Effects of Expert-Determined Reference Standards in Evaluating the Diagnostic Performance of a Deep Learning Model: A Malignant Lung Nodule Detection Task on Chest Radiographs.

Jung Eun Huh Jong Hyuk Lee Eui Jin Hwang Chang Min Park

Published in: Korean journal of radiology (2023)

The LCA model was most similar to the clinical gold standard for malignant pulmonary nodule detection on chest radiographs. Expert-determined standards caused bias in measuring the diagnostic performance of the artificial intelligence model.

Keyphrases

artificial intelligence
deep learning
machine learning
big data
pulmonary hypertension
clinical practice
real time pcr
convolutional neural network