Development of lung cytodiagnosis support system using homology profile method

2023 Fiscal Year Final Research Report

• Project/Area Number: 21H03839
• Research Category: Grant-in-Aid for Scientific Research (B)
• Allocation Type: Single-year Grants
• Section: 一般
• Review Section: Basic Section 90130:Medical systems-related
• Research Institution: Osaka University
• Principal Investigator: Nakane Kazuaki 大阪大学, 大学院医学系研究科, 招へい教授 (10298804)
• Co-Investigator(Kenkyū-buntansha): 山本 浩文 大阪大学, 大学院医学系研究科, 教授 (30322184) ; 横山 雄起 大阪大学, 大学院医学系研究科, 助教 (60615714) ; 橘 理恵 大島商船高等専門学校, 情報工学科, 准教授 (90435462)
• Project Period (FY): 2021-04-01 – 2024-03-31
• Keywords: ホモロジー / クロマチンパターン / 細胞診断 / 遠隔診断

Outline of Final Research Achievements

Homology is a mathematical concept that quantifies the contact of shapes. Recently, it has been recognized as an effective approach for analyzing complex images, such as pathological images. In this study, we further developed this concept through a technique called the Homology Profile Method. Using this method, we quantified the chromatin patterns of lung cell nuclei and attempted to classify lung cancer tissue types based on these measurements. During this period, we were able to distinguish between different types of cancer--small cell carcinoma, adenocarcinoma, and squamous cell carcinoma-- with favorable results. This approach is expected to bring objectivity to the selection of therapeutic drugs and contribute significantly to the standardization of cancer treatment. Notably, it has been demonstrated that analysis is feasible even with 40x magnification images, thereby eliminating the need for oil immersion and enabling potential clinical applications.

Free Research Field

医学

Academic Significance and Societal Importance of the Research Achievements

肺癌の細胞診には、気管支鏡下の擦過・気管支肺胞洗浄・CTガイド下穿刺吸引・EBUS-TBNAならびに喀痰などの細胞採取法があるが、外科切除不能な進行期肺癌では、多くの場合細胞診が確定診断になることが多い。このため小細胞癌・非小細胞癌、腺癌・扁平上皮癌の細胞学的鑑別を精度良く行うことが重要である。
細胞診断の標準化についてはサンプルの多様性から、標準的クライテリアの設定は難しい。しかし、の核クロマチンパターンが診断において大きなウエイトを占めていることは間違いなく、もしこれを客観的な数理的指標を用いて表現できれば、自動診断支援技術開発の手がかりだけでなく、病理現象への理解が進むと考えられる。