A NEW URINE DIPSTICK FOR THE DIGITAL AGE, COMMUNITY SCREENING OF CKD NEED OF THE HOUR : PILOT STUDY

Introduction:

The burden of kidney disease is rising, the Global Burden of Disease Study showed that Chronic Kidney Disease (CKD) prevalence has increased 33% from 1990 to 2017. As CKD is an asymptomatic disease in its early course, majority of CKD patients are unaware. In US, National Health and Nutritional Health and Examination Survey (NHANES) showed 90% of CKD patients were unaware of their CKD diagnosis. This unawareness is likely to be even worse in other parts of the world. These patients are not able to benefit from interventions to slow progression of kidney disease. There is need for community screening to easily identify such patients and refer them for early interventions to decrease the increasing need for renal replacement therapy and mortality associated with CKD. Urine dipstick is an easy tool for screening, however it suffers from the need to train the patients or the community health workers to read the dipstick and there is no easy way to verify the reading and reporting of the dipstick at scale. A new urine dipstick and app has been created that allows taking a picture of the urine dipstick and provides results using image processing & AI/ML algorithms to quantify the results. A pilot study is reported here.

Methods:

A urine dipstick and an app were developed (Neodocs), this allowed a urine dipstick to be photographed and the app reports the results based on this image. the urine test kit has various fiduciry markers and colour controls that allow for correction of lighting and image quality. After this pre processing the image is sent to cloud and software algorithms run on cloud and standardize images to account for different cameras and lighting conditions. Post this the color variation is measured and reported in the form of parameter values to the care provider or user. The steps are depicted in Figure 1 and Figure 2. A Pilot study was done in Sion Hospital, Mumbai. 210 patients presenting to the medicine clinic were screened for albuminuria by both Neodocs urine dipstick with app and the US FDA approved Acon Mission U120 Analyzer with U120 strips. The Analyzer measured urine albumin in 10,30,80,150 mg/l and urine creatinine in 10,50,100,200,300mg/l while the Neodocs measured urine albumin in 10,30,50,80,150 mg/l and urine creatinine in 10,50,100,200,300 mg/l. These are used to get urine Albumin/creatinine ratio and then classify patients as A1 < 30mg/gm creatinine, A2 31-300mg/gm creatinine and A3>300mg/gm creatinine. The accuracy of the Neodocs app was compared to the Acon Analyzer.

Figure 1. NeoDocs App with patented technology

Figure 2. Steps on how to take the test

Results:

210 patients were screened by both methods and the results are in Table 1. Neodocs had correctly placed 91.4% in their A1, A2 and A3 categories. The comparison with the Acon analyzer for urine albumin show good accuracy of 87.6% between the two measurement but more variability for urine creatinine with accuracy of 63.8%.

Conclusions:

This study shows that Neodocs strip with app is an accurate screening tool and can correctly classify the vast majority of patients screened accurately into A1 vs A2 or A3, allowing early referral for those with A2 or A3 for further evaluation and management. This study will need validation with larger cohorts and with measured urine albumin and urine creatinine as well. With its ease of use, being vernacular,  ability to avoid need for training of community health care workers, store images of the tests and allow for validation this could be a game changer for community screening of CKD across the world.    

Table 1 : Comparison of ACR (A1, A2, A3), Urine Albumin and Urine Creatinine 

I have no potential conflict of interest to disclose.

I did not use generative AI and AI-assisted technologies in the writing process.