Impact of a combination of quantitative indices representing uptake intensity, shape, and asymmetry in DAT SPECT using machine learning

comparison of different volume of interest settings

Yu Iwabuchi, Tadaki Nakahara, Masashi Kameyama, Yoshitake Yamada, Masahiro Hashimoto, Yoji Matsusaka, Takashi Osada, Daisuke Ito, Hajime Tabuchi, Masahiro Jinzaki

Research output: Contribution to journalArticle

Abstract

Background: We sought to assess the machine learning-based combined diagnostic accuracy of three types of quantitative indices obtained using dopamine transporter single-photon emission computed tomography (DAT SPECT)—specific binding ratio (SBR), putamen-to-caudate ratio (PCR)/fractal dimension (FD), and asymmetry index (AI)—for parkinsonian syndrome (PS). We also aimed to compare the effect of two different types of volume of interest (VOI) settings from commercially available software packages DaTQUANT (Q) and DaTView (V) on diagnostic accuracy. Methods: Seventy-one patients with PS and 40 without PS (NPS) were enrolled. Using SPECT images obtained from these patients, three quantitative indices were calculated at two different VOI settings each. SBR-Q, PCR-Q, and AI-Q were derived using the VOI settings from DaTQUANT, whereas SBR-V, FD-V, and AI-V were derived using those from DaTView. We compared the diagnostic value of these six indices for PS. We incorporated a support vector machine (SVM) classifier for assessing the combined accuracy of the three indices (SVM-Q: combination of SBR-Q, PCR-Q, and AI-Q; SVM-V: combination of SBR-V, FD-V, and AI-V). A Mann-Whitney U test and receiver-operating characteristics (ROC) analysis were used for statistical analyses. Results: ROC analyses demonstrated that the areas under the curve (AUC) for SBR-Q, PCR-Q, AI-Q, SBR-V, FD-V, and AI-V were 0.978, 0.837, 0.802, 0.906, 0.972, and 0.829, respectively. On comparing the corresponding quantitative indices between the two types of VOI settings, SBR-Q performed better than SBR-V (p = 0.006), whereas FD-V performed better than PCR-Q (p = 0.0003). No significant difference was observed between AI-Q and AI-V (p = 0.56). The AUCs for SVM-Q and SVM-V were 0.988 and 0.994, respectively; the two different VOI settings displayed no significant differences in terms of diagnostic accuracy (p = 0.48). Conclusion: The combination of the three indices obtained using the SVM classifier improved the diagnostic performance for PS; this performance did not differ based on the VOI settings and software used.

Original languageEnglish
Article number7
JournalEJNMMI Research
Volume9
DOIs
Publication statusPublished - 2019 Jan 1

Fingerprint

Dopamine Plasma Membrane Transport Proteins
Fractals
Single-Photon Emission-Computed Tomography
Putamen
Parkinsonian Disorders
ROC Curve
Area Under Curve
Software
Nonparametric Statistics
Support Vector Machine
Machine Learning

Keywords

  • I-FP-CIT
  • I-Ioflupane
  • DAT SPECT
  • Machine learning
  • Parkinson’s syndrome
  • Support vector machine

ASJC Scopus subject areas

  • Radiology Nuclear Medicine and imaging

Cite this

@article{9e426ac290604ab99579ee8e4f8787a6,
title = "Impact of a combination of quantitative indices representing uptake intensity, shape, and asymmetry in DAT SPECT using machine learning: comparison of different volume of interest settings",
abstract = "Background: We sought to assess the machine learning-based combined diagnostic accuracy of three types of quantitative indices obtained using dopamine transporter single-photon emission computed tomography (DAT SPECT)—specific binding ratio (SBR), putamen-to-caudate ratio (PCR)/fractal dimension (FD), and asymmetry index (AI)—for parkinsonian syndrome (PS). We also aimed to compare the effect of two different types of volume of interest (VOI) settings from commercially available software packages DaTQUANT (Q) and DaTView (V) on diagnostic accuracy. Methods: Seventy-one patients with PS and 40 without PS (NPS) were enrolled. Using SPECT images obtained from these patients, three quantitative indices were calculated at two different VOI settings each. SBR-Q, PCR-Q, and AI-Q were derived using the VOI settings from DaTQUANT, whereas SBR-V, FD-V, and AI-V were derived using those from DaTView. We compared the diagnostic value of these six indices for PS. We incorporated a support vector machine (SVM) classifier for assessing the combined accuracy of the three indices (SVM-Q: combination of SBR-Q, PCR-Q, and AI-Q; SVM-V: combination of SBR-V, FD-V, and AI-V). A Mann-Whitney U test and receiver-operating characteristics (ROC) analysis were used for statistical analyses. Results: ROC analyses demonstrated that the areas under the curve (AUC) for SBR-Q, PCR-Q, AI-Q, SBR-V, FD-V, and AI-V were 0.978, 0.837, 0.802, 0.906, 0.972, and 0.829, respectively. On comparing the corresponding quantitative indices between the two types of VOI settings, SBR-Q performed better than SBR-V (p = 0.006), whereas FD-V performed better than PCR-Q (p = 0.0003). No significant difference was observed between AI-Q and AI-V (p = 0.56). The AUCs for SVM-Q and SVM-V were 0.988 and 0.994, respectively; the two different VOI settings displayed no significant differences in terms of diagnostic accuracy (p = 0.48). Conclusion: The combination of the three indices obtained using the SVM classifier improved the diagnostic performance for PS; this performance did not differ based on the VOI settings and software used.",
keywords = "I-FP-CIT, I-Ioflupane, DAT SPECT, Machine learning, Parkinson’s syndrome, Support vector machine",
author = "Yu Iwabuchi and Tadaki Nakahara and Masashi Kameyama and Yoshitake Yamada and Masahiro Hashimoto and Yoji Matsusaka and Takashi Osada and Daisuke Ito and Hajime Tabuchi and Masahiro Jinzaki",
year = "2019",
month = "1",
day = "1",
doi = "10.1186/s13550-019-0477-x",
language = "English",
volume = "9",
journal = "EJNMMI Research",
issn = "2191-219X",
publisher = "Springer Berlin",

}

TY - JOUR

T1 - Impact of a combination of quantitative indices representing uptake intensity, shape, and asymmetry in DAT SPECT using machine learning

T2 - comparison of different volume of interest settings

AU - Iwabuchi, Yu

AU - Nakahara, Tadaki

AU - Kameyama, Masashi

AU - Yamada, Yoshitake

AU - Hashimoto, Masahiro

AU - Matsusaka, Yoji

AU - Osada, Takashi

AU - Ito, Daisuke

AU - Tabuchi, Hajime

AU - Jinzaki, Masahiro

PY - 2019/1/1

Y1 - 2019/1/1

N2 - Background: We sought to assess the machine learning-based combined diagnostic accuracy of three types of quantitative indices obtained using dopamine transporter single-photon emission computed tomography (DAT SPECT)—specific binding ratio (SBR), putamen-to-caudate ratio (PCR)/fractal dimension (FD), and asymmetry index (AI)—for parkinsonian syndrome (PS). We also aimed to compare the effect of two different types of volume of interest (VOI) settings from commercially available software packages DaTQUANT (Q) and DaTView (V) on diagnostic accuracy. Methods: Seventy-one patients with PS and 40 without PS (NPS) were enrolled. Using SPECT images obtained from these patients, three quantitative indices were calculated at two different VOI settings each. SBR-Q, PCR-Q, and AI-Q were derived using the VOI settings from DaTQUANT, whereas SBR-V, FD-V, and AI-V were derived using those from DaTView. We compared the diagnostic value of these six indices for PS. We incorporated a support vector machine (SVM) classifier for assessing the combined accuracy of the three indices (SVM-Q: combination of SBR-Q, PCR-Q, and AI-Q; SVM-V: combination of SBR-V, FD-V, and AI-V). A Mann-Whitney U test and receiver-operating characteristics (ROC) analysis were used for statistical analyses. Results: ROC analyses demonstrated that the areas under the curve (AUC) for SBR-Q, PCR-Q, AI-Q, SBR-V, FD-V, and AI-V were 0.978, 0.837, 0.802, 0.906, 0.972, and 0.829, respectively. On comparing the corresponding quantitative indices between the two types of VOI settings, SBR-Q performed better than SBR-V (p = 0.006), whereas FD-V performed better than PCR-Q (p = 0.0003). No significant difference was observed between AI-Q and AI-V (p = 0.56). The AUCs for SVM-Q and SVM-V were 0.988 and 0.994, respectively; the two different VOI settings displayed no significant differences in terms of diagnostic accuracy (p = 0.48). Conclusion: The combination of the three indices obtained using the SVM classifier improved the diagnostic performance for PS; this performance did not differ based on the VOI settings and software used.

AB - Background: We sought to assess the machine learning-based combined diagnostic accuracy of three types of quantitative indices obtained using dopamine transporter single-photon emission computed tomography (DAT SPECT)—specific binding ratio (SBR), putamen-to-caudate ratio (PCR)/fractal dimension (FD), and asymmetry index (AI)—for parkinsonian syndrome (PS). We also aimed to compare the effect of two different types of volume of interest (VOI) settings from commercially available software packages DaTQUANT (Q) and DaTView (V) on diagnostic accuracy. Methods: Seventy-one patients with PS and 40 without PS (NPS) were enrolled. Using SPECT images obtained from these patients, three quantitative indices were calculated at two different VOI settings each. SBR-Q, PCR-Q, and AI-Q were derived using the VOI settings from DaTQUANT, whereas SBR-V, FD-V, and AI-V were derived using those from DaTView. We compared the diagnostic value of these six indices for PS. We incorporated a support vector machine (SVM) classifier for assessing the combined accuracy of the three indices (SVM-Q: combination of SBR-Q, PCR-Q, and AI-Q; SVM-V: combination of SBR-V, FD-V, and AI-V). A Mann-Whitney U test and receiver-operating characteristics (ROC) analysis were used for statistical analyses. Results: ROC analyses demonstrated that the areas under the curve (AUC) for SBR-Q, PCR-Q, AI-Q, SBR-V, FD-V, and AI-V were 0.978, 0.837, 0.802, 0.906, 0.972, and 0.829, respectively. On comparing the corresponding quantitative indices between the two types of VOI settings, SBR-Q performed better than SBR-V (p = 0.006), whereas FD-V performed better than PCR-Q (p = 0.0003). No significant difference was observed between AI-Q and AI-V (p = 0.56). The AUCs for SVM-Q and SVM-V were 0.988 and 0.994, respectively; the two different VOI settings displayed no significant differences in terms of diagnostic accuracy (p = 0.48). Conclusion: The combination of the three indices obtained using the SVM classifier improved the diagnostic performance for PS; this performance did not differ based on the VOI settings and software used.

KW - I-FP-CIT

KW - I-Ioflupane

KW - DAT SPECT

KW - Machine learning

KW - Parkinson’s syndrome

KW - Support vector machine

UR - http://www.scopus.com/inward/record.url?scp=85060762449&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85060762449&partnerID=8YFLogxK

U2 - 10.1186/s13550-019-0477-x

DO - 10.1186/s13550-019-0477-x

M3 - Article

VL - 9

JO - EJNMMI Research

JF - EJNMMI Research

SN - 2191-219X

M1 - 7

ER -