A Machine Learning-Based Preclinical Osteoporosis Screening Tool (POST): Model Development and Validation Study.

Background: Identifying persons with a high risk of developing osteoporosis and preventing the occurrence of the first fracture is a health care priority. Most existing osteoporosis screening tools have high sensitivity but relatively low specificity. Objective: We aimed to develop an easily accessible and high-performance preclinical risk screening tool for osteoporosis using a machine learning-based method among the Hong Kong Chinese population. Methods: Participants aged 45 years or older were enrolled from 6 clinics in the 3 major districts of Hong Kong. The potential risk factors for osteoporosis were collected through a validated, self-administered questionnaire and then filtered using a machine learning-based method. Bone mineral density was measured with dual-energy x-ray absorptiometry at the clinics; osteoporosis was defined as a t score of -2.5 or lower. We constructed machine learning models, including gradient boosting machines, support vector machines, and naive Bayes, as well as the commonly used logistic regression models, for the prediction of osteoporosis. The best-performing model was chosen as the final tool, named the Preclinical Osteoporosis Screening Tool (POST). Model performance was evaluated by the area under the receiver operating characteristic curve (AUC) and other metrics. Results: Among the 800 participants enrolled in this study, the prevalence of osteoporosis was 10.6% (n=85). The machine learning-based Boruta algorithm identified 15 significantly important predictors from the 113 potential risk factors. Seven variables were further selected based on their accessibility and convenience for daily self-assessment and health care practice, including age, gender, education level, decreased body height, BMI, number of teeth lost, and the intake of vitamin D supplements, to construct the POST. The AUC of the POST was 0.86 and the sensitivity, specificity, and accuracy were all 0.83. The positive predictive value, negative predictive value, and F1-score were 0.41, 0.98, and 0.56, respectively. Conclusions: The machine learning-based POST was conveniently accessible and exhibited accurate discriminative capabilities for the prediction of osteoporosis; it might be useful to guide population-based preclinical screening of osteoporosis and clinical decision-making.

Efficacy and feasibility of a 12-week Tai Chi training for the prophylaxis of episodic migraine in Hong Kong Chinese women: A randomized controlled trial.

Background: Tai Chi has been broadly applied as alternative treatment for many neurological and psychological disorders. Whereas no study using Tai Chi as prophylactic treatment for migraine. The purpose of this study was to preliminarily examine the efficacy and feasibility of a 12-week Tai Chi training on migraine attack prevention in a sample of Chinese women. Methods: A two-arm randomized controlled trial was designed. Women aged 18 to 65 years and diagnosed with episodic migraine were randomized to either Tai Chi group (TC group) or the waiting list control group. A modified 33-short form Yang-style Tai Chi training with 1 h per day, 5 days per week for 12 weeks was implemented in the TC group, with a 12-week follow up period. The control group received a "delayed" Tai Chi training at the end of the trial. The primary outcome was the differences in attack frequency between 4 weeks before baseline and at the 9-12 weeks after randomization. The intensity and duration of headache were also measured. The feasibility was evaluated by the maintenance of Tai Chi practice and satisfactory level of the participants toward training. Results: Eighty-two women were randomized, finally 40 in TC group and 33 in control group were involved in the analysis. On average, women in TC group had 3.0 times (95% CI: -4.0 to -2.0, P < 0.01) and 3.6 days (95% CI: -4.7 to -2.5, P < 0.01) reduction of migraine attack per month. Compared with the control group, the differences were statistically significant (-3.7 attacks/month, 95% CI: -5.4 to -1.9; and -3.0 migraine days/month, 95% CI: -4.5 to -1.5; both P < 0.001). The intensity and duration of headache had 0.6 (95% CI: -1.2 to -0.0, P < 0.05) units and 1.2 (IQR: -5.0 to 1.1, P < 0.05) hours reduction in TC group, respectively. Most of the participants (69.2%-97.4%) were satisfied with the training. At the end of 24 weeks, on average, the participants maintained 1.5 times of practice per week and 20 min for each practice. Conclusion: The 12-week Tai Chi training significantly decreased the frequency of migraine attack. It was acceptable and practicable among female migraineurs. Clinical trial registration: www.ClinicalTrials.gov, identifier: NCT03015753.

Dietary intake of monounsaturated and polyunsaturated fatty acids is related to the reduced risk of esophageal squamous cell carcinoma.

BACKGROUND: The relationship of consumption of dietary fat and fatty acids with esophageal squamous cell carcinoma (ESCC) risk remains unclear. This study aimed to explore the relationship of dietary fat and fatty acids intake with ESCC risk. METHODS: This case-control study included 879 incident cases and 892 community-based controls recruited from Southwest China. A food frequency questionnaire was adopted to collect information about dietary information, and intake of fat, saturated fatty acid (SFA), monounsaturated fatty acid (MUFA), polyunsaturated fatty acid (PUFA), and total fatty acid (TFA) was calculated. Odds ratios (ORs) with 95% confidence intervals (95% CIs) were estimated using the logistic regression model. RESULTS: When comparing the highest with lowest intake quintiles, MUFA (OR: 0.33, 95% CI: 0.21-0.51), PUFA (OR: 0.32, 95% CI: 0.20-0.51), and TFA (OR: 0.44, 95% CI: 0.28-0.70) were related to a reduced risk of ESCC after adjusting for confounders; for non-drinkers rather than drinkers, the intake of SFA was significantly related to a 61% (OR: 0.39, 95% CI: 0.19-0.81) reduced risk of ESCC when comparing the highest with the lowest intake quintiles. Dietary fat was not related to the risk of ESCC. CONCLUSIONS: This study suggested that the more intake of MUFA and PUFA, the lower risk of ESCC, whereas the protective effect of TFA was only observed among non-drinkers. Strategic nutritional programs should consider food rich in unsaturated fatty acids to mitigate the occurrence of ESCC.