Examining the Effect of Color Spaces on Histopathological Image Segmentation with the SHAP Explainable AI Method Renk Uzaylarının Histopatolojik Görüntü Bölütlemeye Etkisinin SHAP Açıklanabilir Yapay Zeka Yöntemiyle Incelenmesi

KARAASLAN Ö. F., BİLGİN G.

32nd IEEE Conference on Signal Processing and Communications Applications, SIU 2024, Mersin, Türkiye, 15 - 18 Mayıs 2024 identifier

  • Yayın Türü: Bildiri / Tam Metin Bildiri
  • Doi Numarası: 10.1109/siu61531.2024.10601125
  • Basıldığı Şehir: Mersin
  • Basıldığı Ülke: Türkiye
  • Anahtar Kelimeler: explainable arti-ficial intelligence (xAI) segmentation, Histopathological image analysis, interpretability, SHAP
  • Yıldız Teknik Üniversitesi Adresli: Evet

Özet

Histopathological images are widely used in the medical field for diagnosis and treatment processes. Despite their rich content in tissue structures, the machine learning algorithms currently employed generally have a black-box structure, limiting their efficient use in segmentation and creating difficulties in examining the obtained results. This study explores SHAP, an explainable artificial intelligence (xAI) method, used for the interpretability of histopathological segmentation processes. With the proposed approach, SHAP value for each color space (RGB, HSV, and L*a*b*) is calculated through the Support Vector Machines (SVM) model, elucidating the impact of these values on segmentation and providing insights into the model. The application of this method suggests its potential to enhance the interpretability of histopathological image segmentation processes using SHAP.