• R. Gurumoorthy
• Dr. M. Kamarasan

Breast cancer (BC) is a common worldwide health issue, with its earlier detection performing an essential function to improve outcomes for patients. The crucial requirement for advanced BC detection and classification approaches stems from the disease's occurrence and possibility for early involvement for significantly enhancing survival rates. Standard diagnostic methods, but valuable, face limitations for efficiency and accuracy. Accordingly, there is a developing imperative to tackle cutting-edge technologies, like artificial intelligence (AI) and machine learning (ML), to refine and increase the diagnostic process of BC. This research offers a new moth flame optimization employing breast cancer detection and classification (MFO-BCDC) algorithm. This research work begins with the application of DCNN, a robustness deep learning (DL) model for image analysis. Convolutional layers in the network extract related features from mammographic images, offering a robust foundation for subsequent classification. Moth Flame Optimization (MFO) has been presented as a meta-heuristic algorithm for hyperparameter tuning of the DCNN. MFO, stimulated by the natural behavior of moths, efficiently analysis the hyperparameter space for enriching the network's performance. To complement the DL method, the study integrates the XGBoost method for classification. XGBoost, known for its ensemble learning abilities can be utilized to effectively handle the extracted features and determine prediction accuracy. The developed combined architecture was assessed employing benchmark BC datasets, and wide-ranging experiments were executed to analysis the effectiveness of all modules. Outcomes represent that the incorporation of DCNN with feature extraction, MFO for hyperparameter tuning, and XGBoost for classification noticeably increases the performances with different features.

Breast cancer; Moth Flame Optimization; Deep Learning; Feature extraction; CAD; Histopathological Image

"Evolutionary Optimization with Deep Convolutional Neural Network based Breast Cancer Detection and Classification", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.10, Issue 12, page no.e145-e155, December-2023, Available :http://www.jetir.org/papers/JETIR2312420.pdf

"Evolutionary Optimization with Deep Convolutional Neural Network based Breast Cancer Detection and Classification", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.10, Issue 12, page no. ppe145-e155, December-2023, Available at : http://www.jetir.org/papers/JETIR2312420.pdf