Meeting Banner
Abstract #2690

In Vivo Prediction of Spermatogenesis in Seminiferous Tubules Using High-Resolution Magnetic Resonance Imaging and Machine-Learning Techniques in Combination

Masayuki Yamaguchi1, Natsumaro Kutsuna2,3, Ryutaro Nakagami1,4, Akira Nabetani5, Atsushi Nozaki5, Mamoru Niitsu4, Seiichiro Hasezawa2,3, Hirofumi Fujii1,3

1Functional Imaging Division, National Cancer Center Hospital East, Kashiwa, Chiba, Japan; 2Graduate School of Frontier Sciences, University of Tokyo, Kashiwa, Chiba, Japan; 3Institute for Bioinformatics Research and Development-Japan Science and Technology Agency, Chiyoda, Tokyo, Japan; 4Graduate School of Human Health Sciences, Tokyo Metropolitan University, Arakawa, Tokyo, Japan; 5GE Healthcare Japan, Hino, Tokyo, Japan

Seminiferous tubules are stratified epithelia composed of germ cells and Sertoli cells. They produce sperm and normally are 200E00μm in diameter. We have succeeded in visualizing rat seminiferous tubules on in vivo MRI using a 3T scanner. In addition, the machine-learning technique allowed automatic classification of testicular regions on MRI into normal and abnormal spermatogenesis in chemotherapy-induced injury in rat testes. If these techniques are implemented in clinics in the future, they will be a helpful tool in reproductive medicine for infertile males.