The effective and precise detection of cancer and neurotransmitter biomarkers including folic acid (FA), dopamine (DA), and epinephrine (EP) are essential for early detection and diagnosis of cancer and neurological disorders and for the development of new drugs. However, it remains challenging to detect FA, DA, and EP with high selectivity and sensitivity with a single material. Herein, we report a photoluminescence (PL)-based selective sensing of FA, DA, and EP with nitrogen-doped graphene quantum dots (NGQDs) synthesized from biocompatible chitosan under ambient conditions using atmospheric pressure microplasmas. By regulating the pH, the selective detection is achieved in broad ranges from 0.8 to 80 μM for FA and 0.4 to 100 μM for both DA and EP with the very low limits of detections of 81.7, 57.8, and 16.7 nM for FA, DA, and EP, respectively. The developed PL sensing method shows the high throughput of 5000 detections per hour. Moreover, highly stable colloidal NGQD dispersion with 100 μg/mL concentration for at least 100 PL detections is produced in 1 h by a single microplasma, and the process is scalable. The mechanisms of the outstanding performance are related to the enhanced, size-dependent π-π stacking attraction between the NGQDs and the pH-regulated chemical states of the analytes and the associated pH-specific photo-induced electron transfer and PL.