This study proposes a hybrid deep learning model, specifically a combination of Convolutional Neural Networks (CNN) and EfficientNetV2B3, for the detection of invasive ductal carcinoma (IDC) in breast cancer whole slide images (WSIs). The model leverages pre-trained models to classify breast cancer images, aiming to support pathologists in making more accurate diagnoses. The proposed model demonstrates superior performance with an accuracy of 96.3%, precision of 93.4%, recall of 86.4%, F1-score of 89.7%, Matthew’s correlation coefficient (MCC) of 87.6%, Area Under the Curve (AUC) of the Receiver Operating Characteristic (ROC) curve of 97.5%, and Area Under the Curve of the Precision-Recall Curve (AUPRC) of 96.8%. The model outperforms other deep learning models such as MobileNet+DenseNet121, MobileNetV2+EfficientNetV2B0, and other contemporary machine learning approaches. The study also evaluates the effectiveness of data augmentation and early stopping techniques to enhance model robustness and generalization. The proposed model is validated using a dataset from Kaggle, which includes 162 breast histopathology samples with a highly imbalanced class ratio. The results demonstrate the model's ability to accurately detect IDC tissues, contributing to advancements in breast cancer screening and diagnosis.This study proposes a hybrid deep learning model, specifically a combination of Convolutional Neural Networks (CNN) and EfficientNetV2B3, for the detection of invasive ductal carcinoma (IDC) in breast cancer whole slide images (WSIs). The model leverages pre-trained models to classify breast cancer images, aiming to support pathologists in making more accurate diagnoses. The proposed model demonstrates superior performance with an accuracy of 96.3%, precision of 93.4%, recall of 86.4%, F1-score of 89.7%, Matthew’s correlation coefficient (MCC) of 87.6%, Area Under the Curve (AUC) of the Receiver Operating Characteristic (ROC) curve of 97.5%, and Area Under the Curve of the Precision-Recall Curve (AUPRC) of 96.8%. The model outperforms other deep learning models such as MobileNet+DenseNet121, MobileNetV2+EfficientNetV2B0, and other contemporary machine learning approaches. The study also evaluates the effectiveness of data augmentation and early stopping techniques to enhance model robustness and generalization. The proposed model is validated using a dataset from Kaggle, which includes 162 breast histopathology samples with a highly imbalanced class ratio. The results demonstrate the model's ability to accurately detect IDC tissues, contributing to advancements in breast cancer screening and diagnosis.