ABSTRACT
Artificial intelligence (AI) in radiology is a rapidly developing field with several prospective clinical studies demonstrating its benefits in clinical practice. In 2022, the Korean Society of Radiology held a forum to discuss the challenges and drawbacks in AI development and implementation. Various barriers hinder the successful application and widespread adoption of AI in radiology, such as limited annotated data, data privacy and security, data heterogeneity, imbalanced data, model interpretability, overfitting, and integration with clinical workflows. In this review, some of the various possible solutions to these challenges are presented and discussed; these include training with longitudinal and multimodal datasets, dense training with multitask learning and multimodal learning, self-supervised contrastive learning, various image modifications and syntheses using generative models, explainable AI, causal learning, federated learning with large data models, and digital twins.
ABSTRACT
Deep learning-based applications have great potential to enhance the quality of medical services. The power of deep learning depends on open databases and innovation. Radiologists can act as important mediators between deep learning and medicine by simultaneously playing pioneering and gatekeeping roles. The application of deep learning technology in medicine is sometimes restricted by ethical or legal issues, including patient privacy and confidentiality, data ownership, and limitations in patient agreement. In this paper, we present an open platform, MI2RLNet, for sharing source code and various pre-trained weights for models to use in downstream tasks, including education, application, and transfer learning, to encourage deep learning research in radiology. In addition, we describe how to use this open platform in the GitHub environment. Our source code and models may contribute to further deep learning research in radiology, which may facilitate applications in medicine and healthcare, especially in medical imaging, in the near future. All code is available at https://github.com/mi2rl/MI2RLNet.
ABSTRACT
PURPOSE: Our study aimed to examine psychometric properties and cross-cultural utility of the Behavior Assessment System for Children-2, Parent Rating Scale-Child (BASC-2 PRS-C) in Korean children. MATERIALS AND METHODS: Two study populations were recruited: a general population sample (n=2115) of 1st to 6th graders from 16 elementary schools and a clinical population (n=219) of 6–12 years old from 5 child psychiatric clinics and an epidemiological sample of autism spectrum disorder. We assessed the validity and reliability of the Korean version of BASC-2 PRS-C (K-BASC-2 PRS-C) and compared subscales with those used for US populations. RESULTS: Our results indicate that the K-BASC-2 PRS-C is a valuable instrument with reliability and validity for measuring developmental psychopathology that is comparable to those in Western population. However, there were some differences noted in the mean scores of BASC-2 PRS-C between Korean and US populations. CONCLUSION: K-BASC-2 PRS-C is an effective and useful instrument with psychometric properties that permits measurement of general developmental psychopathology. Observed Korean-US differences in patterns of parental reports of children's behaviors indicate the importance of the validation, standardization and cultural adaptation for tools assessing psychopathology especially when used in populations different from those for which the instrument was originally created.
Subject(s)
Child , Humans , Autism Spectrum Disorder , Behavior Rating Scale , Cross-Cultural Comparison , Parents , Psychometrics , Psychopathology , Reproducibility of ResultsABSTRACT
PURPOSE: To assess the effectiveness and tolerability of cyclosporine ophthalmic emulsion (CsA) 0.05% in patients with moderate to severe dry eye disease in Korea. METHODS: This was a prospective, multicenter, open-label, surveillance study of 392 Korean patients with moderate to severe dry eye disease who were treated with CsA 0.05% for three months. An assessment of effectiveness was performed at baseline, and after 1, 2, and 3 months. The primary effectiveness outcomes were changes in ocular symptoms and Schirmer score. The secondary effectiveness outcomes were a change in conjunctival staining, use of artificial tears, global evaluation of treatment, and patient satisfaction. The primary safety outcome was the incidence and nature of adverse events. RESULTS: A total of 362 patients completed the study. After three months, all ocular symptom scores were significantly reduced compared to the baseline values, while the Schirmer scores were significantly increased relative to baseline (p 50%) experienced at least a 25% to 50% improvement in symptoms from baseline at each follow-up visit. The majority of patients (72.0%) were satisfied with the treatment results, and 57.2% reported having no or mild symptoms after treatment. The most common adverse events were ocular pain (11.0%). CONCLUSIONS: Our findings indicate that CsA 0.05% is an effective and tolerable treatment for dry eye disease in Korean clinical practice.