Today we take a look at a review of other papers (a way to get a paper published without having to do any actual research yourself, and the drudge work can be delegated to grad students, someone once told me), but nevertheless, there’s much to be learned from looking at the big picture instead of focusing on individual studies, so we’ll forge ahead. Next article will be on rehabilitation therapies that address motor and cognitive issues in PD.

Interestingly enough, this article comes from Indonesian Journal of Electrical Engineering and Computer Science – not the first place you’d probably look for information n Parkinson’s (PD). And, unless you’re a computer science major or are developing applications using AI, I wouldn’t recommend looking this one up.

Most of the studies reviewed used various techniques to analyze datasets of speech recordings, and tried to differentiate between “healthy”people and people with PD. Some of the approaches were fairly accurate (in the 90% plus range) while others were not that much better than reliance on the four cardinal symptoms to be observed clinically (resting tremor, rigidity, bradykinesia and postural instability). 

This one will either be good reading before going to sleep (unless you’re into neural networks and Bayesian prediction) or a good way to induce a headache. The authors do call for more research (naturally) in this area, especially in regard to developing biomarkers using either neuro-imaging or biofluids. And they note that fine-tuning of biomarkers must be  done, so that there is an acceptable balance between false positives and false negatives in predicting whether a person will likely get PD in the near or distant future.

Citation for this is:

A comparative and comprehensive study of prediction of Parkinson’s Disease, (2021) Pandi, V., et al., Indonesian Journal of Electrical Engineering and Computer Systems, vol. 23, No. 2, September 21, pp. 1748-1760.

It’s an open access article, so if you are interested in computer systems. go ahead and search for it.