-
Something wrong with this record ?
Automatic multi-class intertrochanteric femur fracture detection from CT images based on AO/OTA classification using faster R-CNN-BO method
Sun-Jung Yoon, Tae Hyong Kim, Su-Bin Joo, Seung Eel Oh
Language English Country Czech Republic
Document type Clinical Study, Research Support, Non-U.S. Gov't
- MeSH
- Deep Learning MeSH
- Hip Fractures * diagnostic imaging classification MeSH
- Humans MeSH
- Neural Networks, Computer MeSH
- Tomography, X-Ray Computed * methods MeSH
- Statistics as Topic MeSH
- Check Tag
- Humans MeSH
- Male MeSH
- Female MeSH
- Publication type
- Clinical Study MeSH
- Research Support, Non-U.S. Gov't MeSH
Intertrochanteric (IT) femur fractures are the most common fractures in elderly people, and they lead to significant morbidity, mortality, and reduced quality of life. The different types of fractures require a careful definition to ensure accurate surgical planning and reduce the operation time, healing time, and number of surgical failures. In this study, a deep learning-based automatic multi-class IT fracture detection model was developed using computed tomography (CT) images and based on the AO/OTA classification method. The original CT image was resized and rearranged according to the fracture location and an unsharp masking filter was applied. A multi-class classification of nine different types of IT fractures and no fracture was performed using the faster regional-convolutional neural network (R-CNN). Bayesian optimization was also implemented to determine the optimal hyperparameter values for the faster R-CNN algorithm. In our proposed model, IT fractures classified into two classes showed an average accuracy of 0.97 ± 0.02, which was 0.90 ± 0.02 when classified into ten classes. Additionally, the detected region of interest from our proposed model showed minimum root mean square error and intersection over union values of 16.34 ± 47.01 pixels and 0.87 ± 0.12, respectively. In the future, our proposed automatic multi-class IT femur fracture detection model could allow clinicians to identify the fracture region and diagnose different types of femur fractures faster and more accurately. This will increase the probability of correct surgical treatment and minimize postoperative complications.
Korea Food Research Institute Research Group of Consumer Safety Wanju South Korea
Sungkyunkwan University College of Biotechnology and Bioengineering Suwon South Korea
References provided by Crossref.org
Literatura
- 000
- 00000naa a2200000 a 4500
- 001
- bmc21009070
- 003
- CZ-PrNML
- 005
- 20211022221136.0
- 007
- ta
- 008
- 210331s2020 xr ad f 000 0|eng||
- 009
- AR
- 024 7_
- $a 10.32725/jab.2020.013 $2 doi
- 040 __
- $a ABA008 $d ABA008 $e AACR2 $b cze
- 041 0_
- $a eng
- 044 __
- $a xr
- 100 1_
- $a Yoon, Sun-Jung $u Jeonbuk National University, Research Institute of Clinical Medicine, Department of Orthopedic Surgery, Jeonju, South Korea
- 245 10
- $a Automatic multi-class intertrochanteric femur fracture detection from CT images based on AO/OTA classification using faster R-CNN-BO method / $c Sun-Jung Yoon, Tae Hyong Kim, Su-Bin Joo, Seung Eel Oh
- 504 __
- $a Literatura
- 520 9_
- $a Intertrochanteric (IT) femur fractures are the most common fractures in elderly people, and they lead to significant morbidity, mortality, and reduced quality of life. The different types of fractures require a careful definition to ensure accurate surgical planning and reduce the operation time, healing time, and number of surgical failures. In this study, a deep learning-based automatic multi-class IT fracture detection model was developed using computed tomography (CT) images and based on the AO/OTA classification method. The original CT image was resized and rearranged according to the fracture location and an unsharp masking filter was applied. A multi-class classification of nine different types of IT fractures and no fracture was performed using the faster regional-convolutional neural network (R-CNN). Bayesian optimization was also implemented to determine the optimal hyperparameter values for the faster R-CNN algorithm. In our proposed model, IT fractures classified into two classes showed an average accuracy of 0.97 ± 0.02, which was 0.90 ± 0.02 when classified into ten classes. Additionally, the detected region of interest from our proposed model showed minimum root mean square error and intersection over union values of 16.34 ± 47.01 pixels and 0.87 ± 0.12, respectively. In the future, our proposed automatic multi-class IT femur fracture detection model could allow clinicians to identify the fracture region and diagnose different types of femur fractures faster and more accurately. This will increase the probability of correct surgical treatment and minimize postoperative complications.
- 650 _2
- $a ženské pohlaví $7 D005260
- 650 _2
- $a lidé $7 D006801
- 650 _2
- $a mužské pohlaví $7 D008297
- 650 12
- $a fraktury kyčle $x diagnostické zobrazování $x klasifikace $7 D006620
- 650 12
- $a počítačová rentgenová tomografie $x metody $7 D014057
- 650 _2
- $a statistika jako téma $7 D013223
- 650 _2
- $a neuronové sítě $7 D016571
- 650 _2
- $a deep learning $7 D000077321
- 655 _2
- $a klinická studie $7 D000068397
- 655 _2
- $a práce podpořená grantem $7 D013485
- 700 1_
- $a Kim, Tae Hyong $u Sungkyunkwan University, College of Biotechnology and Bioengineering, Suwon, South Korea
- 700 1_
- $a Joo, Su-Bin $u Korea Institute of Machinery & Materials, Daegu Research Center for Medical Devices and Green Energy, Daegu, South Korea
- 700 1_
- $a Oh, Seung Eel $u Korea Food Research Institute, Research Group of Consumer Safety, Wanju, South Korea
- 773 0_
- $t Journal of applied biomedicine $x 1214-021X $g Roč. 18, č. 4 (2020), s. 97-105 $w MED00012667
- 910 __
- $a ABA008 $b B 2301 $c 1249 $y p $z 0
- 990 __
- $a 20210331111905 $b ABA008
- 991 __
- $a 20211022221143 $b ABA008
- 999 __
- $a ok $b bmc $g 1637292 $s 1129432
- BAS __
- $a 3
- BMC __
- $a 2020 $b 18 $c 4 $d 97-105 $i 1214-021X $m Journal of Applied Biomedicine $x MED00012667
- LZP __
- $c NLK189 $d 20211022 $a NLK 2021-15/dk
