Harnessing Deep Learning Techniques for Identification of Breast Tumors

Main Article Content

Hariharaganesh M, Saurabh Premlal Themburane

Abstract

Breast tumor is the main disease in women around the world, 25% of all cancer cases. Traditional detection methods such as mammography, ultrasound, and MRI, have interference with time-consuming to determination and potential mistakes. To address the challenges, analysts introduce intelligent learning model methods as a dependable arrangement for breast tumor location. These strategies include the utilization of artificial neural network systems that can able to learn the data patterns and recognize them in datasets. One such dataset is the MINI-DDSM (Mini-Mammography) Dataset for Breast Cancer Screening, which contains digitized mammography pictures of kinds and dangerous tumors, utilizing the learning calculations to recognize breast cancer from mammography pictures. The information was preprocessed, and the models were assessed utilizing different convolutional neural network structures. These methods may lead radiologists to analyze breast cancer rapidly and effectively to drive earlier to identify and progress the treatment. Intelligent learning model methods have a good preference, counting learning and adjusting to correct spaces, and identifying covered-up patterns and connections not mistakable to the human eye. With information on the properties of the tumor, such as estimate, shape, and surface, a radiologist can make a more educated choice approximately the best treatment arranged for a persistent. Here the mini datasets were preprocessed, data augmented, optimized, and passed on to the CNN models and some pre-trained models such as xception, and ResNet50. Profound learning methods have illustrated better results in the recognizable proof of breast tumors, moving forward in breast cancer determination and treatment. The integration of personalized pharmaceuticals and the utilization of standardized datasets such as the Mini-DDSM altogether improve the general adequacy of breast cancer treatment plans.

Article Details

Section
Articles