دورية أكاديمية
Application of machine learning model to predict osteoporosis based on abdominal computed tomography images of the psoas muscle: a retrospective study.
العنوان: | Application of machine learning model to predict osteoporosis based on abdominal computed tomography images of the psoas muscle: a retrospective study. |
---|---|
المؤلفون: | Huang CB; Department of Orthopaedic Surgery, The Second Affiliated Hospital and Yuying Childrens Hospital of Wenzhou Medical University, Wenzhou, 325000, China.; Key Laboratory of Orthopaedics of Zhejiang Province, Wenzhou, 325000, China., Hu JS; Department of Orthopaedic Surgery, The Second Affiliated Hospital and Yuying Childrens Hospital of Wenzhou Medical University, Wenzhou, 325000, China., Tan K; Department of Orthopaedic Surgery, The Second Affiliated Hospital and Yuying Childrens Hospital of Wenzhou Medical University, Wenzhou, 325000, China.; Key Laboratory of Orthopaedics of Zhejiang Province, Wenzhou, 325000, China., Zhang W; Department of Orthopaedic Surgery, The Second Affiliated Hospital and Yuying Childrens Hospital of Wenzhou Medical University, Wenzhou, 325000, China., Xu TH; Department of Orthopaedic Surgery, The Second Affiliated Hospital and Yuying Childrens Hospital of Wenzhou Medical University, Wenzhou, 325000, China.; Key Laboratory of Orthopaedics of Zhejiang Province, Wenzhou, 325000, China., Yang L; Department of Orthopaedic Surgery, The Second Affiliated Hospital and Yuying Childrens Hospital of Wenzhou Medical University, Wenzhou, 325000, China. yanglei@wmu.edu.cn.; Key Laboratory of Orthopaedics of Zhejiang Province, Wenzhou, 325000, China. yanglei@wmu.edu.cn. |
المصدر: | BMC geriatrics [BMC Geriatr] 2022 Oct 13; Vol. 22 (1), pp. 796. Date of Electronic Publication: 2022 Oct 13. |
نوع المنشور: | Journal Article |
اللغة: | English |
بيانات الدورية: | Publisher: BioMed Central Country of Publication: England NLM ID: 100968548 Publication Model: Electronic Cited Medium: Internet ISSN: 1471-2318 (Electronic) Linking ISSN: 14712318 NLM ISO Abbreviation: BMC Geriatr Subsets: MEDLINE |
أسماء مطبوعة: | Original Publication: London : BioMed Central, [2001]- |
مواضيع طبية MeSH: | Osteoporosis*/diagnostic imaging , Psoas Muscles*/diagnostic imaging, Bayes Theorem ; Humans ; Machine Learning ; Retrospective Studies ; Tomography, X-Ray Computed/methods |
مستخلص: | Background: With rapid economic development, the world's average life expectancy is increasing, leading to the increasing prevalence of osteoporosis worldwide. However, due to the complexity and high cost of dual-energy x-ray absorptiometry (DXA) examination, DXA has not been widely used to diagnose osteoporosis. In addition, studies have shown that the psoas index measured at the third lumbar spine (L3) level is closely related to bone mineral density (BMD) and has an excellent predictive effect on osteoporosis. Therefore, this study developed a variety of machine learning (ML) models based on psoas muscle tissue at the L3 level of unenhanced abdominal computed tomography (CT) to predict osteoporosis. Methods: Medical professionals collected the CT images and the clinical characteristics data of patients over 40 years old who underwent DXA and abdominal CT examination in the Second Affiliated Hospital of Wenzhou Medical University database from January 2017 to January 2021. Using 3D Slicer software based on horizontal CT images of the L3, the specialist delineated three layers of the region of interest (ROI) along the bilateral psoas muscle edges. The PyRadiomics package in Python was used to extract the features of ROI. Then Mann-Whitney U test and the least absolute shrinkage and selection operator (LASSO) algorithm were used to reduce the dimension of the extracted features. Finally, six machine learning models, Gaussian naïve Bayes (GNB), random forest (RF), logistic regression (LR), support vector machines (SVM), Gradient boosting machine (GBM), and Extreme gradient boosting (XGBoost), were applied to train and validate these features to predict osteoporosis. Results: A total of 172 participants met the inclusion and exclusion criteria for the study. 82 participants were enrolled in the osteoporosis group, and 90 were in the non-osteoporosis group. Moreover, the two groups had no significant differences in age, BMI, sex, smoking, drinking, hypertension, and diabetes. Besides, 826 radiomic features were obtained from unenhanced abdominal CT images of osteoporotic and non-osteoporotic patients. Five hundred fifty radiomic features were screened out of 826 by the Mann-Whitney U test. Finally, 16 significant radiomic features were obtained by the LASSO algorithm. These 16 radiomic features were incorporated into six traditional machine learning models (GBM, GNB, LR, RF, SVM, and XGB). All six machine learning models could predict osteoporosis well in the validation set, with the area under the receiver operating characteristic (AUROC) values greater than or equal to 0.8. GBM is more effective in predicting osteoporosis, whose AUROC was 0.86, sensitivity 0.70, specificity 0.92, and accuracy 0.81 in validation sets. Conclusion: We developed six machine learning models to predict osteoporosis based on psoas muscle images of abdominal CT, and the GBM model had the best predictive performance. GBM model can better help clinicians to diagnose osteoporosis and provide timely anti-osteoporosis treatment for patients. In the future, the research team will strive to include participants from multiple institutions to conduct external validation of the ML model of this study. (© 2022. The Author(s).) |
References: | World J Orthop. 2021 Jul 18;12(7):456-466. (PMID: 34354934) Transl Vis Sci Technol. 2020 Feb 27;9(2):14. (PMID: 32704420) Lancet Diabetes Endocrinol. 2017 Nov;5(11):908-923. (PMID: 28689768) Circulation. 2015 Nov 17;132(20):1920-30. (PMID: 26572668) Osteoporos Int. 2019 Jan;30(1):3-44. (PMID: 30324412) Br J Radiol. 2017 Feb;90(1070):20160665. (PMID: 27936886) Endocrinol Metab Clin North Am. 2014 Mar;43(1):233-43. (PMID: 24582100) BMC Cancer. 2022 Mar 11;22(1):258. (PMID: 35277130) Int J Environ Res Public Health. 2021 Aug 18;18(16):. (PMID: 34444459) J Cell Biochem. 2019 Sep;120(9):14262-14273. (PMID: 31106446) Nutrients. 2021 Dec 16;13(12):. (PMID: 34960050) Osteoporos Int. 2020 Jan;31(1):119-130. (PMID: 31654084) JAMA Netw Open. 2021 Aug 2;4(8):e2121106. (PMID: 34398202) Osteoporos Int. 2011 Feb;22(2):391-420. (PMID: 21184054) J Clin Invest. 2019 May 23;129(8):3214-3223. (PMID: 31120440) Subcell Biochem. 2019;91:453-476. (PMID: 30888662) Theranostics. 2019 Feb 12;9(5):1303-1322. (PMID: 30867832) Eur Radiol. 2019 Apr;29(4):2157-2165. (PMID: 30306329) J Appl Physiol (1985). 2000 Jul;89(1):81-8. (PMID: 10904038) Biomarkers. 2019 Mar;24(2):120-126. (PMID: 30442069) BMC Geriatr. 2021 Jan 12;21(1):43. (PMID: 33435869) Arch Osteoporos. 2020 Oct 23;15(1):169. (PMID: 33097976) Comput Biol Med. 2013 Nov;43(11):1910-9. (PMID: 24209936) Cancer Res. 2017 Nov 1;77(21):e104-e107. (PMID: 29092951) Front Med (Lausanne). 2021 Mar 09;8:635771. (PMID: 33768105) Nutrients. 2018 Aug 16;10(8):. (PMID: 30115856) Comput Struct Biotechnol J. 2019 Jul 16;17:1009-1015. (PMID: 31406557) Comput Methods Programs Biomed. 2019 Aug;177:9-15. (PMID: 31319965) BMC Geriatr. 2022 Apr 11;22(1):318. (PMID: 35410173) Eur Radiol. 2020 Jul;30(7):4107-4116. (PMID: 32072260) Maturitas. 2019 Apr;122:51-56. (PMID: 30797530) Nutrients. 2020 May 01;12(5):. (PMID: 32370051) Nutrition. 2022 Jan;93:111428. (PMID: 34474186) Diagnostics (Basel). 2020 Apr 28;10(5):. (PMID: 32353924) Curr Opin Rheumatol. 2016 Jul;28(4):433-41. (PMID: 27163858) J Digit Imaging. 2020 Oct;33(5):1209-1217. (PMID: 32583277) Med Clin North Am. 2020 Sep;104(5):873-884. (PMID: 32773051) |
فهرسة مساهمة: | Keywords: Computed tomography; Machine learning; Middle-aged and aged people; Osteoporosis; Psoas; Radiomics |
تواريخ الأحداث: | Date Created: 20221013 Date Completed: 20221017 Latest Revision: 20221019 |
رمز التحديث: | 20221213 |
مُعرف محوري في PubMed: | PMC9563158 |
DOI: | 10.1186/s12877-022-03502-9 |
PMID: | 36229793 |
قاعدة البيانات: | MEDLINE |
تدمد: | 1471-2318 |
---|---|
DOI: | 10.1186/s12877-022-03502-9 |