Table 2.

Summary of motors symptoms reviewed in this study

Ref	Detection	Task	Data set	Accuracy	Sensitivity	Specificity
Kuresan et al. [45]	Speech signals	MFCC + WPT with HMM	40 subjects	95.16%	93.55%	91.67%
Gürüler [54]	Speech signals	KMCFW – CVANN tenfold CV	23 PD,8 HC	99.52%	–	–
Benba [55]	Speech data	SVM-linear kernel	17 PD,17 HC	91.17%	–	–
Devarajan et al. [49]	Speech disorder	Fog and FKNN-CBR	23 PD, 8 HC	94.87%	97.28%	87.50%
Karaman et al. [48]	Speech signals	DenseNet 161	mPower Voice database	89.75%	91.50%	–
Senturk [58]	Speech signals	SVM	23 PD, 8 HC	93.84%	–	–
Vikas [63]	Speech signals	Gradient Boosting (Ensemble Approach)	147 PD, 48 HC	97.43%	–	–
Ouhmida [66]	Speech signals	KNN	40 PD, 40 HC	97.22%	100%	94.44%
Thakur et al. [84]	Handwriting dynamics	RBM + MLP	62 PD, 15 HC	95.32%	–	–
Drotár et al. [74]	Handwriting dynamics	Mann–Whitney U Test. SVM along with tenfold CV	37 PD, 38 HC	81.3%	87.4%	80.9%
Naseer A et al. [76]	Handwriting dynamics	AlexNet	37 PD, 38 HC	98.28%	–	–
Gil-Martín et al. [69]	Handwriting dynamics	CNN architecture	62 PD, 15 HC	96.5%	–	–
Kurt [78]	Handwriting dynamics	SVM (Static Spiral test)	57 PD, 15 HC	97.52%	–	–
Kotsavaloglou [80]	Handwriting dynamics	Naïve Bayes	24 PD, 20 HC	91%	88%	95%
Saravanan [83]	Handwriting dynamics	VGG19-Inception Model	175 PD, 192 HC	98.45%	–	–
Kamran [85]	Handwriting dynamics	AlexNet (Fine Tuned)	HandPD, NewHandPD, Parkinson’s Drawing	99.22%	–	–
El Maachi et al. [96]	Gait movements	Deep 1D convnet	93 PD, 73 HC	98.7%	98.1%	100%
Balaji et al. [89]	Gait movements	Adam-LSTM (1) multi class (2) binary	PhysioNet	(1) 96.6% (2) 98.6%	(1) 96.20 (2) 98.23	(1) 98.08 (2) 99.10
Richa [95]	Gait movements	Modified KNN	93 PD, 73 HC	99.60%	–	–
Tong et al. [87]	Gait movements	PVI, persistent entropy of topological imprints, SVM, RF, Borderline-SMOTE	29 PD, 18 HC	98.08%	–	–