A two-stage variable selection and classification approach for Parkinson’s disease detection by using voice recording replications

Naranjo, Lizbeth; Prez, Carlos J.; Martn, Jacinto; Campos-Roca, Y.

doi:10.1016/j.cmpb.2017.02.019

Cited by 96 publications

(69 citation statements)

References 39 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…We also compared the output of our method with other unsupervised dimension reduction methods such as t-SNE, non-metric MDS and Sammon Mapping. As shown in the Figure 1 for Naranjo et al data-set [12] the visualisation itself shows two separate clusters for normal and Parkinson's disease. Sammon mapping and MDS have similar separability of two classes like t-SNE.…”

Section: Resultsmentioning

confidence: 81%

“…In order to evaluate TPDS we used data-sets where both classification and visualisation is needed for making useful inference. The first data-set used for evaluation was generated by Naranjo et al [12] and it consists of feature extracted from speech recording of normal individuals and Parkinon's disease patients. For every patient or individual there are 4 replicate of data points which have 44 features.…”

Section: Resultsmentioning

confidence: 99%

“…Overall, we have presented here a technique to utilise method for distance preservation based dimension reduction. Figure 1 Dimension reduction of Parkinson's data-set by Naranjo et al [12] using four different method. The distance stress based cost is represent termed as MDS-cost…”

Section: Discussionmentioning

confidence: 99%

“…The 7 types of cells are shown with different colors.The adjusted Rand Index (ARI) and Normalized mutual information (NMI) were calculated after k-mean clustering (k=7). The outlier cells (shown in red color) seem to have different effect in different dimension reduction methods Dimension reduction of Parkinson's data-set by Naranjo et al[12] using four different method. The distance stress based cost is represent termed as MDS-cost.…”

mentioning

confidence: 99%

See 3 more Smart Citations

Distance preserving dimension reduction with local-topology based scaling for improved classification of Biomedical data-sets

Khosla

Jha²,

Kumar³

2019

Preprint

View full text Add to dashboard Cite

Dimension reduction is often used for several procedures of analysis of high dimensional biomedical data-sets such as classification or outlier detection. To improve performance of such data-mining steps, preserving both distance information and local topology among data-points could be more useful than giving priority to visualisation in low dimension. Therefore, we introduce topology preserving distance scaling (TPDS) to augment dimension reduction method meant to reproduce distance information in higher dimension. Our approach involves distance inflation to preserve local topology to avoid collapse during distance preservation based optimisation. Applying TPDS on diverse biomedical data-sets revealed that besides providing better visualisation than typical distance preserving methods, TPDS leads to better classification of data points in reduced dimension. For data-sets with outliers, the approach of TPDS also proves to be useful, even for purely distance-preserving method for achieving better convergence.

show abstract

Section: Resultsmentioning

confidence: 81%

Section: Resultsmentioning

confidence: 99%

Section: Discussionmentioning

confidence: 99%

mentioning

confidence: 99%

See 2 more Smart Citations

Distance preserving dimension reduction with local-topology based scaling for improved classification of Biomedical data-sets

Khosla

Jha²,

Kumar³

2019

Preprint

View full text Add to dashboard Cite

show abstract

“…Analysis of voice patterns could prove a rich source of information about patients’ emotional states or may detect early features of illness or cognitive deficits 2324. Some pioneers are already exploring the detection of disease by analysing speech patterns in patients 25.…”

Section: What Might Be Possible?mentioning

confidence: 99%