Prediction of osteoporosis using MRI and CT scans with unimodal and multimodal deep-learning models
PURPOSE: Osteoporosis is the systematic degeneration of the human skeleton, with consequences ranging from a reduced quality of life to mortality. Therefore, the prediction of osteoporosis reduces risks and supports patients in taking precautions. Deep-learning and specific models achieve highly acc...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Galenos Publishing House
2024-01-01
|
| Series: | Diagnostic and Interventional Radiology |
| Subjects: | |
| Online Access: |
http://www.dirjournal.org/archives/archive-detail/article-preview/prediction-of-osteoporosis-using-mr-and-ct-scans-w/60592
|
| _version_ | 1827386733473300480 |
|---|---|
| author | Yasemin Küçükçiloğlu Boran Şekeroğlu Terin Adalı Niyazi Şentürk |
| author_facet | Yasemin Küçükçiloğlu Boran Şekeroğlu Terin Adalı Niyazi Şentürk |
| author_sort | Yasemin Küçükçiloğlu |
| collection | DOAJ |
| description | PURPOSE: Osteoporosis is the systematic degeneration of the human skeleton, with consequences ranging from a reduced quality of life to mortality. Therefore, the prediction of osteoporosis reduces risks and supports patients in taking precautions. Deep-learning and specific models achieve highly accurate results using different imaging modalities. The primary purpose of this research was to develop unimodal and multimodal deep-learning-based diagnostic models to predict bone mineral loss of the lumbar vertebrae using magnetic resonance (MR) and computed tomography (CT) imaging. METHODS: Patients who received both lumbar dual-energy X-ray absorptiometry (DEXA) and MRI (n = 120) or CT (n = 100) examinations were included in this study. Unimodal and multimodal convolutional neural networks (CNNs) with dual blocks were proposed to predict osteoporosis using lumbar vertebrae MR and CT examinations in separate and combined datasets. Bone mineral density values obtained by DEXA were used as reference data. The proposed models were compared with a CNN model and six benchmark pre-trained deep-learning models. RESULTS: The proposed unimodal model obtained 96.54%, 98.84%, and 96.76% balanced accuracy for MRI, CT, and combined datasets, respectively, while the multimodal model achieved 98.90% balanced accuracy in 5-fold cross-validation experiments. Furthermore, the models obtained 95.68%–97.91% accuracy with a hold-out validation dataset. In addition, comparative experiments demonstrated that the proposed models yielded superior results by providing more effective feature extraction in dual blocks to predict osteoporosis. CONCLUSION: This study demonstrated that osteoporosis was accurately predicted by the proposed models using both MR and CT images, and a multimodal approach improved the prediction of osteoporosis. With further research involving prospective studies with a larger number of patients, there may be an opportunity to implement these technologies into clinical practice. |
| first_indexed | 2024-03-08T15:49:52Z |
| format | Article |
| id | doaj.art-7d66c132116b4d20a8de8df79c1107a1 |
| institution | Directory Open Access Journal |
| issn | 1305-3612 |
| language | English |
| last_indexed | 2024-03-08T15:49:52Z |
| publishDate | 2024-01-01 |
| publisher | Galenos Publishing House |
| record_format | Article |
| series | Diagnostic and Interventional Radiology |
| spelling | doaj.art-7d66c132116b4d20a8de8df79c1107a12024-01-09T06:45:02ZengGalenos Publishing HouseDiagnostic and Interventional Radiology1305-36122024-01-0130192010.4274/dir.2023.23211613049054Prediction of osteoporosis using MRI and CT scans with unimodal and multimodal deep-learning modelsYasemin Küçükçiloğlu0Boran Şekeroğlu1Terin Adalı2Niyazi Şentürk3 Near East University Faculty of Medicine, Department of Radiology, Nicosia, Cyprus Near East University, Applied Artificial Intelligence Research Center, Nicosia, Cyprus Near East University, Center of Excellence, Tissue Engineering and Biomaterials Research Center, Nicosia, Cyprus Near East University, Center of Excellence, Tissue Engineering and Biomaterials Research Center, Nicosia, Cyprus PURPOSE: Osteoporosis is the systematic degeneration of the human skeleton, with consequences ranging from a reduced quality of life to mortality. Therefore, the prediction of osteoporosis reduces risks and supports patients in taking precautions. Deep-learning and specific models achieve highly accurate results using different imaging modalities. The primary purpose of this research was to develop unimodal and multimodal deep-learning-based diagnostic models to predict bone mineral loss of the lumbar vertebrae using magnetic resonance (MR) and computed tomography (CT) imaging. METHODS: Patients who received both lumbar dual-energy X-ray absorptiometry (DEXA) and MRI (n = 120) or CT (n = 100) examinations were included in this study. Unimodal and multimodal convolutional neural networks (CNNs) with dual blocks were proposed to predict osteoporosis using lumbar vertebrae MR and CT examinations in separate and combined datasets. Bone mineral density values obtained by DEXA were used as reference data. The proposed models were compared with a CNN model and six benchmark pre-trained deep-learning models. RESULTS: The proposed unimodal model obtained 96.54%, 98.84%, and 96.76% balanced accuracy for MRI, CT, and combined datasets, respectively, while the multimodal model achieved 98.90% balanced accuracy in 5-fold cross-validation experiments. Furthermore, the models obtained 95.68%–97.91% accuracy with a hold-out validation dataset. In addition, comparative experiments demonstrated that the proposed models yielded superior results by providing more effective feature extraction in dual blocks to predict osteoporosis. CONCLUSION: This study demonstrated that osteoporosis was accurately predicted by the proposed models using both MR and CT images, and a multimodal approach improved the prediction of osteoporosis. With further research involving prospective studies with a larger number of patients, there may be an opportunity to implement these technologies into clinical practice. http://www.dirjournal.org/archives/archive-detail/article-preview/prediction-of-osteoporosis-using-mr-and-ct-scans-w/60592 osteoporosisdual-energy x-ray absorptiometrylumbar vertebraedeep learningmultimodal cnn |
| spellingShingle | Yasemin Küçükçiloğlu Boran Şekeroğlu Terin Adalı Niyazi Şentürk Prediction of osteoporosis using MRI and CT scans with unimodal and multimodal deep-learning models Diagnostic and Interventional Radiology osteoporosis dual-energy x-ray absorptiometry lumbar vertebrae deep learning multimodal cnn |
| title | Prediction of osteoporosis using MRI and CT scans with unimodal and multimodal deep-learning models |
| title_full | Prediction of osteoporosis using MRI and CT scans with unimodal and multimodal deep-learning models |
| title_fullStr | Prediction of osteoporosis using MRI and CT scans with unimodal and multimodal deep-learning models |
| title_full_unstemmed | Prediction of osteoporosis using MRI and CT scans with unimodal and multimodal deep-learning models |
| title_short | Prediction of osteoporosis using MRI and CT scans with unimodal and multimodal deep-learning models |
| title_sort | prediction of osteoporosis using mri and ct scans with unimodal and multimodal deep learning models |
| topic | osteoporosis dual-energy x-ray absorptiometry lumbar vertebrae deep learning multimodal cnn |
| url |
http://www.dirjournal.org/archives/archive-detail/article-preview/prediction-of-osteoporosis-using-mr-and-ct-scans-w/60592
|
| work_keys_str_mv | AT yaseminkucukciloglu predictionofosteoporosisusingmriandctscanswithunimodalandmultimodaldeeplearningmodels AT boransekeroglu predictionofosteoporosisusingmriandctscanswithunimodalandmultimodaldeeplearningmodels AT terinadalı predictionofosteoporosisusingmriandctscanswithunimodalandmultimodaldeeplearningmodels AT niyazisenturk predictionofosteoporosisusingmriandctscanswithunimodalandmultimodaldeeplearningmodels |