Optimal settings and clinical validation for automated Ki67 calculation in neuroendocrine tumors with open source informatics (QuPath)

Early Breast Cancer Risk Assessment: Integrating Histopathology with Artificial Intelligence

“…The utilization of DIA tools has the potential to facilitate a faster and more standardized evaluation of Ki67 [ 86 , 87 ]. Several freeware image analysis instruments for nuclear staining algorithms implemented in WSI analysis are available for free, with no additional equipment needed, and show excellent agreement with the manual count [ 71 , 83 , 86 , 89 ]. The improved agreement among pathologists in assessing Ki67 in breast tumors has also been reported when employing digital AI tools, as opposed to the subjective naked-eye assessment, especially in heterogeneous marker expression patterns [ 83 , 87 , 90 ].…”

“…Nevertheless, the absence of data from recent randomized trials using modern imaging suggests that surveillance protocols should consider patient needs, costs, and healthcare system burdens [ 125 ]. Additionally, recent advancements in DIA are found to be efficient and time-saving for pathological diagnosis, demonstrating good agreement, particularly in cases of intratumor heterogeneity [ 71 , 86 , 89 ]. The main issues in the integration of AI into the diagnostic pathology process are related to the user dependence of the algorithms with a need for external control [ 25 , 64 ].…”