Deep learning approach for accurate prostate cancer identification and stratification using combined immunostaining of cytokeratin, p63, and racemase
Abstract: Highlights•Prostate cancer diagnosis is improved using immunostaining and deep learning.•A segmentation network is proposed to enhance functional information of cancer.•Achieved 1.64 % mean absolute error in specific-class gland quantification.•Valuable support for pathologists to reduce workload and inter-observer variability.•Potential for broad applicability in other areas beyond prostate cancer.
Loading