Differentiation of Active Corneal Infections from Healed Scars Using Deep Learning.

Purpose: To develop and evaluate an automated, portable algorithm to differentiate active corneal ulcers from healed scars using only external photographs.

Design: A convolutional neural network was trained and tested using photographs of corneal ulcers and scars.

Participants: De-identified photographs of corneal ulcers were obtained from the Steroids for Corneal Ulcers Trial (SCUT), Mycotic Ulcer Treatment Trial (MUTT), and Byers Eye Institute at Stanford University.

The International Science and Evidence-based Education Assessment.

Education is indispensable for the flourishing of people from all backgrounds and stages of life. However, given the accelerating demographic, environmental, economical, socio-political, and technological changes—and their associated risks and opportunities—there is increasing consensus that our current educational systems are falling short and that we need to repurpose education and rethink the organization of learning to meet the challenges of the 21st century. The United Nations Educational Scientific and Cultural Organization (UNESCO) “Futures of Education” initiative was formally launched at the United Nations General Assembly in 2019 to provide such a vision of education for the future.

Impact of non-medical switching on clinical and economic outcomes, resource utilization and medication-taking behavior: a systematic literature review.

Objective: To evaluate current knowledge of the impact of non-medical switching on clinical and economic outcomes, resource utilization and medication-taking behavior.

Methods: The literature was searched (Medline and Web of Science, January 2000-November 2015) to identify United States' studies evaluating ≥25 patients and measuring the impact of non-medical switching of drugs (switching to a chemically distinct but similar medication for reasons other than lack of clinical efficacy/response, side effects or poor adherence) on ≥1 clinical, economic, resource utilization or medication-taking behavior outcome. The direction of association between non-medical switching and outcomes was classified as negative or positive if a statistically significant worsening or improvement was reported, or neutral if no significant difference was observed.

Hematologic outcomes and blood utilization in cancer patients with chemotherapy-induced anemia (CIA) pre- and post-national coverage determination (NCD): results from a multicenter chart review.

Purpose: In July 2007, the Centers for Medicare and Medicaid Services (CMS) limited coverage of erythropoiesis-stimulating agents (ESAs) in cancer patients with chemotherapy-induced anemia (CIA) through a National Coverage Determination (NCD). The primary objective of this study was to compare transfusion rates in patients with CIA with lung, breast, or colorectal cancer before and after the NCD.

Methods: Adult Medicare patients with CIA treated at 49 community oncology clinics were selected from two time periods based on clinics' NCD implementation date.