Engineering of Generative Artificial Intelligence and Natural Language Processing Models to Accurately Identify Arrhythmia Recurrence.

Background: Large language models (LLMs), such as ChatGPT, excel at interpreting unstructured data from public sources, yet are limited when responding to queries on private repositories, such as electronic health records (EHRs). We hypothesized that prompt engineering could enhance the accuracy of LLMs for interpreting EHR data without requiring domain knowledge, thus expanding their utility for patients and personalized diagnostics.

Methods: We designed and systematically tested prompt engineering techniques to improve the ability of LLMs to interpret EHRs for nuanced diagnostic questions, referenced to a panel of medical experts.

Treatment of effluent from municipal wastewater treatment plants using electrochemically produced Caro's acid.

This work describes the application of electrochemically produced Caro's acid in a divided electrochemical flow cell for the removal of fourteen CECs from real effluent at a municipal wastewater treatment plant in Ciudad Real, Spain. The results are compared with direct dosing of the reagent (with an ionic/molecular oxidant) and radical-assisted oxidation (activated sulfate radical via photochemical oxidation or hydrogen peroxide-induced radical oxidation). This study sheds light on the underlying mechanisms of these processes.

New Approach in the Interpretation of Complex Triple-negative Breast Cancer Immunohistochemistry Specimens Processed With VENTANA PD-L1 (SP142) Assay.

Triple-negative breast cancer (TNBC) is challenging to treat because of its lack of specific molecular targets. The IMMUNOPEG study aimed to evaluate a novel structured method for interpreting TNBC immunohistochemistry specimens processed with VENTANA PD-L1 (SP142) assay. The study involved 10 pathologists who evaluated 50 different immunohistochemistry specimens of TNBC with programmed death ligand 1 (PD-L1) expression considered challenging and that were previously evaluated by the scientific committee, using the NAVIFY Digital Pathology platform.

Proteomics in Acute Heart Transplant Rejection, On Behalf of the GRAfT Investigators.

Background: Proteomic phenotyping can provide insights into rejection pathophysiology, novel biomarkers, and therapeutic targets.

Methods: Within the prospective, multicenter Genomic Research Alliance for Transplantation study, 181 proteins were evaluated from blood drawn at the time of endomyocardial biopsy; protein fold change, logistic regression, and pathway analyses were conducted, with protein discovery adjusted for a 5% false discovery rate.

Results: Among 104 adult heart transplant patients (31% female sex, 53% Black race, median age 52 y), 74 had no rejection, 18 developed acute cellular rejection (ACR), and 12 developed antibody-mediated rejection (AMR).