The digital pathology accuracy compared to traditional microscopic pathology: A meta-analysis

Background: Digital pathology has emerged as a transformative technology in diagnostic medicine. It offers various advantages over traditional light microscopy. However, evidence regarding diagnostic accuracy remains fragmented across various studies. Objective: To evaluate and compare the diagnostic accuracy of digital pathology systems versus traditional microscopic pathology through meta-analysis. Methods: We undertook a literature search across PubMed, Embase, Cochrane Library, and Web of Science databases from inception to December 2024. Various studies that compared digital pathology and traditional microscopy for histopathological diagnosis were included. We also performed statistical analysis using R software with the 'meta' package. Additionally, the risk of bias was assessed using the QUADAS-2 tool. We also used random-effects models for pooled sensitivity and specificity calculations. Results: Twenty-eight studies were included. Notably, pooled sensitivity for digital pathology was 94.2% (95% CI: 91.8-96.1%), and specificity was 96.8% (95% CI: 94.9-98.1%). Additionally, Substantial heterogeneity was observed at I² = 78% for sensitivity and 82% for specificity. Subgroup analysis recorded increased accuracy in dermatopathology with a sensitivity of 96.1% compared to general surgical pathology, which recorded a sensitivity of 92.8%. Funnel plot analysis suggested minimal publication bias with an Egger's test of p = 0.12. Conclusion: Digital pathology demonstrated comparable diagnostic accuracy to traditional microscopy with high sensitivity and specificity. However, significant heterogeneity exists across various studies. This suggested the need for various standardized validation protocols and continued quality assurance measures.