Deep Learning-Assisted Quantitative Measurement of Thoracolumbar Fracture Features on Lateral Radiographs
Objective This study aimed to develop and validate a deep learning (DL) algorithm for the quantitative measurement of thoracolumbar (TL) fracture features, and to evaluate its efficacy across varying levels of clinical expertise. Methods Using the pretrained Mask Region-Based Convolutional Neural Ne...
| Main Authors: | , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Korean Spinal Neurosurgery Society
2024-03-01
|
| Series: | Neurospine |
| Subjects: | |
| Online Access: | http://www.e-neurospine.org/upload/pdf/ns-2347366-683.pdf |
| _version_ | 1797237740472893440 |
|---|---|
| author | Woon Tak Yuh Eun Kyung Khil Yu Sung Yoon Burnyoung Kim Hongjun Yoon Jihe Lim Kyoung Yeon Lee Yeong Seo Yoo Kyeong Deuk An |
| author_facet | Woon Tak Yuh Eun Kyung Khil Yu Sung Yoon Burnyoung Kim Hongjun Yoon Jihe Lim Kyoung Yeon Lee Yeong Seo Yoo Kyeong Deuk An |
| author_sort | Woon Tak Yuh |
| collection | DOAJ |
| description | Objective This study aimed to develop and validate a deep learning (DL) algorithm for the quantitative measurement of thoracolumbar (TL) fracture features, and to evaluate its efficacy across varying levels of clinical expertise. Methods Using the pretrained Mask Region-Based Convolutional Neural Networks model, originally developed for vertebral body segmentation and fracture detection, we fine-tuned the model and added a new module for measuring fracture metrics—compression rate (CR), Cobb angle (CA), Gardner angle (GA), and sagittal index (SI)—from lumbar spine lateral radiographs. These metrics were derived from six-point labeling by 3 radiologists, forming the ground truth (GT). Training utilized 1,000 nonfractured and 318 fractured radiographs, while validations employed 213 internal and 200 external fractured radiographs. The accuracy of the DL algorithm in quantifying fracture features was evaluated against GT using the intraclass correlation coefficient. Additionally, 4 readers with varying expertise levels, including trainees and an attending spine surgeon, performed measurements with and without DL assistance, and their results were compared to GT and the DL model. Results The DL algorithm demonstrated good to excellent agreement with GT for CR, CA, GA, and SI in both internal (0.860, 0.944, 0.932, and 0.779, respectively) and external (0.836, 0.940, 0.916, and 0.815, respectively) validations. DL-assisted measurements significantly improved most measurement values, particularly for trainees. Conclusion The DL algorithm was validated as an accurate tool for quantifying TL fracture features using radiographs. DL-assisted measurement is expected to expedite the diagnostic process and enhance reliability, particularly benefiting less experienced clinicians. |
| first_indexed | 2024-04-24T17:24:33Z |
| format | Article |
| id | doaj.art-224ff86ece7e4e739bd6d35ee02b37ef |
| institution | Directory Open Access Journal |
| issn | 2586-6583 2586-6591 |
| language | English |
| last_indexed | 2024-04-24T17:24:33Z |
| publishDate | 2024-03-01 |
| publisher | Korean Spinal Neurosurgery Society |
| record_format | Article |
| series | Neurospine |
| spelling | doaj.art-224ff86ece7e4e739bd6d35ee02b37ef2024-03-28T07:08:10ZengKorean Spinal Neurosurgery SocietyNeurospine2586-65832586-65912024-03-01211304310.14245/ns.2347366.6831545Deep Learning-Assisted Quantitative Measurement of Thoracolumbar Fracture Features on Lateral RadiographsWoon Tak Yuh0Eun Kyung Khil1Yu Sung Yoon2Burnyoung Kim3Hongjun Yoon4Jihe Lim5Kyoung Yeon Lee6Yeong Seo Yoo7Kyeong Deuk An8 Department of Neurosurgery, Hallym University Dongtan Sacred Heart Hospital, Hwaseong, Korea Department of Radiology, Hallym University Dongtan Sacred Heart Hospital, Hwaseong, Korea Department of Radiology, Kyungpook National University Hospital, School of Medicine, Kyungpook National University, Daegu, Korea DEEPNOID Inc., Seoul, Korea DEEPNOID Inc., Seoul, Korea Department of Radiology, Hallym University Dongtan Sacred Heart Hospital, Hwaseong, Korea Department of Radiology, Hallym University Dongtan Sacred Heart Hospital, Hwaseong, Korea Department of Radiology, Hallym University Dongtan Sacred Heart Hospital, Hwaseong, Korea Department of Neurosurgery, Hallym University Dongtan Sacred Heart Hospital, Hwaseong, KoreaObjective This study aimed to develop and validate a deep learning (DL) algorithm for the quantitative measurement of thoracolumbar (TL) fracture features, and to evaluate its efficacy across varying levels of clinical expertise. Methods Using the pretrained Mask Region-Based Convolutional Neural Networks model, originally developed for vertebral body segmentation and fracture detection, we fine-tuned the model and added a new module for measuring fracture metrics—compression rate (CR), Cobb angle (CA), Gardner angle (GA), and sagittal index (SI)—from lumbar spine lateral radiographs. These metrics were derived from six-point labeling by 3 radiologists, forming the ground truth (GT). Training utilized 1,000 nonfractured and 318 fractured radiographs, while validations employed 213 internal and 200 external fractured radiographs. The accuracy of the DL algorithm in quantifying fracture features was evaluated against GT using the intraclass correlation coefficient. Additionally, 4 readers with varying expertise levels, including trainees and an attending spine surgeon, performed measurements with and without DL assistance, and their results were compared to GT and the DL model. Results The DL algorithm demonstrated good to excellent agreement with GT for CR, CA, GA, and SI in both internal (0.860, 0.944, 0.932, and 0.779, respectively) and external (0.836, 0.940, 0.916, and 0.815, respectively) validations. DL-assisted measurements significantly improved most measurement values, particularly for trainees. Conclusion The DL algorithm was validated as an accurate tool for quantifying TL fracture features using radiographs. DL-assisted measurement is expected to expedite the diagnostic process and enhance reliability, particularly benefiting less experienced clinicians.http://www.e-neurospine.org/upload/pdf/ns-2347366-683.pdfartificial intelligencedeep learningspinal fracturesspinal injuriesspinal curvaturesradiography |
| spellingShingle | Woon Tak Yuh Eun Kyung Khil Yu Sung Yoon Burnyoung Kim Hongjun Yoon Jihe Lim Kyoung Yeon Lee Yeong Seo Yoo Kyeong Deuk An Deep Learning-Assisted Quantitative Measurement of Thoracolumbar Fracture Features on Lateral Radiographs Neurospine artificial intelligence deep learning spinal fractures spinal injuries spinal curvatures radiography |
| title | Deep Learning-Assisted Quantitative Measurement of Thoracolumbar Fracture Features on Lateral Radiographs |
| title_full | Deep Learning-Assisted Quantitative Measurement of Thoracolumbar Fracture Features on Lateral Radiographs |
| title_fullStr | Deep Learning-Assisted Quantitative Measurement of Thoracolumbar Fracture Features on Lateral Radiographs |
| title_full_unstemmed | Deep Learning-Assisted Quantitative Measurement of Thoracolumbar Fracture Features on Lateral Radiographs |
| title_short | Deep Learning-Assisted Quantitative Measurement of Thoracolumbar Fracture Features on Lateral Radiographs |
| title_sort | deep learning assisted quantitative measurement of thoracolumbar fracture features on lateral radiographs |
| topic | artificial intelligence deep learning spinal fractures spinal injuries spinal curvatures radiography |
| url | http://www.e-neurospine.org/upload/pdf/ns-2347366-683.pdf |
| work_keys_str_mv | AT woontakyuh deeplearningassistedquantitativemeasurementofthoracolumbarfracturefeaturesonlateralradiographs AT eunkyungkhil deeplearningassistedquantitativemeasurementofthoracolumbarfracturefeaturesonlateralradiographs AT yusungyoon deeplearningassistedquantitativemeasurementofthoracolumbarfracturefeaturesonlateralradiographs AT burnyoungkim deeplearningassistedquantitativemeasurementofthoracolumbarfracturefeaturesonlateralradiographs AT hongjunyoon deeplearningassistedquantitativemeasurementofthoracolumbarfracturefeaturesonlateralradiographs AT jihelim deeplearningassistedquantitativemeasurementofthoracolumbarfracturefeaturesonlateralradiographs AT kyoungyeonlee deeplearningassistedquantitativemeasurementofthoracolumbarfracturefeaturesonlateralradiographs AT yeongseoyoo deeplearningassistedquantitativemeasurementofthoracolumbarfracturefeaturesonlateralradiographs AT kyeongdeukan deeplearningassistedquantitativemeasurementofthoracolumbarfracturefeaturesonlateralradiographs |