DESIGN: Artificial intelligence (neural network) study.
METHODS: We assessed 1400 OCT scans of patients with neovascular AMD. Fifteen physical features for each eligible OCT, as well as patient age, were used as input data and corresponding recorded visual acuity as the target data to train, validate, and test a supervised neural network. We then applied this network to model the impact on acuity of defined OCT changes in subretinal fluid, subretinal hyperreflective material, and loss of external limiting membrane (ELM) integrity.
RESULTS: A total of 1210 eligible OCT scans were analyzed, resulting in 1210 data points, which were each 16-dimensional. A 10-layer feed-forward neural network with 1 hidden layer of 10 neurons was trained to predict acuity and demonstrated a root mean square error of 8.2 letters for predicted compared to actual visual acuity and a mean regression coefficient of 0.85. A virtual model using this network demonstrated the relationship of visual acuity to specific, programmed changes in OCT characteristics. When ELM is intact, there is a shallow decline in acuity with increasing subretinal fluid but a much steeper decline with equivalent increasing subretinal hyperreflective material. When ELM is not intact, all visual acuities are reduced. Increasing subretinal hyperreflective material or subretinal fluid in this circumstance reduces vision further still, but with a smaller gradient than when ELM is intact.
CONCLUSIONS: The supervised machine learning neural network developed is able to generate an estimated visual acuity value from OCT images in a population of patients with AMD. These findings should be of clinical and research interest in macular degeneration, for example in estimating visual prognosis or highlighting the importance of developing treatments targeting more visually destructive pathologies.
METHODS: YouTube videos were systematically acquired with 4 search terms. The top 50 videos per search term by the number of views were stored in a YouTube account. A set of inclusion/exclusion criteria were applied, videos were assessed for viewing characteristics, a 4-point scoring system (0-3) was applied to evaluate QOI in 10 predetermined domains, and a 3-point scoring system (0-2) was applied to evaluate COI. Descriptive statistical analyses and intrarater and interrater reliability tests were performed.
RESULTS: Strong intrarater and interrater reliability scores were observed. Sixty-three videos from the top 58 most-viewed DPs were viewed 1,395,471 times (range, 414-124,939). Most DPs originated from the United States (20%), and orthodontists (62%) uploaded most of the videos. The mean number of reported domains was 2.03 ± 2.40 (out of 10). The mean overall QOI score per domain was 0.36 ± 0.79 (out of 3). The "Placement of miniscrews" domain scored highest (1.23 ± 0.75). The "Cost of miniscrews placement" domain scored the lowest (0.03 ± 0.25). The mean overall QOI score per DP was 3.59 ± 5.64 (out of 30). The COI in 32 videos was immeasurable, and only 2 avoided using technical words.
CONCLUSIONS: The QOI related to temporary anchorage devices contained within videos provided by DPs through the YouTube Web site is deficient, particularly in the cost of placement. Orthodontists should be aware of the importance of YouTube as an information resource and ensure that videos related to temporary anchorage devices contain comprehensive and evidence-based information.
METHODS: A cross-sectional survey was conducted via a validated 49-item questionnaire, administered immediately after all students completed the examination. The questionnaire comprised of questions to evaluate the content and structure of the examination, perception of OSCE validity and reliability, and rating of OSCE in relation to other assessment methods. Open-ended follow-up questions were included to generate qualitative data.
RESULTS: Over 80% of the students found the OSCE to be helpful in highlighting areas of weaknesses in their clinical competencies. Seventy-eight percent agreed that it was comprehensive and 66% believed it was fair. About 46% felt that the 15 minutes allocated per station was inadequate. Most importantly, about half of the students raised concerns that personality, ethnicity, and/or gender, as well as interpatient and inter-assessor variability were potential sources of bias that could affect their scores. However, an overwhelming proportion of the students (90%) agreed that the OSCE provided a useful and practical learning experience.
CONCLUSIONS: Students' perceptions and acceptance of the new method of assessment were positive. The survey further highlighted for future refinement the strengths and weaknesses associated with the development and implementation of an OSCE in the International Islamic University Malaysia's pharmacy curriculum.
METHODS: The 2-unit leadership course was piloted among second- and third-year students in a public college of pharmacy with a 4-year doctor of pharmacy curriculum. The participating students completed the LABS-III during the first and last classes as part of a quality improvement measure for course enhancement. Rasch analysis was then used to assess the reliability and validity evidence for the LABS-III.
RESULTS: A total of 24 students participated in the pilot course. The pre and postcourse surveys had 100% and 92% response rates, respectively. After Rasch analysis model fit was achieved, the item separation for the 14 nonextreme items was 2.19 with an item reliability of 0.83. The person separation index was 2.16 with a person reliability of 0.82.
CONCLUSION: The Rasch analysis revealed that the number of LABS-III items should be decreased and that the 3-point response scale should be used to improve functionality and use in classroom settings for PharmD students in the United States. Further research is needed to augment the reliability and validity evidence of the modified instrument for use at other United States colleges of pharmacy.