Algorithm 4. Ellipse-based RBC Morphology classification and clustering

RBC_Ellipse_Fitting_Clustering(data_list, image_path)   for each folder in data_list do     define folder_path     if folder_path exists then       for each image_file in folder_path do         load image          → image_input         convert image to grayscale   → gray         apply Otsu thresholding    → binary         find contours from binary   → contours         for each contour cnt in contours do           if contour length ≥ 5 then             fit ellipse to contour  → ellipse             extract ellipse parameters: center, major_ax, minor_ax, angle             compute major/minor axis lines and endpoints             convert axis lengths to micrometers (µm)             determine aspect ratio (AR)             generate contour and ellipse masks             compute overlap region (intersection) → inter_contours             calculate area ratio (ER)             annotate image with ellipse, axes, ratio, and area metrics             classify cell morphology using classify_cell() function             define output directories based on classification             save annotated and raw images into their respective folders             log extracted metrics for statistical analysis             append classification results to CSV for later clustering review       end for     else: print warning (folder not found)   end for export full metrics dataset and classification summary   return ellipse_classification_results