Abstract
173
Objectives To evaluate whether the diagnoses of cognitively normal, mild cognitive impairment(MCI) and Alzheimer's disease(AD) based on clinical assessment correlate with findings on FDG PET imaging using CortexID analysis of data from ADNI.
Methods Two radiology residents and a staff radiologist reviewed CortexID screenshots of FDG PET brain images of ADNI patients and categorized them as: 1. Normal 2. AD 3. Frontotemporal dementia 4. Dementia with Lewy bodies. This data was then compared to the diagnoses based on clinical assessment. GE PET CortexID uses 3-dimensional stereotactic surface projections to visualize FDG PET brain studies with respect to age-matched normal cohorts, thereby standardizing the diagnosis of cerebral metabolic deficiencies.
Results CortexID analysis was performed on FDG PET brain images of 126 patients with AD, 164 patients with MCI and 98 cognitively normal patients based on clinical assessment. The results are:
Conclusions Significant discrepancy exists between the diagnoses of ADNI patients based on clinical assessment and the FDG PET findings. The greatest discrepancy was found in the AD group. Discrepancy was also seen in the cognitively normal patients with 8.9% having FDG PET findings of AD. Long term follow up and further investigation is needed to establish the significance of these findings and to determine if patients should be grouped according to FDG PET findings or clinical assessment. Conclusions drawn from studies using grouping based on clinical assessment alone should be interpreted with caution