A Novelty Detection Approach to Effectively Predict Conversion from Mild Cognitive Impairment to Alzheimer’s Disease

Shuo Liu; Yi Cao; Junxiu Liu; Xuemei Ding; Damien Coyle

doi:10.1007/s13042-022-01570-2

A Novelty Detection Approach to Effectively Predict Conversion from Mild Cognitive Impairment to Alzheimer’s Disease

Shuo Liu
, Yi Cao
, Junxiu Liu
, Xuemei Ding
, Damien Coyle

University of Edinburgh

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

11 Citations (Scopus)

69 Downloads (Pure)

Abstract

Accurately recognising patients with progressive mild cognitive impairment (pMCI) who will develop Alzheimer’s disease (AD) in subsequent years is very important, as early identification of those patients will enable interventions to potentially reduce the number of those transitioning from MCI to AD. Most studies in this area have concentrated on high-dimensional neuroimaging data with supervised binary/multi-class classification algorithms. However, neuroimaging data is more costly to obtain than non-imaging, and healthcare datasets are normally imbalanced which may reduce classification performance and reliability. To address these challenges, we proposed a new strategy that employs unsupervised novelty detection (ND) techniques to predict pMCI from the AD Neuroimaging Initiative non-imaging data. ND algorithms, including the k-nearest neighbours (kNN), k-means, Gaussian mixture model (GMM), isolation forest (IF) and extreme learning machine (ELM), were employed and compared with supervised binary Support Vector Machine (SVM) and Random Forest (RF). We introduced optimisation with nested cross-validation and focused on maximising the adjusted F measure to ensure maximum generalisation of the proposed system by minimising false negative rates. Our extensive experimental results show that ND algorithms (0.727±0.029 kNN, 0.7179±0.0523 GMM, 0.7276±0.0281 ELM) obtained comparable performance to supervised binary SVM (0.7359±0.0451) with 20% stable MCI misclassification tolerance and were significantly better than RF (0.4771±0.0167). Moreover, we found that the non-invasive, readily obtainable, and cost-effective cognitive and functional assessment was the most efficient predictor for predicting the pMCI within 2 years with ND techniques. Importantly, we presented an accessible and cost-effective approach to pMCI prediction, which does not require labelled data.

Original language	English
Pages (from-to)	213-228
Number of pages	16
Journal	International Journal of Machine Learning and Cybernetics
Volume	14
Early online date	16 Jun 2022
DOIs	https://doi.org/10.1007/s13042-022-01570-2
Publication status	Published online - 16 Jun 2022

Bibliographical note

Funding Information:
This project was supported by Vice Chancellor Research Scholarship, Ulster University, the Alzheimer’s Research UK NI Networking, and the Global Challenges Research Fund Networking.

Funding Information:
This project was supported by Vice Chancellor Research Scholarship, Ulster University, the Alzheimer’s Research UK NI Networking, and the Global Challenges Research Fund Networking. We are grateful for access to the Tier 2 High Performance Computing resources provided by the Northern Ireland High Performance Computing (NI-HPC) facility, funded by the UK Engineering and Physical Sciences Research Council (EPSRC), Grant No. EP/T022175/1. Damien Coyle is grateful for a UKRI Turing AI Fellowship 2021-2025, funded by the The Alan Turing Institute and EPSRC, Grant No. EP/V025724/1. Data collection and sharing for this project was funded by the Alzheimer’s Disease Neuroimaging Initiative (ADNI) (National Institutes of Health Grant U01 AG024904) and DOD ADNI (Department of Defense award number W81XWH-12-2-0012). ADNI is funded by the National Institute on Aging, the National Institute of Biomedical Imaging and Bioengineering, and through generous contributions from the following: AbbVie, Alzheimer’s Association; Alzheimer’s Drug Discovery Foundation; Araclon Biotech; BioClinica, Inc.; Biogen; Bristol-Myers Squibb Company; CereSpir, Inc.; Cogstate; Eisai Inc.; Elan Pharmaceuticals, Inc.; Eli Lilly and Company; EuroImmun; F. Hoffmann-La Roche Ltd and its affiliated company Genentech, Inc.; Fujirebio; GE Healthcare; IXICO Ltd.; Janssen Alzheimer Immunotherapy Research & Development, LLC.; Johnson & Johnson Pharmaceutical Research & Development LLC.; Lumosity; Lundbeck; Merck & Co., Inc.; Meso Scale Diagnostics, LLC.; NeuroRx Research; Neurotrack Technologies; Novartis Pharmaceuticals Corporation; Pfizer Inc.; Piramal Imaging; Servier; Takeda Pharmaceutical Company; and Transition Therapeutics. The Canadian Institutes of Health Research is providing funds to support ADNI clinical sites in Canada. Private sector contributions are facilitated by the Foundation for the National Institutes of Health ( www.fnih.org ). The grantee organisation is the Northern California Institute for Research and Education, and the study is coordinated by the Alzheimer’s Therapeutic Research Institute at the University of Southern California. ADNI data are disseminated by the Laboratory for Neuro Imaging at the University of Southern California. Data used in preparation of this article were obtained from the Alzheimer’s Disease Neuroimaging Initiative (ADNI) database (adni.loni.usc.edu). As such, the investigators within the ADNI contributed to the design and implementation of ADNI and/or provided data but did not participate in analysis or writing of this report. A complete listing of ADNI investigators can be found at: http://adni.loni.usc.edu/wp-content/uploads/how_to_apply/ADNI_Acknowledgement_List.pdf

Publisher Copyright:
© 2022, The Author(s).

Funding

Funding Information: This project was supported by Vice Chancellor Research Scholarship, Ulster University, the Alzheimer’s Research UK NI Networking, and the Global Challenges Research Fund Networking. Funding Information: This project was supported by Vice Chancellor Research Scholarship, Ulster University, the Alzheimer’s Research UK NI Networking, and the Global Challenges Research Fund Networking. We are grateful for access to the Tier 2 High Performance Computing resources provided by the Northern Ireland High Performance Computing (NI-HPC) facility, funded by the UK Engineering and Physical Sciences Research Council (EPSRC), Grant No. EP/T022175/1. Damien Coyle is grateful for a UKRI Turing AI Fellowship 2021-2025, funded by the The Alan Turing Institute and EPSRC, Grant No. EP/V025724/1. Data collection and sharing for this project was funded by the Alzheimer’s Disease Neuroimaging Initiative (ADNI) (National Institutes of Health Grant U01 AG024904) and DOD ADNI (Department of Defense award number W81XWH-12-2-0012). ADNI is funded by the National Institute on Aging, the National Institute of Biomedical Imaging and Bioengineering, and through generous contributions from the following: AbbVie, Alzheimer’s Association; Alzheimer’s Drug Discovery Foundation; Araclon Biotech; BioClinica, Inc.; Biogen; Bristol-Myers Squibb Company; CereSpir, Inc.; Cogstate; Eisai Inc.; Elan Pharmaceuticals, Inc.; Eli Lilly and Company; EuroImmun; F. Hoffmann-La Roche Ltd and its affiliated company Genentech, Inc.; Fujirebio; GE Healthcare; IXICO Ltd.; Janssen Alzheimer Immunotherapy Research & Development, LLC.; Johnson & Johnson Pharmaceutical Research & Development LLC.; Lumosity; Lundbeck; Merck & Co., Inc.; Meso Scale Diagnostics, LLC.; NeuroRx Research; Neurotrack Technologies; Novartis Pharmaceuticals Corporation; Pfizer Inc.; Piramal Imaging; Servier; Takeda Pharmaceutical Company; and Transition Therapeutics. The Canadian Institutes of Health Research is providing funds to support ADNI clinical sites in Canada. Private sector contributions are facilitated by the Foundation for the National Institutes of Health ( www.fnih.org ). The grantee organisation is the Northern California Institute for Research and Education, and the study is coordinated by the Alzheimer’s Therapeutic Research Institute at the University of Southern California. ADNI data are disseminated by the Laboratory for Neuro Imaging at the University of Southern California. Data used in preparation of this article were obtained from the Alzheimer’s Disease Neuroimaging Initiative (ADNI) database (adni.loni.usc.edu). As such, the investigators within the ADNI contributed to the design and implementation of ADNI and/or provided data but did not participate in analysis or writing of this report. A complete listing of ADNI investigators can be found at: http://adni.loni.usc.edu/wp-content/uploads/how_to_apply/ADNI_Acknowledgement_List.pdf Publisher Copyright: © 2022, The Author(s).

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

SDG 3 Good Health and Well-being

Keywords

Novelty Detection
Alzheimer disease
mild cognitive impairment
coversion
One-class classification
Alzheimer’s disease
Mild cognitive impairment
Conversion
Novelty detection
Alzheimer's disease

Access to Document

10.1007/s13042-022-01570-2

s13042-022-01570-2Final published version, 1.28 MBLicence: CC BY

Cite this

@article{6a94363f8b5745628ec0f3086ca2693b,

title = "A Novelty Detection Approach to Effectively Predict Conversion from Mild Cognitive Impairment to Alzheimer{\textquoteright}s Disease",

abstract = "Accurately recognising patients with progressive mild cognitive impairment (pMCI) who will develop Alzheimer{\textquoteright}s disease (AD) in subsequent years is very important, as early identification of those patients will enable interventions to potentially reduce the number of those transitioning from MCI to AD. Most studies in this area have concentrated on high-dimensional neuroimaging data with supervised binary/multi-class classification algorithms. However, neuroimaging data is more costly to obtain than non-imaging, and healthcare datasets are normally imbalanced which may reduce classification performance and reliability. To address these challenges, we proposed a new strategy that employs unsupervised novelty detection (ND) techniques to predict pMCI from the AD Neuroimaging Initiative non-imaging data. ND algorithms, including the k-nearest neighbours (kNN), k-means, Gaussian mixture model (GMM), isolation forest (IF) and extreme learning machine (ELM), were employed and compared with supervised binary Support Vector Machine (SVM) and Random Forest (RF). We introduced optimisation with nested cross-validation and focused on maximising the adjusted F measure to ensure maximum generalisation of the proposed system by minimising false negative rates. Our extensive experimental results show that ND algorithms (0.727±0.029 kNN, 0.7179±0.0523 GMM, 0.7276±0.0281 ELM) obtained comparable performance to supervised binary SVM (0.7359±0.0451) with 20\% stable MCI misclassification tolerance and were significantly better than RF (0.4771±0.0167). Moreover, we found that the non-invasive, readily obtainable, and cost-effective cognitive and functional assessment was the most efficient predictor for predicting the pMCI within 2 years with ND techniques. Importantly, we presented an accessible and cost-effective approach to pMCI prediction, which does not require labelled data.",

keywords = "Novelty Detection, Alzheimer disease, mild cognitive impairment, coversion, One-class classification, Alzheimer{\textquoteright}s disease, Mild cognitive impairment, Conversion, Novelty detection, Alzheimer's disease",

author = "Shuo Liu and Yi Cao and Junxiu Liu and Xuemei Ding and Damien Coyle",

note = "Funding Information: This project was supported by Vice Chancellor Research Scholarship, Ulster University, the Alzheimer{\textquoteright}s Research UK NI Networking, and the Global Challenges Research Fund Networking. Funding Information: This project was supported by Vice Chancellor Research Scholarship, Ulster University, the Alzheimer{\textquoteright}s Research UK NI Networking, and the Global Challenges Research Fund Networking. We are grateful for access to the Tier 2 High Performance Computing resources provided by the Northern Ireland High Performance Computing (NI-HPC) facility, funded by the UK Engineering and Physical Sciences Research Council (EPSRC), Grant No. EP/T022175/1. Damien Coyle is grateful for a UKRI Turing AI Fellowship 2021-2025, funded by the The Alan Turing Institute and EPSRC, Grant No. EP/V025724/1. Data collection and sharing for this project was funded by the Alzheimer{\textquoteright}s Disease Neuroimaging Initiative (ADNI) (National Institutes of Health Grant U01 AG024904) and DOD ADNI (Department of Defense award number W81XWH-12-2-0012). ADNI is funded by the National Institute on Aging, the National Institute of Biomedical Imaging and Bioengineering, and through generous contributions from the following: AbbVie, Alzheimer{\textquoteright}s Association; Alzheimer{\textquoteright}s Drug Discovery Foundation; Araclon Biotech; BioClinica, Inc.; Biogen; Bristol-Myers Squibb Company; CereSpir, Inc.; Cogstate; Eisai Inc.; Elan Pharmaceuticals, Inc.; Eli Lilly and Company; EuroImmun; F. Hoffmann-La Roche Ltd and its affiliated company Genentech, Inc.; Fujirebio; GE Healthcare; IXICO Ltd.; Janssen Alzheimer Immunotherapy Research \& Development, LLC.; Johnson \& Johnson Pharmaceutical Research \& Development LLC.; Lumosity; Lundbeck; Merck \& Co., Inc.; Meso Scale Diagnostics, LLC.; NeuroRx Research; Neurotrack Technologies; Novartis Pharmaceuticals Corporation; Pfizer Inc.; Piramal Imaging; Servier; Takeda Pharmaceutical Company; and Transition Therapeutics. The Canadian Institutes of Health Research is providing funds to support ADNI clinical sites in Canada. Private sector contributions are facilitated by the Foundation for the National Institutes of Health ( www.fnih.org ). The grantee organisation is the Northern California Institute for Research and Education, and the study is coordinated by the Alzheimer{\textquoteright}s Therapeutic Research Institute at the University of Southern California. ADNI data are disseminated by the Laboratory for Neuro Imaging at the University of Southern California. Data used in preparation of this article were obtained from the Alzheimer{\textquoteright}s Disease Neuroimaging Initiative (ADNI) database (adni.loni.usc.edu). As such, the investigators within the ADNI contributed to the design and implementation of ADNI and/or provided data but did not participate in analysis or writing of this report. A complete listing of ADNI investigators can be found at: http://adni.loni.usc.edu/wp-content/uploads/how\_to\_apply/ADNI\_Acknowledgement\_List.pdf Publisher Copyright: {\textcopyright} 2022, The Author(s).",

year = "2022",

month = jun,

day = "16",

doi = "10.1007/s13042-022-01570-2",

language = "English",

volume = "14",

pages = "213--228",

journal = "International Journal of Machine Learning and Cybernetics",

issn = "1868-8071",

publisher = "Springer Science + Business Media",

}

TY - JOUR

T1 - A Novelty Detection Approach to Effectively Predict Conversion from Mild Cognitive Impairment to Alzheimer’s Disease

AU - Liu, Shuo

AU - Cao, Yi

AU - Liu, Junxiu

AU - Ding, Xuemei

AU - Coyle, Damien

N1 - Funding Information: This project was supported by Vice Chancellor Research Scholarship, Ulster University, the Alzheimer’s Research UK NI Networking, and the Global Challenges Research Fund Networking. Funding Information: This project was supported by Vice Chancellor Research Scholarship, Ulster University, the Alzheimer’s Research UK NI Networking, and the Global Challenges Research Fund Networking. We are grateful for access to the Tier 2 High Performance Computing resources provided by the Northern Ireland High Performance Computing (NI-HPC) facility, funded by the UK Engineering and Physical Sciences Research Council (EPSRC), Grant No. EP/T022175/1. Damien Coyle is grateful for a UKRI Turing AI Fellowship 2021-2025, funded by the The Alan Turing Institute and EPSRC, Grant No. EP/V025724/1. Data collection and sharing for this project was funded by the Alzheimer’s Disease Neuroimaging Initiative (ADNI) (National Institutes of Health Grant U01 AG024904) and DOD ADNI (Department of Defense award number W81XWH-12-2-0012). ADNI is funded by the National Institute on Aging, the National Institute of Biomedical Imaging and Bioengineering, and through generous contributions from the following: AbbVie, Alzheimer’s Association; Alzheimer’s Drug Discovery Foundation; Araclon Biotech; BioClinica, Inc.; Biogen; Bristol-Myers Squibb Company; CereSpir, Inc.; Cogstate; Eisai Inc.; Elan Pharmaceuticals, Inc.; Eli Lilly and Company; EuroImmun; F. Hoffmann-La Roche Ltd and its affiliated company Genentech, Inc.; Fujirebio; GE Healthcare; IXICO Ltd.; Janssen Alzheimer Immunotherapy Research & Development, LLC.; Johnson & Johnson Pharmaceutical Research & Development LLC.; Lumosity; Lundbeck; Merck & Co., Inc.; Meso Scale Diagnostics, LLC.; NeuroRx Research; Neurotrack Technologies; Novartis Pharmaceuticals Corporation; Pfizer Inc.; Piramal Imaging; Servier; Takeda Pharmaceutical Company; and Transition Therapeutics. The Canadian Institutes of Health Research is providing funds to support ADNI clinical sites in Canada. Private sector contributions are facilitated by the Foundation for the National Institutes of Health ( www.fnih.org ). The grantee organisation is the Northern California Institute for Research and Education, and the study is coordinated by the Alzheimer’s Therapeutic Research Institute at the University of Southern California. ADNI data are disseminated by the Laboratory for Neuro Imaging at the University of Southern California. Data used in preparation of this article were obtained from the Alzheimer’s Disease Neuroimaging Initiative (ADNI) database (adni.loni.usc.edu). As such, the investigators within the ADNI contributed to the design and implementation of ADNI and/or provided data but did not participate in analysis or writing of this report. A complete listing of ADNI investigators can be found at: http://adni.loni.usc.edu/wp-content/uploads/how_to_apply/ADNI_Acknowledgement_List.pdf Publisher Copyright: © 2022, The Author(s).

PY - 2022/6/16

Y1 - 2022/6/16

N2 - Accurately recognising patients with progressive mild cognitive impairment (pMCI) who will develop Alzheimer’s disease (AD) in subsequent years is very important, as early identification of those patients will enable interventions to potentially reduce the number of those transitioning from MCI to AD. Most studies in this area have concentrated on high-dimensional neuroimaging data with supervised binary/multi-class classification algorithms. However, neuroimaging data is more costly to obtain than non-imaging, and healthcare datasets are normally imbalanced which may reduce classification performance and reliability. To address these challenges, we proposed a new strategy that employs unsupervised novelty detection (ND) techniques to predict pMCI from the AD Neuroimaging Initiative non-imaging data. ND algorithms, including the k-nearest neighbours (kNN), k-means, Gaussian mixture model (GMM), isolation forest (IF) and extreme learning machine (ELM), were employed and compared with supervised binary Support Vector Machine (SVM) and Random Forest (RF). We introduced optimisation with nested cross-validation and focused on maximising the adjusted F measure to ensure maximum generalisation of the proposed system by minimising false negative rates. Our extensive experimental results show that ND algorithms (0.727±0.029 kNN, 0.7179±0.0523 GMM, 0.7276±0.0281 ELM) obtained comparable performance to supervised binary SVM (0.7359±0.0451) with 20% stable MCI misclassification tolerance and were significantly better than RF (0.4771±0.0167). Moreover, we found that the non-invasive, readily obtainable, and cost-effective cognitive and functional assessment was the most efficient predictor for predicting the pMCI within 2 years with ND techniques. Importantly, we presented an accessible and cost-effective approach to pMCI prediction, which does not require labelled data.

AB - Accurately recognising patients with progressive mild cognitive impairment (pMCI) who will develop Alzheimer’s disease (AD) in subsequent years is very important, as early identification of those patients will enable interventions to potentially reduce the number of those transitioning from MCI to AD. Most studies in this area have concentrated on high-dimensional neuroimaging data with supervised binary/multi-class classification algorithms. However, neuroimaging data is more costly to obtain than non-imaging, and healthcare datasets are normally imbalanced which may reduce classification performance and reliability. To address these challenges, we proposed a new strategy that employs unsupervised novelty detection (ND) techniques to predict pMCI from the AD Neuroimaging Initiative non-imaging data. ND algorithms, including the k-nearest neighbours (kNN), k-means, Gaussian mixture model (GMM), isolation forest (IF) and extreme learning machine (ELM), were employed and compared with supervised binary Support Vector Machine (SVM) and Random Forest (RF). We introduced optimisation with nested cross-validation and focused on maximising the adjusted F measure to ensure maximum generalisation of the proposed system by minimising false negative rates. Our extensive experimental results show that ND algorithms (0.727±0.029 kNN, 0.7179±0.0523 GMM, 0.7276±0.0281 ELM) obtained comparable performance to supervised binary SVM (0.7359±0.0451) with 20% stable MCI misclassification tolerance and were significantly better than RF (0.4771±0.0167). Moreover, we found that the non-invasive, readily obtainable, and cost-effective cognitive and functional assessment was the most efficient predictor for predicting the pMCI within 2 years with ND techniques. Importantly, we presented an accessible and cost-effective approach to pMCI prediction, which does not require labelled data.

KW - Novelty Detection

KW - Alzheimer disease

KW - mild cognitive impairment

KW - coversion

KW - One-class classification

KW - Alzheimer’s disease

KW - Mild cognitive impairment

KW - Conversion

KW - Novelty detection

KW - Alzheimer's disease

UR - https://www.scopus.com/pages/publications/85132244080

U2 - 10.1007/s13042-022-01570-2

DO - 10.1007/s13042-022-01570-2

M3 - Article

SN - 1868-8071

VL - 14

SP - 213

EP - 228

JO - International Journal of Machine Learning and Cybernetics

JF - International Journal of Machine Learning and Cybernetics

ER -

A Novelty Detection Approach to Effectively Predict Conversion from Mild Cognitive Impairment to Alzheimer’s Disease

Abstract

Bibliographical note

Funding

UN SDGs

Keywords

Access to Document

Other files and links

Fingerprint

Efficient level set method for novelty detection

Cite this

A Novelty Detection Approach to Effectively Predict Conversion from Mild Cognitive Impairment to Alzheimer’s Disease

Abstract

Bibliographical note

Funding

UN SDGs

Keywords

Access to Document

Other files and links

Fingerprint

Student theses

Efficient level set method for novelty detection

Cite this