Detection of urinary tract calculi on CT images reconstructed with deep learning algorithms

Abstract

Deep learning Computed Tomography (CT) reconstruction (DLR) algorithms promise to improve image quality but the impact on clinical diagnostic performance remains to be demonstrated. We aimed to compare DLR to standard iterative reconstruction for detection of urolithiasis by unenhanced CT in children and young adults. This was an IRB approved retrospective study involving post-hoc reconstruction of clinically acquired unenhanced abdomen/pelvis CT scans. Images were reconstructed with six different manufacturer-standard DLR algorithms and reformatted in 3 planes (axial, sagittal, and coronal) at 3 mm intervals. De-identified reconstructions were loaded as independent examinations for review by 3 blinded radiologists (R1, R2, R3) tasked with identifying and measuring all stones. Results were compared to the clinical iterative reconstruction images as a reference standard. IntraClass correlation coefficients and kappa (k) statistics were used to quantify agreement. CT data for 14 patients (mean age: 17.3\u2009±\u20093.4 years, 5 males and 9 females, weight class: 31-70 kg (n\u2009=\u20096), 71-100 kg (n\u2009=\u20097),\u2009>\u2009100 kg (n\u2009=\u20091)) were reconstructed into 84 total exams. 7 patients had urinary tract calculi. Interobserver agreement on the presence of any urinary tract calculus was substantial to almost perfect (k\u2009=\u20090.71–1) for all DLR algorithms. Agreement with the reference standard on number of calculi was excellent (ICC\u2009=\u20090.78–0.96) and agreement on the size of the largest calculus was fair to excellent (ICC\u2009=\u20090.51–0.97) depending on reviewer and DLR algorithm. Deep learning reconstruction of unenhanced CT images allows similar renal stone detectability compared to iterative reconstruction.