The Potential of a CT-Based Machine Learning Radiomics Analysis to Differentiate Brucella and Pyogenic Spondylitis
Parhat Yasin,1 Muradil Mardan,2 Dilxat Abliz,3 Tao Xu,1 Nuerbiyan Keyoumu,4 Abasi Aimaiti,4 Xiaoyu Cai,1 Weibin Sheng,1 Mardan Mamat1 1Department of Spine Surgery, The First Affiliated Hospital of Xinjiang Medical University, Urumqi, Xinjiang, 830054, People’s Republic of China; 2School of Medicine,...
| Main Authors: | , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Dove Medical Press
2023-11-01
|
| Series: | Journal of Inflammation Research |
| Subjects: | |
| Online Access: | https://www.dovepress.com/the-potential-of-a-ct-based-machine-learning-radiomics-analysis-to-dif-peer-reviewed-fulltext-article-JIR |
| _version_ | 1797486298135527424 |
|---|---|
| author | Yasin P Mardan M Abliz D Xu T Keyoumu N Aimaiti A Cai X Sheng W Mamat M |
| author_facet | Yasin P Mardan M Abliz D Xu T Keyoumu N Aimaiti A Cai X Sheng W Mamat M |
| author_sort | Yasin P |
| collection | DOAJ |
| description | Parhat Yasin,1 Muradil Mardan,2 Dilxat Abliz,3 Tao Xu,1 Nuerbiyan Keyoumu,4 Abasi Aimaiti,4 Xiaoyu Cai,1 Weibin Sheng,1 Mardan Mamat1 1Department of Spine Surgery, The First Affiliated Hospital of Xinjiang Medical University, Urumqi, Xinjiang, 830054, People’s Republic of China; 2School of Medicine, Tongji University, Shanghai, 200092, People’s Republic of China; 3Department of Orthopedic, The Eighth Affiliated Hospital of Xinjiang Medical University, Urumqi, Xinjiang, 830054, People’s Republic of China; 4Department of Anesthesiology, The First Affiliated Hospital of Xinjiang Medical University, Urumqi, Xinjiang, 830054, People’s Republic of ChinaCorrespondence: Mardan Mamat, Tel +86 0991-4365316, Email mardanmmtmx@163.comBackground: Pyogenic spondylitis (PS) and Brucella spondylitis (BS) are common spinal infections with similar manifestations, making their differentiation challenging. This study aimed to explore the potential of CT-based radiomics features combined with machine learning algorithms to differentiate PS from BS.Methods: This retrospective study involved the collection of clinical and radiological information from 138 patients diagnosed with either PS or BS in our hospital between January 2017 and December 2022, based on histopathology examination and/or germ isolations. The region of interest (ROI) was defined by two radiologists using a 3D Slicer open-source platform, utilizing blind analysis of sagittal CT images against histopathological examination results. PyRadiomics, a Python package, was utilized to extract ROI features. Several methods were performed to reduce the dimensionality of the extracted features. Machine learning algorithms were trained and evaluated using techniques like the area under the receiver operating characteristic curve (AUC; confusion matrix-related metrics, calibration plot, and decision curve analysis to assess their ability to differentiate PS from BS. Additionally, permutation feature importance (PFI; local interpretable model-agnostic explanations (LIME; and Shapley additive explanation (SHAP) techniques were utilized to gain insights into the interpretabilities of the models that are otherwise considered opaque black-boxes.Results: A total of 15 radiomics features were screened during the analysis. The AUC value and Brier score of best the model were 0.88 and 0.13, respectively. The calibration plot and decision curve analysis displayed higher clinical efficiency in the differential diagnosis. According to the interpretation results, the most impactful features on the model output were wavelet LHL small dependence low gray-level emphasis (GLDN).Conclusion: The CT-based radiomics models that we developed have proven to be useful in reliably differentiating between PS and BS at an early stage and can provide a reliable explanation for the classification results.Keywords: Brucella spondylitis, Pyogenic spondylitis, machine learning, radiomics, model interpretation |
| first_indexed | 2024-03-09T23:32:10Z |
| format | Article |
| id | doaj.art-6f73e5a574cc42ce8c10e6f153d85cf2 |
| institution | Directory Open Access Journal |
| issn | 1178-7031 |
| language | English |
| last_indexed | 2024-03-09T23:32:10Z |
| publishDate | 2023-11-01 |
| publisher | Dove Medical Press |
| record_format | Article |
| series | Journal of Inflammation Research |
| spelling | doaj.art-6f73e5a574cc42ce8c10e6f153d85cf22023-11-23T17:07:56ZengDove Medical PressJournal of Inflammation Research1178-70312023-11-01Volume 165585560088472The Potential of a CT-Based Machine Learning Radiomics Analysis to Differentiate Brucella and Pyogenic SpondylitisYasin PMardan MAbliz DXu TKeyoumu NAimaiti ACai XSheng WMamat MParhat Yasin,1 Muradil Mardan,2 Dilxat Abliz,3 Tao Xu,1 Nuerbiyan Keyoumu,4 Abasi Aimaiti,4 Xiaoyu Cai,1 Weibin Sheng,1 Mardan Mamat1 1Department of Spine Surgery, The First Affiliated Hospital of Xinjiang Medical University, Urumqi, Xinjiang, 830054, People’s Republic of China; 2School of Medicine, Tongji University, Shanghai, 200092, People’s Republic of China; 3Department of Orthopedic, The Eighth Affiliated Hospital of Xinjiang Medical University, Urumqi, Xinjiang, 830054, People’s Republic of China; 4Department of Anesthesiology, The First Affiliated Hospital of Xinjiang Medical University, Urumqi, Xinjiang, 830054, People’s Republic of ChinaCorrespondence: Mardan Mamat, Tel +86 0991-4365316, Email mardanmmtmx@163.comBackground: Pyogenic spondylitis (PS) and Brucella spondylitis (BS) are common spinal infections with similar manifestations, making their differentiation challenging. This study aimed to explore the potential of CT-based radiomics features combined with machine learning algorithms to differentiate PS from BS.Methods: This retrospective study involved the collection of clinical and radiological information from 138 patients diagnosed with either PS or BS in our hospital between January 2017 and December 2022, based on histopathology examination and/or germ isolations. The region of interest (ROI) was defined by two radiologists using a 3D Slicer open-source platform, utilizing blind analysis of sagittal CT images against histopathological examination results. PyRadiomics, a Python package, was utilized to extract ROI features. Several methods were performed to reduce the dimensionality of the extracted features. Machine learning algorithms were trained and evaluated using techniques like the area under the receiver operating characteristic curve (AUC; confusion matrix-related metrics, calibration plot, and decision curve analysis to assess their ability to differentiate PS from BS. Additionally, permutation feature importance (PFI; local interpretable model-agnostic explanations (LIME; and Shapley additive explanation (SHAP) techniques were utilized to gain insights into the interpretabilities of the models that are otherwise considered opaque black-boxes.Results: A total of 15 radiomics features were screened during the analysis. The AUC value and Brier score of best the model were 0.88 and 0.13, respectively. The calibration plot and decision curve analysis displayed higher clinical efficiency in the differential diagnosis. According to the interpretation results, the most impactful features on the model output were wavelet LHL small dependence low gray-level emphasis (GLDN).Conclusion: The CT-based radiomics models that we developed have proven to be useful in reliably differentiating between PS and BS at an early stage and can provide a reliable explanation for the classification results.Keywords: Brucella spondylitis, Pyogenic spondylitis, machine learning, radiomics, model interpretationhttps://www.dovepress.com/the-potential-of-a-ct-based-machine-learning-radiomics-analysis-to-dif-peer-reviewed-fulltext-article-JIRbrucella spondylitispyogenic spondylitismachine learningradiomicsmodel interpretation |
| spellingShingle | Yasin P Mardan M Abliz D Xu T Keyoumu N Aimaiti A Cai X Sheng W Mamat M The Potential of a CT-Based Machine Learning Radiomics Analysis to Differentiate Brucella and Pyogenic Spondylitis Journal of Inflammation Research brucella spondylitis pyogenic spondylitis machine learning radiomics model interpretation |
| title | The Potential of a CT-Based Machine Learning Radiomics Analysis to Differentiate Brucella and Pyogenic Spondylitis |
| title_full | The Potential of a CT-Based Machine Learning Radiomics Analysis to Differentiate Brucella and Pyogenic Spondylitis |
| title_fullStr | The Potential of a CT-Based Machine Learning Radiomics Analysis to Differentiate Brucella and Pyogenic Spondylitis |
| title_full_unstemmed | The Potential of a CT-Based Machine Learning Radiomics Analysis to Differentiate Brucella and Pyogenic Spondylitis |
| title_short | The Potential of a CT-Based Machine Learning Radiomics Analysis to Differentiate Brucella and Pyogenic Spondylitis |
| title_sort | potential of a ct based machine learning radiomics analysis to differentiate brucella and pyogenic spondylitis |
| topic | brucella spondylitis pyogenic spondylitis machine learning radiomics model interpretation |
| url | https://www.dovepress.com/the-potential-of-a-ct-based-machine-learning-radiomics-analysis-to-dif-peer-reviewed-fulltext-article-JIR |
| work_keys_str_mv | AT yasinp thepotentialofactbasedmachinelearningradiomicsanalysistodifferentiatebrucellaandpyogenicspondylitis AT mardanm thepotentialofactbasedmachinelearningradiomicsanalysistodifferentiatebrucellaandpyogenicspondylitis AT ablizd thepotentialofactbasedmachinelearningradiomicsanalysistodifferentiatebrucellaandpyogenicspondylitis AT xut thepotentialofactbasedmachinelearningradiomicsanalysistodifferentiatebrucellaandpyogenicspondylitis AT keyoumun thepotentialofactbasedmachinelearningradiomicsanalysistodifferentiatebrucellaandpyogenicspondylitis AT aimaitia thepotentialofactbasedmachinelearningradiomicsanalysistodifferentiatebrucellaandpyogenicspondylitis AT caix thepotentialofactbasedmachinelearningradiomicsanalysistodifferentiatebrucellaandpyogenicspondylitis AT shengw thepotentialofactbasedmachinelearningradiomicsanalysistodifferentiatebrucellaandpyogenicspondylitis AT mamatm thepotentialofactbasedmachinelearningradiomicsanalysistodifferentiatebrucellaandpyogenicspondylitis AT yasinp potentialofactbasedmachinelearningradiomicsanalysistodifferentiatebrucellaandpyogenicspondylitis AT mardanm potentialofactbasedmachinelearningradiomicsanalysistodifferentiatebrucellaandpyogenicspondylitis AT ablizd potentialofactbasedmachinelearningradiomicsanalysistodifferentiatebrucellaandpyogenicspondylitis AT xut potentialofactbasedmachinelearningradiomicsanalysistodifferentiatebrucellaandpyogenicspondylitis AT keyoumun potentialofactbasedmachinelearningradiomicsanalysistodifferentiatebrucellaandpyogenicspondylitis AT aimaitia potentialofactbasedmachinelearningradiomicsanalysistodifferentiatebrucellaandpyogenicspondylitis AT caix potentialofactbasedmachinelearningradiomicsanalysistodifferentiatebrucellaandpyogenicspondylitis AT shengw potentialofactbasedmachinelearningradiomicsanalysistodifferentiatebrucellaandpyogenicspondylitis AT mamatm potentialofactbasedmachinelearningradiomicsanalysistodifferentiatebrucellaandpyogenicspondylitis |