Improving Breast Cancer Detection with Deep Learning Techniques: A Study Using CBIS-DDMS Dataset

Breast cancer (BC) is a disease with a significant global impact, and early identification is essential to better patient outcomes. The objective of this study is to create a sophisticated Deep Learning (DL) model using cloud computing for BC diagnosis. A large collection of medical images called the CBIS-DDMS dataset is used to evaluate the proposed approach. Data collection, pre-processing, and diversity augmentation of the CBIS-DDMS dataset are considered for this study approach. Using some modifications and advancements made to conventional deep learning approaches, improved model architecture increases accuracy and efficiency. The model goes through a thorough training process that includes hyperparameter adjustment and optimization. Researchers assess the sensitivity, specificity, and accuracy of various techniques. Furthermore, this study looks into how the performance of the suggested model in a cloud environment can be enhanced by changing a few nodes in the hidden layer. Based on experimental data with an accuracy of 99.48%, it can be said that the suggested methods improved compared to other competing methods. Furthermore, in a cloud computing environment, the proposed technique performs better, with an accuracy level of 98.60%.