AY2023 Otsuka Pharmaceutical × Keio AIC Event Activity Report

From February 27 to March 12, 2024, Otsuka Pharmaceutical collaborated with Keio AIC to host a workshop titled “Extracting Information from Medical Dissertations on Blood Cancer Using Generative AI.”
Students from across a wide range of faculties attended the workshop, mainly from the Faculty of Science and Technology, but also including those from the Faculty of Economics, the School of Medicine, the Faculty of Pharmacy, and others. A total of 20 participants organized into 4 teams participated.

During the workshop

Pre-workshop training session

In preparation for this workshop, an optional introductory course to prompt engineering was held on February 13, 2024, as a pre-workshop training session.
AIC students served as instructors for the course, teaching the fundamentals of prompt engineering. The participants were deeply engaged in the lecture as this field is a hot topic lately.

Kick-off event

Hidenori Mikami of Otsuka Pharmaceutical gave an explanation on the theme and background of the subject matter. Although the field of cancer gene panel testing is an unfamiliar topic to non-medical students, he explained it in a simple and easy-to-understand manner, garnering the focused attention of the participants.
He also provided a sample exercise for participants to learn about prompt engineering, using a Wikipedia article as the subject.

Q&A session

An online Q&A session was held to provide an avenue for participants to clarify any doubts and resolve other issues they faced during the course of the workshop. Participants seemed to have worked on the workshop tasks enthusiastically, resulting in many of them asking questions.

Presentation session

A final presentation session was held based on the kick-off event and the Q&A session. Although faced with a tight schedule, with only two weeks from the formation of teams to their presentations, all teams performed admirably and delivered excellent presentations.
Otsuka Pharmaceutical’s Diagnostics Division and Planning Department evaluated the accuracy while Otsuka Holdings’ IT Planning Department evaluated the ideas that were presented. In addition, Otsuka Pharmaceutical’s Diagnostics Division and Planning Department, Vice President Masanori Udagawa, who is also Director of Otsuka Holdings’ IT Planning Department, and Makoto Sano, assistant to the director, gave their feedback on the presentations.

Results of the workshop:

• Accuracy Award ― Team F　Hidetoshi Takashima, Tomoya Kono, Ayano Komatsubara
• Idea Award ― Team A　Kohei Shinozaki, Shunta Sato, Hayato Nakamura, Shinji Yamamoto

Comments from participants:

• I’m glad that I was able to participate in this workshop on the subject of ChatGPT and medical dissertations, which is something that we don’t really get to touch on in the humanities faculties. It was really difficult to achieve the desired results through trial and error. Still, it was a valuable experience. I want to try applying what I have learnt on economics-related dissertations that I’m sure I will often have to deal with in the future.
• I really appreciated the strong support given, such as responding to my queries. It was very helpful. Thank you very much.
• I thought it was good that the social significance of the things we worked on during this program was highlighted.

Acknowledgments

We would like to express our sincere gratitude to Makoto Sano, Shinsuke Yokoyama, Hidenori Mikami, and many others at Otsuka Pharmaceutical for their tremendous support.
Keio AIC representative: Takahiro Yakoh, Project Professor: Mari Kobayashi
AIC student administrators: Koki Obayashi, Takeru Kaji, Shunpei Hatanaka