Department of Medical AI Research

Overview

The use of artificial intelligence in medicine continues to expand rapidly, bringing significant benefits not only to patients but also to healthcare providers. Working in close collaboration with various clinical departments across the hospital, our department leverages deep learning and large language models alongside medical big data to contribute to both life sciences research and the development of next-generation medical care.

Research

Our research spans a wide range of topics, from basic life science to clinical medicine, with deep learning and large language models as our core technologies. We are also actively embracing AI-driven development methodologies. Projects include genomic and epigenomic analyses of gene expression regulation, AI-assisted pathological diagnosis, identification of causative pathogens in infectious diseases, screening for inborn errors of metabolism and pediatric ophthalmic conditions, quality assessment of regenerative medicine products from glycan data, immune response prediction from genomic variant data, and the development of life-stage-spanning medical AI using data from across national centers. Beyond AI, we also contribute to digital transformation, DX, at the institute and hospital through systems such as a pre-consultation interview system and an animal experiment management system. We are equipped to support the entire AI development pipeline, from data collection and annotation to GPU computing, machine learning frameworks, web services, containerization, and serverless cloud deployment.

Education

Our educational activities in data science, including programming, extend beyond the institute and hospital to collaborating institutions via online platforms. These include technical reading groups and departmental colloquia. Educational content is regularly recorded and made publicly available through the Introduction to Clinical Research website.

About MIRAI

MIRAI (Medical Intelligence Research for Advanced Innovation), which means "future" in Japanese, was the former name of this laboratory. Its mission was to save the lives of children across the globe through AI in clinical medicine, public health, and molecular biology, with a particular focus on solutions applicable in low- and middle-income countries.

Members

Kohji Okamura, Ph.D.
Tsunehiko Takamura, Ph.D.
Michihiro Yamamura, M.D.
Noriko Kawasaki
Miki Nobe
Natsuhiko Yamada, M.D.
Hiroaki Fukuse, M.D.
Hiroshi Yamanaka, M.D.
Iku Nakajima

Publications

Machine learning approach to analyzing complex care coordination patterns for medically complex children: Aoto S, Ito Y, Morooka S, Ito A, Umezawa A, Okamura K, Nakamura T
BMC Med. Inform. Decis. Mak. [doi: 10.1186/s12911-026-03691-y] (2026)
Automated detection of mulberry bodies in urinary sediment for non-invasive Fabry disease screening: Yamanaka H, So T, Sakamoto N, Aoto S, Li XK, Wang Y, Shen Q, Migita O, Kosuga M, Okamura K
Clin. Chem. Lab. Med. 64, e227–e231 (2026)
Detection of pediatric cataracts through non-invasive facial photography using deep convolutional neural networks for early diagnosis: Hayashi H, Kashizuka E, Sakata K, Motomiya N, Asakawa Y, Hayasaki M, Yokoi T, Yoshida T, Nishina S, Okamura K
BMC Ophthalmol. 26, 282 (2026)
Biomimetic microfibers for myelin-enhancer screening and neural regeneration: Quan L, Uyeda A, Sekiguchi A, Zhang Z, Sakai K, Takamura T, Zhang R, Ichinohe N, Umezu S Nuramatsu R
Cyborg Bionic Syst. 7, 0565 (2026)
CYCS-related thrombocytopenia in three Japanese families with a novel variant in one family: Yamada N, Sakamoto A, Nagoshi R, Endo S, Yamamoto M, Saito S, Yamada Y, Uchiyama T, Yanagi K, Kaname T, Kunishima S, Ishiguro A
Int. J. Hematol. 123, 611–616 (2026)
Exploring deep learning and data requirements through image classification of Erigeron annuus and Erigeron philadelphicus: Yamanaka H, Okamura K
BMC Res. Notes 19, 127 (2026)
Chromosomal and hormonal factors involved in human sexual dimorphism: Fukami M, Okamura K, Sasaki S, Kagami M, Dateki S
Endocr. J. 73, 175–181 (2026)
Privacy-preserving retrieval-augmented generation on local devices for regenerative medicine applications: Takamura T, Umezawa A
medRχiv 2025.10.20.25337146 (2025)
Removing extracellular matrix from the cell surface prior to seeding enhances the adhesion of primary human hepatocytes to the culture vessel: Miyai M, Tanaka-Yachi R, Aizawa K, Okamura K, Kusuhara H, Akutsu H, Nakamura K
Genes Cells 30, e70059 (2025)
Applicability of the regression approach for histological multi-class grading in clear cell renal cell carcinoma: Shibata M, Umezawa A, Aoto S, Okamura K, Nasu M, Mizuno R, Oya M, Yura K, Mikami S
Regen. Ther. 28, 431–437 (2025)
Supervised machine learning of outbred mouse genotypes to predict hepatic immunological tolerance of individuals: Morita-Nakagawa M, Okamura K, Nakabayashi K, Inanaga Y, Shimizu S, Guo WZ, Fujino M, Li XK
Sci. Rep. 14, 24399 (2024)
Proof of mechanism investigation of Transcutaneous auricular vagus nerve stimulation through simultaneous measurement of autonomic functions: a randomized controlled trial protocol: Katsunuma R, Takamura T, Yamada M, Sekiguchi A
Biopsychosoc. Med. 18, 15 (2024)
Machine learning trial to detect sex differences in simple sticker arts of 1606 preschool children: Matsubara K, Ohgami Y, Okamura K, Aoto S, Fukami M, Shimada Y
Minerva Pediatr. 76, 343–349 (2024)
Systematic reduction of gray matter volume in anorexia nervosa, but relative enlargement with clinical symptoms in the prefrontal and posterior insular cortices: a multicenter neuroimaging study: Tose K, Takamura T, Isobe M, Hirano Y, Sato Y, Kodama N, Yoshihara K, Maikusa N, Moriguchi Y, Noda T, Mishima R, Kawabata M, Noma S, Takakura S, Gondo M, Kakeda S, Takahashi M, Ide S, Adachi H, Hamatani S, Kamashita R, Sudo Y, Matsumoto K, Nakazato M, Numata N, Hamamoto Y, Shoji T, Muratsubaki T, Sugiura M, Murai T, Fukudo S, Sekiguchi A
Mol. Psychiatry 29, 891–901 (2024)
Differentiation of large extracellular vesicles in oral fluid: combined protocol of small force centrifugation and sedimentation pattern analysis: Kawano T, Okamura K, Shinchi H, Ueda K, Nomura T, Shiba K
J. Extracell. Biol. 3, e1143 (2024)
Integrator complex subunit 15 controls mRNA splicing and is critical for eye development: Azuma N, Yokoi T, Tanaka T, Matsuzaka E, Saida Y, Nishina S, Terao M, Takada S, Fukami M, Okamura K, Maehara K, Yamasaki T, Hirayama J, Nishina H, Handa H, Yamaguchi Y
Hum. Mol. Genet. 32, 2032–2045 (2023)
Automated urinary sediment detection for Fabry disease using deep-learning algorithms: Uryu H, Migita O, Ozawa M, Kamijo C, Aoto S, Okamura K, Hasegawa F, Okuyama T, Kosuga M, Hata K
Mol. Genet. Metab. Rep. 33, 100921 (2022)
Collection of 2429 constrained headshots of 277 volunteers for deep learning: Aoto S, Hangai M, Ueno-Yokohata H, Ueda A, Igarashi M, Ito Y, Tsukamoto M, Jinno T, Sakamoto M, Okazaki Y, Hasegawa F, Ogata-Kawata H, Namura S, Kojima K, Kikuya M, Matsubara K, Taniguchi K, Okamura K
Sci. Rep. 12, 3730 (2022)