Search

BackgroundAlthough most current medication error prevention systems are rule-based, these systems may result in alert fatigue because of poor accuracy. Previously, we had developed a machine learning (ML) model based on Taiwan’s local databases (TLD) to address this issue. However, the international transferability of this model is unclear. ObjectiveThis study examines the international transferability of a machine learning model for detecting medication errors and whether the federated learning approach could further improve the accuracy of the model. MethodsThe study cohort included 667,572 outpatient prescriptions from 2 large US academic medical centers. Our ML model was applied to build the original model (O model), the local model (L model), and the hybrid model (H model). The O model was built using the data of 1.34 billion outpatient prescriptions from TLD. A validation set with 8.98% (60,000/667,572) of the prescriptions was first randomly sampled, and the remaining 91.02% (607,572/667,572) of the prescriptions served as the local training set for the L model. With a federated learning approach, the H model used the association values with a higher frequency of co-occurrence among the O and L models. A testing set with 600 prescriptions was classified as substantiated and unsubstantiated by 2 independent physician reviewers and was then used to assess model performance. ResultsThe interrater agreement was significant in terms of classifying prescriptions as substantiated and unsubstantiated (κ=0.91; 95% CI 0.88 to 0.95). With thresholds ranging from 0.5 to 1.5, the alert accuracy ranged from 75%-78% for the O model, 76%-78% for the L model, and 79%-85% for the H model. ConclusionsOur ML model has good international transferability among US hospital data. Using the federated learning approach with local hospital data could further improve the accuracy of the model.

Purpose: This study aimed to investigate the effects of subconjunctival injection of aflibercept, a soluble protein decoy for VEGFR-1 and VEGFR-2, on corneal angiogenesis and VEGFR-expressing CD11b+ cells in a mouse model of suture-induced corneal neovascularization. Methods: Corneal neovascularization was induced in BALB/c mice by placing three sutures on the cornea. Immediately after surgery, either 200 µg aflibercept (5 µL) or an equal volume of phosphate-buffered saline (PBS) was administered into the subconjunctival space. Seven days after later, corneal new vessels were quantified through clinical examination and measurement of the CD31-stained area in corneal flat mounts. The levels of pro-angiogenic and inflammatory markers in the cornea were evaluated using RT-qPCR. The percentages of VEGFR-2+CD11b+ cells and VEGFR-3+CD11b+ cells were analyzed in the cornea, blood, and draining cervical lymph nodes (DLNs) using flow cytometry. Results: Subconjunctival injection of aflibercept significantly reduced the growth of corneal new vessels compared to subconjunctival PBS injection. The mRNA levels of Cd31, vascular growth factors (Vegfc and Angpt1), and pro-angiogenic/inflammatory markers (Tek/Tie2, Mrc1, Mrc2, and Il6) in the cornea were downregulated by subconjunctival aflibercept. Also, the percentage of VEGFR-3+CD11b+ cells in the cornea, blood, and DLNs was decreased by aflibercept, whereas that of VEGFR-2+CD11b+ cells was unaffected. Conclusion: Subconjunctival aflibercept administration inhibits inflammatory angiogenesis in the cornea and reduces the numbers of cornea-infiltrating and circulating VEGFR-3+CD11b+ cells. Key messages: What is known: • Aflibercept inhibits hemangiogenesis in the eye by interfering with the VEGF-A/VEGFR-2 signaling. What is new: • Subconjunctival aflibercept downregulates the VEGF-C and angiopoietin signaling in the cornea. • Subconjunctival aflibercept reduces both circulating and cornea-infiltrating VEGFR-3+CD11b+ myeloid cells. [ABSTRACT FROM AUTHOR]

The paper estimates the determinants of the growing volume of bilateral environmental aid for the mitigation of climate change using an empirically testable multilateral framework in which both donors and recipient countries compete in world export markets. As the potential donors weigh environmental benefits against the economic costs of providing aid, strategic interactions between the donors and the recipient countries as well as among the donors, influence the evolution of environmental aid. The paper shows that while the volume of bilateral environmental aid increases with the recipient country's credible environmental commitment and bilateral trade volume, the competitive pressure in the export market reduces bilateral environmental aid. Free-riding incentives prevail among the individual donors, whereas the multilateral environmental aids that aim to restore the loss of global environmental resources without altering individual trade competitiveness can increase bilateral environmental aids. [ABSTRACT FROM AUTHOR]