Combining PET with MRI to improve predictions of progression from mild cognitive impairment to Alzheimer's disease: an exploratory radiomic analysis study

Abstract

Background: This study aimed to explore the potential of a combination of 18F-fluorodeoxyglucose positron emission tomography (¹⁸F-FDG PET) and magnetic resonance imaging (MRI) to improve predictions of conversion from mild cognitive impairment (MCI) to Alzheimer's disease (AD). The predictive performances and specific associated biomarkers of these imaging techniques used alone (single-modality imaging) and in combination (dual-modality imaging) were compared.

Methods: This study enrolled 377 patients with MCI and 94 healthy control participants from 2 medical centers. Enrolment was based on the patients' brain MRI and PET images. Radiomic analysis was performed to evaluate the predictive performance of dual-modality ¹⁸F-FDG PET and MRI scans. Regions of interest (ROIs) were determined using an a priori brain atlas. Radiomic features in these ROIs were extracted from the MRI and ¹⁸F-FDG PET scan data. These features were either concatenated or used separately to select features and construct Cox regression models for prediction in each modality. Harrell's concordance index (C-index) was then used to assess the predictive accuracies of the resulting models, and correlations between the MRI and ¹⁸F-FDG PET features were evaluated.

Results: The C-indices for the two test datasets were 0.77 and 0.80 for dual-modality ¹⁸F-FDG PET/MRI, 0.75 and 0.73 for single-modality ¹⁸F-FDG PET, and 0.74 and 0.76 for single-modality MRI. In addition, there was a significant correlation between the crucial image signatures of the different modalities.

Conclusions: These results indicate the value of imaging features in monitoring the progress of MCI in populations at high risk of developing AD. However, the incremental benefit of combining ¹⁸F-FDG PET and MRI is limited, and radiomic analysis of a single modality may yield acceptable predictive results.

Keywords: Cox model; Mild cognitive impairment (MCI); magnetic resonance imaging (MRI); positron emission tomography (PET); radiology.

Figure 1

The framework of the experimental design in our study. PET, positron emission tomography; MRI, magnetic resonance imaging; AAL, anatomical automatic labeling.

Figure 2

The different brain regions of the features conserved by different radiomic models: (A) the radiomic PET model, (B) the radiomic MRI model, (C) the dual-modality model. PET, positron emission tomography; MRI, magnetic resonance imaging; L, left; R, right; MTG, middle temporal gyrus; HIP, hippocampus; PHG, parahippocampal gyrus; MFG, middle frontal gyrus; PCUN, precuneus; DCG, median cingulate and paracingulate gyrus; SMG, supramarginal gyrus; SFGmed, superior frontal gyrus, medial; IFGtriang, inferior frontal gyrus, triangular part; IPL, inferior parietal; SMA, supplementary motor area; ITG, inferior temporal gyrus; MOG, middle occipital gyrus.

Figure 3

Hazard ratios for different predictors and Kaplan-Meier survival curves for each model. (A) Hazard ratios for different predictors in the PET model with 4 features from the hippocampus and parahippocampal gyrus on PET images as predictors (Fea. 6, Fea. 7, Fea. 8, and Fea. 9, corresponding to features 6, 7, 8, and 9, respectively, in Table S2). Global P value (log-rank), 1.5514e-20; Akaike information criterion (AIC), 1,584.59; C-index, 0.75. (B) Risk stratification of the test dataset in the PET model (log-rank test, P=0.00027). (C) Hazard ratios in the MRI model with 4 features from the hippocampus and parahippocampal gyrus in MRI images as predictors (Fea. 20, Fea. 21, Fea. 22, and Fea. 23, corresponding to features 20, 21, 22, and 23, respectively, in Table S2). Global P value (log-rank), 1.5955e-16; AIC, 1,603.48; C-index, 0.73. (D) Risk stratification of the test dataset in the MRI model (log-rank test, P=0.007). (E) Hazard ratios in the combined model. Global P value (log-rank), 4.4302e-22; AIC, 1,572.27; C-index, 0.77. Predictors were a combination of features from the hippocampus and parahippocampal gyrus on the PET and MRI images. (F) Risk stratification of the test dataset in the combined model (log-rank test, P=0.00073). **, P<0.01; ***, P<0.001. PET, positron emission tomography; MRI, magnetic resonance imaging.

Figure 4

Correlations between 13 conserved features in the PET model and 12 conserved features in the MRI model in cohort A (A) and cohort B (B). The color bar scale represents -log P values. Vertical axis numbers 1–13 represent features in the PET model [consistent with Table S1 (a)], and horizontal axis numbers 1–12 represent features in the MRI model [consistent with Table S1 (b)]. PET, positron emission tomography; MRI, magnetic resonance imaging.