Neural computational models, driven by brain-inspired cognitive systems, have transformed the modeling of complex biological processes, offering advanced tools such as artificial and deep neural networks to probe intricate signaling pathways in disease contexts. These methods enable the simulation and in-depth analysis of dynamic interactions within critical signaling pathways, facilitating the identification of key disease drivers as therapeutic targets. Hence, this study investigated the JAK/STAT pathway in acute myeloid leukemia (AML), utilizing a Swarm-based deep neural network with transcriptomic data to predict gene‒ gene interactions and identify critical drivers as potential therapeutic targets. High-throughput analysis of transcriptomic data, integrated with advanced techniques such as net-work inference filtration, interaction matrix visualization, differential driver analysis, and pathway adherence, revealed key age-specific and pan-age disease drivers via the JAK/STAT signaling pathway as potential therapeutic targets in AML. This approach can reliably identify gene‒gene interactions and novel disease drivers as potential therapeutic targets, further, investigate established ones, and guide future research in diseases while highlighting the transformative role of Swarm-based deep neural computation in advancing personalized therapeutic strategies and offering a pathway to more effective, tailored interventions in oncology.