We aimed to use deep learning to detect tuberculosis in chest radiographs in annual workers’ health examination data and compare the performances of convolutional neural networks (CNNs) based on images only (I-CNN) and CNNs including demographic variables (D-CNN). The I-CNN and D-CNN models were trained on 1000 chest X-ray images, both positive and negative, for tuberculosis. Feature extraction was conducted using VGG19, InceptionV3, ResNet50, DenseNet121, and InceptionResNetV2. Age, weight, height, and gender were recorded as demographic variables. The area under the receiver operating characteristic (ROC) curve (AUC) was calculated for model comparison. The AUC values of the D-CNN models were greater than that of I-CNN. The AUC values for VGG19 increased by 0.0144 (0.957 to 0.9714) in the training set, and by 0.0138 (0.9075 to 0.9213) in the test set (both p < 0.05). The D-CNN models show greater sensitivity than I-CNN models (0.815 vs. 0.775, respectively) at the same cut-off point for the same specificity of 0.962. The sensitivity of D-CNN does not attenuate as much as that of I-CNN, even when specificity is increased by cut-off points. Conclusion: Our results indicate that machine learning can facilitate the detection of tuberculosis in chest X-rays, and demographic factors can improve this process.

There are several different proposed data mining methods for the postmarketing surveillance of drug safety. Adverse events are often classified into a hierarchical structure. Our objective was to compare the performance of several of these different data mining methods for adverse drug events data with a hierarchical structure. We generated datasets based on the World Health Organization’s Adverse Reaction Terminology (WHO-ART) hierarchical structure. We evaluated different data mining methods for signal detection, including several frequentist methods such as reporting odds ratio (ROR), proportional reporting ratio (PRR), information component (IC), the likelihood ratio test-based method (LRT), and Bayesian methods such as gamma Poisson shrinker (GPS), Bayesian confidence propagating neural network (BCPNN), the new IC method, and the simplified Bayesian method (sB), as well as the tree-based scan statistic through an extensive simulation study. We also applied the methods to real data on two diabetes drugs, voglibose and acarbose, from the Korea Adverse event reporting system. Only the tree-based scan statistic method maintained the type I error rate at the desired level. Likelihood ratio test-based methods and Bayesian methods tended to be more conservative than other methods in the simulation study and detected fewer signals in the real data example. No method was superior to the others in terms of the statistical power and sensitivity of detecting true signals. It is recommended that those conducting drug‒adverse event surveillance use not just one method, but make a decision based on several methods.

Although behavioral and psychological symptoms of dementia are a global public health challenge, non-pharmacological interventions using information and communication technologies can be an affordable, cost-effective, and innovative solution. This study aimed to examine the effectiveness of non-pharmacological interventions using information and communication technologies on the behavioral and psychological symptoms of dementia and identify potential moderators of intervention effects. Systematic review and meta-analysis of randomized controlled trials. A systematic literature review was conducted using PubMed, CINAHL, PsycINFO, Embase, and the Cochrane Library from May 2022. Randomized controlled trials that examined the effects of non-pharmacological interventions using information and communication technologies on the behavioral and psychological symptoms of dementia were included. A meta-analysis using a random-effects model was performed to calculate the pooled standardized mean differences between overall symptoms and each type of symptom. For moderator analyses, subgroup and meta-regression analyses were performed. Sixteen trials (15 articles) met the eligibility criteria. The interventions were grouped into activity engagement interventions using digital health that provided music and reminiscence therapy, physical exercise, social interaction interventions using social robots, and telehealth-based care aid interventions that provided coaching or counseling programs. Pooled evidence demonstrated that non-pharmacological interventions using information and communication technologies exerted a large effect on depression (SMD = -1.088, 95% CI -1.983 to -0.193, p = 0.017), a moderate effect on overall behavioral and psychological symptoms of dementia (SMD = -0.664, 95% CI -0.990 to -0.338, p < 0.001), and agitation (SMD = -0.586, 95% CI -1.130 to -0.042, p = 0.035). No effects on neuropsychiatric symptoms (SMD = -0.251, 95% CI -0.579 to 0.077, p = 0.133), anxiety (SMD = -0.541, 95% CI -1.270 to 0.188, p = 0.146), and apathy (SMD = -0.830, 95% CI -1.835 to 0.176, p = 0.106) were reported. Moderator analyses identified the mean age of the participants as a potential moderator of intervention effects. Evidence from this systematic review and meta-analysis suggests that non-pharmacological interventions, using information and communication technologies, were an applicable approach to managing behavioral and psychological symptoms among older adults with dementia, with moderate to large effect sizes. However, evidence on anxiety and apathy is inconclusive due to the limited number of existing randomized controlled trials. Future studies with subgroup analyses are warranted to conclude the most effective types of intervention using information and communication technologies for each type of symptom. CRD42021258498.

Among biological agents for the treatment of atopic dermatitis (AD), dupilumab is a front-runner. Although many studies have been conducted on the real-world use of dupilumab, the sample size is often small and data is primarily on Western people. Therefore, we investigated the efficacy and safety of dupilumab in patients with moderate-to-severe AD in Korea. All patients with moderate-to-severe AD treated with dupilumab from September 2018 to June 2019 in this institution were included and analyzed by medical records. They were evaluated using the Eczema Area and Severity Index (EASI), Numerical Rating Scale (NRS), Patient Oriented Eczema Measure (POEM), and Dermatology Quality of Life Index (DLQI), respectively on admission, after two weeks (only EASI and NRS) and after 16 weeks. Laboratory tests were measured before and 16 weeks after treatment. A total of 101 patients were included. All efficacy tools showed a significant decrease after 16 weeks; EASI 77.4%, NRS 70.0%, POEM 60.7%, and DLQI 65.0%. EASI was characterized by a marked improvement of 51.5% in just two weeks. The treatment response was not significantly different according to the interval of treatment. Elevated Lactate Dehydrogenase (LDH) at 16 weeks was associated with poor treatment response. Moreover, a high eosinophil count was related to a lower change in EASI and POEM. In the correlation analysis, EASI was not correlated to DLQI before treatment. For changes after 16 weeks, POEM showed the highest correlation with DLQI. (R = 0.66, p < 0.001) In the additional analysis for factors affecting treatment response, the female gender was associated with good treatment response. (odds ratio = 5.4, p = 0.04) Adverse events from treatment included facial erythema (9.9%) and conjunctivitis (5.0%). Overall, it was confirmed that the efficacy of dupilumab in the real-world is similar to that of the existing clinical trials. We suggest that POEM is a useful tool for identifying the quality of life. The female gender was associated with a good treatment response. Both an elevated LDH and a high eosinophil count could be a therapeutic biomarker. Further research will be needed for a long-term period.