Publications by authors named "Matthew Silbergleit"

Article Synopsis
  • The study evaluated the ability of four language models (ChatGPT-3.5, ChatGPT-4o, Google Gemini, and Google Gemini Advanced) to accurately generate CAD-RADS scores from coronary CT angiography reports without any fine-tuning.
  • ChatGPT-4o had the highest accuracy at 87%, while ChatGPT-3.5 was the fastest but only achieved 50.5% accuracy, and Gemini had a higher failure rate at 12%.
  • Overall, the findings indicate that while model performance showed promise, further refinement of these AI tools is needed before they can be reliably used in clinical decision-making for CAD-RADS scoring.
View Article and Find Full Text PDF

Syncope is a commonly encountered symptom that accounts for up to 3% of all emergency department visits in the United States [1]. The differential diagnosis for patients presenting with syncope is often broad and can involve multiple distinct organ systems. This case details the hospital course of an 83-year-old male with a complicated past medical history who initially presented for syncope.

View Article and Find Full Text PDF

Growing evidence suggests that imbalances in resident microbes (dysbiosis) can promote chronic inflammation, immune-subversion, and production of carcinogenic metabolites, thus leading to neoplasia. Yet, evidence to support a direct link of individual bacteria species to human sporadic cancer is still limited. This chapter focuses on several emerging bacterial toxins that have recently been characterized for their potential oncogenic properties toward human orodigestive cancer and the presence of which in human tissue samples has been documented.

View Article and Find Full Text PDF

Purpose: Formal agreement studies on interpretation of the videofluoroscopic swallowing study (VFSS) procedure among speech-language pathologists, radiology house officers, and staff radiologists have not been pursued. Each of these professions participates in the procedure, interprets the examination, and writes separate reports on the findings. The aim of this study was to determine reliability of interpretation between and within the disciplines and to determine if structured training improved reliability.

View Article and Find Full Text PDF