An Image Analysis using Statistical Features of Brain Tumor and Breast Tumor Digital Images

Abstract: The analysis of medical images, specifically those related to brain and breast tumors, plays a vital role in early diagnosis and treatment planning. This paper presents a comprehensive study of tumor detection using statistical features extracted from digital images. The focus is on brain and breast tumor images, where statistical methods such as mean, standard deviation, entropy, skewness, and kurtosis are utilized to quantify image characteristics. These statistical features are essential in distinguishing between malignant and benign tumors based on texture patterns within the images. Image processing techniques, including preprocessing, segmentation, and feature extraction, are applied to tumor images to enhance the accuracy of classification models. The results demonstrate that statistical features, combined with machine learning algorithms, can significantly improve tumor classification accuracy compared to traditional methods. A comparative analysis is conducted between the statistical features for both brain and breast tumors, offering insights into their distinctive characteristics and the potential for cross-disease analysis. This research highlights the significance of statistical image features in tumor analysis and suggests further advancements in automated diagnostic systems for clinical applications. Keywords - Statistical Feature Extraction, Medical Image Analysis, Tumor Classification, Gray Level Co- occurrence Matrix (GLCM), Support Vector Machine (SVM), Brain and Breast Tumor Images