Deploying Hybrid VGG19-BiGRU Model for Kidney Disease Segmentation

The kidney is an essential organ in the human body, and kidney disease is a widespread and sometimes fatal problem. Timely detection of Kidney illness is vital for preventing rapid mortality. Automation in identifying Kidney illness improves its efficacy in facilitating prompt intervention. To tackle this significant issue, it is necessary to implement effective and automated diagnostic solutions. This study focuses on automating the segmentation of Kidney disease. The dataset consists of 12,446 photos categorized into four classes: cyst, normal, stone, and malignancy. We can identify these four diseases using a deep learning technique. We utilize a hybrid VGG19-BiGRu model to segment kidney disease. This model achieves an impressive training accuracy of 99.77% and a validation accuracy of 99.98%. In addition, we incorporate precision, recall, and F1 scores to assess the performance of our model. These results demonstrate that our segmentation criteria are highly effective and positively impact the healthcare system.