Matthew J. Betts

Background: The substantia nigra (SN) is susceptible to neurodegeneration in Alzheimer’s disease. Automatic segmentation of the SN using MRI permits in vivo morphological assessment of the SN in cohort studies. Here we evaluate the multi‐brain segmentation (MB, Brudfors, 2020) to learn a probabilistic tissue prior of the SN using neuromelanin‐sensitive MRI while systematically studying effects of (A) SN tissue contrast; and (B) volumetric differences between healthy subjects and memory patients.Method: We studied 50 3T single‐site T₁w FLASH MRI from cognitively healthy older adults from the DZNEDELCODE cohort and a second ageing cohort in Magdeburg and 25 multi‐site diagnosed memory patients (MCI or AD). We quantified local SN signal contrast and image noise as a quality metric (obtained from CAT/SPM toolbox). We learnt tissue priors for each 10, 20, 30 and 50 older adults with the highest SN tissue contrast. For each tissue prior, we obtained model‐based SN tissue segments and calculated (integrated) SN volume. We compared SN volume between tissue prior segmentations and correlated it to age and quality. We coregistered all patients to the N = 50 tissue prior and compared volumetric differences between healthy adults and patients using voxel‐based morphometry (VBM).Result: The template‐space N = 50 SN prior (in red) was visually larger than all other priors (Fig 1A). Patients visually had a lower SN volume than healthy from N = 50 (Fig 1B). The N = 50 native space SN tissue prior volume was significantly higher than all other tissue prior volumes (p>0.01, Fig 2A). SN volume correlated with quality for N = 50 (p = 0.02, Fig 2B). The N = 50 sample SN contrast (Fig 2C) did not differ between patients and healthy adults. Patients had a decreased volume in a cluster overlapping with the SN (N = 268, p_{FDR corrected} = 0.02 (Fig 2D).Conclusion: Here we demonstrate that model‐based learning of a probabilistic SN tissue prior using intensity‐based generative approach to segmentation is feasable in a heterogenous cohort. Additionally, using this prior, we showed that while there is no difference of the SN contrast, there is a significant decrease in SN volume consistent with previous analyses (Krohn et al 2022).

Background: Transcutaneous auricular vagus nerve stimulation (taVNS) may affect the locus coeruleus‐norepinephrine system (LC‐NE system) via modulation of the vagus nerve. The LC is our main source of NE in the brainstem and declines with age and is a potential epicentre of tau pathology in Alzheimer´s diseases (AD). Besides the excessive NE release leading to the formation of amyloid plaques, the severity and duration of AD are also related to LC cell death, decreased cortical NE levels, and accelerated tau spread. An increase in NE release in the LC accompanied by reduced inflammatory markers was achieved in animal models using phasic invasive (i)VNS. In humans, phasic taVNS leads to an increase in pupil dilation, which can serve as a proxy for LC activation. Therefore, taVNS may have the potential to increase NE release and thereby impact LC function and related links to tau pathology.Method: Twenty‐four subjects (50% women; 22.96 +/‐ 2.24 years) received time‐synchronous phasic taVNS during a fixation task (∼ 18min) while recording pupillary dilation changes, which allowed systematic testing of phasic taVNS (3 sec) with low (3 mA) and high (5 mA) intensities and low (10 Hz) and high (25 Hz) frequencies at constant pulse width (250 µs).Result: During on stimulation pupil dilation was higher during real (0.18±0.03) as compared to sham (0.09±0.03) stimulation (χ² = 10.29, p = 0.001) (see Fig. 1). Likewise higher (0.21±0.03) compared to lower (0.06±0.03) intensity (χ² = 23.02, p < 0.001) led to higher pupil dilation, while there was no significant difference between higher (0.13±0.03) and lower (0.14±0.03) frequencies (χ² = 0.11, p = 0.74).Conclusion: Systematic testing of different frequencies and intensities, while keeping the pulse width constant, showed that phasic real taVNS and higher intensity lead to higher pupillary dilation, which is consistent with animal and human studies. Furthermore, the results suggest that the influence of intensity on pupil dilation might be stronger than that of frequency. The increased pupillary dilation during phasic taVNS could be an indication of LC involvement, however, more detailed studies e.g., functional magnetic resonance imaging (fMRI) during taVNS are needed to confirm this association.

Background: The dopaminergic system plays a crucial role in modulating learning and memory and undergoes age‐related changes that contribute to neurological disorders (Volkow et al., 1998; Pan et al., 2019) and reduced cognitive maintenance in ageing (Nyberg et al., 2016). Advanced imaging techniques can assess dopaminergic function in vivo in humans. In our study, we used dynamic concurrent magnetic resonance‐positron emission tomography (MR‐PET) which allows to infer dopamine release concurrent to functional MR imaging to investigate the relationship between dopaminergic neuromodulation in memory and healthy ageing. These measurements provide valuable insights into the critical role of the dopaminergic system in sustaining cognitive function during ageing and may help us elucidate the consequences of neuromodulatory decline in neurodegenerative disorders such as Parkinson’s disease and Alzheimer’s dementia.Method: 27 healthy adults over 65 underwent two dynamic MR‐PET imaging sessions with [¹⁸F]fallypride to measure D2/3 receptor occupancy and changes in nondisplaceable binding potential (∆BP_ND) in the striatum and extrastriatal regions. MRI scans were acquired concurrently, along with functional scans during a reward‐based memory task. Transient endogenous dopamine release was captured using a linear parametric neurotransmitter PET model.Result: ∆BP_ND of active PET (high dopamine condition) session was negatively correlated with the T‐static values in the bilateral hippocampus activation clusters in the reward versus neutral scene stimuli contrast (rho = ‐.41, p<.05), while the correlation was nonsignificant in passive PET (low dopamine condition) session.Conclusion: We used concurrent MR‐PET imaging to investigate the role of endogenous dopamine release and hemodynamic activity in memory consolidation. Higher endogenous dopamine release in the hippocampus during memory encoding was associated with greater functional activation in response to reward‐associated stimuli. Future analyses will examine links between brain health indicators and individual differences in memory consolidation and cognitive function. These findings may have implications for improving memory function in ageing and neurodegenerative diseases.

Background: Degeneration of the Locus coeruleus (LC) is associated with Alzheimer’s disease (AD) pathology and cognitive decline. Recent evidence from animal models indicates that this relationship may be mediated by neuroinflammation. Here we examined how LC MRI contrast is related to CSF biomarkers of neuroinflammation in humans.Method: The LC as well as the Substantia nigra (SN) were visualized and segmented using neuromelanin‐sensitive T₁‐weighted FLASH‐MRI images acquired at 3T and MRI contrast values were calculated as previously described (Betts et al., 2019b). A cohort of 65 participants with MRI contrast and CSF data from the DZNE longitudinal cognitive impairment and dementia (DELCODE) study (20 HC, 22 SCD, 16 MCI, 7 AD dementia; 53.8% female) were included in the analyses. We performed correlational analyses to test for associations between LC and SN contrast with a panel of 4 Principal Components (PC) created from a Principial Component Analysis on 16 individual inflammation‐related CSF biomarkers (Fig. 1). Both bivariate and partial correlation analyses were performed between PCs and MRI contrast values using sex, age, site and CSF levels of AD pathology, namely Aß42/40‐ratio and phosphorylated tau, as covariates (see Fig. 2).Result: We found a significant positive association between LC MRI contrast and the inflammatory component PC4 including the proinflammatory complement factors C3 and C3b across the entire sample (r = 0.365, p = 0.003). Additional post‐hoc correlational analyses showed this association with LC MRI contrast was mainly due to a correlation with C3b (r = 0.349, p = 0.004). A significant positive association was still observed (r = 0.347, p = 0.008) in a subgroup of individuals comprising healthy older adults and individuals with a risk of AD dementia (HC, SCD and MCI) as shown in Fig. 2. In contrast, no significant associations between SN MRI contrast and any neurodegenerative or inflammatory measure were observed indicating effects were specific to the LC.Conclusion: The results demonstrate LC MRI contrast is associated with complement factors in CSF, in particular complement factor C3b in healthy older adults and individuals with a risk of AD dementia. Further analyses are required to determine if elevated complement factors may predict degeneration of the LC and subsequent cognitive decline.

Introduction: Artificial intelligence (AI) and neuroimaging offer new opportunities for diagnosis and prognosis of dementia.Methods: We systematically reviewed studies reporting AI for neuroimaging in diagnosis and/or prognosis of cognitive neurodegenerative diseases.Results: A total of 255 studies were identified. Most studies relied on the Alzheimer's Disease Neuroimaging Initiative dataset. Algorithmic classifiers were the most commonly used AI method (48%) and discriminative models performed best for differentiating Alzheimer's disease from controls. The accuracy of algorithms varied with the patient cohort, imaging modalities, and stratifiers used. Few studies performed validation in an independent cohort.Discussion: The literature has several methodological limitations including lack of sufficient algorithm development descriptions and standard definitions. We make recommendations to improve model validation including addressing key clinical questions, providing sufficient description of AI methods and validating findings in independent datasets. Collaborative approaches between experts in AI and medicine will help achieve the promising potential of AI tools in practice.Highlights: There has been a rapid expansion in the use of machine learning for diagnosis and prognosis in neurodegenerative disease Most studies (71%) relied on the Alzheimer's Disease Neuroimaging Initiative (ADNI) dataset with no other individual dataset used more than five times There has been a recent rise in the use of more complex discriminative models (e.g., neural networks) that performed better than other classifiers for classification of AD vs healthy controls We make recommendations to address methodological considerations, addressing key clinical questions, and validation We also make recommendations for the field more broadly to standardize outcome measures, address gaps in the literature, and monitor sources of bias

Background: The locus coeruleus (LC) plays an important role in numerous cognitive functions such as working memory, learning and attention. Novel magnetic resonance imaging (MRI) techniques now provide an opportunity to quantify structural features of the LC in vivo which may help to understand the contribution of LC neurodegeneration to clinical and pathological manifestations in Alzheimer’s disease (AD) and Parkinson’s disease (PD). In this study we aimed to use in vivo T₁‐weighted Fast Low Angle Shot (FLASH) imaging to visualize and quantify LC MRI contrast in AD and PD and to determine the relationship between LC MRI contrast and cognition.Method: Baseline data from the DZNE Longitudinal Cognitive Impairment and Dementia (DELCODE) study comprising 48 healthy elderly adults, 27 individuals with subjective cognitive decline (SCD), 16 individuals with mild cognitive impairment (MCI), 11 individuals with AD together with 30 patients with idiopathic PD were analysed. All subjects were scanned at 3T using a 3D T₁‐weighted FLASH sequence (0.75mm isotropic resolution whole brain acquisition). FLASH images were sinc interpolated and upsampled to 0.375mm³ resolution prior to manual segmentation of the LC. Median LC signal intensity was determined relative to reference regions delineated in the rostral pontomesencephalic area.Result: A significant decrease in LC MRI contrast in AD and PD was observed compared to healthy controls. However no significant decrease in LC MRI contrast was observed in SCD and MCI compared to healthy controls or between AD and PD patients. In AD, a linear regression analysis across groups (n=102), revealed a significant negative correlation between LC MRI contrast and performance in the backward digit span (DS)/total (DS) test. In PD, a significant positive correlation between LC MRI contrast and slower performance in the Trail Making Test (TMT)‐A and B, was observed.Conclusion: The results confirm that T₁‐weighted FLASH imaging using 3 T MRI can identify substantial LC degeneration in vivo and is associated with impaired working memory and attention in AD and PD, respectively. Collectively these results indicate that in vivo LC imaging may help to track LC degeneration and understand the pathological underpinnings of cognitive dysfunction in neurodegenerative disease.

Background: The panel of fluid- and imaging-based biomarkers available for neurodegenerative disease research is growing and has the potential to close important gaps in research and the clinic. With this growth and increasing use, appropriate implementation and interpretation are paramount. Various biomarkers feature nuanced differences in strengths, limitations, and biases that must be considered when investigating disease etiology and clinical utility. For example, neuropathological investigations of Alzheimer’s disease pathogenesis can fall in disagreement with conclusions reached by biomarker-based investigations. Considering the varied strengths, limitations, and biases of different research methodologies and approaches may help harmonize disciplines within the neurodegenerative disease field.Purpose of review: Along with separate review articles covering fluid and imaging biomarkers in this issue ofAlzheimer’s Research and Therapy, we present the result of a discussion from the 2019 Biomarkers in Neurodegenerative Diseases course at the University College London. Here, we discuss themes of biomarker use in neurodegenerative disease research, commenting on appropriate use, interpretation, and considerations for implementation across different neurodegenerative diseases. We also draw attention to areas where biomarker use can be combined with other disciplines to understand issues of pathophysiology and etiology underlying dementia. Lastly, we highlight novel modalities that have been proposed in the landscape of neurodegenerative disease research and care.

The caption in Figure 1 mistakenly omits that the figure was adapted from Richter et al., 2014 (Front Syst Neurosci 2014; 8:140). The authors apologize for this error. The full citation is reported below.