A team led by researchers at MUSC Hollings Cancer Center developed an artificial intelligence tool that uses natural language processing to identify the original cancer type of patients with brain metastases by analyzing their doctors’ clinical notes. The model outperformed standard diagnostic codes, accurately determining the primary cancer type in over 90% of cases, which can improve research efforts and enable more precise care. The lightweight, scalable tool could streamline cancer care and research by unlocking critical insights buried in electronic health records.

In this study, an automated method for lumen, vessel wall, and plaque segmentation for quantitative assessment based on MR vessel wall images and a CNN model with different priors was developed and evaluated on many patients with ischemic stroke.

Identifying embryos with the highest likelihood of successful implantation is a critical component of the in vitro fertilization (IVF) process. Visual assessments are limited by the subjectivity of embryologists, making consistent evaluation of embryo health challenging with traditional methods. Recent advances in artificial intelligence (AI)—particularly in computer vision and deep learning—have enabled the automated analysis of embryo morphology images, reducing subjectivity and improving evaluation efficiency. Through an extensive literature search using keywords such as “embryo health assessment” and “artificial intelligence,” the present review focuses on AI-driven approaches for automated embryo evaluation. It examines AI techniques applied to embryo assessment across the early development, blastocyst, and full developmental stages. This review indicated the promising potential of AI technologies in enhancing the precision, consistency, and speed of embryo selection. AI models have been reported to outperform manual evaluations across several parameters, offering promising opportunities to improve success rates and operational efficiency in reproductive medicine. Additionally, this review discusses the current limitations of AI implementation in clinical settings and explores future research directions. Overall, the review provides insight into AI’s growing role in advancing embryo selection and highlights the path toward fully automated evaluation systems in assisted reproductive technology. 

Salk Institute researchers launch machine learning framework ShortStop, which explores overlooked DNA regions in the "dark side" of the human genome in search of microproteins that may play roles in health or disease. They already used ShortStop to analyze a lung cancer dataset and find microproteins that may serve as biomarkers or therapeutic targets.