Comparative diagnostic performance of digital and conventional tests for congenital red–green colour vision deficiency

PURPOSE: To compare the diagnostic performance of the DIVE Colour Test for detecting congenital red-green (RG) colour vision deficiency, using the Colour Assessment and Diagnosis (CAD) test as the reference standard, with established conventional colour vision tests providing within-sample benchmarks. METHODS: Sixty participants were enrolled in an observational study: 30 CAD-confirmed normal trichromats and 30 participants with congenital RG deficiency. Each completed the CAD test, DIVE Colour Test (touch and eye-tracking), Nagel anomaloscope, Ishihara 38 Plates Test and Roth 28-Hue Test. Diagnostic accuracy against CAD was assessed using sensitivity, specificity and Cohen's kappa (k), both in right-eye and binocular testing. RESULTS: In right-eye testing, DIVE Touch achieved a sensitivity of 78.1% and a specificity of 100.0%, demonstrating substantial agreement with CAD (κ = 0.77). In comparison, DIVE eye-tracking showed lower sensitivity (53.1%) while maintaining high specificity (96.4%), yielding moderate agreement with CAD (κ = 0.48). Regarding the binocular analysis: DIVE Touch yielded a sensitivity of 69.0%, a specificity of 100.0% and substantial agreement with CAD (κ = 0.69). Among conventional tests versus CAD, Ishihara showed the highest sensitivity with complete specificity, the Nagel anomaloscope achieved substantial agreement, and the Roth 28-Hue Test showed the lowest sensitivity for RG classification. CONCLUSION: Diagnostic performance varied across colour vision tests. DIVE Touch showed high specificity, moderate sensitivity and substantial agreement with CAD, whereas DIVE Eye-Tracking maintained high specificity but showed lower sensitivity and requires further optimisation.