Prediction of amyloid PET positivity via machine learning algorithms trained with EDTA-based blood amyloid-β oligomerization data

Table 1 Demographic characteristics of participants with amyloid PET positive and negative findings

Amyloid PET	Participants (n)		Age (y) (mean ± SD)	EDTA-based MDS-OAβ value * (mean ± SD)	APOE 4 allele (no. of APOE 4 allele-positive patients/total no. of patients)	MMSE ** [mean score ± SD (no. of patients)]
Positive	226	87:139	68.4 ± 9.7	1.057 ± 0.242	111/193	17.9 ± 7.4 (213)
Negative	86	43:43	70.8 ± 8.8	0.972 ± 0.226	28/70	22.1 ± 5.6 (76)
Total	312	130:182	69.0 ± 9.5	1.034 ± 0.240	139/263	19.0 ± 7.3 (289)

Of the 312 participants, 263 participants were tested for apolipoprotein E (APOE) and 289 participants were administered the Mini-Mental Status Examination (MMSE). APOE was considered “positive” when at least one APOE epsilon 4 allele existed
PET, positron emission tomography; APOE, apolipoprotein E; SD, standard deviation; EDTA, ethylenediaminetetraacetic acid; MDS-OAβ, Multimer Detection System-Oligomeric Amyloid-β; MMSE, Mini-Mental State Examination
*The difference in MDS-OAβ values (p-value = 0.005)
**MMSE scores (p-value < 0.0001) between the amyloid PET positive and negative groups, based on the Student t-test

ISSN: 1472-6947