Background: In this ongoing multicenter prospective cohort study on active surveillance (AS) in low-risk papillary thyroid microcarcinoma (PTMC), we aimed to compare the quality of life (QoL) of participants based on their choice of treatment, that is, AS or immediate surgery (OP). Methods: QoL of 203 participants who chose AS and 192 participants who underwent OP was evaluated using a thyroid-specific QoL questionnaire at diagnosis and during follow-up (median 8 months). Results: The mean ages of the participants in the AS and OP groups were 47.3 ± 11.7 and 45.6 ± 10.5 years ( p = 0.138), respectively, and the mean tumor sizes were 5.7 ± 1.6 and 6.5 ± 2.1 mm ( p = 0.065), respectively. At baseline, significantly better psychological health (7.1 ± 1.3 vs. 6.8 ± 1.6, p = 0.023) and overall health (6.8 ± 1.2 vs. 6.5 ± 1.3, p = 0.018) were observed in the AS group than in the OP group. During follow-up, significantly better physical (7.9 ± 1.1 vs. 7.4 ± 1.2, p < 0.001), psychological (7.4 ± 1.3 vs. 6.9 ± 1.6, p = 0.004), and overall health (6.9 ± 1.0 vs. 6.5 ± 1.1, p = 0.002) were observed in the AS group than in the OP group, whereas spiritual health was comparable between the two groups. Compared with the AS group, the OP group experienced more fatigue, changes in voice and appearance, less satisfaction, and low fear of recurrence. The self-assessed financial burden was similar at baseline and follow-up in both groups. Conclusion: The QoL of PTMC patients is different according to the type of treatment. Better psychological health at baseline and physical and psychological health at follow-up were observed in the AS group than in the OP group. However, studies with longer follow-up periods are needed.

The prediction of fracture risk in osteoporotic patients has been a topic of interest for decades, and models have been developed for the accurate prediction of fracture, including the fracture risk assessment tool (FRAX). As machine‐learning methodologies have recently emerged as a potential model for medical prediction tools, we aimed to develop a novel fracture prediction model using machine‐learning methods in a prospective community‐based cohort. In this study, 2227 participants (1257 females) with a baseline bone mineral density (BMD) and trabecular bone score were enrolled from the Ansung cohort. The primary endpoint was the fragility fractures reported by patients or confirmed by X‐rays. We used 3 different models: CatBoost, support vector machine (SVM), and logistic regression. During a mean 7.5‐year follow‐up (range, 2.5 to 10 years), fragility fractures occurred in 537 (25.6%) of participants. In predicting total fragility fractures, the area under the curve (AUC) values of the CatBoost, SVM, and logistic regression models were 0.688, 0.500, and 0.614, respectively. The AUC value of CatBoost was significantly better than that of FRAX (0.663; p < 0.001), whereas the the SVM and logistic regression models were not. Compared with the conventional models such as SVM and logistic regression, the CatBoost model had the best performance in predicting total fragility fractures (p < 0.001). According to feature importance in the CatBoost model, the top predicting factors (listed in order) were total hip, lumbar spine, and femur neck BMD, subjective arthralgia score, serum creatinine, and homocysteine. The latter three factors were listed higher than conventional predictors such as age or previous fracture history. In summary, we hereby report the development of a prediction model for fragility fractures using a machine‐learning method, CatBoost, which outperforms the FRAX model as well as two conventional machine‐learning models. The model was also able to propose novel high‐ranking predictors. © 2020 The Authors. JBMR Plus published by Wiley Periodicals, Inc. on behalf of American Society for Bone and Mineral Research.

Background: Although recent studies comparing various dosages and intervals of vitamin D supplementation have been published, it is yet to be elucidated whether there is an appropriate dose or interval to provide benefit regarding fracture risk. We aimed to assess the published evidence available to date regarding the putative beneficial effects of vitamin D supplements on fractures and falls according to various dosages and intervals.Methods: We performed a meta-analysis of randomized controlled studies reporting associations between vitamin D supplementation and the risks of fractures and falls in PubMed, EMBASE, and Cochrane library. Studies with supplements of ergocalciferol or calcitriol, those with a number of event ≤10, or those with a follow-up duration of less than 6 months were also excluded.Results: Thirty-two studies were included in the final analysis. Vitamin D supplementation with daily dose of 800 to 1,000 mg was associated with lower risks of osteoporotic fracture and fall (pooled relative risk [RR], 0.87; 95% confidence interval [CI], 0.78 to 0.97 and RR, 0.91; 95% CI, 0.85 to 0.98), while studies with 1,000 mg/day did not. Also, among intervals, daily administration of vitamin D was associated with the reduced risk of falls, while intermittent dose was not. Also, patients with vitamin D deficiency showed a significant risk reduction of falls after vitamin D supplementation.Conclusion: Daily vitamin D dose of 800 to 1,000 IU was the most probable way to reduce the fracture and fall risk. Further studies designed with various regimens and targeted vitamin D levels are required to elucidate the benefits of vitamin D supplements.

Objective: Recently, the European Society of Endocrinology (ESE), in collaboration with the European Network for the Study of Adrenal Tumors (ENSAT), asserted that adrenal incidentalomas (AIs) <4 cm and ≤10 Hounsfield units (HU) do not require further follow-up imaging. To validate the clinical application of the follow-up strategies suggested by the 2016 ESE-ENSAT guidelines, we explored the clinical characteristics and natural course of AIs in a single center over 13 years. Design and methods: This retrospective cohort study included a total of 1149 patients diagnosed with AIs between 2000 and 2013 in a single tertiary center. Hormonal examination and radiological evaluations were performed at the initial diagnosis of AI and during the follow-up according to the appropriate guidelines. Results: The mean age at diagnosis was 54.2 years, and the majority of AIs (68.0%) were nonfunctional lesions. Receiver operating curve analysis was used to discriminate malignant from benign lesions; the optimal cut-off value for mass size was 3.4 cm (sensitivity: 100%; specificity: 95.0%), and that for the pre-contrast HU was 19.9 (sensitivity: 100%; specificity: 67.4%). The majority of nonfunctional lesions did not change in size during the 4-year follow-up period. Applying a cut-off value of 1.8 μg/dL after a 1-mg overnight dexamethasone suppression test, 28.0% of all nonfunctional AIs progressed to autonomous cortisol secretion during the follow-up period. However, we observed no development of overt Cushing’s syndrome in the study. Conclusions: We advocate that no follow-up imaging is required if the detected adrenal mass is <4 cm and has clear benign features. However, prospective studies with longer follow-up are needed to confirm the appropriate follow-up strategies.

Background: Since image-based fracture prediction models using deep learning are lacking, we aimed to develop an X-ray-based fracture prediction model using deep learning with longitudinal data.Methods: This study included 1,595 participants aged 50 to 75 years with at least two lumbosacral radiographs without baseline fractures from 2010 to 2015 at Seoul National University Hospital. Positive and negative cases were defined according to whether vertebral fractures developed during follow-up. The cases were divided into training (n=1,416) and test (n=179) sets. A convolutional neural network (CNN)-based prediction algorithm, DeepSurv, was trained with images and baseline clinical information (age, sex, body mass index, glucocorticoid use, and secondary osteoporosis). The concordance index (C-index) was used to compare performance between DeepSurv and the Fracture Risk Assessment Tool (FRAX) and Cox proportional hazard (CoxPH) models.Results: Of the total participants, 1,188 (74.4%) were women, and the mean age was 60.5 years. During a mean follow-up period of 40.7 months, vertebral fractures occurred in 7.5% (120/1,595) of participants. In the test set, when DeepSurv learned with images and clinical features, it showed higher performance than FRAX and CoxPH in terms of C-index values (DeepSurv, 0.612; 95% confidence interval [CI], 0.571 to 0.653; FRAX, 0.547; CoxPH, 0.594; 95% CI, 0.552 to 0.555). Notably, the DeepSurv method without clinical features had a higher C-index (0.614; 95% CI, 0.572 to 0.656) than that of FRAX in women.Conclusion: DeepSurv, a CNN-based prediction algorithm using baseline image and clinical information, outperformed the FRAX and CoxPH models in predicting osteoporotic fracture from spine radiographs in a longitudinal cohort.

Background: Osteoporosis is one of the diseases that requires early screening and detection for its management. Common clinical tools and machine-learning (ML) models for screening osteoporosis have been developed, but they show limitations such as low accuracy. Moreover, these methods are confined to limited risk factors and lack individualized explanation. Objective: The aim of this study was to develop an interpretable deep-learning (DL) model for osteoporosis risk screening with clinical features. Clinical interpretation with individual explanations of feature contributions is provided using an explainable artificial intelligence (XAI) technique. Methods: We used two separate data sets: the National Health and Nutrition Examination Survey data sets from the United States (NHANES) and South Korea (KNHANES) with 8274 and 8680 respondents, respectively. The study population was classified according to the T-score of bone mineral density at the femoral neck or total femur. A DL model for osteoporosis diagnosis was trained on the data sets and significant risk factors were investigated with local interpretable model-agnostic explanations (LIME). The performance of the DL model was compared with that of ML models and conventional clinical tools. Additionally, contribution ranking of risk factors and individualized explanation of feature contribution were examined. Results: Our DL model showed area under the curve (AUC) values of 0.851 (95% CI 0.844-0.858) and 0.922 (95% CI 0.916-0.928) for the femoral neck and total femur bone mineral density, respectively, using the NHANES data set. The corresponding AUC values for the KNHANES data set were 0.827 (95% CI 0.821-0.833) and 0.912 (95% CI 0.898-0.927), respectively. Through the LIME method, significant features were induced, and each feature’s integrated contribution and interpretation for individual risk were determined. Conclusions: The developed DL model significantly outperforms conventional ML models and clinical tools. Our XAI model produces high-ranked features along with the integrated contributions of each feature, which facilitates the interpretation of individual risk. In summary, our interpretable model for osteoporosis risk screening outperformed state-of-the-art methods.