Key points are not available for this paper at this time.
BACKGROUND Background: Visual judgment is susceptible to human errors, such as evaluation item errors and judgment time variations. Automated Urine Test Strip Colorimetric testing may be used by practitioners and small-scale medical check-up facilities to test urine samples. OBJECTIVE Objective: We develop an automated urine test strip colorimetric program using images taken with a smartphone camera to judge the color of the urinalysis test strips objectively. METHODS Methods: To verify the program's reliability, we compare urine samples from real patients with the results of an automated urine analyzer. RESULTS Results: The concordance rate with the automated urine analyzer is 78.6% on average for 12 items. The concordance rate for the ±1 rank test is 95.4%. In addition, determining urinary albumin level results in a sensitivity of 100% and specificity of 58.6%. The area under the curve derived from the ROC curve is 0.907. CONCLUSIONS Conclusions: The results clarified that colorimetric analysis in a fixed imaging environment could be used to determine urine test strips without calibration. This method may be used to screen for microalbuminuria, helping prevent exacerbation of diabetic nephropathy.
Inagaki et al. (Mon,) studied this question.