Development of Double-Speed Movie Drive Processing Technology for Cardiac Catheterography

2022 Fiscal Year Final Research Report

• Project/Area Number: 17K17738
• Research Category: Grant-in-Aid for Young Scientists (B)
• Allocation Type: Multi-year Fund
• Research Field: Medical Physics and Radiological Technology; Radiation science
• Research Institution: Niigata University of Health and Welfare (2019-2022); Niigata University (2017-2018)
• Principal Investigator: Hasegawa Akira 新潟医療福祉大学, 医療技術学部, 講師 (20749999)
• Project Period (FY): 2017-04-01 – 2023-03-31
• Keywords: 人工知能 / ディープラーニング / U-Net / 不鋭除去 / 心臓カテーテル撮影 / 冠動脈 / 動画

Outline of Final Research Achievements

The result of this study is a deep neural network (U-Net) for removing unsharpness due to motion. First, the U-Net was trained using still movies of metronome oscillations as the teacher images and movies with a frame rate of 7.5 fps and a metronome frequency of 160 bpm as the training images. Then, from the 50 cases of movies of the right coronary artery, additional U-Net training was conducted using the image from the cardiac phase with small coronary artery motion as the teacher image and the training image as the image in which insensitivity could be confirmed. As a result, sharpness of coronary arteries was significantly eliminated. The processed images were stitched together as frame images and played back as a pseudo-movie, resulting in a video in which the coronary arteries can be clearly observed.

Free Research Field

放射線技術学

Academic Significance and Societal Importance of the Research Achievements

本研究の結果は心血管撮影の画像処理に応用でき、これにより被ばくを増やすことや患者負担を増やすことなく、心血管撮影において冠動脈がより明瞭に描出されることが期待される。冠動脈の描出の向上は冠動脈病変の診断や経皮的冠動脈インターベンションにおけるデバイスの位置確認精度の向上が期待される。更にデバイスの位置確認精度の向上は安全なインターベンションに寄与することが期待される。