Background In Alzheimer’s disease, amyloid- β (A β) peptides aggregate in the lowering CSF amyloid levels - a key pathological hallmark of the disease. However, lowered CSF amyloid levels may also be present in cognitively unimpaired elderly individuals. Therefore, it is of great value to explain the variance in disease progression among patients with A β pathology. Methods A cohort of n=2293 participants, of whom n=749 were A β positive, was selected from the Alzheimer’s Disease Neuroimaging Initiative (ADNI) database to study heterogeneity in disease progression for individuals with A β pathology. The analysis used baseline clinical variables including demographics, genetic markers, and neuropsychological data to predict how the cognitive ability and AD diagnosis of subjects progressed using statistical models and machine learning. Due to the relatively low prevalence of A β pathology, models fit only to A β-positive subjects were compared to models fit to an extended cohort including subjects without established A β pathology, adjusting for covariate differences between the cohorts. Results A β pathology status was determined based on the A β42/A β40 ratio. The best predictive model of change in cognitive test scores for A β-positive subjects at the 2-year follow-up achieved an R2 score of 0.388 while the best model predicting adverse changes in diagnosis achieved a weighted F1 score of 0.791. A β-positive subjects declined faster on average than those without A β pathology, but the specific level of CSF A β was not predictive of progression rate. When predicting cognitive score change 4 years after baseline, the best model achieved an R2 score of 0.325 and it was found that fitting models to the extended cohort improved performance. Moreover, using all clinical variables outperformed the best model based only on a suite of cognitive test scores which achieved an R2 score of 0.228. Conclusion Our analysis shows that CSF levels of A β are not strong predictors of the rate of cognitive decline in A β-positive subjects when adjusting for other variables. Baseline assessments of cognitive function accounts for the majority of variance explained in the prediction of 2-year decline but is insufficient for achieving optimal results in longer-term predictions. Predicting changes both in cognitive test scores and in diagnosis provides multiple perspectives of the progression of potential AD subjects.

Missing values are unavoidable in many applications of machine learning and present challenges both during training and at test time. When variables are missing in recurring patterns, fitting separate pattern submodels have been proposed as a solution. However, fitting models independently does not make efficient use of all available data. Conversely, fitting a single shared model to the full data set relies on imputation which often leads to biased results when missingness depends on unobserved factors. We propose an alternative approach, called sharing pattern submodels (SPSM), which i) makes predictions that are robust to missing values at test time, ii) maintains or improves the predictive power of pattern submodels, and iii) has a short description, enabling improved interpretability. Parameter sharing is enforced through sparsity-inducing regularization which we prove leads to consistent estimation. Finally, we give conditions for when a sharing model is optimal, even when both missingness and the target outcome depend on unobserved variables. Classification and regression experiments on synthetic and real-world data sets demonstrate that our models achieve a favorable tradeoff between pattern specialization and information sharing.

BackgroundIn Alzheimer’s disease (AD), amyloid- β (Aβ) peptides aggregate in the brain forming amyloid plaques, which are a key pathological hallmark of the disease. However, plaques may also be present in cognitively unimpaired elderly individuals. Therefore, it is of great value to explain the variance in disease progression among patients with Aβ pathology. MethodsA cohort of n= 2293 participants from the Alzheimer’s Disease Neuroimaging Initiative (ADNI) database was selected to study heterogeneity in disease progression for individuals with Aβ plaque pathology. The analysis used baseline clinical variables including demographics, genetic markers and neuropsychological data to predict how the cognitive ability and AD diagnosis of subjects progressed using statistical models and machine learning. Due to the limited prevalence of Aβ pathology, models fit only to Aβ-positive subjects were compared to models fit to an extended cohort including subjects without established Aβ pathology, adjusting for covariate differences between the cohorts. ResultsAβ pathology status was determined based the Aβ 42 /Aβ 40 ratio. The best predictive model of change in cognitive test scores for Aβ-positive subjects at the two-year follow-up achieved an R 2 score of 0.388 while the best model predicting adverse changes in diagnosis achieved a weighted F1 score of 0.791. Conforming to expectations, Aβ-positive subjects declined faster on average than those without Aβ pathology, but the specific level of Aβ plaques was not predictive of progression rate. For the four-year prediction task of cognitive score change, the best model achieved an R 2 score of 0.325 and it was found that fitting models to the extended cohort substantially improved performance. Moreover, using all clinical variables outperformed the best model based only on baseline cognitive test scores which achieved an R 2 score of 0.228. ConclusionOur analysis shows that levels of Aβ plaques are not strong predictors of the rate of cognitive decline in Aβ-positive subjects. Baseline assessments of cognitive function accounts for the majority of variance explained in the prediction of two-year decline but is insufficient for achieving optimal results in longer-term predictions. Predicting changes both in cognitive test scores and in diagnosis provides multiple perspectives of the progression of potential AD subjects.