Fracture risk assessment by a novel bone structure analysis method using X-ray images

• Project/Area Number: 17K10420
• Research Category: Grant-in-Aid for Scientific Research (C)
• Allocation Type: Multi-year Fund
• Section: 一般
• Research Field: Radiation science
• Research Institution: University of Occupational and Environmental Health, Japan
• Principal Investigator: Aoki Takatoshi 産業医科大学, 医学部, 准教授 (40299631)
• Co-Investigator(Kenkyū-buntansha): 興梠 征典 産業医科大学, 医学部, 教授 (60195691) ; 金 亨燮 九州工業大学, 大学院工学研究院, 教授 (80295005)
• Project Period (FY): 2017-04-01 – 2020-03-31
• Project Status: Completed (Fiscal Year 2019)
• Budget Amount: ¥3,250,000 (Direct Cost: ¥2,500,000、Indirect Cost: ¥750,000); Fiscal Year 2019: ¥650,000 (Direct Cost: ¥500,000、Indirect Cost: ¥150,000); Fiscal Year 2018: ¥650,000 (Direct Cost: ¥500,000、Indirect Cost: ¥150,000); Fiscal Year 2017: ¥1,950,000 (Direct Cost: ¥1,500,000、Indirect Cost: ¥450,000)
• Keywords: 骨強度 / 画像診断 / トモシンセシス / エックス線画像 / 骨粗鬆症 / 骨折 / 糖尿病

Outline of Final Research Achievements

Tomosynthesis is a recent technology that obtains images on arbitrary parallel cross-sections through image reconstruction of multiple projected images taken by a single X-ray exposure. The combination of the bone mineral density (BMD) measurements by dual X-ray absorptiometry (DXA) with tomosynthesis-based trabecular bone analysis improved the correlation to the bone strength in comparison with BMD alone. In addition, the correlation between the bone strength and the BMD with the tomosynthesis-based trabecular bone analysis was higher than that between the bone strength and the various established estimations for fracture risk assessment. Tomosynthesis-based trabecular bone analysis provide bone quality information, and can potentially give additional information to BMD in bone strength prediction of lifestyle-related disease patients.

Academic Significance and Societal Importance of the Research Achievements

エックス線画像を用いた新しい骨構造解析法を用いることで、臨床用CTよりもはるかに少ない被ばくで詳細な骨構造解析を行うことができ、DXAによる骨密度測定では予測できない骨折リスクを推定し得ると予想される。骨粗鬆症の治療では、まず初発骨折を予防することが大切であり、生活習慣病症例では本手法を用いることで、従来よりも正確な薬物治療開始の判断が可能になる可能性がある。

Research Products

• [Journal Article] Classification of Osteoporosis from Phalanges Computed Radiography Images Based on Convolutional Neural Network (2019)
  • Author(s): 畠野 和裕, 村上 誠一, 植村 知規, 陸 慧敏, 金 亨燮, 青木 隆敏
  • Journal Title: Medical Imaging and Information Sciences Volume: 36 Issue: 2 Pages: 72-76
  • DOI: 10.11318/mii.36.72
  • NAID: 130007668726
  • ISSN: 0910-1543, 1880-4977
  • Year and Date: 2019-06-30
  • Peer Reviewed
• [Journal Article] Automatic identification of bone erosions in rheumatoid arthritis from hand radiographs based on deep convolutional neural network (2018)
  • Author(s): Seiichi Murakami, Kazuhiro Hatano, JooKooi Tan, Hyoungseop Kim, Takatoshi Aoki
  • Journal Title: Multimedia Tools Applications Volume: 印刷中 Issue: 9 Pages: 107-115
  • DOI: 10.1007/s11042-017-5449-4
  • Peer Reviewed / Int'l Joint Research
• [Presentation] Automatic Segmentation Method of Phalange Regions Based on Residual U-Net and MSGVF Snakes (2019)
  • Author(s): Kohei Kawagoe, Kazuhiro Hatano, Seiichi Murakami, Huimin Lu, Hyoungseop Kim, Takatoshi Aoki
  • Organizer: Proc. of the 19th International Conference on Control, Automation and Systems
  • Int'l Joint Research
• [Presentation] Segmentation of Bone Metastasis in CT Images Based on Modified HED (2019)
  • Author(s): Yuchan Song, Huimin Lu, Hyoungseop Kim, Seiichi Murakami, Aoi Ueno, Takashi Terasawa, Takatoshi Aoki
  • Organizer: Proc. of the 19th International Conference on Control, Automation and Systems
  • Int'l Joint Research
• [Presentation] Fracture risk assessment of diabetes mellitus patients: comparison of trabecular bone analysis by tomosynthesis with various established estimations (2018)
  • Author(s): Yo Todoroki, Takatoshi Aoki, Masami Fujii, Akitaka Fujisaki, Yoshiko Hayashida, Yosuke Okada, Yoshiya Tanaka, Yukunori Korogi
  • Organizer: RSNA 104th SCIENTIFIC ASSEMBLY & ANNUAL MEETING (第104回 北米放射線学会）
  • Int'l Joint Research
• [Presentation] Bone erosions detection on hand CR images based on DCNN (2018)
  • Author(s): Seiichi Murakami, Kazuhiro Hatano, Huimin Lu, JooKooi Tan, Hyoungseop Kim, Takatoshi Aoki
  • Organizer: The 23rd International Symposium on Artificial Life and Robotics
  • Int'l Joint Research
• [Presentation] Femoral neck strength prediction in osteoporosis patients: trabecular bone analysis using tomosynthesis images (2017)
  • Author(s): Yo Todoroki, Takatoshi Aoki, Masami Fujii, Shinpei Yamaguchi, Akitaka Fujisaki, Yoshiko Hayashida, Kazuhisa Nakano, Yosuke Okada, Kazuyoshi Saito, Yoshiya Tanaka, Yukunori Korogi
  • Organizer: 第35回日本骨代謝学会学術集会