新しい低価格分裂中期細胞検出装置プロジェクト（第7報）

Biological dosimetry is used to estimate individual absorbed radiation dose by quantifying an appropriate biological marker. The most popular gold-standard marker is the appearance of dicentric chromosomes in metaphase. The metaphase finder is a tool for automation of biological dosimetry that finds metaphase cells on glass slides. The author and a software company have designed a new system and are now preparing to produce the system commercially. This system was capable of identifying not only normal chromosomes but also PCC cells. The metaphase finder consists of an automated microscope, an auto-focus system, an X-Y stage, a camera, and a computer. To enhance the accuracy of the system, an artificial intelligence (AI) with deep learning was tested. The pre-selection of metaphases using mathematical morphology before the AI process was enabled the AI classification of true metaphases or not. A total of 1709 images of the metaphase finder detected as \u27metaphases\u27 were read into a nine-layer artificial neural network to detect true metaphases. A total of 456 images were used for training, and the rest of the images were used for validation. The accuracy of AI was 0.89 for metaphases and 0.90 for non-metaphases. This year, The computers for metaphase finding and for AI are connected by file sharing, then the metaphases found by the machine are immediately processed by the AI.日本放射線影響学会第64回大