Hybrid DenseNet-U-Net framework for automated grading of renal cell carcinoma

1

Renal cell carcinoma (RCC) is the most common kidney cancer, accounting for nearly 90% of cases and significantly impacting cancer-related mortality.
2

The International Society of Urological Pathology grading system for RCC is subjective and time-consuming, leading to variability in diagnoses.
3

Recent advancements in artificial intelligence and digital pathology have created a demand for automated and reproducible RCC grading systems.
4

Hybrid models combining convolutional neural networks with segmentation techniques show promise in improving RCC grading accuracy and interpretability.
5

Optimized convolutional pipelines may achieve performance comparable to transformer-based methods while reducing computational complexity and memory usage.

1