Visualization of Decision Tree State for the Classification of Parkinson's Disease

David Williams; Deborah Mudali; Hugo Buddelmeijer; Parisa Noorishad; Sanne Meles; Remco Renken; Klaus Leenders; E Valentijn; Johannes Roerdink

doi:10.14738/jbemi.33.1858

Visualization of Decision Tree State for the Classification of Parkinson's Disease

David Williams, Deborah Mudali, Hugo Buddelmeijer, Parisa Noorishad, Sanne Meles, Remco Renken, Klaus Leenders, E Valentijn, Johannes Roerdink

Research output: Contribution to journal › Article › Academic › peer-review

46 Downloads (Pure)

Abstract

Decision trees have been shown to be effective at classifying subjects with Parkinson's disease when provided with features (subject scores) derived from FDG-PET data. Such subject scores have strong discriminative power but are not intuitive to understand. We therefore augment each decision node with thumbnails of the principal component (PC) images from which the subject scores are computed, and also provide labeled scatter plots of the distribution of scores. These plots allow the progress of individual subjects to be traced through the tree and enable the user to focus on complex or unexpected classifications. In addition, we present a visual representation of a typical brain activity pattern arriving at each leaf node, and show how this can be compared to a known reference to validate the behaviour of the tree.

Original language	English
Pages (from-to)	25-41
Number of pages	25
Journal	Journal of Biomedical Engineering and Medical Imaging
Volume	3
Issue number	3
DOIs	https://doi.org/10.14738/jbemi.33.1858
Publication status	Published - 1-Aug-2016

Keywords

Decision tree; Visualization; Medical imaging; Parkinson's Disease; Computer aided diagnosis;

Access to Document

10.14738/jbemi.33.1858

Visualization of Decision Tree State for the Classification of Parkinson’s DiseaseFinal publisher's version, 1.01 MBLicence: Taverne

Handle.net

Persistent link

Cite this

@article{4fbcef25131141b6b2dde4e8ced3bced,

title = "Visualization of Decision Tree State for the Classification of Parkinson's Disease",

abstract = "Decision trees have been shown to be effective at classifying subjects with Parkinson's disease when provided with features (subject scores) derived from FDG-PET data. Such subject scores have strong discriminative power but are not intuitive to understand. We therefore augment each decision node with thumbnails of the principal component (PC) images from which the subject scores are computed, and also provide labeled scatter plots of the distribution of scores. These plots allow the progress of individual subjects to be traced through the tree and enable the user to focus on complex or unexpected classifications. In addition, we present a visual representation of a typical brain activity pattern arriving at each leaf node, and show how this can be compared to a known reference to validate the behaviour of the tree.",

keywords = "Decision tree; Visualization; Medical imaging; Parkinson's Disease; Computer aided diagnosis;",

author = "David Williams and Deborah Mudali and Hugo Buddelmeijer and Parisa Noorishad and Sanne Meles and Remco Renken and Klaus Leenders and E Valentijn and Johannes Roerdink",

year = "2016",

month = aug,

day = "1",

doi = "10.14738/jbemi.33.1858",

language = "English",

volume = "3",

pages = "25--41",

journal = "Journal of Biomedical Engineering and Medical Imaging",

issn = "2055-1266",

publisher = "Society for Science and Education",

number = "3",

}

TY - JOUR

T1 - Visualization of Decision Tree State for the Classification of Parkinson's Disease

AU - Williams, David

AU - Mudali, Deborah

AU - Buddelmeijer, Hugo

AU - Noorishad, Parisa

AU - Meles, Sanne

AU - Renken, Remco

AU - Leenders, Klaus

AU - Valentijn, E

AU - Roerdink, Johannes

PY - 2016/8/1

Y1 - 2016/8/1

N2 - Decision trees have been shown to be effective at classifying subjects with Parkinson's disease when provided with features (subject scores) derived from FDG-PET data. Such subject scores have strong discriminative power but are not intuitive to understand. We therefore augment each decision node with thumbnails of the principal component (PC) images from which the subject scores are computed, and also provide labeled scatter plots of the distribution of scores. These plots allow the progress of individual subjects to be traced through the tree and enable the user to focus on complex or unexpected classifications. In addition, we present a visual representation of a typical brain activity pattern arriving at each leaf node, and show how this can be compared to a known reference to validate the behaviour of the tree.

AB - Decision trees have been shown to be effective at classifying subjects with Parkinson's disease when provided with features (subject scores) derived from FDG-PET data. Such subject scores have strong discriminative power but are not intuitive to understand. We therefore augment each decision node with thumbnails of the principal component (PC) images from which the subject scores are computed, and also provide labeled scatter plots of the distribution of scores. These plots allow the progress of individual subjects to be traced through the tree and enable the user to focus on complex or unexpected classifications. In addition, we present a visual representation of a typical brain activity pattern arriving at each leaf node, and show how this can be compared to a known reference to validate the behaviour of the tree.

KW - Decision tree; Visualization; Medical imaging; Parkinson's Disease; Computer aided diagnosis;

UR - http://scholarpublishing.org/index.php/JBEMi/article/view/1858

U2 - 10.14738/jbemi.33.1858

DO - 10.14738/jbemi.33.1858

M3 - Article

SN - 2055-1266

VL - 3

SP - 25

EP - 41

JO - Journal of Biomedical Engineering and Medical Imaging

JF - Journal of Biomedical Engineering and Medical Imaging

IS - 3

ER -

Visualization of Decision Tree State for the Classification of Parkinson's Disease

Abstract

Keywords

Access to Document

Handle.net

Other files and links

Embargoed Document

Fingerprint

Cite this