Application of machine learning method to diagnosis of alzheimer diseases using SPECT brain image

M. Takahashi; T. Idani; H. Fukuda; R. Kawashima; M. Kitamura

Application of machine learning method to diagnosis of alzheimer diseases using SPECT brain image

M. Takahashi, T. Idani, H. Fukuda, R. Kawashima, M. Kitamura

Tohoku University

Research output: Contribution to conference › Paper › peer-review

1 Citation (Scopus)

Abstract

The automatic diagnosis system for dementia with SPECT brain image has been developed. Previous work had analyzed the average of blood flow with respect to each functional region. However, the accuracy of diagnosis was lower than that of expert radiologists'. This is partially because that previous work couldn't retrieve some of important information for discrimination from SPECT images. In the present study, new attribute values have been added, which are symmetry index of cerebral flow and rate of blood flow change in the functional regions based on the diagnostic strategy of the expert radiologists'. The accuracy has been improved by introducing these additional attributes.

Original language	English
Pages	2420-2423
Number of pages	4
Publication status	Published - 2005
Event	SICE Annual Conference 2005 - Okayama, Japan Duration: 2005 Aug 8 → 2005 Aug 10

Conference

Conference	SICE Annual Conference 2005
Country/Territory	Japan
City	Okayama
Period	05/8/8 → 05/8/10

Keywords

Diagnosis support
Dimentia
Functional brain image
SPECT

Cite this

@conference{924d30cf0f984697962723e22f51cb83,

title = "Application of machine learning method to diagnosis of alzheimer diseases using SPECT brain image",

abstract = "The automatic diagnosis system for dementia with SPECT brain image has been developed. Previous work had analyzed the average of blood flow with respect to each functional region. However, the accuracy of diagnosis was lower than that of expert radiologists'. This is partially because that previous work couldn't retrieve some of important information for discrimination from SPECT images. In the present study, new attribute values have been added, which are symmetry index of cerebral flow and rate of blood flow change in the functional regions based on the diagnostic strategy of the expert radiologists'. The accuracy has been improved by introducing these additional attributes.",

keywords = "Diagnosis support, Dimentia, Functional brain image, SPECT",

author = "M. Takahashi and T. Idani and H. Fukuda and R. Kawashima and M. Kitamura",

year = "2005",

language = "English",

pages = "2420--2423",

note = "SICE Annual Conference 2005 ; Conference date: 08-08-2005 Through 10-08-2005",

}

TY - CONF

T1 - Application of machine learning method to diagnosis of alzheimer diseases using SPECT brain image

AU - Takahashi, M.

AU - Idani, T.

AU - Fukuda, H.

AU - Kawashima, R.

AU - Kitamura, M.

PY - 2005

Y1 - 2005

N2 - The automatic diagnosis system for dementia with SPECT brain image has been developed. Previous work had analyzed the average of blood flow with respect to each functional region. However, the accuracy of diagnosis was lower than that of expert radiologists'. This is partially because that previous work couldn't retrieve some of important information for discrimination from SPECT images. In the present study, new attribute values have been added, which are symmetry index of cerebral flow and rate of blood flow change in the functional regions based on the diagnostic strategy of the expert radiologists'. The accuracy has been improved by introducing these additional attributes.

AB - The automatic diagnosis system for dementia with SPECT brain image has been developed. Previous work had analyzed the average of blood flow with respect to each functional region. However, the accuracy of diagnosis was lower than that of expert radiologists'. This is partially because that previous work couldn't retrieve some of important information for discrimination from SPECT images. In the present study, new attribute values have been added, which are symmetry index of cerebral flow and rate of blood flow change in the functional regions based on the diagnostic strategy of the expert radiologists'. The accuracy has been improved by introducing these additional attributes.

KW - Diagnosis support

KW - Dimentia

KW - Functional brain image

KW - SPECT

UR - http://www.scopus.com/inward/record.url?scp=33645292335&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=33645292335&partnerID=8YFLogxK

M3 - Paper

AN - SCOPUS:33645292335

SP - 2420

EP - 2423

T2 - SICE Annual Conference 2005

Y2 - 8 August 2005 through 10 August 2005

ER -

Application of machine learning method to diagnosis of alzheimer diseases using SPECT brain image

Abstract

Conference

Keywords

Other files and links

Fingerprint

Cite this