One of the main motivations to look beyond the SM is the discrepancy between the theoretical prediction and observation of anomalous magnetic moment of muon. Alleviating this tension between theory and experiment and satisfying the bounds from lepton flavor violation data simultaneously is a challenging task. In this paper, we consider generalised Two Higgs Doublet Model, with a Yukawa structure as a perturbation of Type X Two Higgs Doublet Model. In view of this model, we explore muon anomaly and lepton flavor violation along with constraints coming from B-physics, theoretical constraints, electroweak observables and collider data which can restrict the model parameter space significantly. We find that within the framework of this model it is possible to obtain regions allowed by all constraints, that can provide an explanation for the observed muon anomaly and at the same time predicts interesting signatures of lepton flavor violation. Furthermore, we consider the flavor violating decay of low-mass CP-odd scalar to probe the allowed parameter space at future runs of the LHC. With simple cut-based analysis we show that part of that parameter space can be probed with significance $> 5 sigma$. We also provide Artificial Neural Network analysis which definitely improves our cut-based results significantly.