A USC-led team has developed an innovative way to predict blood pressure outcomes after bariatric surgery that outperforms the current standard practice. In 108 adolescents who underwent bariatric surgery, the researchers used blood tests taken before the procedure to predict which patients would have improved blood pressure five years later. The study is one of the first to use cutting-edge “omics” techniques, which analyze the body’s small molecules and proteins, to predict long-term health outcomes. Using a combination of metabolomics and proteomics, which analyze thousands of small molecules and proteins in the body to give a snapshot of cellular function, researchers identified the top 10 molecules linked to improved blood pressure five years after surgery. These 10 molecules gave better predictions of which patients would improve than the current predictive approach, which uses a combination of demographic factors and clinical tests. Five of the 10 molecules were also linked to blood pressure in another group of youth with above average BMI, suggesting the findings may be broadly applicable. The results were just published in the journal Hypertension. 

Mass General Brigham-led study found that large language models (LLMs) often fail to challenge illogical medical prompts due to sycophantic behavior, posing risks for misinformation. The researchers showed that targeted fine-tuning can significantly improve model reasoning.