This study proposes a hybrid deep learning model (CNN+EfficientNetV2B3) for breast cancer detection using whole slide images (WSIs). The model combines convolutional neural networks (CNNs) with pre-trained models to classify breast cancer, supporting pathologists in making more accurate diagnoses. The proposed model achieved high performance metrics, including 96.3% accuracy, 93.4% precision, 86.4% recall, 89.7% F1-score, 87.6% Matthew's correlation coefficient (MCC), 97.5% Area Under the Curve (AUC) of the Receiver Operating Characteristic (ROC) curve, and 96.8% Area Under the Curve of the Precision-Recall Curve (AUPRC). These results outperformed other models in accuracy, precision, recall, F1-score, MCC, AUC, and AUPRC. The model was tested against other deep learning models, demonstrating superior performance. The study also evaluated the performance of three hybrid models: CNN+EfficientNetV2B3, MobileNet+DenseNet121, and MobileNetV2+EfficientNetV2_b0. The CNN+EfficientNetV2B3 model achieved the highest accuracy, precision, recall, F1-score, MCC, AUC, and AUPRC. The model's performance was assessed using various metrics, including accuracy, precision, recall, F1-score, MCC, AUC, and AUPRC. The study also included a confusion matrix and a comparison of the models' performance. The results showed that the CNN+EfficientNetV2B3 model outperformed other models in accuracy, precision, recall, F1-score, MCC, AUC, and AUPRC. The study concluded that the hybrid model provides a robust and efficient solution for breast cancer detection, with high accuracy and performance metrics. The model's performance was evaluated using various metrics, including accuracy, precision, recall, F1-score, MCC, AUC, and AUPRC. The study also discussed the advantages and limitations of the proposed model, including its effectiveness in classifying histopathology images and the need for pathologists to validate final diagnoses. Future work could involve incorporating additional image features and exploring methods to handle variations in image sizes. The study concludes that the hybrid model offers a novel and effective solution for breast cancer detection, with high accuracy and performance metrics.This study proposes a hybrid deep learning model (CNN+EfficientNetV2B3) for breast cancer detection using whole slide images (WSIs). The model combines convolutional neural networks (CNNs) with pre-trained models to classify breast cancer, supporting pathologists in making more accurate diagnoses. The proposed model achieved high performance metrics, including 96.3% accuracy, 93.4% precision, 86.4% recall, 89.7% F1-score, 87.6% Matthew's correlation coefficient (MCC), 97.5% Area Under the Curve (AUC) of the Receiver Operating Characteristic (ROC) curve, and 96.8% Area Under the Curve of the Precision-Recall Curve (AUPRC). These results outperformed other models in accuracy, precision, recall, F1-score, MCC, AUC, and AUPRC. The model was tested against other deep learning models, demonstrating superior performance. The study also evaluated the performance of three hybrid models: CNN+EfficientNetV2B3, MobileNet+DenseNet121, and MobileNetV2+EfficientNetV2_b0. The CNN+EfficientNetV2B3 model achieved the highest accuracy, precision, recall, F1-score, MCC, AUC, and AUPRC. The model's performance was assessed using various metrics, including accuracy, precision, recall, F1-score, MCC, AUC, and AUPRC. The study also included a confusion matrix and a comparison of the models' performance. The results showed that the CNN+EfficientNetV2B3 model outperformed other models in accuracy, precision, recall, F1-score, MCC, AUC, and AUPRC. The study concluded that the hybrid model provides a robust and efficient solution for breast cancer detection, with high accuracy and performance metrics. The model's performance was evaluated using various metrics, including accuracy, precision, recall, F1-score, MCC, AUC, and AUPRC. The study also discussed the advantages and limitations of the proposed model, including its effectiveness in classifying histopathology images and the need for pathologists to validate final diagnoses. Future work could involve incorporating additional image features and exploring methods to handle variations in image sizes. The study concludes that the hybrid model offers a novel and effective solution for breast cancer detection, with high accuracy and performance metrics.