Identification of risk of developing aortic dissection using artificial intelligence and establishment of a high-risk group sharing system

人工智能识别主动脉夹层发生风险并建立高危人群共享系统

• 批准号: 22K07803
• 负责人: 折居 誠
• 金额: $ 2.58万
• 依托单位: Iwate Medical University
• 依托单位国家: 日本
• 项目类别: Grant-in-Aid for Scientific Research (C)
• 财政年份: 2022
• 资助国家: 日本
• 起止时间: 2022-04-01 至 2025-03-31
• 项目状态: 未结题
• 来源: https://kaken.nii.ac.jp/grant/KAKENHI-PROJECT-22K07803/
• 关键词: 大動脈解離; 人工知能; CT

申請者および研究分担者は，人工知能（AI）による腹部大動脈瘤の同定および瘤径計測，過去画像との対比，診療ガイドラインを参照した治療方針の提示という画像診断補助システムの開発に成功した（Orii M, Fully Automatic Detection And Measurement Of Abdominal Aortic Aneurysm Using Artificial Intelligence, RSNA 2021）．大動脈解離の発症危険因子を同定し，発症高リスク群を抽出するためには，本システムを腹部大動脈から胸部および骨盤部の動脈にも応用する必要があると考えた．特に腸骨動脈では近接する腸管などとの判別が困難で開発に難航したが，最終的には高い診断精度（大動脈セグメンテーションにおけるDice score：全大動脈で0.90，胸部大動脈で0.94，腹部大動脈で0.93，腸骨動脈で0.83）で開発に成功した．AIが検出できた217例の大動脈瘤に対し，放射線診断レポートとの最大短径の誤差は0.83 ± 2.68mm，相関係数が0.974であった．本結果については，JCS2023-第87回日本循環器学会学術集会で成果報告を行った（Fujiwara J, Orii M, Yoshioka K. Diagnostic accuracy of an automatic detection and measurement system of aortic aneurysm using deep learning-based artificial intelligence）．

The applicant and the research contributor are interested in the successful development of the diagnosis system Of Abdominal Aortic Aneurysm Using Artificial Intelligence (RSNA 2021). The risk factors of arterial dissociation are determined by the method of extracting the high risk factors from the abdominal artery and the pelvic artery. In particular, it is difficult to distinguish between iliac artery and intestinal tract, and it is difficult to develop and navigate the final diagnosis.(Dice score: 0.90 for whole artery, 0.94 for thoracic artery, 0.93 for abdominal artery and 0.83 for iliac artery). The development was successful.AI detected 217 cases of arterial tumor. The error of maximum short diameter of radiographic diagnosis was 0.83 ± 2.68mm, and the correlation coefficient was 0.974. The results of this study were published in JCS2023- 87th Japan Circulator Society Academic Conference (Fujiwara J, Orii M, Yoshioka K. Diagnostic accuracy of an automatic detection and measurement system of aortic aneurysm using deep learning-based artificial intelligence）.

项目成果

Diagnostic accuracy of an automatic detection and measurement system of aortic aneurysm using deep learning-based artificial intelligence

基于深度学习的人工智能主动脉瘤自动检测和测量系统的诊断准确性

• 发表时间: 2023
• 作者: Nakao Y;Nakamura S;Htun Y;Mitsuie T;Koyano K;Ohta K;Konishi Y;Miki T;Ueno M;Kusaka T.;Kusaka Takashi;Kusaka Takashi;日下 隆;日下 隆;藤原純平，折居誠，吉岡邦浩
• 通讯作者: 藤原純平，折居誠，吉岡邦浩