METHODS: The study was done at the International Medical University, Malaysia, and the University of Lahore, Pakistan. Semi-structured interviews were conducted from 1st August 2021 to 17th March 2022. Demographic information was noted. Themes were identified, and a summary of the information under each theme was created.
RESULTS: A total of 24 (14 from Malaysia and 10 from Pakistan) faculty participated. Most agreed that undergraduate medical students can progress (at a differential rate) if they attain the required competencies. Among the major advantages mentioned were that students may graduate faster, learn at a pace comfortable to them, and develop an individualized learning pathway. Several logistical challenges must be overcome. Providing assessments on demand will be difficult. Significant regulatory hurdles were anticipated. Artificial intelligence (AI) can play an important role in creating an individualized learning pathway and supporting time-independent progression. The course may be (slightly) cheaper than a traditional one.
CONCLUSION: This study provides a foundation to further develop and strengthen flexible-length competency-based medical education modules. Further studies are required among educators at other medical schools and in other countries. Online learning and AI will play an important role.
METHODS: A comprehensive search was conducted in the Web of Science (WOS) electronic database to identify the top 100 most-cited articles on AI in orthodontics and orthognathic surgery. Publication and citation data were obtained and further analyzed and visualized using R Biblioshiny. The key domains of the 100 articles were also identified.
RESULTS: The top 100 most-cited articles were published between 2005 and 2022, contributed by 458 authors, with an average citation count of 22.09. South Korea emerged as the leading contributor with the highest number of publications (28) and citations (595), followed by China (16, 373), and the United States (7, 248). Notably, six South Korean authors ranked among the top 10 contributors, and three South Korean institutions were listed as the most productive. International collaborations were predominantly observed between the United States, China, and South Korea. The main domains of the articles focused on automated imaging assessment (42%), aiding diagnosis and treatment planning (34%), and the assessment of growth and development (10%). Besides, a positive correlation was observed between the testing sample size and citation counts (P = 0.010), as well as between the time of publication and citation counts (P
OBJECTIVE: To develop and validate a deep learning model using readily available clinical information to predict treatment success with the first ASM for individual patients.
DESIGN, SETTING, AND PARTICIPANTS: This cohort study developed and validated a prognostic model. Patients were treated between 1982 and 2020. All patients were followed up for a minimum of 1 year or until failure of the first ASM. A total of 2404 adults with epilepsy newly treated at specialist clinics in Scotland, Malaysia, Australia, and China between 1982 and 2020 were considered for inclusion, of whom 606 (25.2%) were excluded from the final cohort because of missing information in 1 or more variables.
EXPOSURES: One of 7 antiseizure medications.
MAIN OUTCOMES AND MEASURES: With the use of the transformer model architecture on 16 clinical factors and ASM information, this cohort study first pooled all cohorts for model training and testing. The model was trained again using the largest cohort and externally validated on the other 4 cohorts. The area under the receiver operating characteristic curve (AUROC), weighted balanced accuracy, sensitivity, and specificity of the model were all assessed for predicting treatment success based on the optimal probability cutoff. Treatment success was defined as complete seizure freedom for the first year of treatment while taking the first ASM. Performance of the transformer model was compared with other machine learning models.
RESULTS: The final pooled cohort included 1798 adults (54.5% female; median age, 34 years [IQR, 24-50 years]). The transformer model that was trained using the pooled cohort had an AUROC of 0.65 (95% CI, 0.63-0.67) and a weighted balanced accuracy of 0.62 (95% CI, 0.60-0.64) on the test set. The model that was trained using the largest cohort only had AUROCs ranging from 0.52 to 0.60 and a weighted balanced accuracy ranging from 0.51 to 0.62 in the external validation cohorts. Number of pretreatment seizures, presence of psychiatric disorders, electroencephalography, and brain imaging findings were the most important clinical variables for predicted outcomes in both models. The transformer model that was developed using the pooled cohort outperformed 2 of the 5 other models tested in terms of AUROC.
CONCLUSIONS AND RELEVANCE: In this cohort study, a deep learning model showed the feasibility of personalized prediction of response to ASMs based on clinical information. With improvement of performance, such as by incorporating genetic and imaging data, this model may potentially assist clinicians in selecting the right drug at the first trial.