2021 Fiscal Year Final Research Report
Mass spectrometry and artificial intelligence-based real-time endoscopic diagnosis system
| Project/Area Number |
19K07728
|
|---|
| Research Category |
Grant-in-Aid for Scientific Research (C)
|
| Allocation Type | Multi-year Fund |
|---|
| Section | 一般 |
|---|
| Review Section |
Basic Section 50020:Tumor diagnostics and therapeutics-related
|
|---|
| Research Institution | University of Yamanashi |
|---|
Principal Investigator |
|
|---|
| Project Period (FY) |
2019-04-01 – 2022-03-31
|
| Keywords | がん診断 / メタボロミクス / 質量分析 |
|---|
| Outline of Final Research Achievements |
To detect malignant tumors in real-time during endoscopy, we constructed a machine to collect tissue using a string and then transfer the sample to the outside of the patient's body. The tissue is transported by the string to an ion inlet of the mass spectrometer, where biological molecules are extracted and ionized to analyze the molecular composition of the tissue. To perform this process, we also constructed a novel electrospray ionization-based ion source. In addition, we constructed a tissue composition database (47 normal mucosal membrane samples and 44 colorectal cancer tissue samples). The database was learned with a support vector machine, a type of machine learning, to use as a diagnostic algorithm for malignant tumors. When the discrimination ability was evaluated by leave-one-out cross-validation, the concordance rate with the pathological diagnosis was 86%.
|
| Free Research Field |
分子生物学
|
| Academic Significance and Societal Importance of the Research Achievements |
本研究で構築された「AI型質量分析内視鏡がん診断支援システム」は、数ミリの細いチューブを介して、生きている人間の体内や既存の器具では到達しにくい狭い場所（特に消化管内腔）へアクセスし、組織をわずかに採取して迅速に体外へと移送し、リアルタイムで詳細な成分分析を行うことを可能とする。内視鏡検査において実用化されている計測機器のほとんどは光学カメラによって画像を取得するものであり、その場では詳細な分析結果を得られないが、当該システムを利用すれば、検査で得られる情報量が一気に拡大するため、悪性腫瘍を始めとした各種疾患の初期スクリーニングの確度を向上させることが可能となる。
|
