Early Diagnosis of Alzheimer's Disease: A Multi-class Deep Learning Framework with Modified k-sparse Autoencoder Classification

Pushkar Bhatkoti, Manoranjan Paul

Research output: Book chapter/Published conference paperConference paper

5 Citations (Scopus)

Abstract

Successful, timely diagnosis of neuropsychiatric diseases is key to management. Research efforts in the area of diagnosis of Alzheimer's disease have used various aspects of computer-aided multi-class diagnosis approaches with varied degrees of success. However, there is still need for more efficient and reliable approaches to successful diagnosis of the disease. This research used deep learning framework with modified k-sparse autoencoder (mKSA) classification to locate neutrally degenerated areas of the brain MRI, low amyloid beta 1-42 imaging in cerebrospinal fluid (CSF) and positron emission tomography (PET) imaging of amyloid; each with a sample of 150 images. Results show a correlation between computational demarcation of infected regions and the images. Degeneration in the studied areas was evidenced by high phosphorylated t-/p-tau levels in CSF, regional hypometabolism fluorodeoxyglucose PET, and the presence of atrophy patterns. The use of mKSA algorithm in boosting classification helped to improve the classifier performance. The KSA method with deep learning framework is used for the first time to produce accurate results in diagnosis of Alzheimer's disease.
Original languageEnglish
Title of host publicationProceedings of the 2016 International Conference on Image and Vision Computing New Zealand (IVCNZ)
EditorsDonald Bailey, Gourab Sen Gupta, Stephen Marsland
Place of PublicationUnited States
PublisherIEEE, Institute of Electrical and Electronics Engineers
Pages1-6
Number of pages6
ISBN (Electronic)9781509027484
DOIs
Publication statusPublished - 05 Jan 2017
Event2016 International Conference on Image and Vision Computing New Zealand: IVCNZ - Massey University, Palmerston North, New Zealand
Duration: 21 Nov 201622 Nov 2016

Conference

Conference2016 International Conference on Image and Vision Computing New Zealand
CountryNew Zealand
CityPalmerston North
Period21/11/1622/11/16

Fingerprint

Cerebrospinal fluid
Positron emission tomography
Imaging techniques
Magnetic resonance imaging
Brain
Classifiers
Deep learning
Amyloid

Cite this

Bhatkoti, P., & Paul, M. (2017). Early Diagnosis of Alzheimer's Disease: A Multi-class Deep Learning Framework with Modified k-sparse Autoencoder Classification. In D. Bailey, G. S. Gupta, & S. Marsland (Eds.), Proceedings of the 2016 International Conference on Image and Vision Computing New Zealand (IVCNZ) (pp. 1-6). [7804459] United States: IEEE, Institute of Electrical and Electronics Engineers. https://doi.org/10.1109/IVCNZ.2016.7804459
Bhatkoti, Pushkar ; Paul, Manoranjan. / Early Diagnosis of Alzheimer's Disease : A Multi-class Deep Learning Framework with Modified k-sparse Autoencoder Classification. Proceedings of the 2016 International Conference on Image and Vision Computing New Zealand (IVCNZ) . editor / Donald Bailey ; Gourab Sen Gupta ; Stephen Marsland. United States : IEEE, Institute of Electrical and Electronics Engineers, 2017. pp. 1-6
@inproceedings{ee183a8a784546f78bf97459a19e67d6,
title = "Early Diagnosis of Alzheimer's Disease: A Multi-class Deep Learning Framework with Modified k-sparse Autoencoder Classification",
abstract = "Successful, timely diagnosis of neuropsychiatric diseases is key to management. Research efforts in the area of diagnosis of Alzheimer's disease have used various aspects of computer-aided multi-class diagnosis approaches with varied degrees of success. However, there is still need for more efficient and reliable approaches to successful diagnosis of the disease. This research used deep learning framework with modified k-sparse autoencoder (mKSA) classification to locate neutrally degenerated areas of the brain MRI, low amyloid beta 1-42 imaging in cerebrospinal fluid (CSF) and positron emission tomography (PET) imaging of amyloid; each with a sample of 150 images. Results show a correlation between computational demarcation of infected regions and the images. Degeneration in the studied areas was evidenced by high phosphorylated t-/p-tau levels in CSF, regional hypometabolism fluorodeoxyglucose PET, and the presence of atrophy patterns. The use of mKSA algorithm in boosting classification helped to improve the classifier performance. The KSA method with deep learning framework is used for the first time to produce accurate results in diagnosis of Alzheimer's disease.",
keywords = "aKSA, Alzheimer's, deep learning, diagnosis, k-sparse, neuroimaging",
author = "Pushkar Bhatkoti and Manoranjan Paul",
note = "Imported on 03 May 2017 - DigiTool details were: publisher = USA: IEEE, 2016. Event dates (773o) = 21-22 November 2016; Parent title (773t) = Image and Vision Computing New Zealand (IVCNZ) International Conference.",
year = "2017",
month = "1",
day = "5",
doi = "10.1109/IVCNZ.2016.7804459",
language = "English",
pages = "1--6",
editor = "Donald Bailey and Gupta, {Gourab Sen} and Stephen Marsland",
booktitle = "Proceedings of the 2016 International Conference on Image and Vision Computing New Zealand (IVCNZ)",
publisher = "IEEE, Institute of Electrical and Electronics Engineers",
address = "United States",

}

Bhatkoti, P & Paul, M 2017, Early Diagnosis of Alzheimer's Disease: A Multi-class Deep Learning Framework with Modified k-sparse Autoencoder Classification. in D Bailey, GS Gupta & S Marsland (eds), Proceedings of the 2016 International Conference on Image and Vision Computing New Zealand (IVCNZ) ., 7804459, IEEE, Institute of Electrical and Electronics Engineers, United States, pp. 1-6, 2016 International Conference on Image and Vision Computing New Zealand, Palmerston North, New Zealand, 21/11/16. https://doi.org/10.1109/IVCNZ.2016.7804459

Early Diagnosis of Alzheimer's Disease : A Multi-class Deep Learning Framework with Modified k-sparse Autoencoder Classification. / Bhatkoti, Pushkar; Paul, Manoranjan.

Proceedings of the 2016 International Conference on Image and Vision Computing New Zealand (IVCNZ) . ed. / Donald Bailey; Gourab Sen Gupta; Stephen Marsland. United States : IEEE, Institute of Electrical and Electronics Engineers, 2017. p. 1-6 7804459.

Research output: Book chapter/Published conference paperConference paper

TY - GEN

T1 - Early Diagnosis of Alzheimer's Disease

T2 - A Multi-class Deep Learning Framework with Modified k-sparse Autoencoder Classification

AU - Bhatkoti, Pushkar

AU - Paul, Manoranjan

N1 - Imported on 03 May 2017 - DigiTool details were: publisher = USA: IEEE, 2016. Event dates (773o) = 21-22 November 2016; Parent title (773t) = Image and Vision Computing New Zealand (IVCNZ) International Conference.

PY - 2017/1/5

Y1 - 2017/1/5

N2 - Successful, timely diagnosis of neuropsychiatric diseases is key to management. Research efforts in the area of diagnosis of Alzheimer's disease have used various aspects of computer-aided multi-class diagnosis approaches with varied degrees of success. However, there is still need for more efficient and reliable approaches to successful diagnosis of the disease. This research used deep learning framework with modified k-sparse autoencoder (mKSA) classification to locate neutrally degenerated areas of the brain MRI, low amyloid beta 1-42 imaging in cerebrospinal fluid (CSF) and positron emission tomography (PET) imaging of amyloid; each with a sample of 150 images. Results show a correlation between computational demarcation of infected regions and the images. Degeneration in the studied areas was evidenced by high phosphorylated t-/p-tau levels in CSF, regional hypometabolism fluorodeoxyglucose PET, and the presence of atrophy patterns. The use of mKSA algorithm in boosting classification helped to improve the classifier performance. The KSA method with deep learning framework is used for the first time to produce accurate results in diagnosis of Alzheimer's disease.

AB - Successful, timely diagnosis of neuropsychiatric diseases is key to management. Research efforts in the area of diagnosis of Alzheimer's disease have used various aspects of computer-aided multi-class diagnosis approaches with varied degrees of success. However, there is still need for more efficient and reliable approaches to successful diagnosis of the disease. This research used deep learning framework with modified k-sparse autoencoder (mKSA) classification to locate neutrally degenerated areas of the brain MRI, low amyloid beta 1-42 imaging in cerebrospinal fluid (CSF) and positron emission tomography (PET) imaging of amyloid; each with a sample of 150 images. Results show a correlation between computational demarcation of infected regions and the images. Degeneration in the studied areas was evidenced by high phosphorylated t-/p-tau levels in CSF, regional hypometabolism fluorodeoxyglucose PET, and the presence of atrophy patterns. The use of mKSA algorithm in boosting classification helped to improve the classifier performance. The KSA method with deep learning framework is used for the first time to produce accurate results in diagnosis of Alzheimer's disease.

KW - aKSA

KW - Alzheimer's

KW - deep learning

KW - diagnosis

KW - k-sparse

KW - neuroimaging

UR - http://www.scopus.com/inward/record.url?scp=85010868068&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85010868068&partnerID=8YFLogxK

U2 - 10.1109/IVCNZ.2016.7804459

DO - 10.1109/IVCNZ.2016.7804459

M3 - Conference paper

SP - 1

EP - 6

BT - Proceedings of the 2016 International Conference on Image and Vision Computing New Zealand (IVCNZ)

A2 - Bailey, Donald

A2 - Gupta, Gourab Sen

A2 - Marsland, Stephen

PB - IEEE, Institute of Electrical and Electronics Engineers

CY - United States

ER -

Bhatkoti P, Paul M. Early Diagnosis of Alzheimer's Disease: A Multi-class Deep Learning Framework with Modified k-sparse Autoencoder Classification. In Bailey D, Gupta GS, Marsland S, editors, Proceedings of the 2016 International Conference on Image and Vision Computing New Zealand (IVCNZ) . United States: IEEE, Institute of Electrical and Electronics Engineers. 2017. p. 1-6. 7804459 https://doi.org/10.1109/IVCNZ.2016.7804459