2025/10/23
Keio University
Kindai University
Fuso Pharmaceutical Industries, Ltd.
A research group including Mikitomo Kanazawa (second-year student in the Doctoral Programs at the Keio University Graduate School of Science and Technology), Professor Akira Funahashi (Department of Biosciences and Informatics, Faculty of Science and Technology, Keio University), Researcher Yudai Tokuoka (Keio Leading-edge Laboratory of Science and Technology), Sora Takeshita (second-year student in the Master's Program at the Graduate School of Biology-Oriented Science and Technology, Kindai University), Ryo Suenaga (second-year student in the same program), Professor Kazuo Yamagata (Department of Genetic Engineering, Faculty of Biology-Oriented Science and Technology, Kindai University), and Researchers Ryoma Yao and Tatsuki Hirai (Fuso Pharmaceutical Industries, Ltd.) has successfully developed an algorithm (FL2-Net) that accurately identifies cell nuclei from bright-field microscope images of mouse fertilized eggs (embryos). FL2-Net outperformed all four cell nucleus detection algorithms previously considered the best in the world in this field. Furthermore, when predicting the birth potential of mouse embryos based on features extracted by FL2-Net, it achieved accuracy significantly higher than existing methods and expert predictions. This method, which allows for the accurate evaluation of embryo quality, is expected to support expert decision-making in fertility treatments and contribute to the standardization of evaluations and reduction of workload. The results of this research were published in the online preliminary version of the academic journal "Computers in Biology and Medicine" on October 11 (UK time).
Please see below for the full press release.