ABSTRACT
Alzheimer Disease (AD) is a progressive neurodegenerative disease that can significantly impair cognition and memory. AD is the leading cause of dementia and affects one in ten people age 65 and older. Current diagnoses methods of AD heavily rely on the use of Magnetic Resonance Imaging (MRI) since non-imaging methods can vary widely leading to inaccurate diagnoses. Furthermore, recent research has revealed a substage of AD, Mild Cognitive Impairment (MCI), that is characterized by symptoms between normal cognition and dementia which makes it more prone to misdiagnosis.
A large battery of clinical variables are currently used to detect cognitive impairment and classify early mild cognitive impairment (EMCI), late mild cognitive impairment (LMCI), and AD from cognitive normal (CN) patients. The goal of this study was to derive a simplified risk-stratification algorithm for diagnosis and identify a few significant clinical variables that can accurately classify these four groups using an empirical deep learning approach. Over 100 variables that included neuropsychological/neurocognitive tests, demographics, genetic factors, and blood biomarkers were collected from EMCI, LMCI, AD, and CN patients from the Alzheimer’s Disease Neuroimaging Initiative (ADNI) database. Feature engineering was performed with 5 different methods and a neural network was trained on 90% of the data and tested on 10% using 10-fold cross validation. Prediction performance used area under the curve (AUC) of the receiver operating characteristic analysis.
The five different feature selection methods consistently yielded the top classifiers to be the Clinical Dementia Rating Scale - Sum of Boxes (CDRSB), Delayed total recall (LDELTOTAL), Modified Preclinical Alzheimer Cognitive Composite with Trails test (mPACCtrailsB), the Modified Preclinical Alzheimer Cognitive Composite with Digit test (mPACCdigit), and Mini-Mental State Examination (MMSE). The best classification model yielded an AUC of 0.984, and the simplified risk-stratification score yielded an AUC of 0.963 on the test dataset.
Our results show that this deep-learning algorithm and simplified risk score derived from our deep-learning algorithm accurately diagnose EMCI, LMCI, AD and CN patients using a few commonly available neurocognitive tests. The project was successful in creating an accurate, clinically translatable risk-stratified scoring aid for primary care providers to diagnose AD in a fast and inexpensive manner.
Competing Interest Statement
The authors have declared no competing interest.
Funding Statement
None
Author Declarations
I confirm all relevant ethical guidelines have been followed, and any necessary IRB and/or ethics committee approvals have been obtained.
Yes
The details of the IRB/oversight body that provided approval or exemption for the research described are given below:
1) Approval was received by the ADNI data sharing and publications committee (DPC) after registration. 2) No Registration ID or approval code was given. 3) This non-human subject research study was exempt from IRB approval. However, the original data collection procedures were in accordance with an IRB. According to ADNI: "All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards.”
All necessary patient/participant consent has been obtained and the appropriate institutional forms have been archived.
Yes
I understand that all clinical trials and any other prospective interventional studies must be registered with an ICMJE-approved registry, such as ClinicalTrials.gov. I confirm that any such study reported in the manuscript has been registered and the trial registration ID is provided (note: if posting a prospective study registered retrospectively, please provide a statement in the trial ID field explaining why the study was not registered in advance).
Yes
I have followed all appropriate research reporting guidelines and uploaded the relevant EQUATOR Network research reporting checklist(s) and other pertinent material as supplementary files, if applicable.
Yes
Footnotes
Funding None
Disclaimer NA
Updated Figure 3 and made Figure 2 clearer.
Data Availability
All data used in this study is publicly available on the ADNI (Alzheimer's Disease Neuroimaging Initiative website. Due to the public and de-identified nature of the data, an IRB was not needed. This is non-human subject research.