AUTHOR=Bakoglu Nilay , Cesmecioglu Emine , Sakamoto Hirotsugu , Yoshida Masao , Ohnishi Takashi , Lee Seung-Yi , Smith Lindsey , Yagi Yukako TITLE=Artificial intelligence-based automated determination in breast and colon cancer and distinction between atypical and typical mitosis using a cloud-based platform JOURNAL=Pathology and Oncology Research VOLUME=Volume 30 - 2024 YEAR=2024 URL=https://www.por-journal.com/journals/pathology-and-oncology-research/articles/10.3389/pore.2024.1611815 DOI=10.3389/pore.2024.1611815 ISSN=1532-2807 ABSTRACT=Artificial intelligence (AI) technology in pathology has been utilized in many areas and requires supervised machine learning. Notably, annotations that define the ground truth for the identification of different confusing process pathologies, vary based on the study. In the following studies, we present our findings in invasive breast cancer detection for the IHC/ISH assessment system, as well as automated analysis of each tissue layer, cancer type, and so on in colorectal specimens. Additionally, models for the detection of atypical and typical mitosis in several organs were developed using existing whole-slide image (WSI) sets for other AI projects. All H&E slides were scanned by different scanners with 0.12-0.50μm/pixel resolution, and then uploaded to a cloud-based AI platform. Training series of convolutional neural networks (CNN) consist of invasive carcinoma, atypical and typical mitosis, colon tissue elements (mucosa-epithelium, lamina propria, muscularis mucosa-, submucosa, muscularis propria, subserosa, vessels, and lymph nodes). 59 WSIs of 59 breast cases, 217 WSIs of 54 colon cases, and 28 WSIs of 23 different types of tumor cases with relatively higher amounts of mitosis were annotated for the training. The harmonic average of precision and sensitivity was scored as F1 by AI. The final AI models of the Breast Project showed F1 score of 94.49% for Invasive carcinoma. Mitosis project showed F1 score of 80.18%, 97.40%, and 97.68% for mitosis, atypical and typical mitosis layers, respectively. Overall F1 scores for the current results of the colon project were 90.02% for invasive carcinoma, 94.81% for the submucosa layer, and 98.02% for vessels and lymph nodes. After the training & optimization of AI models and validation of each model, external validators evaluated the results of the AI models via blind-reader tasks. The AI models developed in this study were able to identify tumor foci, distinguish in situ areas, define colon layers, detect vessels and lymph nodes, and catch the difference between atypical and typical mitosis. All results were exported for integration to our in-house applications for breast cancer and AI model development for both whole-block and whole-slide image-based 3D imaging assessment