Published Paper Details:

A global challenge that has become a menace is the example of cancers such as lung and colon cancer (LC AND CC) Colon cancer(CC). The path of this research, therefore, in this very important area, is early detection through the use of AI in histological image analysis. The study introduces a novel hybrid feature set aiming to improve classification accuracy by integrating DenseNet201 with color histogram techniques. The validation of this feature set works with eight major ML algorithms-KNN, SVM, Light GBM, CatBoost, XGBoost, decision trees, random forests, and multinomial naive Bayes. This comprehensive study, therefore, highlights an exotic model that achieved an accuracy of 99.683% on the LC AND CC25000 dataset. Taking the same concept to breast cancer detection using the Break His dataset shows a great accuracy of 94.808%. These results highlight the revolutionary potential of AI in the challenging area of histopathological analysis and therefore stand to become a transformative player in accelerating diagnostic accuracy. A thorough comparative analysis showcases the strengths and weaknesses of current AI practices in medical imaging, outlining a pathway for improvement and future clinical application.