Annual Report 2024

Division of Medical AI Research and Development

Ryuji Hamamoto, Syuzo Kaneko, Kazuma Kobayashi, Masaaki Komatsu, Ken Asada, Hidenori Machino, Satoshi Takahashi, Masayoshi Yamada, Kyoko Fujioka, Noriko Ikawa, Hiroko Kondo, Shigemi Yamada, Amina Bolatkan, Norio Shinkai, Nobuji Kouno, Naoki Teraya, Yusuke Sakaguchi, Ryota Shibaki, Takashi Natsume, Eri Suzuki, Natsumi Tsuboyama, Yuri Jonouchi

Introduction

 This division was established as a pioneering large-scale project in medical AI within Japan, following the decision to advance artificial intelligence (AI) research and development as a national policy based on the Fifth Science and Technology Basic Plan approved by the Cabinet in January 2016. It functions as the core for medical AI research and development at the Tsukiji Campus and actively promotes research and development as Japan's premier hub for medical AI research and development. The project particularly emphasizes two key points in its research activities: 1) Aiming for practical clinical application—advancing research for patients, not research for research's sake; and 2) Building an integrated database containing high-quality, abundant clinical information.

Research Activities

1.  Development of Novel Cancer Diagnostic Methods Utilizing AI Technology

 In collaboration with various departments and sections at the National Cancer Center Hospital (NCCH), we developed AI diagnostic support systems for endoscopy, radiology, ultrasound, and dermatology imaging. Through joint research with the Department of Endoscopy, we completed a colonoscopy AI that automatically predicts pathological diagnoses based on the revised Vienna Classification. This AI obtained regulatory approval as a Class III medical device in March 2025. In a joint study with the Department of Gastric Surgery, we constructed a model using Bayesian methods to predict splenic hilum lymph node (SLN) metastasis in upper gastric cancer, reporting the findings in npj Digital Medicine.

2.  Identification of New Therapeutic Targets in Lung Adenocarcinoma: Integrated Whole-Genome Sequencing Contributes to the Development of Personalized Medicine for Lung Adenocarcinoma

 We identified super-enhancers in 174 driver mutation/translocation-negative lung adenocarcinoma cases using H3K27Ac ChIP-seq and performed an integrated analysis with whole-genome data. Overlapping regions between super-enhancers and structural abnormalities were limited to approximately 1% of the total, but were observed in about 40% of cases. We confirmed increased expression associated with overlap in HER2 and EGFR, and suggested driver candidate potential for FRS2, CAV2, FGF3, FGF4, and FGF19. This finding advances the understanding of gene clusters involved in lung adenocarcinoma progression and treatment response, and was published in the international journal Molecular Cancer.

3.  Clinical Application of Ultrasound Diagnosis Support AI

 We jointly developed an AI system to assist in fetal echocardiography screening and obtained regulatory approval from the Ministry of Health, Labour and Welfare on July 29, 2024, as an AI-equipped medical device program. This system helps prevent the oversight of severe and complex congenital heart diseases, supports early diagnosis and treatment planning, and contributes to addressing disparities caused by physician shortages and regional differences.

Education

 A total of seven individuals—six graduate students affiliated with the University of Tokyo, Kyoto University, and Showa University, and one cancer training physician from the NCCH—were assigned to this division and received research guidance. Members of this division also actively participated in the Young Researchers Seminar hosted by the institute.

Future Prospects

1. Collaborate with departments and divisions at the NCCH and the NCCHE to continuously advance research and development of new AI for cancer diagnosis support, aiming for practical clinical application.

2. Advance the utilization of the AI development platform across multiple themes within the National Cancer Center, promoting development with an eye toward obtaining regulatory approval as a medical device.

3. In addition to expanding the Lung Cancer Integrated Database, work on constructing integrated databases for other cancer types and continue advancing the development of an omics analysis platform utilizing AI technology.

List of papers published in 2024

Journal

1. Hamamoto R, Komatsu M, Yamada M, Kobayashi K, Takahashi M, Miyake M, Jinnai S, Koyama T, Kouno N, Machino H, Takahashi S, Asada K, Ueda N, Kaneko S. Current status and future direction of cancer research using artificial intelligence for clinical application. Cancer science, 116:297-307, 2025

2. Tanimoto S, Sone K, Jonouchi Y, Hachijo R, Suzuki E, Tsuboyama N, Toyohara Y, Inoue F, Honjoh H, Fukuda T, Taguchi A, Miyamoto Y, Iriyama T, Mori M, Asada K, Komatsu M, Kaneko S, Hamamoto R, Wada-Hiraike O, Oda K, Hirota Y, Osuga Y. BET inhibitor JQ1 induces apoptosis of ovarian and endometrial endometrioid carcinoma cells by downregulating c‑Myc. Oncology letters, 29:106, 2025

3. Fujii E, Kato MK, Ono H, Yamaguchi M, Higuchi D, Koyama T, Komatsu M, Hamamoto R, Ishikawa M, Kato T, Kohno T, Shiraishi K, Yoshida H. TP53 Mutations and PD-L1 Amplification in Vulvar Adenocarcinoma of the Intestinal Type: Insights From Whole Exome Sequencing of 2 Cases. International journal of gynecological pathology, 44:358-363, 2025

4. Komatsu M, Teraya N, Natsume T, Harada N, Takeda K, Hamamoto R. Clinical Application of Artificial Intelligence in Ultrasound Imaging for Oncology. JMA journal, 8:18-25, 2025

5. Kitadai R, Yazaki S, Kuchiba A, Yamanaka T, Shiino S, Yamauchi C, Harano K, Saito M, Hirotsu Y, Aiba H, Yoshida T, Hamamoto R, Shimizu C, Shimomura A, Kojima Y, Shimoi T, Momozawa Y, Sudo K, Yoshida M, Sunami K, Hori M, Katanoda K, Shimada Y, Yamashita Y, Kogawa T, Murata T, Fujiwara S, Miyagi Y, Nakagomi H, Tachibana K, Omata M, Ohtake T, Suto A, Onishi T, Naito Y, Yamashita T, Yonemori K, Kohno T, Shiraishi K. Germline Pathogenic Variants and Clinical Outcomes in Asian Patients With Breast Cancer. Cancer science, 116:1048-1058, 2025

6. Ishizu K, Takahashi S, Kouno N, Takasawa K, Takeda K, Matsui K, Nishino M, Hayashi T, Yamagata Y, Matsui S, Yoshikawa T, Hamamoto R. Establishment of a machine learning model for predicting splenic hilar lymph node metastasis. NPJ digital medicine, 8:93, 2025

7. Takeuchi T, Horinouchi H, Takasawa K, Mukai M, Masuda K, Shinno Y, Okuma Y, Yoshida T, Goto Y, Yamamoto N, Ohe Y, Miyake M, Watanabe H, Kusumoto M, Aoki T, Nishimura K, Hamamoto R. A series of natural language processing for predicting tumor response evaluation and survival curve from electronic health records. BMC medical informatics and decision making, 25:85, 2025

8. Torasawa M, Yoshida T, Shiraishi K, Yagishita S, Ono H, Uehara Y, Miyakoshi J, Tateishi A, Igawa YS, Higashiyama RI, Mochizuki A, Masuda K, Matsumoto Y, Shinno Y, Okuma Y, Goto Y, Horinouchi H, Hamamoto R, Yamamoto N, Watanabe SI, Yatabe Y, Takahashi K, Kohno T, Ohe Y. Implications of EGFR expression on EGFR signaling dependency and adaptive immunity against EGFR-mutated lung adenocarcinoma. Lung cancer (Amsterdam, Netherlands), 202:108494, 2025

9. Kato MK, Fujii E, Asami Y, Momozawa Y, Hiranuma K, Komatsu M, Hamamoto R, Ebata T, Matsumoto K, Ishikawa M, Kohno T, Kato T, Yoshida H, Shiraishi K. Clinical features and impact of p53 status on sporadic mismatch repair deficiency and Lynch syndrome in uterine cancer. Cancer science, 115:1646-1655, 2024

10. Fujii E, Kato MK, Yamaguchi M, Higuchi D, Koyama T, Komatsu M, Hamamoto R, Ishikawa M, Kato T, Kohno T, Shiraishi K, Yoshida H. Genomic profiles of Japanese patients with vulvar squamous cell carcinoma. Scientific reports, 14:13058, 2024

11. Kaneko S, Takasawa K, Asada K, Shiraishi K, Ikawa N, Machino H, Shinkai N, Matsuda M, Masuda M, Adachi S, Takahashi S, Kobayashi K, Kouno N, Bolatkan A, Komatsu M, Yamada M, Miyake M, Watanabe H, Tateishi A, Mizuno T, Okubo Y, Mukai M, Yoshida T, Yoshida Y, Horinouchi H, Watanabe SI, Ohe Y, Yatabe Y, Saloura V, Kohno T, Hamamoto R. Mechanism of ERBB2 gene overexpression by the formation of super-enhancer with genomic structural abnormalities in lung adenocarcinoma without clinically actionable genetic alterations. Molecular cancer, 23:126, 2024

12. Kobayashi K, Takamizawa Y, Miyake M, Ito S, Gu L, Nakatsuka T, Akagi Y, Harada T, Kanemitsu Y, Hamamoto R. Can physician judgment enhance model trustworthiness? A case study on predicting pathological lymph nodes in rectal cancer. Artificial intelligence in medicine, 154:102929, 2024

13. Kato MK, Fujii E, Yamaguchi M, Higuchi D, Asami Y, Hiranuma K, Komatsu M, Hamamoto R, Matumoto K, Kato T, Kohno T, Ishikawa M, Shiraishi K, Yoshida H. Excellent concordance of the molecular classification between preoperative biopsy and final hysterectomy in endometrial carcinoma. Gynecologic oncology, 190:139-145, 2024

14. Asada K, Kaneko S, Takasawa K, Shiraishi K, Shinkai N, Shimada Y, Takahashi S, Machino H, Kobayashi K, Bolatkan A, Komatsu M, Yamada M, Miyake M, Watanabe H, Tateishi A, Mizuno T, Okubo Y, Mukai M, Yoshida T, Yoshida Y, Horinouchi H, Watanabe SI, Ohe Y, Yatabe Y, Kohno T, Hamamoto R. Multi-omics and clustering analyses reveal the mechanisms underlying unmet needs for patients with lung adenocarcinoma and identify potential therapeutic targets. Molecular cancer, 23:182, 2024

15. Takahashi S, Sakaguchi Y, Kouno N, Takasawa K, Ishizu K, Akagi Y, Aoyama R, Teraya N, Bolatkan A, Shinkai N, Machino H, Kobayashi K, Asada K, Komatsu M, Kaneko S, Sugiyama M, Hamamoto R. Comparison of Vision Transformers and Convolutional Neural Networks in Medical Image Analysis: A Systematic Review. Journal of medical systems, 48:84, 2024

16. Kouno N, Takahashi S, Komatsu M, Sakaguchi Y, Ishiguro N, Takeda K, Fujioka K, Matsuoka A, Fujimori M, Hamamoto R. Introduction of AI Technology for Objective Physical Function Assessment. Bioengineering (Basel, Switzerland), 11:1154, 2024

17. Kouno N, Takahashi S, Takasawa K, Komatsu M, Ishiguro N, Takeda K, Matsuoka A, Fujimori M, Yokoyama K, Yamamoto S, Honma Y, Kato K, Obama K, Hamamoto R. Analysis of Inertial Measurement Unit Data for an AI-Based Physical Function Assessment System Using In-Clinic-like Movements. Bioengineering (Basel, Switzerland), 11:1232, 2024

18. Aoyama R, Komatsu M, Harada N, Komatsu R, Sakai A, Takeda K, Teraya N, Asada K, Kaneko S, Iwamoto K, Matsuoka R, Sekizawa A, Hamamoto R. Automated Assessment of the Pulmonary Artery-to-Ascending Aorta Ratio in Fetal Cardiac Ultrasound Screening Using Artificial Intelligence. Bioengineering (Basel, Switzerland), 11:1256, 2024