A Bone Segmentation for the Knee Joint in MDCT Image Based on Anatomical Information

Abstract

The purpose of this study is to propose an automated segmentation about the knee bones (femur, tibia, patellar, and fibula) in MDCT images. The proposed method was applied for six patients (Age 33 ± 13, four males / two females). The proposed method segments the knee joint into each bone by using anatomical structure for the knee joint. The experiments calculate accuracy of the proposed method by comparing with the manual delineation result. As a result, The accuracy of the femur, tibia, patellar, and fibula were 95.84 ± 0.57%, 94.12 ± 1.01%, 94.49 ± 0.83%, 86.37 ± 4.28%, respectively. This study concluded that the proposed method is enough to segment the knee bones.

Keywords::

• MDCT Image
• Segmentation
• Knee

Additional information

Notes on contributors

Yosuke Uozumi

Yosuke Uozumi receive M.E. degree (human and artificial intelligent systems) from University of Fukui, 2013. He is a student of Doctor course in Graduate School of Engineering, University of Fukui. His research interests include biomedical engineering, image analysis and soft computing.

Kouki Nagamune

Kouki Nagamune received the Ph.D. degree (Computer Engineering) in 2004 from Himeji Institute of Technology. He worked as researcher (2004–2005) and lecturer (2005–2007) at Department of Orthopaedic Surgery, Graduate School of Medicine, Kobe University. He worked as lecturer (2007–2010) at Department of Human and Artificial Intelligent Systems Graduate School of Engineering, University of Fukui. He is currently associate professor at Department of Human and Artificial Intelligent Systems Graduate School of Engineering, University of Fukui. He is a member of the Institute of Electronics, Information and Communication Engineers (IEICE) and the IEEE.

Daisuke Araki

Daisuke Araki received the Ph. D degree from the Kobe University School of Medicine, Kobe, Japan in 2011. He is currently served as a VI sitting researcher at Pittsburg University. His interest includes sports traumatology and knee joint.

Yuichi Hoshino

Yuichi Hoshino received the Ph. D degree from the Kobe University School of Medicine, Kobe, Japan in 2008. He is currently served as a researcher at Kobe University. His interest includes sports traumatology and knee joint.

Takehiko Matsushita

Takehiko Matsushita received the Ph. D degree from the Kobe University School of Medicine, Kobe, Japan in 2004. He is currently served as a researcher at Kobe University. His interest includes sports traumatology and knee joint.

Ryosuke Kuroda

Ryosuke Kuroda received the Ph. D degree from the Kobe University School of Medicine, Kobe, Japan in 2000. He is currently served as an associate professor at Kobe University. His interest includes sports traumatology and knee joint.

Masahiro Kurosaka

Masahiro Kurosaka received the M.D. degree from the Kobe University School of Medicine, Kobe, Japan in 1977. He is currently a Professor with Orthopaedic Laboratory of the Kobe University Graduate School of Medicine. His research interests include knee joint, and rheumatism. He is a member of the American Orthopaedic Society for Sports Medicine, the Orthopaedic Research Society, the European Society of Sports Traumatology, Knee Surgery and Arthroscopy, the International Society of Arthroscopy, Knee Surgery and Orthopaedic Sports Medicine, and the Arthroscopy Association of North America.