Radiomics and Machine Learning Can Differentiate Transient Osteoporosis from Avascular Necrosis of the Hip
Differentiation between transient osteoporosis (TOH) and avascular necrosis (AVN) of the hip is a longstanding challenge in musculoskeletal radiology. The purpose of this study was to utilize MRI-based radiomics and machine learning (ML) for accurate differentiation between the two entities. A total...
| Main Authors: | , , , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2021-09-01
|
| Series: | Diagnostics |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2075-4418/11/9/1686 |
| _version_ | 1797519657296461824 |
|---|---|
| author | Michail E. Klontzas Georgios C. Manikis Katerina Nikiforaki Evangelia E. Vassalou Konstantinos Spanakis Ioannis Stathis George A. Kakkos Nikolas Matthaiou Aristeidis H. Zibis Kostas Marias Apostolos H. Karantanas |
| author_facet | Michail E. Klontzas Georgios C. Manikis Katerina Nikiforaki Evangelia E. Vassalou Konstantinos Spanakis Ioannis Stathis George A. Kakkos Nikolas Matthaiou Aristeidis H. Zibis Kostas Marias Apostolos H. Karantanas |
| author_sort | Michail E. Klontzas |
| collection | DOAJ |
| description | Differentiation between transient osteoporosis (TOH) and avascular necrosis (AVN) of the hip is a longstanding challenge in musculoskeletal radiology. The purpose of this study was to utilize MRI-based radiomics and machine learning (ML) for accurate differentiation between the two entities. A total of 109 hips with TOH and 104 hips with AVN were retrospectively included. Femoral heads and necks with segmented radiomics features were extracted. Three ML classifiers (XGboost, CatBoost and SVM) using 38 relevant radiomics features were trained on 70% and validated on 30% of the dataset. ML performance was compared to two musculoskeletal radiologists, a general radiologist and two radiology residents. XGboost achieved the best performance with an area under the curve (AUC) of 93.7% (95% CI from 87.7 to 99.8%) among ML models. MSK radiologists achieved an AUC of 90.6% (95% CI from 86.7% to 94.5%) and 88.3% (95% CI from 84% to 92.7%), respectively, similar to residents. The general radiologist achieved an AUC of 84.5% (95% CI from 80% to 89%), significantly lower than of XGboost (<i>p</i> = 0.017). In conclusion, radiomics-based ML achieved a performance similar to MSK radiologists and significantly higher compared to general radiologists in differentiating between TOH and AVN. |
| first_indexed | 2024-03-10T07:45:54Z |
| format | Article |
| id | doaj.art-dafbd2c6e37b404684c355644fc23215 |
| institution | Directory Open Access Journal |
| issn | 2075-4418 |
| language | English |
| last_indexed | 2024-03-10T07:45:54Z |
| publishDate | 2021-09-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Diagnostics |
| spelling | doaj.art-dafbd2c6e37b404684c355644fc232152023-11-22T12:40:51ZengMDPI AGDiagnostics2075-44182021-09-01119168610.3390/diagnostics11091686Radiomics and Machine Learning Can Differentiate Transient Osteoporosis from Avascular Necrosis of the HipMichail E. Klontzas0Georgios C. Manikis1Katerina Nikiforaki2Evangelia E. Vassalou3Konstantinos Spanakis4Ioannis Stathis5George A. Kakkos6Nikolas Matthaiou7Aristeidis H. Zibis8Kostas Marias9Apostolos H. Karantanas10Department of Medical Imaging, University Hospital of Heraklion, 71110 Heraklion, Crete, GreeceComputational BioMedicine Laboratory, Institute of Computer Science, Foundation for Research and Technology (FORTH), 70013 Heraklion, Crete, GreeceComputational BioMedicine Laboratory, Institute of Computer Science, Foundation for Research and Technology (FORTH), 70013 Heraklion, Crete, GreeceDepartment of Medical Imaging, University Hospital of Heraklion, 71110 Heraklion, Crete, GreeceDepartment of Medical Imaging, University Hospital of Heraklion, 71110 Heraklion, Crete, GreeceDepartment of Medical Imaging, University Hospital of Heraklion, 71110 Heraklion, Crete, GreeceDepartment of Medical Imaging, University Hospital of Heraklion, 71110 Heraklion, Crete, GreeceDepartment of Medical Imaging, University Hospital of Heraklion, 71110 Heraklion, Crete, GreeceDepartment of Anatomy, Medical School, University of Thessaly, 41334 Larissa, GreeceComputational BioMedicine Laboratory, Institute of Computer Science, Foundation for Research and Technology (FORTH), 70013 Heraklion, Crete, GreeceDepartment of Medical Imaging, University Hospital of Heraklion, 71110 Heraklion, Crete, GreeceDifferentiation between transient osteoporosis (TOH) and avascular necrosis (AVN) of the hip is a longstanding challenge in musculoskeletal radiology. The purpose of this study was to utilize MRI-based radiomics and machine learning (ML) for accurate differentiation between the two entities. A total of 109 hips with TOH and 104 hips with AVN were retrospectively included. Femoral heads and necks with segmented radiomics features were extracted. Three ML classifiers (XGboost, CatBoost and SVM) using 38 relevant radiomics features were trained on 70% and validated on 30% of the dataset. ML performance was compared to two musculoskeletal radiologists, a general radiologist and two radiology residents. XGboost achieved the best performance with an area under the curve (AUC) of 93.7% (95% CI from 87.7 to 99.8%) among ML models. MSK radiologists achieved an AUC of 90.6% (95% CI from 86.7% to 94.5%) and 88.3% (95% CI from 84% to 92.7%), respectively, similar to residents. The general radiologist achieved an AUC of 84.5% (95% CI from 80% to 89%), significantly lower than of XGboost (<i>p</i> = 0.017). In conclusion, radiomics-based ML achieved a performance similar to MSK radiologists and significantly higher compared to general radiologists in differentiating between TOH and AVN.https://www.mdpi.com/2075-4418/11/9/1686hipavascular necrosis of boneosteoporosismachine learningartificial intelligencetransient osteoporosis |
| spellingShingle | Michail E. Klontzas Georgios C. Manikis Katerina Nikiforaki Evangelia E. Vassalou Konstantinos Spanakis Ioannis Stathis George A. Kakkos Nikolas Matthaiou Aristeidis H. Zibis Kostas Marias Apostolos H. Karantanas Radiomics and Machine Learning Can Differentiate Transient Osteoporosis from Avascular Necrosis of the Hip Diagnostics hip avascular necrosis of bone osteoporosis machine learning artificial intelligence transient osteoporosis |
| title | Radiomics and Machine Learning Can Differentiate Transient Osteoporosis from Avascular Necrosis of the Hip |
| title_full | Radiomics and Machine Learning Can Differentiate Transient Osteoporosis from Avascular Necrosis of the Hip |
| title_fullStr | Radiomics and Machine Learning Can Differentiate Transient Osteoporosis from Avascular Necrosis of the Hip |
| title_full_unstemmed | Radiomics and Machine Learning Can Differentiate Transient Osteoporosis from Avascular Necrosis of the Hip |
| title_short | Radiomics and Machine Learning Can Differentiate Transient Osteoporosis from Avascular Necrosis of the Hip |
| title_sort | radiomics and machine learning can differentiate transient osteoporosis from avascular necrosis of the hip |
| topic | hip avascular necrosis of bone osteoporosis machine learning artificial intelligence transient osteoporosis |
| url | https://www.mdpi.com/2075-4418/11/9/1686 |
| work_keys_str_mv | AT michaileklontzas radiomicsandmachinelearningcandifferentiatetransientosteoporosisfromavascularnecrosisofthehip AT georgioscmanikis radiomicsandmachinelearningcandifferentiatetransientosteoporosisfromavascularnecrosisofthehip AT katerinanikiforaki radiomicsandmachinelearningcandifferentiatetransientosteoporosisfromavascularnecrosisofthehip AT evangeliaevassalou radiomicsandmachinelearningcandifferentiatetransientosteoporosisfromavascularnecrosisofthehip AT konstantinosspanakis radiomicsandmachinelearningcandifferentiatetransientosteoporosisfromavascularnecrosisofthehip AT ioannisstathis radiomicsandmachinelearningcandifferentiatetransientosteoporosisfromavascularnecrosisofthehip AT georgeakakkos radiomicsandmachinelearningcandifferentiatetransientosteoporosisfromavascularnecrosisofthehip AT nikolasmatthaiou radiomicsandmachinelearningcandifferentiatetransientosteoporosisfromavascularnecrosisofthehip AT aristeidishzibis radiomicsandmachinelearningcandifferentiatetransientosteoporosisfromavascularnecrosisofthehip AT kostasmarias radiomicsandmachinelearningcandifferentiatetransientosteoporosisfromavascularnecrosisofthehip AT apostoloshkarantanas radiomicsandmachinelearningcandifferentiatetransientosteoporosisfromavascularnecrosisofthehip |