Diagnostic accuracy of smartphone technology for capturing videos and images of semen analysis samples: A cross-sectional study.

Background: Semen analysis is an essential indicator of male infertility potential. The study aims to assess the diagnostic accuracy of smartphones compared to manual microscopy for semen analysis.

Method: It is a cross-sectional analytical study with investigator blinding. Data were collected from August to September 2023. Pictures of semen analysis were captured through light microscopy and stored in a coded format on a smartphone. Later, the results of both methods were compared.

Results: A total of 50 adequate semen samples were included. The age of enrolled males was Mean±SD 29.4 ± 5.9 years. The sensitivity and specificity were 100% (95% CI 83.9%-100%) and (95% CI 88.1%-100%), respectively, for smartphones compared to light microscopy for total sperm counts. The positive predictive value (PPV) and negative predictive value (NPV) were both 100%, with (95% CI 83.9%-100%) and (95% CI 88.1%-100%), respectively. Sensitivity, specificity, PPV, and NPV for total sperm motility were 97.9%, 100%, 100%, and 66%, respectively. For normal morphology sperm, the sensitivity, specificity, PPV, and NPV of smartphones were 72.7%, 82.1%, 53.3%, and 91.4%, respectively, and for abnormal morphology sperm, they were 100%, 98%, 50%, and 100%, respectively. Smartphones exhibited a sensitivity of 98%, specificity of 100%, PPV of 100%, and NPV of 50% for assessing sperm vitality. The diagnostic agreement between smartphones and light microscopy was very good (κ value -0.6-1) for the detection of total count, vitality, and total motility of sperm.

Conclusion: Smartphone technology demonstrates high sensitivity and specificity for semen analysis compared to manual microscopy. It also shows excellent agreement with manual microscopy for most parameters in semen analysis. We recommend smartphone reporting for semen analysis in remote areas and poor resource settings.