Effectiveness of a digital quantitative assessment of protein-to-creatinine ratio using computer-vision technologies in proteinuria screening
Keywords: Digital health intervention, computer vision, proteinuria, screening, chronic kidney disease
TL;DR: This study proposed a digital solution of quantitative urine protein-to-creatinine ratio analysis based on advanced computer-vision technologies, which has been evaluated with good performance in proteinuria screening.
Abstract: Objective: Proteinuria is an effective indicator for early kidney damage, and is recommended to be tested routinely in population at high risk of chronic kidney disease (CKD). This study proposed a novel digital approach, the quantitative urine protein-to-creatinine ratio (UPCR) home-testing kit (UTK), for proteinuria screening based on advanced computer-vision algorithms, and the screening effectiveness was evaluated.
Method: A total of 199 participants who have visited the Peking University First Hospital in October 2023, were included in our study. Randomly selected spot urine samples were collected and measured using laboratory and UTK methods respectively. The UTK method utilized the contour detection and image segmentation algorithms to extract the topological information of the urinalysis strip for color calibration, and employed the three-dimensional color space interpolation algorithm for quantitative readings of UPCR. With the laboratory results as golden criteria, the diagnostic performance of UTK in proteinuria screening was evaluated in terms of validity, reliability, predictive values, and area under the receiver operating characteristic curve (AUC). We conducted the Bland-Altman analysis to assess the agreement of the quantitative UPCR results between UTK and laboratory methods across different levels of UPCR.
Result: The mean age of the 199 participants was 51.6 ± 16.2 years and 88 (44.2%) of them were male. The median UPCR was 222.5 mg/g (interquartile range: 106.5-844.6), and 20 (10.1%) participants were identified as proteinuria. The UTK method performed well in proteinuria screening, with a high accuracy of 91.0%, an 85.0% sensitivity, a 91.6% specificity, and a 0.996 AUC. The Bland-Altman plot showed high agreement of the quantitative UPCR results between the UTK and laboratory methods, especially for participants with a relatively low level of UPCR.
Conclusion: The proposed digital solution for quantitative UPCR analysis based on advanced computer-vision technologies showed good performance in proteinuria screening. This user-friendly and cost-effective UPCR monitoring method can be a promising new strategy to enhance the efficiency of primary prevention and management of CKD.
Submission Number: 10
Loading