AI for the prediction of early stages of Alzheimer's disease from neuroimaging biomarkers - A narrative review of a growing field.

OBJECTIVES: The objectives of this narrative review are to summarize the current state of AI applications in neuroimaging for early Alzheimer's disease (AD) prediction and to highlight the potential of AI techniques in improving early AD diagnosis, prognosis, and management. METHODS: We conducted a narrative review of studies using AI techniques applied to neuroimaging data for early AD prediction. We examined single-modality studies using structural MRI and PET imaging, as well as multi-modality studies integrating multiple neuroimaging techniques and biomarkers. Furthermore, they reviewed longitudinal studies that model AD progression and identify individuals at risk of rapid decline. RESULTS: Single-modality studies using structural MRI and PET imaging have demonstrated high accuracy in classifying AD and predicting progression from mild cognitive impairment (MCI) to AD. Multi-modality studies, integrating multiple neuroimaging techniques and biomarkers, have shown improved performance and robustness compared to single-modality approaches. Longitudinal studies have highlighted the value of AI in modeling AD progression and identifying individuals at risk of rapid decline. However, challenges remain in data standardization, model interpretability, generalizability, clinical integration, and ethical considerations. CONCLUSION: AI techniques applied to neuroimaging data have the potential to improve early AD diagnosis, prognosis, and management. Addressing challenges related to data standardization, model interpretability, generalizability, clinical integration, and ethical considerations is crucial for realizing the full potential of AI in AD research and clinical practice. Collaborative efforts among researchers, clinicians, and regulatory agencies are needed to develop reliable, robust, and ethical AI tools that can benefit AD patients and society.

Fatigue and perceived fatigability, not objective fatigability, are prevalent in people with post-COVID-19.

Persistent symptoms after acute COVID-19 infection, termed post-COVID-19 fatigue, occur in 44-70% of patients. Characterizing fatigue in this population is vital to determine the etiology of post-COVID-19 fatigue symptoms and to assess the effectiveness of potential interventions. The purpose of this study was to assess differences in perceived and objective fatigability between people with post-COVID-19 symptoms (N = 29, 20 females) and people who had COVID-19 but are not experiencing persistent symptoms (N = 20, 12 females). Perceived fatigability, fatigue, pain, and quality of life were assessed with the Fatigue Severity Scale (FSS), Fatigue Assessment Scale (FAS), Visual Analog Scale for Pain (VAS), and the EQ-5D-5L, respectively. Objective fatigability was evaluated with torque and work fatigue indices (FI-T and FI-W), calculated via an isokinetic fatigue task. The results revealed that, the subjects with post-COVID-19 symptoms had significantly higher FAS (p < 0.01), FSS (p < 0.01), VAS (p < 0.01), and EQ-5D-5L VAS (p < 0.01) scores compared to subjects without post-COVID-19 symptoms, indicating greater fatigue and perceived fatigability, increased pain, and worse quality of life. However, there were no differences between the two groups for the FI-Ts (all p ≥ 0.07) or FI-W (all p ≥ 0.08), indicating no differences in objective fatigability. This study found that people with post-COVID-19 symptoms have increased fatigue and perceived fatigability, but not objective fatigability, compared to subjects without post-COVID-19 symptoms.

Cerebellar Contributions to Motor Impairments in People with Multiple Sclerosis.

Although Charcot characterized classic cerebellar symptoms in people with multiple sclerosis (PwMS) in 1877, the impact of cerebellar dysfunction on MS symptoms has predominately been evaluated in the last two decades. Recent studies have clearly demonstrated the association between cerebellar pathology, including atrophy and reduced fractional anisotropy in the peduncles, and motor impairments, such as reduced gait velocity and time to complete walking tasks. However, future studies using novel imaging techniques are needed to elucidate all potential pathophysiology that is associated with disability in PwMS. Additionally, future studies are required to determine the most effective treatments for motor impairments in PwMS, including the specific type and duration of exercise interventions, and potential means to amplify their effects, such as transcranial direct current stimulation (tDCS). This mini-review critically discusses the distinct role of cerebellar dysfunction in motor impairments in PwMS, potential treatments, and directions for future studies.

Women report more severe sensations from 2 mA and 4 mA transcranial direct current stimulation than men.

Interest in transcranial direct current stimulation (tDCS) to alter cortical excitability, facilitate neural plasticity, and improve performance is increasing. Subjects often report temporary stimulation-related sensations, which might distract from the task being performed or compromise blinding. tDCS is also prone to high outcome irregularity and one potential variability source is the biological sex of the subject. The purpose of this study was to re-analyze existing tolerability data to ascertain any sex differences in sensation severity and blinding guesses from tDCS at 2 mA and 4 mA. Each subject underwent tDCS at three randomly ordered intensities (sham, 2 mA, 4 mA), reported the severity sensations experienced, and guessed which tDCS condition they underwent (blinding). Women reported higher sensation severities than men from 2 mA and 4 mA tDCS and higher severities with increasing intensity (sham < 2 mA < 4 mA). Men reported similar severities in all stimulation conditions. Both sexes distinguished sham from 2 mA and 4 mA, and neither were able to discriminate between 2 mA from 4 mA. This study highlights differences in severity reports between women and men and adds to the growing body of literature, indicating that current sham methodologies might be inadequate to maintain blinding.

Increased leg muscle fatigability during 2 mA and 4 mA transcranial direct current stimulation over the left motor cortex.

Transcranial direct current stimulation (tDCS) using intensities ≤ 2 mA on physical and cognitive outcomes has been extensively investigated. Studies comparing the effects of different intensities of tDCS have yielded mixed results and little is known about how higher intensities (> 2 mA) affect outcomes. This study examined the effects of tDCS at 2 mA and 4 mA on leg muscle fatigability. This was a double-blind, randomized, sham-controlled study. Sixteen healthy young adults underwent tDCS at three randomly ordered intensities (sham, 2 mA, 4 mA). Leg muscle fatigability of both legs was assessed via isokinetic fatigue testing (40 maximal reps, 120°/s). Torque- and work-derived fatigue indices (FI-T and FI-W, respectively), as well as total work performed (TW), were calculated. FI-T of the right knee extensors indicated increased fatigability in 2 mA and 4 mA compared with sham (p = 0.01, d = 0.73 and p < 0.001, d = 1.61, respectively). FI-W of the right knee extensors also indicated increased fatigability in 2 mA and 4 mA compared to sham (p = 0.01, d = 0.57 and p < 0.001, d = 1.12, respectively) and 4 mA compared with 2 mA (p = 0.034, d = 0.37). tDCS intensity did not affect TW performed. The 2 mA and 4 mA tDCS intensities increased the fatigability of the right knee extensors in young, healthy participants, potentially from altered motor unit recruitment/discharge rate or cortical hyperexcitability. Despite this increase in fatigability, the TW performed in both these conditions was not different from sham.

FDG-PET detects nonuniform muscle activity in the lower body during human gait.

INTRODUCTION: Nonuniform muscle activity has been partially explained by anatomically defined neuromuscular compartments. The purpose of this study was to investigate the uniformity of skeletal muscle activity during walking. METHODS: Eight participants walked at a self-selected speed, and muscle activity was quantified using [18 F]-fluorodeoxyglucose positron emission tomography imaging. Seventeen muscles were divided into 10 equal length sections, and within muscle activity was compared. RESULTS: Nonuniform activity was detected in 12 of 17 muscles (ƒ > 4.074; P < 0.046), which included both uni- and multi-articular muscles. Greater proximal activity was detected in 6 muscles (P < 0.049), and greater distal versus medial activity was found in the iliopsoas (P < 0.042). CONCLUSIONS: Nonuniform muscle activity is likely related to recruitment of motor units located within separate neuromuscular compartments. These findings indicate that neuromuscular compartments are recruited selectively to allow for efficient energy transfer, and these patterns may be task-dependent. Muscle Nerve 54: 959-966, 2016.

Cognitive-Motor Related Brain Activity During Walking: Differences Between Men and Women With Multiple Sclerosis.

OBJECTIVE: To determine if sex differences in glucose uptake, a marker of brain activity, are present in brain regions that facilitate walking performance in persons with multiple sclerosis (MS). DESIGN: Cross-sectional, observational pilot. SETTING: University laboratory. PARTICIPANTS: Positron emission tomography with fluorine-18-labeled deoxyglucose (FDG) was performed on persons with MS and healthy controls (4 men and 4 women per group; N=16) after a 15-minute walking test. INTERVENTIONS: Not applicable. MAIN OUTCOME MEASURE: Brain activity was quantified as the mean standardized uptake value (SUV). RESULTS: The mean SUV was significantly lower in the thalamus (P=.029) and cerebellum (P=.029) for men with MS compared with women with MS, but not for the prefrontal (P=.057) or frontal (P=.057) cortices. Similar nonsignificant trends were found for healthy controls. No mean SUV group × sex interaction effects were found between the MS and healthy control groups (all P>.05). CONCLUSIONS: To our knowledge, this is the first study of brain activity sex differences based on FDG uptake in persons with MS during walking. Significantly less FDG uptake in the thalamus and cerebellum brain regions important for walking performance was found in men with MS compared with women with MS; however, these comparisons were not significantly different in the healthy control group. No differences in FDG uptake were found between the MS and healthy control groups in any of the brain regions examined. Results from this study provide pilot data for larger studies aimed at identifying underlying mechanisms responsible for accelerated disability in men with MS.

Brown adipose and central nervous system glucose uptake is lower during cold exposure in older compared to young men: a preliminary PET study.

The purpose of this study was to determine the activity of brown adipose tissue (BAT) and the central nervous system (CNS) during cold exposure in young and older men. Two young, 24 and 21 years, and two older, 76 and 74 years, men participated in the study. Positron emission tomography images showed cold-induced BAT activity was absent in older men but clearly present in the clavicular region of the young men (Standardized Uptake Value: SUVmean: 3.12 and 3.71). Statistical parametric mapping revealed cortical brain activity was lower in the older men within areas of the frontal, parietal, temporal, and occipital lobes, and the thalamus (peak-level p uncorr  < 0.036). Cervical spinal cord SUVmean values tended to be lower for older (SUVmean: 1.64 and 1.61) compared to young men (SUVmean: 1.91 and 1.71). These preliminary findings suggest lower BAT activity in older men may in part be due to lower CNS activity.

Positron emission tomography detects greater blood flow and less blood flow heterogeneity in the exercising skeletal muscles of old compared with young men during fatiguing contractions.

The purpose of this study was to investigate blood flow and its heterogeneity within and among the knee muscles in five young (26 ± 6 years) and five old (77 ± 6 years) healthy men with similar levels of physical activity while they performed two types of submaximal fatiguing isometric contraction that required either force or position control. Positron emission tomography (PET) and [(15)O]-H2O were used to determine blood flow at 2 min (beginning) and 12 min (end) after the start of the tasks. Young and old men had similar maximal forces and endurance times for the fatiguing tasks. Although muscle volumes were lower in the older subjects, total muscle blood flow was similar in both groups (young men: 25.8 ± 12.6 ml min(-1); old men: 25.1 ± 15.4 ml min(-1); age main effect, P = 0.77) as blood flow per unit mass of muscle in the exercising knee extensors was greater in the older (12.5 ± 6.2 ml min(-1) (100 g)(-1)) than the younger (8.6 ± 3.6 ml min(-1) (100 g)(-1)) men (age main effect, P = 0.001). Further, blood flow heterogeneity in the exercising knee extensors was significantly lower in the older (56 ± 27%) than the younger (67 ± 34%) men. Together, these data show that although skeletal muscles are smaller in older subjects, based on the intact neural drive to the muscle and the greater, less heterogeneous blood flow per gram of muscle, old fit muscle achieves adequate exercise hyperaemia.

Sex differences in time to task failure during early pubertal development.

INTRODUCTION: We compared fatigability and activation of elbow flexor muscles in children at 3 pubertal stages during a sustained submaximal contraction. METHODS: In 72 healthy children (39 boys) aged 11 ± 3 years (range, 8-14 years), differences in fatigability (time to task failure) and muscle activation were compared at 3 Tanner stages (T1-T3). RESULTS: Time to task failure and muscle activation were similar between boys and girls at prepubertal Tanner stage 1. Time to task failure was briefer for girls than boys at Tanner stages 2 and 3 and was predicted by the coactivation indices and percent body fat in girls. Muscle torque was the only predictor for the time to task failure in boys. CONCLUSIONS: Differences in fatigability and muscle coactivation were evident during the initial pubertal stages (T2 and T3), but not before the onset of puberty (T1).

Artificial Intelligence's Transformative Role in Illuminating Brain Function in Long COVID Patients Using PET/FDG.

Cutting-edge brain imaging techniques, particularly positron emission tomography with Fluorodeoxyglucose (PET/FDG), are being used in conjunction with Artificial Intelligence (AI) to shed light on the neurological symptoms associated with Long COVID. AI, particularly deep learning algorithms such as convolutional neural networks (CNN) and generative adversarial networks (GAN), plays a transformative role in analyzing PET scans, identifying subtle metabolic changes, and offering a more comprehensive understanding of Long COVID's impact on the brain. It aids in early detection of abnormal brain metabolism patterns, enabling personalized treatment plans. Moreover, AI assists in predicting the progression of neurological symptoms, refining patient care, and accelerating Long COVID research. It can uncover new insights, identify biomarkers, and streamline drug discovery. Additionally, the application of AI extends to non-invasive brain stimulation techniques, such as transcranial direct current stimulation (tDCS), which have shown promise in alleviating Long COVID symptoms. AI can optimize treatment protocols by analyzing neuroimaging data, predicting individual responses, and automating adjustments in real time. While the potential benefits are vast, ethical considerations and data privacy must be rigorously addressed. The synergy of AI and PET scans in Long COVID research offers hope in understanding and mitigating the complexities of this condition.

Long COVID in Brain Health Research: A Call to Action.

The COVID-19 pandemic has brought attention to the long-term consequences of the virus, particularly the persistent symptoms that characterize long COVID. This syndrome, which can last for months after the initial infection, includes a range of neurological and neuropsychiatric manifestations that have significant implications for brain health and dementia research. This review explores the current understanding of long COVID's cognitive, neurological, and psychiatric symptoms and their potential impact on brain stimulation and neuroimaging studies. It argues that researchers must adapt their study designs and screening processes to account for the confounding effects of long COVID and ensure the accuracy and reliability of their findings. To advance the understanding of this condition and its long-term effects on brain health, the review proposes a series of strategies, including the development of standardized screening tools, the investigation of underlying mechanisms, and the identification of risk factors and protective factors. It also emphasizes the importance of collaborative research efforts and international data sharing platforms in accelerating the pace of discovery and developing targeted interventions for individuals with long COVID. As the prevalence of this condition continues to grow, it is imperative that the neuroscience community comes together to address this challenge and support those affected by long COVID.

Decoding Post-Viral Fatigue: The Basal Ganglia's Complex Role in Long-COVID.

Long-COVID afflicts millions with relentless fatigue, disrupting daily life. The objective of this narrative review is to synthesize current evidence on the role of the basal ganglia in long-COVID fatigue, discuss potential mechanisms, and highlight promising therapeutic interventions. A comprehensive literature search was conducted using PubMed, Scopus, and Web of Science databases. Mounting evidence from PET, MRI, and functional connectivity data reveals basal ganglia disturbances in long-COVID exhaustion, including inflammation, metabolic disruption, volume changes, and network alterations focused on striatal dopamine circuitry regulating motivation. Theories suggest inflammation-induced signaling disturbances could impede effort/reward valuation, disrupt cortical-subcortical motivational pathways, or diminish excitatory input to arousal centers, attenuating drive initiation. Recent therapeutic pilots targeting basal ganglia abnormalities show provisional efficacy. However, heterogeneous outcomes, inconsistent metrics, and perceived versus objective fatigue discrepancies temper insights. Despite the growing research, gaps remain in understanding the precise pathways linking basal ganglia dysfunction to fatigue and validating treatment efficacy. Further research is needed to advance understanding of the basal ganglia's contribution to long-COVID neurological sequelae and offer hope for improving function across the expanding affected population.

Revealing the Complexity of Fatigue: A Review of the Persistent Challenges and Promises of Artificial Intelligence.

Part I reviews persistent challenges obstructing progress in understanding complex fatigue's biology. Difficulties quantifying subjective symptoms, mapping multi-factorial mechanisms, accounting for individual variation, enabling invasive sensing, overcoming research/funding insularity, and more are discussed. Part II explores how emerging artificial intelligence and machine and deep learning techniques can help address limitations through pattern recognition of complex physiological signatures as more objective biomarkers, predictive modeling to capture individual differences, consolidation of disjointed findings via data mining, and simulation to explore interventions. Conversational agents like Claude and ChatGPT also have potential to accelerate human fatigue research, but they currently lack capacities for robust autonomous contributions. Envisioned is an innovation timeline where synergistic application of enhanced neuroimaging, biosensors, closed-loop systems, and other advances combined with AI analytics could catalyze transformative progress in elucidating fatigue neural circuitry and treating associated conditions over the coming decades.

Associations between neuromuscular function and levels of physical activity differ for boys and girls during puberty.

OBJECTIVE: To compare the associations between neuromuscular performance and anthropometric characteristics with habitual levels of physical activity in boys and girls during the initial stages of puberty. STUDY DESIGN: In a cross-sectional study of 72 healthy children (39 boys and 33 girls) ranging in age from 8 to 14 years, sex differences in anthropometric and motor performance characteristics were compared at 3 Tanner stages (T1-T3). Outcome variables included dual-energy x-ray absorptiometry measurements of body composition, assessments of neuromuscular function, and levels of physical activity (steps/day) measured by accelerometry. RESULTS: Physical activity was lower in girls than boys at T2 and T3, but there was no sex difference at T1. Physical activity increased with Tanner stage for boys but did not differ between Tanner stages in girls. Physical activity at each Tanner stage was strongly associated (R(2) > 0.85) with neuromuscular characteristics for both boys and girls, but percentage of body fat also was associated with physical activity for T3 girls. CONCLUSIONS: The attenuated gains in neuromuscular function experienced by girls in early stages of puberty were strongly associated with lower levels of physical activity, whereas the increase in physical activity exhibited by boys was mostly related to increases in the strength and endurance of leg muscles. Because sedentary activity is a known contributor to the development of obesity and type 2 diabetes in youth, this study helps to identify possible contributors to decreases in physical activity in young girls and provides potential targets for early intervention.

Effect of Post-COVID-19 on Brain Volume and Glucose Metabolism: Influence of Time Since Infection and Fatigue Status.

Post-COVID-19 syndrome (PCS) fatigue is typically most severe <6 months post-infection. Combining magnetic resonance imaging (MRI) and positron emission tomography (PET) imaging with the glucose analog [18F]-Fluorodeoxyglucose (FDG) provides a comprehensive overview of the effects of PCS on regional brain volumes and metabolism, respectively. The primary purpose of this exploratory study was to investigate differences in MRI/PET outcomes between people < 6 months (N = 18, 11 female) and > 6 months (N = 15, 6 female) after COVID-19. The secondary purpose was to assess if any differences in MRI/PET outcomes were associated with fatigue symptoms. Subjects > 6 months showed smaller volumes in the putamen, pallidum, and thalamus compared to subjects < 6 months. In subjects > 6 months, fatigued subjects had smaller volumes in frontal areas compared to non-fatigued subjects. Moreover, worse fatigue was associated with smaller volumes in several frontal areas in subjects > 6 months. The results revealed no brain metabolism differences between subjects > 6 and < 6 months. However, both groups exhibited both regional hypo- and hypermetabolism compared to a normative database. These results suggest that PCS may alter regional brain volumes but not metabolism in people > 6 months, particularly those experiencing fatigue symptoms.

Handedness but not dominance influences variability in endurance time for sustained, submaximal contractions.

The purpose of this study was to compare endurance time and accompanying neuromuscular adjustments when left- and right-handed subjects used the dominant and nondominant arms to sustain submaximal contractions that required either force or position control. Ten left-handed and 10 right-handed healthy adults (21 ± 5 yr) participated in the study. Each subject exerted a similar net torque about the elbow joint during the force and position tasks to achieve a target force of 20% maximal voluntary contraction (MVC) force (56 ± 18 N). MVC force declined to a similar level immediately after task failure for left- and right-handed subjects (27 ± 13 vs. 25 ± 15%, P = 0.9). Endurance time for the position task was similar for the dominant and nondominant arms (task × dominance interaction, P = 0.17). Although the difference in endurance time between the two tasks was similar for left-handed (136 ± 165 s) and right-handed individuals (92 ± 73 s, task × handedness interaction, P = 0.38), there was greater variance in the ratio of the endurance times for the force and position tasks for left-handed (0.77) than right-handed subjects (0.13, P < 0.001; see Fig. 2). Furthermore, endurance time for the force and position tasks was significantly correlated for right-handed subjects (r(2) = 0.62, P < 0.001), but not for left-handed subjects (r(2) = 0.004, P = 0.79). Multiple regression analyses identified sets of predictor variables for each endurance time, and these differed with handedness and task. Hand dominance, however, did not influence endurance time for either group of subjects. These findings indicate that endurance times for the elbow flexors when performing submaximal isometric contractions that required either force or position control were not influenced by hand dominance but did depend on handedness.

Potential Factors That Contribute to Post-COVID-19 Fatigue in Women.

Mortality of acute coronavirus disease (COVID-19) is higher in men than in women. On the contrary, women experience more long-term consequences of the disease, such as fatigue. In this perspective article, we proposed a model of the potential factors that might contribute to the higher incidence of post-COVID-19 fatigue in women. Specifically, psycho-physiological factors are features that might increase central factors (e.g., inflammation) and result in greater perceptions of fatigue. Furthermore, pre-existing conditions likely play a prominent role. This model offers a framework for researchers and clinicians, and future research is required to validate our proposed model and elucidate all mechanisms of the increased incidence and prevalence of post-COVID-19 fatigue in women.

Sample entropy discriminates balance performance of older cannabis users from non-users.

BACKGROUND: Maintaining an upright stance involves a complex interaction of sensory processing and motor outputs to adequately perform this fundamental motor skill. Aging and cannabis use independently disrupt balance performance, but our recent data did not find differences in static balance performance between older cannabis Users and older Non-Users using traditional linear measures (i.e., characteristics of the center of pressure sway). The purpose of this analysis was to determine whether an unbiased entropy measure (sample entropy) can differentiate postural control (standing posture) strategies between older cannabis Users and Non-Users when typical linear measures could not. METHODS: Eight medical cannabis Users and eight age- and sex-matched controls completed static posturography testing in an eyes-open condition for 60 s. Linear measures included pathlength of the anterior-posterior and medio-lateral directions and an ellipse that encapsulates 95% of the 2D area explored. The nonlinear measure was the sample entropy of the center of pressure time-series in anterior-posterior and medio-lateral directions. Group comparisons were accomplished via pairwise testing and effect size calculations. FINDINGS: The statistical testing revealed that sample entropy in the anterior-posterior direction was significantly larger in the Users (mean ± SD = 0.29 ± 0.08) compared to the Non-Users (0.19 ± 0.05; P = 0.01, d = 1.55). INTERPRETATION: This finding indicates that the Users had a decreased regularity of their center of pressure signal in the anterior-posterior direction, which might reflect reduced balance adaptability and accompanies the increased fall risk observed in our recent report on these same subjects.

High Estrogen Levels Cause Greater Leg Muscle Fatigability in Eumenorrheic Young Women after 4 mA Transcranial Direct Current Stimulation.

Transcranial direct current stimulation (tDCS) research has shown great outcome variability in motor performance tasks, with one possible source being sex differences. The goal of this study was to evaluate the effects of estrogen levels on leg muscle fatigability during a fatigue task (FT) after 4 mA tDCS over the left motor cortex (M1). Ten young, healthy eumenorrheic women received 4 mA anodal active or sham stimulation over the left M1 during periods of high and low estrogen levels. A fatigue index (FI) was calculated to quantify fatigability, and the electromyography (EMG) of the knee extensors and flexors was recorded during the FT. The findings showed that tDCS applied during high estrogen levels resulted in greater leg muscle fatigability. Furthermore, a significant increase in EMG activity of the right knee extensors was observed during periods of active stimulation, independent of estrogen level. These results suggest that estrogen levels should be considered in tDCS studies with young healthy women.