大腸癌の遺伝子サブタイプを予測するAI内視鏡診断システムの確立

建立预测结直肠癌基因亚型的AI内镜诊断系统

• 批准号: 22K15979
• 负责人: 神山 勇太
• 金额: $ 2.91万
• 依托单位: Showa University
• 依托单位国家: 日本
• 项目类别: Grant-in-Aid for Early-Career Scientists
• 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/en/grant/KAKENHI-PROJECT-22K15979/
• 关键词: 大腸癌; AI; CMS; 内視鏡; 遺伝子発現; 大腸内視鏡

本研究は、人工知能(AI)を用いることで、大腸癌の内視鏡画像から治療選択や予後予測に関連する遺伝子サブタイプを予測するシステムの確立を目指す。大腸癌の治療選択や予後予測に関連する遺伝子サブタイプ(CMS分類)が提唱されている。しかし、遺伝子学的特徴の把握には時間的・金銭的コストや、腫瘍内不均一性が問題となる。大腸内視鏡検査は、診断・治療選択を行う上で必要不可欠な検査であり、短時間で腫瘍全体をくまなくスキャンし組織型や深達度の診断を行うことができる。大腸内視鏡画像から遺伝子サブタイプを予測するAI内視鏡診断システムの確立を目指す。具体的には、画像を取得した大腸癌症例に対し、RNAシーケンス、免疫染色およびVISIUM解析を行い、CMS分類の同定を行う。AIを用い、内視鏡画像とCMS分類を学習することで、内視鏡画像から遺伝子サブタイプを予測可能とする。現在まで、T2癌55例に対し免疫染色によるCMS分類の解析を施行し、論文報告を行った(Molecular and clinicopathological differences between depressed and protruded T2 colorectal cancer, PLoS One. 2022 Oct 20;17(10))。結果は大腸癌の肉眼型毎にCMS分類が異なっており、とりわけ、予後が不良と考えられている、CMS４に分類された症例については、陥凹型大腸癌であり、肉眼型から、大腸癌のCMSが予測可能である可能性が示唆された。更に、早期大腸癌４例に対し、VISIUM解析を行い、遺伝子発現解析を施行し、解析を行った。肉眼型毎に、遺伝子発現が異なり、特に、陥凹型では、CMS４に特徴的な間質反応所見が強く出ていた。

In this study, artificial intelligence (AI) was used to make sure that the eyes were fixed in the picture of large cancer. For the treatment of cancer, you will choose to raise the number of children in the future (CMS category) to sing the song. The special features of computer science and technology have a good grasp of the problems of financial and internal heterogeneity in the time of purchase. It is necessary to make sure that it is necessary to do so, and for a short period of time, it is necessary to do so in a short period of time. The portrait of the inside of the AI is made sure that the eyes are fixed. Specific images and portraits were obtained, including cancer cases, RNA images, immunostaining, VISIUM analysis, and CMS classification. AI portrait, internal portrait, CMS classification, internal portrait, internal portrait, image, image, picture, picture At present, 55 cases of Molecular and clinicopathological differences between depressed and protruded and T2 cancer were treated by immunostaining, CMS classification and Molecular and clinicopathological differences between depressed and protruded T2 colorectal cancer, PLoS One. 2022 Oct 2017 (10). Results the gross CMS classification of large breast cancer showed that the possibility of predicting the possibility of cancer was instigated by the classification of gross cancer, the diagnosis of gross cancer, the classification of gross cancer, the classification of gross cancer, the diagnosis of gross cancer, the diagnosis of gross cancer, the classification of gross cancer, the diagnosis of gross cancer, the classification of gross cancer, the diagnosis of gross cancer, the classification of gross cancer, the classification of gross cancer, the diagnosis of gross cancer, the classification of gross cancer, the diagnosis of gross cancer, the classification of gross cancer, the classification of gross cancer, the possibility of diagnosis, the classification of CMS4, the possibility of cancer, the possibility of cancer, the possibility of diagnosis. In the early stage, 4 cases of large intestine cancer were diagnosed by VISIUM analysis and analysis. The naked eye type, the son of the eye, the special type, the concave type, and the CMS4 type show the strong signs of the disease.

项目成果

Molecular and clinicopathological differences between depressed and protruded T2 colorectal cancer.

• DOI: 10.1371/journal.pone.0273566
• 发表时间: 2022
• 期刊: PloS one
• 影响因子: 3.7

Whole slide image-based prediction of lymph node metastasis in T1 colorectal cancer using unsupervised artificial intelligence

使用无监督人工智能基于全幻灯片图像预测 T1 结直肠癌淋巴结转移

• DOI: 10.1111/den.14547
• 发表时间: 2023
• 期刊: Digestive Endoscopy
• 影响因子: 5.3
• 作者: Yuki Takashina ;Shin-Ei Kudo;Yuta Kouyama ;Katsuro Ichimasa ;Hideyuki Miyachi ;Masashi Misawa;et. al
• 通讯作者: et. al