2022 Fiscal Year Final Research Report

Ultra-High-Resolution CT: Prediction of Therapeutic Induced Complication with Radiomics Appropach

Research Project

PDF

Project/Area Number	20K08037
Research Category	Grant-in-Aid for Scientific Research (C)
Allocation Type	Multi-year Fund
Section	一般
Review Section	Basic Section 52040:Radiological sciences-related
Research Institution	Fujita Health University
Principal Investigator	Ohno Yoshiharu 藤田医科大学, 医学部, 教授 (30324924)
Co-Investigator(Kenkyū-buntansha)	近藤征史藤田医科大学, 医学部, 教授 (00378077) 村山和宏藤田医科大学, 医学部, 准教授 (40622931) 今泉和良藤田医科大学, 医学部, 教授 (50362257) 林真也藤田医科大学, 医学部, 教授 (60313904) 服部秀計藤田医科大学, 医学部, 講師 (70351046) 外山宏藤田医科大学, 医学部, 教授 (90247643) 星川康藤田医科大学, 医学部, 教授 (90333814)
Project Period (FY)	2020-04-01 – 2023-03-31
Keywords	放射線医学 / CT / 人工知能
Outline of Final Research Achievements	In this study, we assessed the influence of reconstruction algorithms including currently available iterative reconstruction techniques as well as deep learning techniques on CT value evaluation and image quality improvements, which were determined as signal-to-noise ratio (SNR) or contrast-to-noise ratio (CNR) on ultra-high-resolution CT (UHR-CT). Then, machine-learning-based artificial intelligence (AI) was also developed to assess lung textures for evaluation of various lung diseases including complication or side effects due to therapy. Moreover, radiomics approach was performed to assessed the primary lesion. According to the above-mentioned data, new AI algorithm was started to be developed and tested for prediction of therapeutic outcome or complications based on conservative therapy in non-small cell lung cancer patients.
Free Research Field	放射線医学
Academic Significance and Societal Importance of the Research Achievements	近年，臨床応用された超高精細CT（Ultra-High-Resolution CT: 以下UHR-CT）は慢性閉塞性肺疾患や間質性肺炎の定量評価においては再構成法や撮像法などに関して様々な影響を受けるとともに、人工知能を用いた定量的評価法やRadiomics解析法の確立が求められている。本研究では世界に先駆けてMachine learningの手法を用いた人工知能を開発し、非小細胞肺癌の保存的治療の予後改善を目的とするため、学問的および社会的意義が高いと考えられる。