Machine learning methods to predict amyloid positivity using domain scores from cognitive tests

Predicting amyloid positivity from FDG-PET images using radiomics: A parsimonious model

BACKGROUND AND OBJECTIVE

Amyloid plaques are one of the physical hallmarks of Alzheimer's disease. The objective of this study is to predict amyloid positivity non-invasively from FDG-PET images using a radiomics approach.

METHODS

We obtained FDG-PET images of 301 individuals from various groups, including control normal (CN), mild cognitive impairment (MCI), and Alzheimer's Disease (AD), from the ADNI database. Following the utilization of the CSF Aβ1-42 (192) and Standardized Uptake Value Ratio (SUVR) (1.11) thresholds derived from Florbetapir scans, the subjects were categorized into two categories: those with a positive amyloid status (n = 185) and those with a negative amyloid status (n = 116). The process of segmenting the entire brain into 95 classes using the DKT-atlas was utilized. Following that, we obtained 120 characteristics for each of the 95 regions of interest (ROIs). We employed eight feature selection methods to analyze the features. Additionally, we utilized eight different classifiers on the 20 most significant features extracted from each feature selection method. Finally, in order to improve interpretability, we selected the most important features and ROIs.

RESULT

We found that the GNB classifier and the LASSO feature selection method had the best performance with an average accuracy of (AUC=0.924) while using 18 features on 15 ROIs. We were then able to reduce the model to three regions (Hippocampus, inferior parietal, and isthmus cingulate) and three gray-level based features (AUC=0.853).

CONCLUSION

The FDG-PET images which serve to study metabolic activity can be used to predict amyloid positivity without the use of invasive methods or another PET tracer and study. The proposed method has superior prediction accuracy with respect to similar studies reported in the literature using other imaging modalities. Only three brain regions had a high impact on amyloid positivity results.

MRI-based Deep Learning Assessment of Amyloid, Tau, and Neurodegeneration Biomarker Status across the Alzheimer Disease Spectrum

Background PET can be used for amyloid-tau-neurodegeneration (ATN) classification in Alzheimer disease, but incurs considerable cost and exposure to ionizing radiation. MRI currently has limited use in characterizing ATN status. Deep learning techniques can detect complex patterns in MRI data and have potential for noninvasive characterization of ATN status. Purpose To use deep learning to predict PET-determined ATN biomarker status using MRI and readily available diagnostic data. Materials and Methods MRI and PET data were retrospectively collected from the Alzheimer's Disease Imaging Initiative. PET scans were paired with MRI scans acquired within 30 days, from August 2005 to September 2020. Pairs were randomly split into subsets as follows: 70% for training, 10% for validation, and 20% for final testing. A bimodal Gaussian mixture model was used to threshold PET scans into positive and negative labels. MRI data were fed into a convolutional neural network to generate imaging features. These features were combined in a logistic regression model with patient demographics, APOE gene status, cognitive scores, hippocampal volumes, and clinical diagnoses to classify each ATN biomarker component as positive or negative. Area under the receiver operating characteristic curve (AUC) analysis was used for model evaluation. Feature importance was derived from model coefficients and gradients. Results There were 2099 amyloid (mean patient age, 75 years ± 10 [SD]; 1110 male), 557 tau (mean patient age, 75 years ± 7; 280 male), and 2768 FDG PET (mean patient age, 75 years ± 7; 1645 male) and MRI pairs. Model AUCs for the test set were as follows: amyloid, 0.79 (95% CI: 0.74, 0.83); tau, 0.73 (95% CI: 0.58, 0.86); and neurodegeneration, 0.86 (95% CI: 0.83, 0.89). Within the networks, high gradients were present in key temporal, parietal, frontal, and occipital cortical regions. Model coefficients for cognitive scores, hippocampal volumes, and APOE status were highest. Conclusion A deep learning algorithm predicted each component of PET-determined ATN status with acceptable to excellent efficacy using MRI and other available diagnostic data. © RSNA, 2023 Supplemental material is available for this article.

Disease progression modelling of Alzheimer's disease using probabilistic principal components analysis

The recent biological redefinition of Alzheimer's Disease (AD) has spurred the development of statistical models that relate changes in biomarkers with neurodegeneration and worsening condition linked to AD. The ability to measure such changes may facilitate earlier diagnoses for affected individuals and help in monitoring the evolution of their condition. Amongst such statistical tools, disease progression models (DPMs) are quantitative, data-driven methods that specifically attempt to describe the temporal dynamics of biomarkers relevant to AD. Due to the heterogeneous nature of this disease, with patients of similar age experiencing different AD-related changes, a challenge facing longitudinal mixed-effects-based DPMs is the estimation of patient-realigning time-shifts. These time-shifts are indispensable for meaningful biomarker modelling, but may impact fitting time or vary with missing data in jointly estimated models. In this work, we estimate an individual's progression through Alzheimer's disease by combining multiple biomarkers into a single value using a probabilistic formulation of principal components analysis. Our results show that this variable, which summarises AD through observable biomarkers, is remarkably similar to jointly estimated time-shifts when we compute our scores for the baseline visit, on cross-sectional data from the Alzheimer's Disease Neuroimaging Initiative (ADNI). Reproducing the expected properties of clinical datasets, we confirm that estimated scores are robust to missing data or unavailable biomarkers. In addition to cross-sectional insights, we can model the latent variable as an individual progression score by repeating estimations at follow-up examinations and refining long-term estimates as more data is gathered, which would be ideal in a clinical setting. Finally, we verify that our score can be used as a pseudo-temporal scale instead of age to ignore some patient heterogeneity in cohort data and highlight the general trend in expected biomarker evolution in affected individuals.

Monte Carlo cross-validation for a study with binary outcome and limited sample size

Cross-validation (CV) is a resampling approach to evaluate machine learning models when sample size is limited. The number of all possible combinations of folds for the training data, known as CV rounds, are often very small in leave-one-out CV. Alternatively, Monte Carlo cross-validation (MCCV) can be performed with a flexible number of simulations when computational resources are feasible for a study with limited sample size. We conduct extensive simulation studies to compare accuracy between MCCV and CV with the same number of simulations for a study with binary outcome (e.g., disease progression or not). Accuracy of MCCV is generally higher than CV although the gain is small. They have similar performance when sample size is large. Meanwhile, MCCV is going to provide reliable performance metrics as the number of simulations increases. Two real examples are used to illustrate the comparison between MCCV and CV.

Continuity corrected score confidence interval for the difference in proportions in paired data

For paired binary data, the hybrid method and the score method are often recommended for use to calculate the confidence interval for risk difference. These asymptotic intervals do not control the coverage probability. We propose to develop a new score interval with continuity correction to further improve the performance of the existing intervals. The traditional correction value may be too large which leads to a wide interval. For that reason, we propose three different correction values to identify the optimal correction interval with balanced coverage probability and interval width. From simulation studies, we find that a small correction value for the score interval has good performance. In addition, we derive the non-iterative solutions for the developed continuity correction score intervals.

Neuroproteomics Chip-Based Mass Spectrometry and Other Techniques for Alzheimer´S Disease Biomarkers – Update

Background:

Alzheimer's disease (AD) is a progressive neurodegenerative disease of growing interest given that there is cognitive damage and symptom onset acceleration. Therefore, it is important to find AD biomarkers for early diagnosis, disease progression, and discrimination of AD and other diseases.

Objective:

To update the relevance of mass spectrometry for the identification of peptides and proteins involved in AD useful as discriminating biomarkers.

Methods:

Proteomics and peptidomics technologies that show the highest possible specificity and selectivity for AD biomarkers are analyzed, together with the biological fluids used. In addition to positron emission tomography and magnetic resonance imaging, MALDI-TOF mass spectrometry is widely used to identify proteins and peptides involved in AD. The use of protein chips in SELDI technology and electroblotting chips for peptides makes feasible small amounts (L) of samples for analysis.

Results:

Suitable biomarkers are related to AD pathology, such as intracellular neurofibrillary tangles; extraneuronal senile plaques; neuronal and axonal degeneration; inflammation and oxidative stress. Recently, peptides were added to the candidate list, which are not amyloid-b or tau fragments, but are related to coagulation, brain plasticity, and complement/neuroinflammation systems involving the neurovascular unit.

Conclusion:

The progress made in the application of mass spectrometry and recent chip techniques is promising for discriminating between AD, mild cognitive impairment, and matched healthy controls. The application of this technique to blood samples from patients with AD has shown to be less invasive and fast enough to determine the diagnosis, stage of the disease, prognosis, and follow-up of the therapeutic response.

Randomized two-stage optimal design for interval-censored data

Interval-censored data occur in a study where the exact event time of each participant is not observed but it is known to be within a certain time interval. Multiple tests were proposed for such data, including the logrank test by Sun, the proportional hazard test by Finkelstein, and the Wilcoxon-type test by Peto and Peto. We propose sample size calculations based on these tests for a parallel one-stage or two-stage design. When the proportional hazard assumption is met, the proportional hazard test and the logrank test need smaller sample sizes than the Wilcoxon-type test, and the sample size savings are substantial. But this trend is reversed when the proportional hazard assumption does not hold, and the sample size savings using the Wilcoxon-type test are sizable. An example from a lung cancer clinical trial is used to illustrate the application of the proposed sample size calculations.

Potential Applications of Artificial Intelligence in Clinical Trials for Alzheimer’s Disease

Clinical trials for Alzheimer’s disease (AD) face multiple challenges, such as the high screen failure rate and the even allocation of heterogeneous participants. Artificial intelligence (AI), which has become a potent tool of modern science with the expansion in the volume, variety, and velocity of biological data, offers promising potential to address these issues in AD clinical trials. In this review, we introduce the current status of AD clinical trials and the topic of machine learning. Then, a comprehensive review is focused on the potential applications of AI in the steps of AD clinical trials, including the prediction of protein and MRI AD biomarkers in the prescreening process during eligibility assessment and the likelihood stratification of AD subjects into rapid and slow progressors in randomization. Finally, this review provides challenges, developments, and the future outlook on the integration of AI into AD clinical trials.