This week's news update features new studies on the use of artificial intelligence in endocrine care. The first study compares DeepSeek-R1 and ChatGPT-4o in interpreting thyroid ultrasound reports. The second study uses explainable machine learning to identify factors associated with exercise-related dysglycemia in children with T1D. The third study examines the use of unauthorized AI tools in healthcare, highlighting both benefits and risks.

Read the articles covered: 

DeepSeek-R1 tops ChatGPT-4o in thyroid nodule classification | AACE Endocrine AI

Machine learning maps exercise-related dysglycemia thresholds in children with T1D | AACE Endocrine AI

Study maps physicians' use of 'Shadow AI' | AACE Endocrine AI