Alzheimer’s Disease (AD), characterized by a gradual onset and a lack of exact therapeutic interventions, underscores the imperative for the development of uncomplicated and cost-effective biosensors capable of detecting its biomarkers. This necessity arises in anticipation of a projected surge in the incidence of AD. Quantum dots (QDs) represent the promising new generation of luminophores owing to their size, composition and desired optical properties. In this study, an Artificial Intelligence-driven paper-based QD aptasensor for the early detection of AD by targeting amyloid beta (Aβ-42) and p-tau181 proteins using Förster Resonance Energy Transfer (FRET) is developed. The sensor employs a Whatman paper with six sensing wells, integrating hydrophobic and hydrophilic regions, hydrophobic parts created through wax. Blood samples are placed in the inlet, dispersing into six sensing wells containing QD-aptamer-AuNP complexes. Target proteins induce conformational changes in aptamers, leading to fluorescence quenching in CdTe QDs due to FRET. Two wells target p-tau181, two target Aβ-42, and two serve as references. Fluorescence emission spectra from each well are recorded showing a linear correlation between fluorescence quenching and analyte concentration. Values from each pair of wells are then averaged, and the average values from the pairs targeting Aβ-42 and p-tau181 are compared to the average value of the reference wells. Subsequently, the performance of the sensor is enhanced with the Convolutional Neural Network algorithm and a user-friendly app for real-time results which takes the sensor images as input and outputs the biomarker levels of patients is developed. This facilitates monitoring and enabling long-term management of the disease and opens the avenue of personalized medicine for the diagnosis of Alzheimer’s.