Traditional Drosophila melanogaster climbing (RING) assays rely on manual scoring or commercial setups, limiting throughput and reproducibility. Behavioural data from these assays is rarely computationally integrated with corresponding histological evidence of neurodegeneration, leaving a critical gap in neurotoxicity characterisation. ASCEND (Automated Screening and Characterization of Neurological Decline) is an open-source computational pipeline designed to automate behavioural tracking in the RING assay and link locomotor metrics directly to neuroimaging based neurodegeneration analysis. The behavioural module uses OpenCV-based background subtraction, Kalman filter state prediction, and Hungarian algorithm assignment to track individual flies across the full observation window. An auto-calibrating, tube-width-based pixel-to-millimetre conversion enables extraction of four locomotor metrics: climbing velocity, total distance traveled, negative geotaxis index (NGI), and trajectory linearity. Per-fly and population-level data are exported to CSV with auto-generated SPSS syntax for statistical analysis. ASCEND's multimodal module consists of: NeuroCount, a deep learning segmentation module utilising StarDist 3D, which automates the counting of dopaminergic (DA) neurons from confocal Z stacks of immunostained brains exposed to neurotoxins such as rotenone, paraquat, and malathion. A Correlator module that bridges these datasets, computing dose response correlation matrices between in-vivo climbing deficits and ex-vivo DA neuron loss at the level of individual neuronal clusters (PPL1, PPM3, PAM). ASCEND will provide a unified, accessible infrastructure linking behavioural phenotypes directly to histological neurodegeneration, enabling high-throughput neurotoxicity screening in Drosophila models of neurodegenerative disease.