Performance improvement of computer-assisted detection/diagnosis system by considering diversity of image data and temporal change of radiologists' reading characteristics

• Project/Area Number: 25750182
• Research Category: Grant-in-Aid for Young Scientists (B)
• Allocation Type: Multi-year Fund
• Research Field: Medical systems
• Research Institution: The University of Tokyo
• Principal Investigator: NOMURA Yukihiro 東京大学, 医学部附属病院, 特任研究員 (60436491)
• Project Period (FY): 2013-04-01 – 2015-03-31
• Project Status: Completed (Fiscal Year 2014)
• Budget Amount: ¥4,160,000 (Direct Cost: ¥3,200,000、Indirect Cost: ¥960,000); Fiscal Year 2014: ¥1,430,000 (Direct Cost: ¥1,100,000、Indirect Cost: ¥330,000); Fiscal Year 2013: ¥2,730,000 (Direct Cost: ¥2,100,000、Indirect Cost: ¥630,000)
• Keywords: 医用画像 / 診断支援システム / 機械学習 / 遠隔読影環境 / 診断傾向 / 脳動脈瘤 / 遠隔読影 / 多施設連携

Outline of Final Research Achievements

In this study, a performance improvement of computer-assisted detection/diagnosis (CAD) software by considering diversity of image data and temporal change of reading characteristics was investigated. We investigated the performance improvement of CAD software for each institution based on retraining through a simulation-based study. According to the results, the performance of CAD software for each institution was improved by retraining. We also investigated a temporal variation of detecting characteristics for cerebral aneurysms in a routine work.

Research Products

• [Journal Article] Performance improvement in computerized detection of cerebral aneurysms by retraining classifier using feedback data collected in routine reading environment (2014)
  • Author(s): Nomura Y, Masutani Y, Miki S, et al.
  • Journal Title: Journal of Biomedical Graphics and Computing Volume: 4(4) Issue: 4 Pages: 12-21
  • DOI: 10.5430/jbgc.v4n4p12
  • Peer Reviewed / Open Access / Acknowledgement Compliant
• [Journal Article] 遠隔読影環境における多施設連携型CAD 開発、実運用、および継続的性能改善 (2014)
  • Author(s): 野村行弘、増谷佳孝、三木聡一郎 他
  • Journal Title: MEDICAL IMAGING TECHNOLOGY Volume: 32
  • NAID: 130004554965
  • Peer Reviewed
• [Presentation] 通常業務におけるCADを併用した放射線技師の脳動脈瘤検出に関する初期検討 (2015)
  • Author(s): 石田敏哉、野村行弘、薗田恭一、他
  • Organizer: 第71回日本放射線技術学会総会学術大会
  • Place of Presentation: 神奈川
  • Year and Date: 2015-04-16 – 2015-04-19
• [Presentation] 画素識別器の再学習による頭部MRA画像の脳動脈瘤自動検出の性能改善 (2014)
  • Author(s): 野村行弘、増谷佳孝、根本充貴、他
  • Organizer: 第33回日本医用画像工学会大会
  • Place of Presentation: 東京
  • Year and Date: 2014-07-24 – 2014-07-26
• [Presentation] Building an integrated CAD development environment in clinical settings (the 7th report): Continuous performance improvement of CAD by multi-institutional collaboration (2014)
  • Author(s): Nomura Y, Masutani Y, Hayashi N, et al.
  • Organizer: 第73回日本医学放射線学会総会
  • Place of Presentation: 横浜
• [Presentation] 臨床における統合的CAD開発環境の構築（第六報）：再学習処理による施設毎のCAD性能改善 (2013)
  • Author(s): 野村行弘、増谷佳孝、林直人 他
  • Organizer: 第72回日本医学放射線学会総会
  • Place of Presentation: 横浜
• [Presentation] 複数画素サイズのデータを用いた画素識別処理による頭部MRA画像の脳動脈瘤自動検出 (2013)
  • Author(s): 野村行弘、増谷佳孝、根本充貴 他
  • Organizer: 第32回日本医用画像工学会大会
  • Place of Presentation: 東京
• [Presentation] 遠隔読影環境における多施設連携型CAD開発、実運用および継続的性能改善 (2013)
  • Author(s): 野村行弘、増谷佳孝、三木聡一郎 他
  • Organizer: 第15回医用画像認知研究会
  • Place of Presentation: 京都
• [Presentation] Training strategy for performance improvement in computer-assisted detection of lesions: based on multi-institutional study in teleradiology environment (2013)
  • Author(s): Nomura Y, Masutani Y, Miki S, et al.
  • Organizer: 1st International Workshop on BioImage Recognition
  • Place of Presentation: 松山
• [Remarks] CIRCUSプロジェクトウェブサイト
  • URL: http://www.ut-radiology.umin.jp/ical/CIRCUS/
• [Remarks] 統合的診断支援開発プラットホーム CIRCUS
  • URL: http://www.ut-radiology.umin.jp/ical/CIRCUS/