Does Informant-Based Reporting of Cognitive Decline Correlate with Age-Adjusted Hippocampal Volume in Mild Cognitive Impairment and Alzheimer's Disease?

Spatial navigation questionnaires as a supportive diagnostic tool in early Alzheimer's disease

Impaired spatial navigation is early marker of Alzheimer's disease (AD). We examined ability of self- and informant-reported navigation questionnaires to discriminate between clinically and biomarker-defined participants, and associations of questionnaires with navigation performance, regional brain atrophy, AD biomarkers, and biomarker status. 262 participants (cognitively normal, with subjective cognitive decline, amnestic mild cognitive impairment [aMCI], and mild dementia) and their informants completed three navigation questionnaires. Navigation performance, magnetic resonance imaging volume/thickness of AD-related brain regions, and AD biomarkers were measured. Informant-reported questionnaires distinguished between cognitively normal and impaired participants, and amyloid-β positive and negative aMCI. Lower scores were associated with worse navigation performance, greater atrophy in AD-related brain regions, and amyloid-β status. Self-reported questionnaire scores did not distinguish between the groups and were weakly associated with navigation performance. Other associations were not significant. Informant-reported navigation questionnaires may be a screening tool for early AD reflecting atrophy of AD-related brain regions and AD pathology.

Comparison of brain imaging and physical health between research and clinical neuroimaging cohorts of ageing

OBJECTIVES

To compare brain MRI measures between Adult Changes in Thought (ACT) participants who underwent research, clinical, or both MRI scans, and clinical health measures across the groups and non-MRI subjects.

METHODS

Retrospective cohort study leveraging MRI, clinical, demographic, and medication data from ACT. Three neuroradiologists reviewed MRI scans using NIH Neuroimaging Common Data Elements (CDEs). Total brain and white matter hyperintensity (WMH) volumes, clinical characteristics, and outcome measures of brain and overall health were compared between groups. 1166 MRIs were included (77 research, 1043 clinical, and 46 both) and an additional 3146 participants with no MRI were compared.

RESULTS

Compared to the group with research MRI only, the clinical MRI group had higher prevalence of the following: acute infarcts, chronic haematoma, subarachnoid haemorrhage, subdural haemorrhage, haemorrhagic transformation, and hydrocephalus (each P < .001). Quantitative WMH burden was significantly lower (P < .001) and total brain volume significantly higher (P < .001) in research MRI participants compared to other MRI groups. Prevalence of hypertension, self-reported cerebrovascular disease, congestive heart failure, dementia, and recent hospitalization (all P < .001) and diabetes (P = .002) differed significantly across groups, with smaller proportions in the research MRI group.

CONCLUSION

In ageing populations, significant differences were observed in MRI metrics between research MRI and clinical MRI groups, and clinical health metric differences between research MRI, clinical MRI, and no-MRI groups.

ADVANCES IN KNOWLEDGE

This questions whether research cohorts can adequately represent the greater ageing population undergoing imaging. These findings may also be useful to radiologists when interpreting neuroimaging of ageing.

FDA-approved machine learning algorithms in neuroradiology: A systematic review of the current evidence for approval

Over the past decade, machine learning (ML) and artificial intelligence (AI) have become increasingly prevalent in the medical field. In the United States, the Food and Drug Administration (FDA) is responsible for regulating AI algorithms as "medical devices" to ensure patient safety. However, recent work has shown that the FDA approval process may be deficient. In this study, we evaluate the evidence supporting FDA-approved neuroalgorithms, the subset of machine learning algorithms with applications in the central nervous system (CNS), through a systematic review of the primary literature. Articles covering the 53 FDA-approved algorithms with applications in the CNS published in PubMed, EMBASE, Google Scholar and Scopus between database inception and January 25, 2022 were queried. Initial searches identified 1505 studies, of which 92 articles met the criteria for extraction and inclusion. Studies were identified for 26 of the 53 neuroalgorithms, of which 10 algorithms had only a single peer-reviewed publication. Performance metrics were available for 15 algorithms, external validation studies were available for 24 algorithms, and studies exploring the use of algorithms in clinical practice were available for 7 algorithms. Papers studying the clinical utility of these algorithms focused on three domains: workflow efficiency, cost savings, and clinical outcomes. Our analysis suggests that there is a meaningful gap between the FDA approval of machine learning algorithms and their clinical utilization. There appears to be room for process improvement by implementation of the following recommendations: the provision of compelling evidence that algorithms perform as intended, mandating minimum sample sizes, reporting of a predefined set of performance metrics for all algorithms and clinical application of algorithms prior to widespread use. This work will serve as a baseline for future research into the ideal regulatory framework for AI applications worldwide.

Development of a simple screening tool for determining cognitive status in Alzheimer's disease

Cognitive screening is often a first step to document cognitive status of patients suspected having Alzheimer's disease (AD). Unfortunately, screening neuropsychological tests are often insensitivity in the detection. The goal of this study was to develop a simple and sensitive screening neuropsychological test to facilitate early detection of AD. This study recruited 761 elderly individuals suspected of having AD and presenting various cognitive statuses (mean age: 77.69 ± 8.45 years; proportion of females: 65%; cognitively unimpaired, CU, n = 133; mild cognitive impairment, MCI, n = 231; dementia of Alzheimer's type, DAT, n = 397). This study developed a novel screening neuropsychological test incorporating assessments of the core memory deficits typical of early AD and an interview on memory function with an informant. The proposed History-based Artificial Intelligence-Show Chwan Assessment of Cognition (HAI-SAC) was assessed in terms of psychometric properties, test time, and discriminative ability. The results were compared with those obtained using other common screening tests, including Cognitive Abilities Screening Instrument (CASI), Montreal Cognitive Assessment (MoCA), and an extracted Mini-Mental State Examination score from CASI. HAI-SAC demonstrated acceptable internal consistency. Factor analysis revealed two factors: memory (semantic and contextual) and cognition-related information from informants. The assessment performance of HAI-SAC was strongly correlated with that of the common screening neuropsychological tests addressed in this study. HAI-SAC outperformed the other tests in differentiating CU individuals from patients with MCI (sensitivity: 0.87; specificity: 0.58; area under the curve [AUC]: 0.78) or DAT (sensitivity: 0.99; specificity: 0.89; AUC: 0.98). Performance of HAI-SAC on differentiating MCI from DAT was on par with performances of other tests (sensitivity: 0.78; specificity: 0.84; AUC: 0.87), while the test time was less than one quarter that of CASI and half that of MoCA. HAI-SAC is psychometrically sound, cost-effective, and sensitive in discriminating the cognitive status of AD.

The Case for a Novel Therapeutic Approach to Dementia: Small Molecule Hepatocyte Growth Factor (HGF/MET) Positive Modulators

An estimated 6.5 million Americans aged 65 years or older have Alzheimer's disease (AD), which will grow to 13.8 million Americans by 2060. Despite the growing burden of dementia, no fundamental change in drug development for AD has been seen in > 20 years. Currently approved drugs for AD produce only modest symptomatic improvements in cognition with small effect sizes. A growing mismatch exists between the urgent need to develop effective drugs for symptomatic AD and the largely failed search for disease modification. The failure rate of clinical trials in AD is high overall, and in particular for disease-modifying therapies. Research efforts in AD have focused predominantly on amyloid-β and tau pathologies, but limiting clinical research to these "classical hallmarks" of the disease does not address the most urgent patient, caregiver, or societal needs. Rather, clinical research should consider the complex pathophysiology of AD. Innovative approaches are needed that provide outside-the-box thinking, and re-imagine trial design, interventions, and outcomes as well as progress in proteomics and fluid biomarker analytics for both diagnostics and disease monitoring. A new approach offering a highly specific, yet multi-pronged intervention that exerts positive modulation on the HGF/MET neurotrophic system is currently being tested in mid-to-late-stage clinical trials in mild to moderate AD. Findings from such trials may provide data to support novel approaches for development of innovative drugs for treating AD at various disease stages, including among patients already symptomatic, and may offer benefits for other neurodegenerative diseases.

Updated Review of the Evidence Supporting the Medical and Legal Use of NeuroQuant® and NeuroGage® in Patients With Traumatic Brain Injury

Over 40 years of research have shown that traumatic brain injury affects brain volume. However, technical and practical limitations made it difficult to detect brain volume abnormalities in patients suffering from chronic effects of mild or moderate traumatic brain injury. This situation improved in 2006 with the FDA clearance of NeuroQuant®, a commercially available, computer-automated software program for measuring MRI brain volume in human subjects. More recent strides were made with the introduction of NeuroGage®, commercially available software that is based on NeuroQuant® and extends its utility in several ways. Studies using these and similar methods have found that most patients with chronic mild or moderate traumatic brain injury have brain volume abnormalities, and several of these studies found-surprisingly-more abnormal enlargement than atrophy. More generally, 102 peer-reviewed studies have supported the reliability and validity of NeuroQuant® and NeuroGage®. Furthermore, this updated version of a previous review addresses whether NeuroQuant® and NeuroGage® meet the Daubert standard for admissibility in court. It concludes that NeuroQuant® and NeuroGage® meet the Daubert standard based on their reliability, validity, and objectivity. Due to the improvements in technology over the years, these brain volumetric techniques are practical and readily available for clinical or forensic use, and thus they are important tools for detecting signs of brain injury.