Objective: The aim of the present work was to investigate the features of the elderly population aged ≥65 yrs and with deteriorative mild cognitive impairment (MCI) due to Alzheimer’s disease (AD) to establish a prediction model.
Methods: A total of 105 patients aged ≥65 yrs and with MCI were followed up, with a collection of 357 features, which were derived from the demographic characteristics, hematological indicators (serum Aβ1-40, Aβ1-42, P-tau and MCP-1 levels, APOE gene), and multimodal brain Magnetic Resonance Imaging (MRI) imaging indicators of 116 brain regions (ADC, FA and CBF values). Cognitive function was followed up for 2 yrs. Based on the Python platform Anaconda, 105 patients were randomly divided into a training set (70%) and a test set (30%) by analyzing all features through a random forest algorithm, and a prediction model was established for the form of rapidly deteriorating MCI.
Results: Of the 105 patients enrolled, 41 deteriorated, and 64 did not come within 2 yrs. Model 1 was established based on demographic characteristics, hematological indicators and multi-modal MRI image features, the accuracy of the training set being 100%, the accuracy of the test set 64%, sensitivity 50%, specificity 67%, and AUC 0.72. Model 2 was based on the first five features (APOE4 gene, FA value of left fusiform gyrus, FA value of left inferior temporal gyrus, FA value of left parahippocampal gyrus, ADC value of right calcarine fissure as surrounding cortex), the accuracy of the training set being 100%, the accuracy of the test set 85%, sensitivity 91%, specificity 80% and AUC 0.96. Model 3 was based on the first four features of Model 1, the accuracy of the training set is 100%, the accuracy of the test set 97%, sensitivity100%, specificity 95% and AUC 0.99. Model 4 was based on the first three characteristics of Model 1, the accuracy of the training set being 100%, the accuracy of the test set 94%, sensitivity 92%, specificity 94% and AUC 0.96. Model 5 was based on the hematological characteristics, the accuracy of the training set is 100%, the accuracy of the test set 91%, sensitivity 100%, specificity 88% and AUC 0.97. The models based on the demographic characteristics, imaging characteristics FA, CBF and ADC values had lower sensitivity and specificity.
Conclusion: Model 3, which has four important predictive characteristics, can predict the rapidly deteriorating MCI due to AD in the community.
Keywords: Machine learning, random forest, Alzheimer’s disease, mild cognitive impairment, prediction model, magnetic resonance imaging.
[http://dx.doi.org/10.1186/s13195-017-0283-5] [PMID: 28793924]
[http://dx.doi.org/10.1016/j.jval.2019.01.006] [PMID: 30975395]
[http://dx.doi.org/10.1016/B978-0-12-804766-8.00009-1] [PMID: 31753130]
[http://dx.doi.org/10.1016/j.jalz.2017.12.006] [PMID: 29433981]
[http://dx.doi.org/10.1016/j.steroids.2019.108548] [PMID: 31805293]
[http://dx.doi.org/10.1016/j.jalz.2018.02.018] [PMID: 29653606]
[http://dx.doi.org/10.1002/ana.26071] [PMID: 33772866]
[http://dx.doi.org/10.1016/bs.pmbts.2019.04.002] [PMID: 31481168]
[http://dx.doi.org/10.1021/acschemneuro.1c00472] [PMID: 34723463]
[http://dx.doi.org/10.1161/STROKEAHA.118.024293] [PMID: 30890116]
[http://dx.doi.org/10.3389/fpubh.2020.00357] [PMID: 32719767]
[http://dx.doi.org/10.1016/j.jbi.2021.103690] [PMID: 33540075]
[http://dx.doi.org/10.1093/europace/euz036] [PMID: 31067312]
[http://dx.doi.org/10.3748/wjg.v27.i21.2910] [PMID: 34135561]
[http://dx.doi.org/10.1176/ajp.139.9.1136] [PMID: 7114305]
[http://dx.doi.org/10.1007/s11325-021-02327-x ] [PMID: 33792886]
[http://dx.doi.org/10.1016/j.media.2007.06.004] [PMID: 17659998]
[http://dx.doi.org/10.1016/j.neuroimage.2015.07.075] [PMID: 26241684]
[http://dx.doi.org/10.21037/apm-21-2013] [PMID: 34628897]
[http://dx.doi.org/10.3390/brainsci11040453] [PMID: 33918453]
[http://dx.doi.org/10.1177/0269881120972331] [PMID: 33586518]
[http://dx.doi.org/10.1136/jnnp.2010.231555] [PMID: 21493755]
[http://dx.doi.org/10.1038/s41598-020-77023-z] [PMID: 33208795]
[http://dx.doi.org/10.1016/j.neuroscience.2020.04.011] [PMID: 32304722]
[http://dx.doi.org/10.1002/ana.26237] [PMID: 34590340]
[http://dx.doi.org/10.1186/s13148-020-00916-3] [PMID: 32854783]
[http://dx.doi.org/10.1002/aur.2609] [PMID: 34529363]
[http://dx.doi.org/10.1007/s12264-020-00498-3] [PMID: 32350798]
[http://dx.doi.org/10.1016/j.wneu.2020.12.136] [PMID: 33412321]