Amyotrophic Lateral Sclerosis (ALS) is a progressive neurological disorder characterized by the gradual degeneration of cells in the spinal cord, brainstem, and motor cortex, leading to manifestations such as weakness, muscle atrophy, and spasticity (Elisen, 2009). A comprehensive understanding of the disease process is currently lacking, making preclinical ALS models crucial to study the mechanisms of pathology insurgence, characterize the disease and test therapeutics approaches. Pathological in vitro models emerged as viable alternative to in vivo studies, consolidating a non-invasive approach that aims to simplify the complexity of the brain while still preserving its fundamental features (Schlachetzki, et al., 2013). In this study, we coupled cortical networks of healthy (WT) and ALS-affected neurons to Micro-Electrode Arrays (MEAs). The cells were isolated from SOD1G93A mouse embryos at the 14th gestational day (E14). We investigated the networks’ electrophysiological activity at 14, 21, 28 days in vitro (DIV). In particular, we focused on analyzing spiking and bursting spontaneous activity and synchronization within networks. To discern ASL and WT neuronal networks, we exploited a deep learning classifier employing a 3-layer neural network structure with a learning process based on error backpropagation. Starting from DIV 21, we were able to classify the two classes with an accuracy higher than 80%. The results highlighted a faster development in WT model with respect to ALS one. Moreover, pathological networks displayed a slower spiking and bursting activity and were characterized by a high percentage of random spikes that reduced the global synchronization of the network.