Augmented prediction of vertebral collapse after osteoporotic vertebral compression fractures through parameter-efficient fine-tuning of biomedical foundation models
Abstract Vertebral collapse (VC) following osteoporotic vertebral compression fracture (OVCF) often requires aggressive treatment, necessitating an accurate prediction for early intervention. This study aimed to develop a predictive model leveraging deep neural networks to predict VC progression aft...
Saved in:
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Nature Portfolio
2024-12-01
|
| Series: | Scientific Reports |
| Subjects: | |
| Online Access: | https://doi.org/10.1038/s41598-024-82902-w |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1841559439961751552 |
|---|---|
| author | Sibeen Kim Inkyeong Kim Woon Tak Yuh Sangmin Han Choonghyo Kim Young San Ko Wonwoo Cho Sung Bae Park |
| author_facet | Sibeen Kim Inkyeong Kim Woon Tak Yuh Sangmin Han Choonghyo Kim Young San Ko Wonwoo Cho Sung Bae Park |
| author_sort | Sibeen Kim |
| collection | DOAJ |
| description | Abstract Vertebral collapse (VC) following osteoporotic vertebral compression fracture (OVCF) often requires aggressive treatment, necessitating an accurate prediction for early intervention. This study aimed to develop a predictive model leveraging deep neural networks to predict VC progression after OVCF using magnetic resonance imaging (MRI) and clinical data. Among 245 enrolled patients with acute OVCF, data from 200 patients were used for the development dataset, and data from 45 patients were used for the test dataset. To construct an accurate prediction model, we explored two backbone architectures: convolutional neural networks and vision transformers (ViTs), along with various pre-trained weights and fine-tuning methods. Through extensive experiments, we built our model by performing parameter-efficient fine-tuning of a ViT model pre-trained on a large-scale biomedical dataset. Attention rollouts indicated that the contours and internal features of the compressed vertebral body were critical in predicting VC with this model. To further improve the prediction performance of our model, we applied the augmented prediction strategy, which uses multiple MRI frames and achieves a significantly higher area under the curve (AUC). Our findings suggest that employing a biomedical foundation model fine-tuned using a parameter-efficient method, along with augmented prediction, can significantly enhance medical decisions. |
| format | Article |
| id | doaj-art-7980ee2a52dc4d52bfa326091963d1d3 |
| institution | Kabale University |
| issn | 2045-2322 |
| language | English |
| publishDate | 2024-12-01 |
| publisher | Nature Portfolio |
| record_format | Article |
| series | Scientific Reports |
| spelling | doaj-art-7980ee2a52dc4d52bfa326091963d1d32025-01-05T12:30:53ZengNature PortfolioScientific Reports2045-23222024-12-0114111110.1038/s41598-024-82902-wAugmented prediction of vertebral collapse after osteoporotic vertebral compression fractures through parameter-efficient fine-tuning of biomedical foundation modelsSibeen Kim0Inkyeong Kim1Woon Tak Yuh2Sangmin Han3Choonghyo Kim4Young San Ko5Wonwoo Cho6Sung Bae Park7School of Biomedical Engineering, Korea UniversityDepartment of Neurosurgery, Kangwon National University HospitalDepartment of Neurosurgery, Hallym University College of MedicineDepartment of Intelligence Convergence, Yonsei UniversityDepartment of Neurosurgery, Kangwon National University HospitalDepartment of Neurosurgery, Kyungpook National University HospitalKim Jaechul Graduate School of Artificial Intelligence, Korea Advanced Institute of Science and TechnologyDepartment of Medical Device Development, Seoul National University College of MedicineAbstract Vertebral collapse (VC) following osteoporotic vertebral compression fracture (OVCF) often requires aggressive treatment, necessitating an accurate prediction for early intervention. This study aimed to develop a predictive model leveraging deep neural networks to predict VC progression after OVCF using magnetic resonance imaging (MRI) and clinical data. Among 245 enrolled patients with acute OVCF, data from 200 patients were used for the development dataset, and data from 45 patients were used for the test dataset. To construct an accurate prediction model, we explored two backbone architectures: convolutional neural networks and vision transformers (ViTs), along with various pre-trained weights and fine-tuning methods. Through extensive experiments, we built our model by performing parameter-efficient fine-tuning of a ViT model pre-trained on a large-scale biomedical dataset. Attention rollouts indicated that the contours and internal features of the compressed vertebral body were critical in predicting VC with this model. To further improve the prediction performance of our model, we applied the augmented prediction strategy, which uses multiple MRI frames and achieves a significantly higher area under the curve (AUC). Our findings suggest that employing a biomedical foundation model fine-tuned using a parameter-efficient method, along with augmented prediction, can significantly enhance medical decisions.https://doi.org/10.1038/s41598-024-82902-wCompression fractureSpineBiomedical foundation modelVision transformerParameter-efficient fine-tuning |
| spellingShingle | Sibeen Kim Inkyeong Kim Woon Tak Yuh Sangmin Han Choonghyo Kim Young San Ko Wonwoo Cho Sung Bae Park Augmented prediction of vertebral collapse after osteoporotic vertebral compression fractures through parameter-efficient fine-tuning of biomedical foundation models Scientific Reports Compression fracture Spine Biomedical foundation model Vision transformer Parameter-efficient fine-tuning |
| title | Augmented prediction of vertebral collapse after osteoporotic vertebral compression fractures through parameter-efficient fine-tuning of biomedical foundation models |
| title_full | Augmented prediction of vertebral collapse after osteoporotic vertebral compression fractures through parameter-efficient fine-tuning of biomedical foundation models |
| title_fullStr | Augmented prediction of vertebral collapse after osteoporotic vertebral compression fractures through parameter-efficient fine-tuning of biomedical foundation models |
| title_full_unstemmed | Augmented prediction of vertebral collapse after osteoporotic vertebral compression fractures through parameter-efficient fine-tuning of biomedical foundation models |
| title_short | Augmented prediction of vertebral collapse after osteoporotic vertebral compression fractures through parameter-efficient fine-tuning of biomedical foundation models |
| title_sort | augmented prediction of vertebral collapse after osteoporotic vertebral compression fractures through parameter efficient fine tuning of biomedical foundation models |
| topic | Compression fracture Spine Biomedical foundation model Vision transformer Parameter-efficient fine-tuning |
| url | https://doi.org/10.1038/s41598-024-82902-w |
| work_keys_str_mv | AT sibeenkim augmentedpredictionofvertebralcollapseafterosteoporoticvertebralcompressionfracturesthroughparameterefficientfinetuningofbiomedicalfoundationmodels AT inkyeongkim augmentedpredictionofvertebralcollapseafterosteoporoticvertebralcompressionfracturesthroughparameterefficientfinetuningofbiomedicalfoundationmodels AT woontakyuh augmentedpredictionofvertebralcollapseafterosteoporoticvertebralcompressionfracturesthroughparameterefficientfinetuningofbiomedicalfoundationmodels AT sangminhan augmentedpredictionofvertebralcollapseafterosteoporoticvertebralcompressionfracturesthroughparameterefficientfinetuningofbiomedicalfoundationmodels AT choonghyokim augmentedpredictionofvertebralcollapseafterosteoporoticvertebralcompressionfracturesthroughparameterefficientfinetuningofbiomedicalfoundationmodels AT youngsanko augmentedpredictionofvertebralcollapseafterosteoporoticvertebralcompressionfracturesthroughparameterefficientfinetuningofbiomedicalfoundationmodels AT wonwoocho augmentedpredictionofvertebralcollapseafterosteoporoticvertebralcompressionfracturesthroughparameterefficientfinetuningofbiomedicalfoundationmodels AT sungbaepark augmentedpredictionofvertebralcollapseafterosteoporoticvertebralcompressionfracturesthroughparameterefficientfinetuningofbiomedicalfoundationmodels |