A Deep Learning Method for Foot Progression Angle Detection in Plantar Pressure Images
Foot progression angle (FPA) analysis is one of the core methods to detect gait pathologies as basic information to prevent foot injury from excessive in-toeing and out-toeing. Deep learning-based object detection can assist in measuring the FPA through plantar pressure images. This study aims to es...
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2022-04-01
|
| Series: | Sensors |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1424-8220/22/7/2786 |
| _version_ | 1797437678421016576 |
|---|---|
| author | Peter Ardhianto Raden Bagus Reinaldy Subiakto Chih-Yang Lin Yih-Kuen Jan Ben-Yi Liau Jen-Yung Tsai Veit Babak Hamun Akbari Chi-Wen Lung |
| author_facet | Peter Ardhianto Raden Bagus Reinaldy Subiakto Chih-Yang Lin Yih-Kuen Jan Ben-Yi Liau Jen-Yung Tsai Veit Babak Hamun Akbari Chi-Wen Lung |
| author_sort | Peter Ardhianto |
| collection | DOAJ |
| description | Foot progression angle (FPA) analysis is one of the core methods to detect gait pathologies as basic information to prevent foot injury from excessive in-toeing and out-toeing. Deep learning-based object detection can assist in measuring the FPA through plantar pressure images. This study aims to establish a precision model for determining the FPA. The precision detection of FPA can provide information with in-toeing, out-toeing, and rearfoot kinematics to evaluate the effect of physical therapy programs on knee pain and knee osteoarthritis. We analyzed a total of 1424 plantar images with three different You Only Look Once (YOLO) networks: YOLO v3, v4, and v5x, to obtain a suitable model for FPA detection. YOLOv4 showed higher performance of the profile-box, with average precision in the left foot of 100.00% and the right foot of 99.78%, respectively. Besides, in detecting the foot angle-box, the ground-truth has similar results with YOLOv4 (5.58 ± 0.10° vs. 5.86 ± 0.09°, <i>p</i> = 0.013). In contrast, there was a significant difference in FPA between ground-truth vs. YOLOv3 (5.58 ± 0.10° vs. 6.07 ± 0.06°, <i>p</i> < 0.001), and ground-truth vs. YOLOv5x (5.58 ± 0.10° vs. 6.75 ± 0.06°, <i>p</i> < 0.001). This result implies that deep learning with YOLOv4 can enhance the detection of FPA. |
| first_indexed | 2024-03-09T11:24:53Z |
| format | Article |
| id | doaj.art-edd673736d7843f08b8c302478ac95f7 |
| institution | Directory Open Access Journal |
| issn | 1424-8220 |
| language | English |
| last_indexed | 2024-03-09T11:24:53Z |
| publishDate | 2022-04-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Sensors |
| spelling | doaj.art-edd673736d7843f08b8c302478ac95f72023-12-01T00:05:38ZengMDPI AGSensors1424-82202022-04-01227278610.3390/s22072786A Deep Learning Method for Foot Progression Angle Detection in Plantar Pressure ImagesPeter Ardhianto0Raden Bagus Reinaldy Subiakto1Chih-Yang Lin2Yih-Kuen Jan3Ben-Yi Liau4Jen-Yung Tsai5Veit Babak Hamun Akbari6Chi-Wen Lung7Department of Visual Communication Design, Soegijapranata Catholic University, Semarang 50234, IndonesiaDepartment of Mathematics, Airlangga University, Surabaya 60115, IndonesiaDepartment of Electrical Engineering, Yuan Ze University, Chung-Li 32003, TaiwanRehabilitation Engineering Lab, University of Illinois at Urbana-Champaign, Champaign, IL 61820, USADepartment of Biomedical Engineering, Hungkuang University, Taichung 433304, TaiwanDepartment of Digital Media Design, Asia University, Taichung 413305, TaiwanDepartment of Creative Product Design, Asia University, Taichung 413305, TaiwanRehabilitation Engineering Lab, University of Illinois at Urbana-Champaign, Champaign, IL 61820, USAFoot progression angle (FPA) analysis is one of the core methods to detect gait pathologies as basic information to prevent foot injury from excessive in-toeing and out-toeing. Deep learning-based object detection can assist in measuring the FPA through plantar pressure images. This study aims to establish a precision model for determining the FPA. The precision detection of FPA can provide information with in-toeing, out-toeing, and rearfoot kinematics to evaluate the effect of physical therapy programs on knee pain and knee osteoarthritis. We analyzed a total of 1424 plantar images with three different You Only Look Once (YOLO) networks: YOLO v3, v4, and v5x, to obtain a suitable model for FPA detection. YOLOv4 showed higher performance of the profile-box, with average precision in the left foot of 100.00% and the right foot of 99.78%, respectively. Besides, in detecting the foot angle-box, the ground-truth has similar results with YOLOv4 (5.58 ± 0.10° vs. 5.86 ± 0.09°, <i>p</i> = 0.013). In contrast, there was a significant difference in FPA between ground-truth vs. YOLOv3 (5.58 ± 0.10° vs. 6.07 ± 0.06°, <i>p</i> < 0.001), and ground-truth vs. YOLOv5x (5.58 ± 0.10° vs. 6.75 ± 0.06°, <i>p</i> < 0.001). This result implies that deep learning with YOLOv4 can enhance the detection of FPA.https://www.mdpi.com/1424-8220/22/7/2786YOLOobject detectionfoot problemsangle parameterfoot clinic |
| spellingShingle | Peter Ardhianto Raden Bagus Reinaldy Subiakto Chih-Yang Lin Yih-Kuen Jan Ben-Yi Liau Jen-Yung Tsai Veit Babak Hamun Akbari Chi-Wen Lung A Deep Learning Method for Foot Progression Angle Detection in Plantar Pressure Images Sensors YOLO object detection foot problems angle parameter foot clinic |
| title | A Deep Learning Method for Foot Progression Angle Detection in Plantar Pressure Images |
| title_full | A Deep Learning Method for Foot Progression Angle Detection in Plantar Pressure Images |
| title_fullStr | A Deep Learning Method for Foot Progression Angle Detection in Plantar Pressure Images |
| title_full_unstemmed | A Deep Learning Method for Foot Progression Angle Detection in Plantar Pressure Images |
| title_short | A Deep Learning Method for Foot Progression Angle Detection in Plantar Pressure Images |
| title_sort | deep learning method for foot progression angle detection in plantar pressure images |
| topic | YOLO object detection foot problems angle parameter foot clinic |
| url | https://www.mdpi.com/1424-8220/22/7/2786 |
| work_keys_str_mv | AT peterardhianto adeeplearningmethodforfootprogressionangledetectioninplantarpressureimages AT radenbagusreinaldysubiakto adeeplearningmethodforfootprogressionangledetectioninplantarpressureimages AT chihyanglin adeeplearningmethodforfootprogressionangledetectioninplantarpressureimages AT yihkuenjan adeeplearningmethodforfootprogressionangledetectioninplantarpressureimages AT benyiliau adeeplearningmethodforfootprogressionangledetectioninplantarpressureimages AT jenyungtsai adeeplearningmethodforfootprogressionangledetectioninplantarpressureimages AT veitbabakhamunakbari adeeplearningmethodforfootprogressionangledetectioninplantarpressureimages AT chiwenlung adeeplearningmethodforfootprogressionangledetectioninplantarpressureimages AT peterardhianto deeplearningmethodforfootprogressionangledetectioninplantarpressureimages AT radenbagusreinaldysubiakto deeplearningmethodforfootprogressionangledetectioninplantarpressureimages AT chihyanglin deeplearningmethodforfootprogressionangledetectioninplantarpressureimages AT yihkuenjan deeplearningmethodforfootprogressionangledetectioninplantarpressureimages AT benyiliau deeplearningmethodforfootprogressionangledetectioninplantarpressureimages AT jenyungtsai deeplearningmethodforfootprogressionangledetectioninplantarpressureimages AT veitbabakhamunakbari deeplearningmethodforfootprogressionangledetectioninplantarpressureimages AT chiwenlung deeplearningmethodforfootprogressionangledetectioninplantarpressureimages |