tracing dopaminergic agents in human urine by a smartphone-based fluorescent electronic tongue

Abstract: parkinson's disease (pd) is a progressive neurological disorder that has affected more than 10 million people worldwide. dopaminergic agents such as levodopa (levo), carbidopa (carb), benserazide (bens) and entacapone (enta) are often prescribed to treat pd by facilitating the function of the neurotransmitter dopamine. despite the therapeutic efficacy of the dopaminergic drugs, these agents' dysfunction contributes to pd progression and disrupts the patient’s quality of life [1]. the importance their dosage control in human biological fluids has boosted the demand for fast, sensitive and real-time multi-analyte detection. herein, a simple and efficient electronic tongue based on fingerprint fluorimetric patterns has been developed for rapid, visual and simultaneous detection and discrimination of these agents by a single fluorometric test. the optical assay principle mainly relies on the different fluorescence quenching of the dual emissive nanoprobe in the absence and presence of cetyltrimethylammonium bromide (ctab) at alkaline conditions, upon the addition of target analytes. the distinct assembly dynamics of the nanoprobes and the oxidized targets in the absence and presence of ctab led to diverse aggregation levels under electron transfer (et) mechanisms. a smartphone camera was used to take photos from the solutions in the wells and extract rgb values on which data analysis was performed. distinct changes in the spectral profiles along with vivid and concentration-dependent color variations led to visual discrimination of dopaminergic agents in a broad concentration range (1 μm to 100 μm). in addition, color difference maps (cdms) were used as distinct visual fingerprints for discrimination between the targets. the linear discriminant analysis (lda) model successfully discriminated among different concentration of targets having similar structures. in addition, excellent results of heat maps and radar plots of the data matrix were visual evidence of the remarkable classification ability of the constructed array. furthermore, the feasibility of the array for quantification and prediction of different concentrations of target analytes in human urine, as a complicated biological media, was satisfactorily appraised by the array under an lda model with discrimination accuracy of 100% [2]. since simultaneous monitoring of agents is essential in evaluating the disease process, the strategy of the designed electronic tongue as a model may be a vital remedy for practical point-of-care diagnostics by any user soon.