Abstract Evaluation of digital ELISA using a high-resolution computational optical system for measurement of total 25-hydroxyvitamin D Background Biomarkers such as vitamin D occur at very low concentrations in blood, and sensitive detection methods are required in order to measure them. It is difficult to accurately measure low-concentration biomarkers using conventional enzyme-linked immunosorbent assays (ELISA) with fluorescence staining. By developing a digital ELISA (DELISA, Small Machines, Seoul, Korea) platform, a low concentration substance was realized by analyzing the digitalized signal through a high-resolution computational optical system in a microfluidic chip. We evaluated DELISA for measurement of total 25-hydroxyvitamin D (25OHD) in a fluidic chip. Methods The sample with monoclonal antibody bead conjugates mixture was incubated for 20 min at 37°C, then 30 uL sample mixture was applied to inlet of the chip. After additional incubation for 2 min at room temperature and 2 min after sample application, 120 uL of washing buffer was applied onto the washing hole and the chip was incubated for 10 min. 25OHD molecules in the sample bond to magnetic beads coupled with anti-25OHD at the end of the membrane. Chips were inserted into DELISA, and 25OHD results were derived by analysis of the computational method of microscopic imaging, proportional to the concentration of 25OHD. Assay performance characteristics of the limit of detection (LoD), precision and analytical measurement interval (AMI) using Vitamin D Depleted, Low and High Serum (Molecular Depot, San Diego, CA) were determined according to the Clinical and Laboratory Standards Institute (CLSI) guidelines. 87 Samples for method comparison studies were obtained from routine clinical samples submitted for 2 OHD determination or from apparently healthy normal volunteers. The study was approved by the Institutional Review Board of Dongguk University Ilsan Hospital (202207024). Results LoD was 10.3 ng/mL and within-laboratory precision was 11, 12, and 10% coefficient of variation at 14.1, 56.6, and 113.2 ng/mL, respectively. The best fit polynomial was a cubic equation from 14.1 to 113.2 ng/mL and the difference between the third-order equation and the first-order equation at Level 3 and 4 concentrations was outside the non-linearity limit. Comparison with chemiluminescent assay (Elecsys Vitamin D total II, Roche Diagnostics, Mannheim, Germany) yielded a 0.904 of correlation coefficient and 0.921 slope in the range from 7.0 to 57.9 ng/mL. Conclusions The DELISA 25OHD assay has a suitable LoD and precision, however, it is necessary to improve the AMI at high concentration values and correlation with the automated analyzer.