Ventral penile curvature estimation using an app

The assessment of penile curvature is a key component of hypospadias surgery, as it often determines if a 1 or 2-stage procedure should be done. The objective of this study was to test the accuracy of penile curvature measurements using an App-based method among non-urologists.

Lateral pictures of six plastic 3-dimensional penile models representing each decile of curvature from 20 to 70° were obtained and stored on the research project iPad. Each picture was labeled with a name (i.e. Jim). Non-urology medical professionals were recruited to estimate the curvature of the six pictures with an App-based method consisting of the Photoblend Pro App and 9 provided graded 2-dimensional penile curvature images (representing each decile of curvature from 10 to 90°). Curvature estimations were done by merging the picture of the 3-dimensional penile model with one of the 9 provided graded 2-dimensional penile curvature images inside the App in an iterative process until the penile model picture matched the graded image (see figure). A research associate taught the research subjects the App-based method and then the subjects were asked to estimate the ventral penile curvature of each of the six penile model pictures. Measurement error was calculated as the absolute difference in between the measured value and the true value for the 6 models. A comparison was then made with previous research where pediatric urologists used either a goniometer or unaided visual inspection (UVI) to measure the ventral curvature of the same plastic 3-dimensional penile models used for the pictures of this study.

Twenty-one subjects were recruited, and all completed the study. Mean errors using the App ranged from 1.9° to 7.1°, compared to 6.5°-15° for UVI and 4.4°-15.9° for goniometry. The median error for the App was 0° compared to 5-10° for both UVI and goniometry. Mean errors were significantly lower (p < 0.05) when using the app compared to UVI/goniometry for all except the 30° and 50° models. Assuming patients with VC ≤ 30° would have had a one stage repair versus a 2-stage repair if curvature was >30°, the number of measurements that could have resulted in the unintended operation was calculated. There was a statistically significant difference in number of potential unintended surgeries in between App (17%) versus UVI + Goniometer (37%) (p = 0.0133).

This pilot study demonstrated better penile curvature estimations using the App compared to the two most common methods currently used by pediatric urologists. Plastic models provide an avenue to test and compare penile curvature measurement techniques.