Combined quantitative amyloid-ß PET and structural MRI features improve Alzheimer's Disease classification in random forest model - A multicenter study.

RATIONALE AND OBJECTIVES: Prior to clinical presentations of Alzheimer's Disease (AD), neuropathological changes, such as amyloid-ß and brain atrophy, have accumulated at the earlier stages of the disease. The combination of such biomarkers assessed by multiple modalities commonly improves the likelihood of AD etiology. We aimed to explore the discriminative ability of Aß PET features and whether combining Aß PET and structural MRI features can improve the classification performance of the machine learning model in older healthy control (OHC) and mild cognitive impairment (MCI) from AD. MATERIAL AND METHODS: We collected 94 AD patients, 82 MCI patients, and 85 OHC from three different cohorts. 17 global/regional Aß features in Centiloid, 122 regional volume, and 68 regional cortical thickness were extracted as imaging features. Single or combined modality features were used to train the random forest model on the testing set. The top 10 features were sorted based on the Gini index in each binary classification. RESULTS: The results showed that AUC scores were 0.81/0.86 and 0.69/0.68 using sMRI/Aß PET features on the testing set in differentiating OHC and MCI from AD. The performance was improved while combining two-modality features with an AUC of 0.89 and an AUC of 0.71 in two classifications. Compared to sMRI features, particular Aß PET features contributed more to differentiating AD from others. CONCLUSION: Our study demonstrated the discriminative ability of Aß PET features in differentiating AD from OHC and MCI. A combination of Aß PET and structural MRI features can improve the RF model performance.

Differential centiloid scale normalization techniques: comparison between hybrid PET/MRI and independently acquired MRI.

OBJECTIVE: Centiloid (CL) scales play an important role in semiquantitative analyses of amyloid-ß (Aß) PET. CLs are derived from the standardized uptake value ratio (SUVR), which needs Aß positron emission tomography (PET) normalization processing. There are two methods to collect the T1-weighted imaging (T1WI) for normalization: (i) anatomical standardization using simultaneously acquired T1WI (PET/MRI), usually adapted to PET images from PET/MRI scanners, and (ii) T1WI from a separate examination (PET + MRI), usually adapted to PET images from PET/CT scanners. This study aimed to elucidate the correlations and differences in CLs between when using the above two T1WI collection methods. METHODS: Among patients who underwent Aß PET/MRI (using 11C-Pittuberg compound B (11C-PiB) or 18F-flutemetamol (18F-FMM)) at our institution from 2015 to 2023, we selected 49 patients who also underwent other additional MRI examinations, including T1WI for anatomic standardization within 3 years. Thirty-one of them underwent 11C-PiB PET/MRI, and 18 participants underwent 18F-FMM PET/MRI. Twenty-five of them, additional MRI acquisition parameters were identical to simultaneous MRI during PET, and 24 participants were different. After normalization using PET/MRI or PET + MRI method each, SUVR was measured using the Global Alzheimer's Association Initiative Network cerebral cortical and striatum Volume of Interest templates (VOI) and whole cerebellum VOI. Subsequently, CLs were calculated using the previously established equations for each Aß PET tracer. RESULTS: Between PET/MRI and PET + MRI methods, CLs correlated linearly in 11C-PiB PET (y = 1.00x - 0.11, R2 = 0.999), 18F-FMM PET (y = 0.97x - 0.12, 0.997), identical additional MRI acquisition (y = 1.00x + 0.33, 0.999), different acquisition (y = 0.98x - 0.43, 0.997), and entire study group (y = 1.00x - 0.24, 0.999). Wilcoxon signed-rank test revealed no significant differences: 11C-PiB (p = 0.49), 18F-FMM (0.08), and whole PET (0.46). However, significant differences were identified in identical acquisition (p = 0.04) and different acquisition (p = 0.02). Bland-Altman analysis documented only a small bias between PET/MRI and PET + MRI in 11C-PiB PET, 18F-FMM PET, identical additional MRI acquisition, different acquisition, and whole PET (- 0.05, 0.67, - 0.30, 0.78, and 0.21, respectively). CONCLUSIONS: Anatomical standardizations using PET/MRI and using PET + MRI can lead to almost equivalent CL. The CL values obtained using PET/MRI or PET + MRI normalization methods are consistent and comparable in clinical studies.

Deep Learning-Driven Estimation of Centiloid Scales from Amyloid PET Images with 11C-PiB and 18F-Labeled Tracers in Alzheimer's Disease.

BACKGROUND: Standard methods for deriving Centiloid scales from amyloid PET images are time-consuming and require considerable expert knowledge. We aimed to develop a deep learning method of automating Centiloid scale calculations from amyloid PET images with 11C-Pittsburgh Compound-B (PiB) tracer and assess its applicability to 18F-labeled tracers without retraining. METHODS: We trained models on 231 11C-PiB amyloid PET images using a 50-layer 3D ResNet architecture. The models predicted the Centiloid scale, and accuracy was assessed using mean absolute error (MAE), linear regression analysis, and Bland-Altman plots. RESULTS: The MAEs for Alzheimer's disease (AD) and young controls (YC) were 8.54 and 2.61, respectively, using 11C-PiB, and 8.66 and 3.56, respectively, using 18F-NAV4694. The MAEs for AD and YC were higher with 18F-florbetaben (39.8 and 7.13, respectively) and 18F-florbetapir (40.5 and 12.4, respectively), and the error rate was moderate for 18F-flutemetamol (21.3 and 4.03, respectively). Linear regression yielded a slope of 1.00, intercept of 1.26, and R2 of 0.956, with a mean bias of -1.31 in the Centiloid scale prediction. CONCLUSIONS: We propose a deep learning means of directly predicting the Centiloid scale from amyloid PET images in a native space. Transferring the model trained on 11C-PiB directly to 18F-NAV4694 without retraining was feasible.

Amyloid brain-dedicated PET images can diagnose Alzheimer's pathology with Centiloid Scale.

PURPOSE: To evaluate whether the Centiloid Scale may be used to diagnose Alzheimer's Disease (AD) pathology effectively with the only use of amyloid PET imaging modality from a brain-dedicated PET scanner. METHODS: This study included 26 patients with amyloid PET images with 3 different radiotracers. All patients were acquired both on a PET/CT and a brain-dedicated PET scanner (CareMiBrain, CMB), from which 4 different reconstructions were implemented. A new pipeline was proposed and used for the PET image analysis based on the original Centiloid Scale processing pipeline, but with only PET images. The Youden's Index was employed to calculate the optimal cutoffs for diagnosis and evaluated by the AUC, accuracy, precision, and recall metrics. RESULTS: The Centiloid Scale (CL) processing pipeline was validated with and without the use of MR images. The CL cutoffs for AD pathology diagnosis on the PET/CT and the 4 CMB reconstructions were 34.4 ±â€¯2.2, 43.5 ±â€¯3.5, 51.9 ±â€¯12.5, 57.5 ±â€¯6.8 and 41.8 ±â€¯1.2 respectively. Overall, for these cutoffs all metrics obtained the maximum score. CONCLUSION: The Centiloid scale applied to PET images allows for AD pathology diagnosis. The CMB scanner can be used with the Centiloid scale to automatically assist in the diagnosis of AD pathology, relieving the large burden of neurodegenerative diseases on a traditional PET/CT.

Comparison of consistency in centiloid scale among different analytical methods in amyloid PET: the CapAIBL, VIZCalc, and Amyquant methods.

OBJECTIVE: The Centiloid (CL) scale is a standardized measure for quantifying amyloid deposition in amyloid positron emission tomography (PET) imaging. We aimed to assess the agreement among 3 CL calculation methods: CapAIBL, VIZCalc, and Amyquant. METHODS: This study included 192 participants (mean age: 71.5 years, range: 50-87 years), comprising 55 with Alzheimer's disease, 65 with mild cognitive impairment, 13 with non-Alzheimer's dementia, and 59 cognitively normal participants. All the participants were assessed using the three CL calculation methods. Spearman's rank correlation, linear regression, Friedman tests, Wilcoxon signed-rank tests, and Bland-Altman analysis were employed to assess data correlations, linear associations, method differences, and systematic bias, respectively. RESULTS: Strong correlations (rho = 0.99, p < .001) were observed among the CL values calculated using the three methods. Scatter plots and regression lines visually confirmed these strong correlations and met the validation criteria. Despite the robust correlations, a significant difference in CL value between CapAIBL and Amyquant was observed (36.1 ± 39.7 vs. 34.9 ± 39.4; p < .001). In contrast, no significant differences were found between CapAIBL and VIZCalc or between VIZCalc and Amyquant. The Bland-Altman analysis showed no observable systematic bias between the methods. CONCLUSIONS: The study demonstrated strong agreement among the three methods for calculating CL values. Despite minor variations in the absolute values of the Centiloid scores obtained using these methods, the overall agreement suggests that they are interchangeable.