Computational strategies have transformed the entire drug design and discovery process. However, the primary concerns associated are time consumption and production costs. Other hurdles include incompetency, inexact target delivery, and insuitable dosage. Such challenges can be eliminated with technological excellence by integrating computer-assisted drug design and AI algorithms. The computational technologies like deep learning and single-cell methods which conquer extensive biological facts from images, can speed-up the discovery process. ML approaches have facilitated the improvement at many steps in drug discovery for analysis of high-dimensional profiling data and technological advances to generate huge data sets. Using Image-based profiling, information in biological images is scaled-down to multidimensional profile, which reveals unanticipated biological activity that is applicable in drug discovery at many steps like recognizing disease-associated phenotypes, perceiving disease mechanisms and predicting drug’s activity, toxicity or MOA. In this review, we have discussed how the Machine Learning & Deep Learning approaches have been applied in functional profiling workflows by recent studies, the use of advanced techniques to optimize the challenges and the potential of emerging techniques in drug discovery which are anticipated to amplify the applicability of ML in drug discovery. Further, we focus on image-based profiling applications to the drug discovery process.