Neuronal dysfunction in specific brain regions or across distributed brain networks is a known feature of Alzheimer’s disease. An often reported finding in the early stage of the disease is the presence of increased functional MRI (fMRI) BOLD signal under task conditions relative to cognitively normal controls, a phenomenon known as “hyperactivation”. However, research in the past decades yielded complex, sometimes conflicting results. The magnitude and topology of fMRI hyperactivation patterns have been found to vary across the preclinical and clinical spectrum of Alzheimer’s disease, including concomitant “hypoactivation” in some cases. These incongruences are likely due to a range of factors, including the disease stage at which the cohort is examined, the brain areas or networks studied, and the fMRI paradigm utilised to evoke these functional abnormalities. Additionally, a perennial question pertains to the nature of hyperactivation in the context of Alzheimer’s disease. Some propose it reflects compensatory mechanisms to sustain cognitive performance while others suggest it is linked to the pathological disruption of a highly regulated homeostatic cycle that contributes to, or even drives, disease progression. Providing a coherent narrative for these empirical and conceptual discrepancies is paramount to develop disease models, understand the synergy between hyperactivation and the Alzheimer’s disease pathological cascade, and tailor effective interventions. We first provide a comprehensive overview of functional brain changes spanning the course from normal ageing to the clinical spectrum of Alzheimer’s disease. We then highlight evidence supporting a close relationship between fMRI hyperactivation and in vivo markers of Alzheimer’s pathology. We primarily focus on task-based fMRI studies in humans, but also consider studies using different functional imaging techniques and animal models. We then discuss the potential mechanisms underlying hyperactivation in the context of Alzheimer’s disease, and provide a testable framework bridging hyperactivation, ageing, cognition, and the Alzheimer’s disease pathological cascade. We conclude with a discussion of future challenges and opportunities to advance our understanding of the fundamental disease mechanisms of Alzheimer’s disease, and the promising development of therapeutic interventions incorporating or aimed at hyperactivation and large-scale functional systems.

Predominant limbic degeneration has been associated with various underlying aetiologies and an older age, predominant impairment of episodic memory and slow clinical progression. However, the neurological syndrome associated with predominant limbic degeneration is not defined. This endeavour is critical to distinguish such a syndrome from those originating from neocortical degeneration, which may differ in underlying aetiology, disease course and therapeutic needs. We propose a set of clinical criteria for a limbic-predominant amnestic neurodegenerative syndrome that is highly associated with limbic-predominant age-related TDP-43 encephalopathy but also other pathologic entities. The criteria incorporate core, standard and advanced features, including older age at evaluation, mild clinical syndrome, disproportionate hippocampal atrophy, impaired semantic memory, limbic hypometabolism, absence of neocortical degeneration and low likelihood of neocortical tau, with degrees of certainty (highest, high, moderate and low). We operationalized this set of criteria using clinical, imaging and biomarker data to validate its associations with clinical and pathologic outcomes. We screened autopsied patients from Mayo Clinic and Alzheimer’s Disease Neuroimaging Initiative cohorts and applied the criteria to those with an antemortem predominant amnestic syndrome (Mayo, n = 165; Alzheimer’s Disease Neuroimaging Initiative, n = 53) and who had Alzheimer’s disease neuropathological change, limbic-predominant age-related TDP-43 encephalopathy or both pathologies at autopsy. These neuropathology-defined groups accounted for 35, 37 and 4% of cases in the Mayo cohort, respectively, and 30, 22 and 9% of cases in the Alzheimer’s Disease Neuroimaging Initiative cohort, respectively. The criteria effectively categorized these cases, with Alzheimer’s disease having the lowest likelihoods, limbic-predominant age-related TDP-43 encephalopathy patients having the highest likelihoods and patients with both pathologies having intermediate likelihoods. A logistic regression using the criteria features as predictors of TDP-43 achieved a balanced accuracy of 74.6% in the Mayo cohort, and out-of-sample predictions in an external cohort achieved a balanced accuracy of 73.3%. Patients with high likelihoods had a milder and slower clinical course and more severe temporo-limbic degeneration compared to those with low likelihoods. Stratifying patients with both Alzheimer’s disease neuropathological change and limbic-predominant age-related TDP-43 encephalopathy from the Mayo cohort according to their likelihoods revealed that those with higher likelihoods had more temporo-limbic degeneration and a slower rate of decline and those with lower likelihoods had more lateral temporo-parietal degeneration and a faster rate of decline. The implementation of criteria for a limbic-predominant amnestic neurodegenerative syndrome has implications to disambiguate the different aetiologies of progressive amnestic presentations in older age and guide diagnosis, prognosis, treatment and clinical trials.

Given the prevalence of dementia and the development of pathology-specific disease-modifying therapies, high-value biomarker strategies to inform medical decision-making are critical. In vivo tau-PET is an ideal target as a biomarker for Alzheimer’s disease diagnosis and treatment outcome measure. However, tau-PET is not currently widely accessible to patients compared to other neuroimaging methods. In this study, we present a convolutional neural network (CNN) model that imputes tau-PET images from more widely available cross-modality imaging inputs. Participants (n = 1192) with brain T₁-weighted MRI (T1w), fluorodeoxyglucose (FDG)-PET, amyloid-PET and tau-PET were included. We found that a CNN model can impute tau-PET images with high accuracy, the highest being for the FDG-based model followed by amyloid-PET and T1w. In testing implications of artificial intelligence-imputed tau-PET, only the FDG-based model showed a significant improvement of performance in classifying tau positivity and diagnostic groups compared to the original input data, suggesting that application of the model could enhance the utility of the metabolic images. The interpretability experiment revealed that the FDG- and T1w-based models utilized the non-local input from physically remote regions of interest to estimate the tau-PET, but this was not the case for the Pittsburgh compound B-based model. This implies that the model can learn the distinct biological relationship between FDG-PET, T1w and tau-PET from the relationship between amyloid-PET and tau-PET. Our study suggests that extending neuroimaging’s use with artificial intelligence to predict protein specific pathologies has great potential to inform emerging care models.

From a complex systems perspective, clinical syndromes emerging from neurodegenerative diseases are thought to result from multiscale interactions between aggregates of misfolded proteins and the disequilibrium of large-scale networks coordinating functional operations underpinning cognitive phenomena. Across all syndromic presentations of Alzheimer’s disease, age-related disruption of the default mode network is accelerated by amyloid deposition. Conversely, syndromic variability may reflect selective neurodegeneration of modular networks supporting specific cognitive abilities. In this study, we leveraged the breadth of the Human Connectome Project-Aging cohort of non-demented individuals (N = 724) as a normative cohort to assess the robustness of a biomarker of default mode network dysfunction in Alzheimer’s disease, the network failure quotient, across the aging spectrum. We then examined the capacity of the network failure quotient and focal markers of neurodegeneration to discriminate patients with amnestic (N = 8) or dysexecutive (N = 10) Alzheimer’s disease from the normative cohort at the patient-level, as well as between Alzheimer’s disease phenotypes. Importantly, all participants and patients were scanned using the Human Connectome Project-Aging protocol, allowing for the acquisition of high-resolution structural imaging and longer resting-state connectivity acquisition time. Using a regression framework, we found that the network failure quotient related to age, global and focal cortical thickness, hippocampal volume, and cognition in the normative Human Connectome Project-Aging cohort, replicating previous results from the Mayo Clinic Study of Aging that used a different scanning protocol. Then, we used quantile curves and group-wise comparisons to show that the network failure quotient commonly distinguished both dysexecutive and amnestic Alzheimer’s disease patients from the normative cohort. In contrast, focal neurodegeneration markers were more phenotype-specific, where the neurodegeneration of parieto-frontal areas associated with dysexecutive Alzheimer’s disease while the neurodegeneration of hippocampal and temporal areas associated with amnestic Alzheimer’s disease. Capitalizing on a large normative cohort and optimized imaging acquisition protocols, we highlight a biomarker of default mode network failure reflecting shared system-level pathophysiological mechanisms across aging and dysexecutive and amnestic Alzheimer’s disease and biomarkers of focal neurodegeneration reflecting distinct pathognomonic processes across the amnestic and dysexecutive Alzheimer’s disease phenotypes. These findings provide evidence that variability in inter-individual cognitive impairment in Alzheimer’s disease may relate to both modular network degeneration and default mode network disruption. These results provide important information to advance complex systems approaches to cognitive aging and degeneration, expand the armamentarium of biomarkers available to aid diagnosis, monitor progression, and inform clinical trials.

Distinguishing neurodegenerative diseases is a challenging task requiring neurologic expertise. Clinical decision support systems (CDSSs) powered by machine learning (ML) and artificial intelligence can assist with complex diagnostic tasks by augmenting user capabilities, but workflow integration poses many challenges. We propose that a modeling framework based on fluorodeoxyglucose PET (FDG-PET) imaging can address these challenges and form the basis of an effective CDSS for neurodegenerative disease. This retrospective study focused on FDG-PET images in a discovery cohort drawn from 3 research studies plus routine clinical patients. When selecting research study participants, the inclusion criterion was the availability of an FDG-PET image from within 2.5 years of diagnosis with 1 of 9 specific neurodegenerative syndromes or designation as unimpaired. Participants from disease groups were recruited from the clinical patient population while unimpaired participants came primarily from a population study. The discovery cohort was used to develop a clinical decision support framework we call StateViewer, which applies a neighbor matching algorithm to detect the presence of 9 different neurodegenerative phenotypes. The ML performance of this framework was evaluated in the discovery cohort by nested cross-validation and externally validated in the Alzheimer's Disease Neuroimaging Initiative. Potential for clinical integration was demonstrated in a radiologic reader study focused on differentiating posterior cortical atrophy from Lewy body dementia. The discovery cohort contained 3,671 individuals with a mean age of 68 years and consisted of 49% reported female. Our model framework was able to detect the presence of 9 different neurodegenerative phenotypes with a sensitivity of 0.89 ± 0.03 and an area under the receiver operating characteristic curve of 0.93 ± 0.02. In the radiologic reader study, readers using our model were found to have 3.3 ± 1.1 times greater odds of making a correct diagnosis than readers using a current standard-of-care workflow. Our proposed framework provides strong classification performance with high interpretability, and it addresses many of the challenges that face clinical integration of ML-based decision support tools. One limitation of this study is a uniform discovery cohort that is not representative of other patient populations in some regards.