Point-of-care fentanyl detection: Combining smartphone imaging and chemometrics for reliable quantitative analysis

In this work, the possibility of using smartphone image analysis in combination with the spot test as a quick and simple method of fentanyl analysis was investigated. A probe composed of silver nanoprism-graphene quantum dots (AgNprs-GQD) has been used to induce color variations at different fentanyl concentrations. The data for these variations were recorded by spectrophotometry, spectrofluorimetry, and the digital image colorimetry methodology for comparison. Image analysis was achieved using the PhotoMetrix smartphone App, which uses univariate calibration of the collected images. For further investigation of image analysis, multivariate calibration was performed using the parallel factor analysis (PARAFAC) method in the MATLAB environment. The method was validated by comparing the results obtained from the image analysis methods and spectrophotometry as a standard method. Under optimized conditions, linear ranges and limits of detection (LOD) were 0.001- 0.3 µg/mL and 0.0007 µg/mL for the spectrophotometry method, 0.01- 0.5 µg/mL and 0.009 µg/mL for the spectrofluorimetry method, 0.001- 3.0 µg/mL and 0.0008 µg/mL for the smartphone colorimetry method, and 0.001- 0.5 µg/mL and 0.0007 µg/mL for PARAFAC. Intra-day and inter-day precisions were obtained at ≤ 6.4 % and ≤ 10.8 %. The method was successfully applied to exhaled breath condensate samples from ventilated neonates, demonstrating a non-invasive, rapid, and field-deployable approach for point-of-care fentanyl quantification.