Wildebeest Habit Optimizer with Deep Learning-Based Histopathological Image Analysis for Breast Cancer Diagnosis
R. Gurumoorthy, Mari Kamarasan · International Journal of Engineering Trends and Technology · 2023
Automatic categorization of Breast Cancer (BC) Histopathological Image (HPI) is a major research study in the biomedical informatics domain because of the great medical eminence of classification in offering prognosis and diagnosis of BC. Computer-Aided Diagnosis (CAD) of BC histopathologic images is a significant tool to improve the efficiency and precision of BC classification and diagnosis. Machine Learning (ML) models are trained on large sets of HPI images to detect features and patterns linked with different stages and types of BC. This could aid pathologists in making more precise detections and developing potential treatment plans. Therefore, this study develops a Wildebeest Habit Optimizer with Deep Learning based Histopathological Image Analysis for Breast Cancer Diagnosis (WHODL-HIABCD) technique. The presented WHODL-HIABCD technique exploits Bilateral Filtering (BF) based noise removal process to get rid of the noise. Followed by the WHODL-HIABCD technique uses an EfficientNet model with a WHO-based hyperparameter optimizer for feature extraction purposes. Finally, the Attention-based Bidirectional Gated Recurrent Unit (ABiGRU) method was utilized for classifying and recognizing BC. The investigational results of the WHODL-HIABCD technique are experimented with the benchmark dataset. The comparative study stated the enhanced performance of the WHODL-HIABCD technique than recent models.