Development and Validation of a Deep Learning Algorithm to Differentiate Colon Carcinoma From Acute Diverticulitis in Computed Tomography Images
Key Points Question Can a deep learning algorithm differentiate between acute diverticulitis and colon cancer on computed tomography images and improve radiologists’ performance under routine clinical conditions? Findings In this diagnostic study, a 3-dimensional convolutional neural network developed on contrast-enhanced computed tomography images of 585 patients with colon cancer and acute diverticulitis was able to predict both entities with a high sensitivity (83%) and specificity (87%). As an artificial intelligence support system, the model significantly improved the sensitivity and specificity and reduced the number of false-negative and false-positive findings. Meaning The findings of this study suggest that, as a support system, a deep learning model may improve the care of patients with large-bowel wall thickening.
Paper
Full text
Development and Validation of a Deep Learning Algorithm to Differentiate Colon Carcinoma From Acute Diverticulitis in Computed Tomography Images
Semantic Scholar · Medicine · 2023
Abstract
Key Points Question Can a deep learning algorithm differentiate between acute diverticulitis and colon cancer on computed tomography images and improve radiologists’ performance under routine clinical conditions? Findings In this diagnostic study, a 3-dimensional convolutional neural network developed on contrast-enhanced computed tomography images of 585 patients with colon cancer and acute diverticulitis was able to predict both entities with a high sensitivity (83%) and specificity (87%). As an artificial intelligence support system, the model significantly improved the sensitivity and specificity and reduced the number of false-negative and false-positive findings. Meaning The findings of this study suggest that, as a support system, a deep learning model may improve the care of patients with large-bowel wall thickening.