1
|
Wang P, Yin Q, Ding K, Zhong H, Jia Q, Xiao Z, Xiong H. Comparing machine learning models for osteoporosis prediction in Tibetan middle aged and elderly women. Sci Rep 2025; 15:10960. [PMID: 40164752 PMCID: PMC11958675 DOI: 10.1038/s41598-025-95707-2] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 09/05/2024] [Accepted: 03/24/2025] [Indexed: 04/02/2025] Open
Abstract
The aim of this study was to establish the optimal prediction model by comparing the prediction effect of 6 kinds of prediction models containing biochemical indexes on the risk of osteoporosis in middle-aged and elderly women in Tibet. This study adopted a multi-stage cluster random sampling cross-sectional survey method. From January 2022 to January 2024, we obtained biochemical and bone mineral density (BMD) data from high altitudes in Tibet. We built a predictive model of osteoporosis in three steps. First, we performed feature selection to identify factors associated with osteoporosis. Next, the eligible participants were randomly divided into a training set and a test set in a ratio of 8:2. Then, the prediction model of osteoporosis was established based on Random Forest, ANN, XGB, and SVM. Finally, we compared the performance of the prediction models using sensitivity, specificity, and the area under the receiver operating characteristic curve (AUC) to select the best prediction model. Correlation analysis was used to screen indicators with statistical differences from T-score. Finally, Age (P < 0.01), LDL-C (P < 0.05), UA (P < 0.01), AST (P < 0.05), CREA (P < 0.01), BMI (P < 0.01), ALT (P < 0.01) were associated with osteoporosis. In train set, the order of AUC from highest to lowest is Random Forest (1.000), XGB (0.887), SVM (0.868), regression (0.801), ANN (0.793) and OSTA (0.739). In test set, the order of AUC from highest to lowest is XGB (0.848), regression (0.801), Random Forest (0.772), SVM (0.755), OSTA (0.739), ANN (0.732). SVM and XGB algorithm models had better screening effect on osteoporosis than OSTA in middle-aged and elderly Tibetan residents in Tibet. Compared with Random Forest, ANN and SVM, the established XGB model had the best prediction ability and can be used to predict the risk of osteoporosis on biochemical indexes. The model needs to be further improved through large sample research.
Collapse
Affiliation(s)
- Peng Wang
- School of Medicine, Tibet University, Lhasa, 850000, China
| | - Qiang Yin
- School of Ecology and Environment, Tibet University, Lhasa, 850000, China
| | - Kangzhi Ding
- School of Medicine, Tibet University, Lhasa, 850000, China
| | - Huaichang Zhong
- Hospital Infection Management Department, The First People's Hospital of Shuangliu District, West China Airport Hospital of Sichuan University, Chengdu, 610200, China
| | - Qundi Jia
- School of Medicine, Tibet University, Lhasa, 850000, China
| | - Zhasang Xiao
- School of Medicine, Tibet University, Lhasa, 850000, China.
| | - Hai Xiong
- School of Medicine, Tibet University, Lhasa, 850000, China.
| |
Collapse
|
2
|
Avsar E, Celik S, Peynirci H, Yilmaz FT, Anataca G. The relationship between bone mineral density, vitamin D level, and sleep quality in postmenopausal women with osteoporosis: a relation-seeker type study. REVISTA DA ASSOCIACAO MEDICA BRASILEIRA (1992) 2024; 70:e20240440. [PMID: 39536248 PMCID: PMC11554326 DOI: 10.1590/1806-9282.20240440] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Subscribe] [Scholar Register] [Received: 07/23/2024] [Accepted: 07/24/2024] [Indexed: 11/16/2024]
Abstract
OBJECTIVE This study was conducted to determine the relationship between bone mineral density, vitamin D level, and sleep quality in female patients with osteoporosis. METHODS This descriptive correlational study included a total of 318 women diagnosed with osteoporosis. The data were collected using a patient identification form, including items for the plasma vitamin D level and bone mineral density score obtained by the dual-energy X-ray absorptiometry method, and the Pittsburgh Sleep Quality Index. RESULTS The mean age of the women was 56.49±5.68 years, and their femoral neck T mean score, an indicator of bone mineral density, was -2.94±0.31. Only 6.3% of the women had adequate vitamin D levels. In addition, according to their Pittsburgh Sleep Quality Index scores, 85.8% of the women had poor sleep quality. In the study, no significant difference was found between the women's bone mineral densities and vitamin D levels according to sleep quality (p>0.05). However, there was a weak negative correlation between the duration of osteoporosis, body mass index, and sleep quality (p<0.05). CONCLUSION There was no association between the vitamin D level, bone mineral density, and sleep quality, but the duration of osteoporosis was negatively associated with sleep quality. Accordingly, it may be recommended to provide education and counseling to postmenopausal women diagnosed with osteoporosis on issues such as sunbathing, vitamin D and calcium preparation intake, weight control, and non-pharmacological treatment approaches by making necessary individual-specific plans to improve sleep quality.
Collapse
Affiliation(s)
- Emir Avsar
- Yeditepe University, Faculty of Health Sciences, Department of Nursing – İstanbul, Turkey
| | - Selda Celik
- University of Health Sciences, Hamidiye Faculty of Nursing – İstanbul, Turkey
| | - Hande Peynirci
- University of Health Sciences, Kanuni Sultan Suleyman Training and Research Hospital, Department of Endocrinology and Metabolic Diseases – İstanbul, Turkey
| | - Feride Taskin Yilmaz
- Sakarya University of Applied Sciences, Faculty of Health Sciences, Department of Nursing – Sakarya, Turkey
| | - Gulden Anataca
- University of Health Sciences, Kanuni Sultan Suleyman Training and Research Hospital, Department of Diabetes Nursing – İstanbul, Turkey
| |
Collapse
|
3
|
Chen KH, Su CM, Liu WJ, Tzeng HE, Lee CL, Tsai CH. The joint effects of physical activity and sleep duration on risk of osteoporosis in Taiwanese adult population: The Taiwan Biobank Study. Osteoporos Int 2024; 35:523-531. [PMID: 37947843 PMCID: PMC10867060 DOI: 10.1007/s00198-023-06947-9] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Received: 10/27/2022] [Accepted: 10/09/2023] [Indexed: 11/12/2023]
Abstract
Most studies investigating the association between physical activity and osteoporosis prevention only focused on specific types of physical activity. This study's evidence regarding the combined effects or interaction of sleep duration and physical activity. The findings emphasize the role of sleep duration and physical activity in association with osteoporosis. PURPOSE The associations between physical activity, sleep duration, and prevalent osteoporosis in Taiwanese adults were studied in this cross-sectional study. METHODS The Taiwan Biobank enrolled a community-based cohort of ~ 120,000 volunteers (as of April 30, 2020) between 30 and 76 years of age with no history of cancer. Amongst, bone mineral density (BMD) measures by dual-energy X-ray absorptiometry (DXA) were available in 22,402 participants. After excluding individuals who had no complete data of BMI (n = 23), MET score (n = 207), T-score (n = 8,826), and sleep duration (n = 16), 13,330 subjects were included as the primary cohort. Univariate and multivariable regression analyses were performed to determine the associations between the presence of osteoporosis, physical activity level, sleep duration, and other variables. RESULTS The results showed that after adjustment, subjects with physical activity < 20 METs/week and ≥ 20 METs/week (aOR = 1.017 and 0.767, respectively) were associated with risk of osteoporosis than those with zero MET. The odds of osteoporosis were not significantly lower in subjects who slept for ≥ 8 h/day (aOR = 0.934,p=0.266). In addition, compared to short sleepers with no physical activity, adults with increased physical activity ≥ 20 METs/week and sleep ≥ 8 h/day had a significantly lowest likelihood of osteoporosis (aOR = 0.702). Those with medium physical activity (< 20 METs/week) plus average sleep duration (6.5-8 h/day) did not have significant higher odds of osteoporosis (aOR = 1.129,p=0.151). CONCLUSION The findings emphasize the joint role of sleep duration and physical activity in association with osteoporosis. Adults with high physical activity plus high sleep hours have the highest BMD and lowest risk of osteoporosis.
Collapse
Affiliation(s)
- Kun-Hui Chen
- Department of Orthopedic Surgery, Taichung Veterans General Hospital, Taichung, 407219, Taiwan
- Department of Post-Baccalaureate Medicine, College of Medicine, National Chung Hsing University, Taichung, 402204, Taiwan
- Department of Computer Science & Information Engineering, College of Computing and Informatics, Providence University, Taichung City, 43301, Taiwan
| | - Chen-Ming Su
- Department of Sports Medicine, China Medical University, Taichung, 406040, Taiwan
| | - Wei-Ju Liu
- Department of Medical Research, Taichung Veterans General Hospital, Taichung, 407219, Taiwan
- Intelligent Data Mining Laboratory, Department of Medical Research, Taichung Veterans General Hospital, Taichung, 407219, Taiwan
| | - Huey-En Tzeng
- Department of Post-Baccalaureate Medicine, College of Medicine, National Chung Hsing University, Taichung, 402204, Taiwan
- Department of Medical Research, Taichung Veterans General Hospital, Taichung, 407219, Taiwan
- School of Medicine, Taipei Medical University, Taipei, 11031, Taiwan
- Division of Transfusion Medicine /Hematology - Oncology, Taichung Veterans General Hospital, Taichung, 40705, Taiwan
| | - Chia-Lin Lee
- Department of Post-Baccalaureate Medicine, College of Medicine, National Chung Hsing University, Taichung, 402204, Taiwan.
- Department of Medical Research, Taichung Veterans General Hospital, Taichung, 407219, Taiwan.
- Intelligent Data Mining Laboratory, Department of Medical Research, Taichung Veterans General Hospital, Taichung, 407219, Taiwan.
- Department of Public Health, College of Public Health, China Medical University, Taichung, 406040, Taiwan.
- Department of Medicine, School of Medicine, National Yang Ming Chiao Tung University, Taipei, 112304, Taiwan.
| | - Chun-Hao Tsai
- Department of Sports Medicine, China Medical University, Taichung, 406040, Taiwan.
- Department of Orthopedics, China Medical University Hospital, Taichung, 406040, Taiwan.
- College of Biomedical Engineering, China Medical University, Taichung, 406040, Taiwan.
| |
Collapse
|
4
|
Zhao H, Zhu L, Fan L, Yang J, Hou J, Zhang G, Wang C, Pan J. Association of nocturnal sleep duration and sleep midpoint with osteoporosis risk in rural adults: a large-scale cross-sectional study. Sleep Breath 2023; 27:2191-2198. [PMID: 37009967 DOI: 10.1007/s11325-023-02818-z] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 10/29/2022] [Revised: 03/11/2023] [Accepted: 03/20/2023] [Indexed: 04/04/2023]
Abstract
BACKGROUND Evidence has suggested the linkage between sleep habits and several metabolic diseases, but the association of sleep factors with bone health remains unclear, especially in regions with low economic levels. Thus, this study aimed to investigate the relationship of nocturnal sleep duration and sleep midpoint with the osteoporosis risk in a rural population. METHOD Eligible subjects were derived from the Henan Rural Cohort Study. The Pittsburgh Sleep Quality Index was applied to collect sleep information including sleep initiating time and wake-up time. The bone mineral density of the calcaneus was measured by the ultrasonic bone density apparatus. Multivariable logistic regression models and restricted cubic splines were utilized to evaluate the odds ratio (OR) and 95% confidence intervals (95% CI). RESULTS For 8033 participants, 1636 subjects suffered from osteoporosis. Compared with the reference group (7 ~ h group), the ORs and 95% CI of osteoporosis associated with duration of nocturnal sleep were 1.32 (1.10, 1.56), 1.59 (1.25, 2.01), and 1.82 (1.25, 2.65) in the 8 ~ h, 9 ~ h, and ≥ 10 h group, respectively. Additionally, the adjusted ORs and 95% CI were 1.20 (1.01, 1.44) in the early sleep midpoint group and 1.09 (0.92, 1.29) in the intermediate sleep midpoint, compared with the late. Furthermore, there was a joint effect of long duration of nocturnal sleep and the early sleep midpoint on osteoporosis. CONCLUSION Long duration of nocturnal sleep and early sleep midpoint were independently and jointly associated with higher risk of osteoporosis in rural areas. TRIAL REGISTRATION The Henan Rural Cohort Study has been registered at Chinese Clinical Trial Register (Registration number: ChiCTR-OOC-15006699). Date of registration: 06 July 2015. http://www.chictr.org.cn/showproj.aspx?proj=11375.
Collapse
Affiliation(s)
- Hongfei Zhao
- Department of Epidemiology and Biostatistics, College of Public Health, Zhengzhou University, 100 Kexue Avenue, Zhengzhou, 450001, Henan, People's Republic of China
| | - Linghui Zhu
- Department of Epidemiology and Biostatistics, College of Public Health, Zhengzhou University, 100 Kexue Avenue, Zhengzhou, 450001, Henan, People's Republic of China
| | - Lin Fan
- Department of Orthopaedic Surgery, The Second Affiliated Hospital and Yuying Children's Hospital of Wenzhou Medical University, Wenzhou, Zhejiang, People's Republic of China
| | - Jing Yang
- Department of Epidemiology and Biostatistics, College of Public Health, Zhengzhou University, 100 Kexue Avenue, Zhengzhou, 450001, Henan, People's Republic of China
| | - Jian Hou
- Department of Epidemiology and Biostatistics, College of Public Health, Zhengzhou University, 100 Kexue Avenue, Zhengzhou, 450001, Henan, People's Republic of China
| | - Gongyuan Zhang
- Department of Epidemiology and Biostatistics, College of Public Health, Zhengzhou University, 100 Kexue Avenue, Zhengzhou, 450001, Henan, People's Republic of China
| | - Chongjian Wang
- Department of Epidemiology and Biostatistics, College of Public Health, Zhengzhou University, 100 Kexue Avenue, Zhengzhou, 450001, Henan, People's Republic of China.
| | - Jun Pan
- Department of Orthopaedic Surgery, The Second Affiliated Hospital and Yuying Children's Hospital of Wenzhou Medical University, Wenzhou, Zhejiang, People's Republic of China.
| |
Collapse
|
5
|
Chung PC, Chan TC. Environmental and personal factors for osteoporosis or osteopenia from a large health check-up database: a retrospective cohort study in Taiwan. BMC Public Health 2022; 22:1531. [PMID: 35948894 PMCID: PMC9367101 DOI: 10.1186/s12889-022-13938-8] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 04/22/2022] [Accepted: 08/02/2022] [Indexed: 11/20/2022] Open
Abstract
Background Osteoporosis is an important public health issue in aging societies because of its associated morbidity, mortality, and decreased quality of life. The study aims to identify the association of low bone mineral density, including osteoporosis and osteopenia, with environmental and personal factors. Methods The data of participants aged ≥ 20 years with multiple visits were obtained from a health check-up database in Taiwan from 2008 to 2016. Multivariable logistic regressions were performed to identify the selected factors associated with low bone mineral density for multiple visit data. Results A total of 194,910 participants with 359,943 visits were included in this study. The prevalence of low bone mineral density (BMD) in the study population was 10.6% (n = 20.615). Older women, ever and current smokers (odds ratio (OR) = 1.04 [95% confidence interval (CI) = 1.01, 1.08]), or participants who were underweight (OR = 1.72 [1.64, 1.81]), consumed a vegetarian diet (OR = 1.32 [1.25, 1.39]), or had higher triglyceride levels (OR = 1.04 [1.01, 1.06]) were significantly associated with a higher risk of low BMD. Participants who had higher educational years (OR = 0.43 [0.41, 0.46]), higher physical activity (OR = 0.93 [0.89, 0.97]), appropriate sleep duration and better quality (OR = 0.98 [0.97, 0.99]), dairy intake (≥ 1 slice of yogurt or cheese/week, OR = 0.97 [0.95, 0.99]), higher uric acid (OR = 0.93 [0.91, 0.95]), higher walkability (OR = 0.997 [0.995,0.999]), and higher solar radiation exposure (OR = 0.997 [0.97,0.99]) were significantly associated with a lower risk of low BMD. Conclusion Interventions in different directions, such as having better health behaviors, increasing sun exposure, and residing in a highly walkable environment, are beneficial for reducing the risk of low BMD.
Collapse
Affiliation(s)
- Ping-Chen Chung
- Department of Dentistry, Puzi Hospital, Ministry of Health and Welfare, Chiayi, Taiwan
| | - Ta-Chien Chan
- Research Center for Humanities and Social Sciences, Academia Sinica, 128 Academia Road, Section 2, Taipei, Taiwan. .,Institute of Public Health, School of Medicine, National Yang Ming Chiao Tung University, Taipei, Taiwan.
| |
Collapse
|