information, recursive feature elimination, and LASSO to recognize the most significant clinical attributes. For classification, a feature tokenization transformer is employed to learn deep feature representations and capture complex relationships among clinical parameters. During training, the Ranger optimizer is utilized to enhance convergence and overall model performance. In addition, explainable artificial intelligence is incorporated using SHAP to provide clear insights into feature contributions, enhancing the interpretability of the model’s predictions for both PCOS and No PCOS cases. The experimental outcomes show that the proposed model accomplishes effective performance in PCOS detection with an accuracy of 95.71% while also offering transparency in decision-making. Therefore, the proposed XAIDRL-PCOSD model can serve as an effective tool in assisting medical professionals in understanding key factors influencing diagnosis.

Zitieren

APA

Rani, V., Venkatachalam, C., & M., K. R. K. (2026). Explainable Artificial Intelligence with Deep Representation Learning Model for Polycystic Ovary Syndrome Detection Using Clinical Features. Journal of Innovative Image Processing, 8(3), 916-937. https://doi.org/10.36548/jiip.2026.3.009

Vancouver

Rani V, Venkatachalam C, M. KRK. Explainable Artificial Intelligence with Deep Representation Learning Model for Polycystic Ovary Syndrome Detection Using Clinical Features. Journal of Innovative Image Processing. 2026;8(3):916-937. doi:10.36548/jiip.2026.3.009

BibTeX
@article{Rani2026,
  author = {Rani, Veena and Venkatachalam, Chandrasekar and M., Karthick Raghunath K},
  title = {Explainable Artificial Intelligence with Deep Representation Learning Model for Polycystic Ovary Syndrome Detection Using Clinical Features},
  journal = {Journal of Innovative Image Processing},
  year = {2026},
  volume = {8},
  number = {3},
  pages = {916--937},
  doi = {10.36548/jiip.2026.3.009}
}
Kostenlos starten →