Artificial Intelligence in the diagnosis of Alzheimer's disease
A minimally invasive blood test along with artificial intelligence (AI) may flag early-stage Alzheimer's disease (AD), raising the prospect of early intervention when effective treatments become available.
In a study, investigators used six AI methodologies, including Deep Learning, to assess blood leukocyte epigenomic biomarkers. They found more than 150 genetic differences among study participants with AD in comparison with participants who did not have AD.
All of the AI platforms were effective in predicting AD. Deep Learning's assessment of intragenic cytosine-phosphate-guanines (CpGs) had sensitivity and specificity rates of 97%.
The researchers note that the findings, if replicated in future studies, may help in providing AD diagnoses "much earlier" in the disease process.
The holy grail is to identify patients in the preclinical stage so effective early interventions, including new medications, can be studied and ultimately used.
The findings were published online March 31 in PLOS ONE.
The study included 24 individuals with late-onset AD (70.8% women; mean age, 83 years); 24 were deemed to be "cognitively healthy" (66.7% women; mean age, 80 years). About 500 ng of genomic DNA was extracted from whole-blood samples from each participant.
Using the Infinium MethylationEPIC BeadChip array, the samples were then examined for markers of methylation that would indicate the disease process has started.
Results showed that the AD group had 152 significantly differentially methylated CpGs in 171 genes in comparison with the non-AD group (false discovery rate P value <.05).
As a whole, using intragenic and intergenic/extragenic CpGs, the AI platforms were effective in predicting who had AD (area under the curve [AUC], ≥0.93). Using intragenic markers, the AUC for deep Learning was 0.99.
Altered genes that were found in the AD group included CR1L, CTSV, S1PR1, and LTB4R ― all of which have been previously linked with AD and dementi.
They also found the methylated genes CTSV and PRMT5, both of which have been previously associated with cardiovascular disease.