Fine Tuning Deep Learning Models For Breast Tumor Classification

This article discusses the potential use of MGTO for hyperparameter optimization tasks and proposes a diagnostic approach using computer vision models to differentiate between benign and malignant BTs. The approach involves preprocessing, data augmentation, feature extraction, and classification tasks using both custom and pretrained models. The BreakHis dataset is used for training and testing, and hyperparameter tuning is done using GWO and MGTO optimizers. The performance is evaluated using different metrics.