Comparative Analysis of CLAHE and VOI LUT for Enhanced Mammogram Contrast in Tumor Detection

This study presents a comparative analysis of Contrast Limited Adaptive Histogram Equalization (CLAHE) and Value of Interest Look-Up Table (VOI LUT) as preprocessing methods for enhancing mammogram contrast to improve tumor detection. Using a dataset of 928 mammograms (464 BI-RADS 1 and 464 BI-RADS 4 and 5), we evaluated the performance of these methods in detecting tumors by extracting texture features using the Gray-Level Co-occurrence Matrix (GLCM) and classifying them with a Support Vector Machine (SVM). The results indicate that CLAHE slightly outperforms VOI LUT, with classification metrics showing precision, recall, F1-score, and accuracy of 0.57 for CLAHE compared to 0.55 for VOI LUT. These findings suggest that CLAHE may be more effective in enhancing mammogram textures, thereby improving the detection and classification of breast tumors. Future work should explore additional preprocessing techniques and classification algorithms to further validate and improve these results.