Development of a deep learning model for the automated detection of green pixels indicative of gout on dual energy CT scan

• Published in: Research in Diagnostic and Interventional Imaging (Online) Vol. 9; p. 100044
• Main Authors: Faghani, Shahriar, Nicholas, Rhodes G., Patel, Soham, Baffour, Francis I., Moassefi, Mana, Rouzrokh, Pouria, Khosravi, Bardia, Powell, Garret M., Leng, Shuai, Glazebrook, Katrina N., Erickson, Bradley J., Tiegs-Heiden, Christin A.
• Format: Journal Article
• Language: English
• Published: France Elsevier Masson SAS 01.03.2024; Elsevier
• Subjects: Convolutional neural network; Deep learning; Dual-energy CT; Gout; Original; Segmentation; Vision transformers; Deep learning; digital imaging and communications in medicine; Segmentation; monosodium urate; Gout; Convolutional neural network; dice similarity score; Dual-energy CT; convolutional neural networks; Vision transformers; neuroimaging informatics technology initiative; standard deviation

Dual-energy CT (DECT) is a non-invasive way to determine the presence of monosodium urate (MSU) crystals in the workup of gout. Color-coding distinguishes MSU from calcium following material decomposition and post-processing. Most software labels MSU as green and calcium as blue. There are limitatio...