One of the most rampant and worst brain disorders, Parkinson’s disease is defined by the death of neuronal cells, which causes motor ailments, memory loss, cognitive problems, and synaptic dysfunction among others. Early diagnosis of the disease is crucial to thwart its progressive symptoms which amplify with age. A lot of research is being done to identify the best performing learning techniques in various modalities. Handwriting analysis is a newer modality for Parkinson’s disease detection whose power needs to be unleashed. It is a reliable, non-invasive, and an affordable approach for the early detection of Parkinson's disease. This paper highlights the latest learning techniques in the field of detection of Parkinson's disease using handwriting analysis. Moreover, it summarizes their pros and cons and provides the detailed summary of the handwriting datasets available, some unanswered questions, and the potential directions for further research.