BMC Neurology

Structured AbstractO_ST_ABSObjectiveC_ST_ABSWe represent primary headaches of the ICHD3 in matrix form and show that this representation allows for automated diagnosis as well as additional insights into headache classification. MethodsEach diagnosis in the ICHD3 is defined by a list of characteristics; combinations of characteristics form phenotypes. Multiple phenotypes may fit a given diagnosis. We first translated all characteristics for primary headache diagnoses in the ICHD3 into true/false statements. We generated a matrix of valid ICHD3 diagnosis as follows: O_LIEach row of the matrix represents a phenotype. C_LIO_LIEach column of the matrix represents a characteristic. C_LIO_LIIf any phenotype contains a characteristic, then that element is encoded as 1. Otherwise, it is encoded as 0. C_LI From this matrix, we calculated its bipartite projection and Markov cluster. We also row reduced to derive the basis vectors that span the space of all headache phenotypes. ResultsChronic migraine diagnoses as well as the characteristics "greater than 15 days per month" and "more than 3 months" have the strongest associations based on bipartite projection. Markov clustering yields 64 clusters. These clusters can be organized by ICHD3 diagnoses and demonstrates the level of fragmentation of individual diagnosis in the classification: Migraine is composed of 1 cluster, for example, whereas paroxysmal hemicrania can be broken down into 9 clusters. Finally, row reduction of our matrix yields 63 basis vectors, implying that all headache diagnoses in the ICHD3 can be represented as linear combinations of 63 characteristics. These 63 characteristics corresponds to the following: duration, frequency, aura characteristics, size/location, laterality, clearly remembered onset, TAC features, total number of episodes, severity, nausea/vomiting, photophobia, pulsating, alleviation by triptans, and association with awakening, sexual activity, physical activity, temperature, compression or traction, coughing. ConclusionOur result demonstrates that ICHD3 is a mathematical entity and that headache diagnoses exist in a 63-dimensional vector space. This mathematical embodiment of classification allows us to conduct 1) large scale systematic investigations of relationships between headache and phenotypes, 2) generate a graphical representation of characteristics and phenotypes and 3) improves diagnostic accuracy and efficiency.

BackgroundChronic autoimmune diseases such as Chronic Inflammatory Demyelinating Polyneuropathy (CIDP), Multifocal Motor Neuropathy (MMN), and Thyroid Eye Disease (TED) impose a considerable burden on affected individuals. Patient-reported outcome measures (PROMs)--both disease-specific and generic--are widely used to assess functioning, quality of life, and treatment effects in these populations. However, most PROMs currently lack reference values derived from the general population, limiting the interpretability of patient scores. ObjectiveThe GENESIS (GENEral population normS--An International Survey) study aims to establish general population norms for a range of PROMs used in CIDP, MMN, and TED across six countries: Germany, Italy, Japan, Spain, the United Kingdom, and the United States. These norms will improve patient score interpretation and help quantify unmet needs in patients with these rare autoimmune diseases. MethodsGENESIS is an observational, cross-sectional, online survey of the adult general population (N=21,000). Participants will be recruited to be representative by age, gender, region, and education. The survey includes validated instruments such as the EQ-5D-5L, I-RODS, MMN-RODS, CAP-PRI, GO-QoL, BPI-SF, RT-FSS, FACIT-Fatigue, HADS, and WPAI, along with items on demographics, caregiver need, and healthcare utilization. To reduce respondent burden, participants will be randomized into two groups, each completing a subset of the full questionnaire. A subset of respondents (n=2,333) will be re-surveyed after two months to support psychometric validation. Data will be analyzed descriptively to generate normative values for each PROM by country and in aggregate. Results and DisseminationData collection is scheduled to begin in August 2025, with results expected by Q4 2025. Findings will be disseminated via peer-reviewed publications and conference presentations. ConclusionGENESIS will provide foundational normative data across six countries for PROMs commonly used in rare autoimmune diseases. These data will support more meaningful interpretation of PROM scores in both clinical practice and research settings.

2- AbstractO_ST_ABSOBJECTIVEC_ST_ABSTo compare, in a Brazilian population, the clinical efficacy and quality-of-life (QoL) impact of one-stage bilateral thoracic sympathectomy (BTS) versus unilateral sympathectomy on the dominant side (UniS), with additional analysis of patients who later underwent contralateral surgery (two-stage bilateral, 2stS). METHODSProspective, randomized, controlled, multicenter trial (11 centers) including 163 adults with primary palmar hyperhidrosis. Participants were randomized 1:1 to BTS or UniS. From 6 months onward, UniS patients could elect contralateral sympathectomy (2stS). Sweating severity was assessed using the Hyperhidrosis Disease Severity Scale (HDSS) across 18 anatomical sites at each visit. Compensatory sweating (CS) was defined as new sweating in previously unaffected areas (preoperative HDSS = 1) and graded by the magnitude of HDSS increase. QoL was measured with two complementary validated instruments: HidroQOL and the Horn questionnaire. RESULTSBaseline characteristics were similar between groups, with most participants presenting severe preoperative disease. Improvement in the operated (dominant) hand was comparable after BTS and UniS, whereas control of the non-operated hand favored BTS. In the UniS group, spontaneous contralateral improvement occurred in approximately one-seventh of untreated hands. The proportion of patients without CS was similar in both groups ([~]25%), but severe CS was more frequent after BTS (40.4% vs 21.0%, p = 0.0344). QoL improved in both groups, with larger and more sustained reductions in Horn and HidroQOL scores after BTS (p < 0.001). In the 2stS subgroup, contralateral surgery produced a consistent HDSS decrease and marked QoL improvement, with predominantly mild additional CS. CONCLUSIONSBTS provides more complete symptom control and greater QoL improvement, but at the cost of more severe CS. UniS offers excellent control on the treated side, may reduce severe CS, and supports a staged strategy in which some patients avoid a second procedure (requested by 22.5% in this study); when needed, contralateral completion tends to restore additional clinical and QoL gains.

Artificial intelligence (AI) is increasingly being integrated into healthcare, particularly in data-intensive chronic diseases that rely on longitudinal monitoring and shared decision-making. Multiple sclerosis is a prototypical example of such care, but real-world benefit will depend on whether people accept AI support in different clinical roles. We conducted a cross-sectional, web-based survey among 241 people with MS (pwMS) to assess comfort with AI across eight clinical domains and to identify predictors of acceptance. We derived an artificial-intelligence attitudes composite with high internal consistency (Cronbach alpha = 0.90). Overall acceptance was moderate (mean 3.39 {+/-} 0.78). Acceptance differed across domains, demonstrating a responsibility gradient: comfort was highest for supportive applications such as chronic management (54.4%) and symptom screening (50.2%), but lower for treatment selection (38.6%) and diagnosis (35.3%; P < 0.001). In multivariable models, frequent general AI use (at least weekly; 30.7%) was the strongest independent predictor of acceptance (P < 0.001). Acceptance also differed by region (Eastern vs Western Germany, P = 0.025), whereas clinical disability was not significantly associated. Older age was associated with lower acceptance of AI-supported management. Most participants viewed AI as a logistical support tool but, assuming equal diagnostic accuracy, 78.8% preferred joint artificial-intelligence-clinician decision-making with clinician final responsibility. These findings indicate that acceptance is context-dependent and aligns more closely with prior familiarity than with disease severity. Implementation should move beyond technical validation to transparent, clinician-ledhuman-in-the-loop workflows with explicit accountability and staged adoption beginning with low-risk use cases. Author SummaryWe use artificial intelligence more and more in everyday life, and similar tools are now being introduced into medical care. For long-term conditions such as multiple sclerosis, digital systems could help manage large amounts of clinical information and support monitoring between visits. At the same time, these tools will only be useful if the people receiving care are willing to use them and understand what role they play. In this study, we asked 241 people living with multiple sclerosis in Germany how comfortable they would feel with artificial intelligence in different parts of care. We found that comfort depended strongly on the task. Participants were most open to artificial intelligence when it supported practical, lower-risk functions such as ongoing monitoring or symptom screening, and they were more cautious when it was described as influencing diagnosis or treatment choices. Most participants wanted clinicians to remain responsible for final decisions. Acceptance was higher among people who already used artificial intelligence frequently in everyday life, and it differed by age and by region. Our findings suggest that successful implementation will require more than technical performance: it should be introduced transparently, with clinician oversight, and in a stepwise way that builds familiarity without shifting responsibility away from the clinical team.

Neuropathic pain (NP) affects approximately 60% of individuals with spinal cord injury (SCI). Existing pharmacological treatments provide only modest relief and are often limited by adverse effects, while non-pharmacological options show small effects at best. As such, there remains a need for accessible, mechanism-informed treatments for SCI-NP. This protocol describes a trial evaluating two promising home-based neuromodulatory interventions for SCI-NP - electroencephalography neurofeedback (EEG-NF) and transcranial direct current stimulation (tDCS) - tested both independently and when applied in combination. We will employ a partially double-blinded (i.e. 1 treatment blinded, the other not), 2x2 factorial randomised controlled trial. Adults with chronic SCI-NP (N=192) will be randomised to: (1) EEG-NF + active tDCS, (2) EEG-NF + sham tDCS, (3) active tDCS alone, or (4) sham tDCS alone, in addition to treatment as usual. Participants will complete 20 home-based sessions over 5 weeks. The primary outcome is change in overall pain severity with the primary endpoint being 6 weeks post-randomisation, with secondary endpoints at 16, 26 and 52 weeks post-randomisation. Secondary outcomes (worst pain intensity, pain interference, sleep, depressive symptoms, health-related quality of life) will be assessed at 6 weeks, 16 weeks, 26 weeks and 52 weeks post-randomisation. This will be the first large-scale trial of home-based EEG-NF and tDCS for SCI-NP. If found to be effective, these scalable interventions could be integrated into routine care and inform further optimisation of neuromodulation strategies for managing SCI-NP.

Background Venous thromboembolism (VTE) remains a major cause of perioperative morbidity in cranial neurosurgery, yet clinical practice varies widely, and formal guidelines are inconsistent. Understanding internationally sampled neurosurgical practice is essential for informing consensus and future trials. Methods An international, 2-stage cross-sectional, internet-based survey was conducted. Practising neurosurgeons performing elective adult cranial surgery were eligible. Descriptive statistics were used to summarise practice. Responses covered patterns of pre-operative haemostasis decision making, use and timing of mechanical and/or chemical prophylaxis, use of perioperative imaging prior to anticoagulation, and frequency of clinical assessment for VTE. Associations with geographical income status, subspecialty, and years post-certification were statistically tested. Practice heterogeneity was quantified and contextual influence was summarised using mean effect sizes across stratifying variables in order to determine domains of true equipoise. Results Of 585 responses, 456 (78%) met criteria for inclusion: representing 322 units across 78 countries (71% high-income). Thirteen per cent reported no departmental VTE plan; 23% followed no guidelines and 12% used multiple. Routine pre-operative testing almost universally included haemoglobin/platelets/haematocrit, with fibrinogen more common in high-income settings. Compared with high-income country respondents, low- and middle-income respondents reported higher haemoglobin transfusion thresholds (>90 g/dL; p<0.001) and shorter antiplatelet interruption (p[&le;]0.03), and less frequent outpatient VTE assessment (p<0.001). Mechanical prophylaxis was common (TEDs 81%, IPC 62%), typically started pre- or intra-operatively. Among those completing the chemoprophylaxis section (n=310), 57% required a CT or MRI scan before LMWH which was then initiated on average 31.4 hours after surgery. 1% of respondents did not routinely use LMWH. Many clinical decisions demonstrated statistical equipoise ie. high heterogeneity with low contextual influence. Conclusion Peri-operative haemostasis and VTE prophylaxis practices in adult elective cranial neurosurgery vary substantially worldwide, with some decisions reflecting geographical or socioeconomic differences and many others reflecting true clinical equipoise rather than contextual determinants. By mapping contemporary real-world practice across diverse health-system contexts, this study provides a necessary empirical foundation for rational trial design and future guideline development.

BackgroundPostoperative delirium is a common complication in surgical patients, and is associated with a multitude of negative outcomes, including mortality, dementia, and increased healthcare costs. Therefore, a better understanding of what factors contribute to postoperative delirium, especially those that can be easily obtained, is important. MethodsWe conducted a retrospective cohort study using patients from the Medical Information Mart for Intensive Care (MIMIC)-IV database. Adult patients undergoing procedures in abdominal surgery who did not have pre-existing delirium were included in the study. Overall, we included 8022 procedures across 7212 patients. For each admission, we extracted values obtained from common blood tests, the Charlson and Elixhauser comorbidity score, and patient demographic information. We used stepwise logistic regression to identify predictive factors of postoperative delirium in this cohort. ResultsThe model isolated factors well known to be associated with postoperative delirium, such as age, comorbidity (as represented by the Elixhauser comorbidity score), and Parkinsons disease. The model also selected variables that are less studied, such as minimum preoperative platelets and maximum preoperative sodium levels. We hypothesize that the former is associated with postoperative delirium as a surrogate marker for inflammation as an acute phase reactant, and the second due to it being a marker for cerebral edema and altered neurotransmission. ConclusionPreoperative blood tests contain valuable information that can be used alongside patient demographics and past medical history to better predict the risk of postoperative delirium.

BackgroundFunctional neurological disorder (FND) is one of the most common, but least researched, conditions in neurology. Debate exists as to whether the clinical entity referred to as FND is truly a single disorder or is in fact multiple entities which have been erroneously amalgamated into the same condition. We sought to provide empirical evidence on this question by treating it as a problem of model comparison. MethodsWe formulated statistical models equivalent to: (1) FND being a single entity with variation in phenotype, represented by latent trait (binary factor/item response theory) models, and (2) FND being multiple discrete entities, represented by latent class analysis (LCA) models. We fitted these models to data on the symptoms experienced by 697 people with FND from the FND Research Connect database (fnd-research.org) and used Bayesian model comparison methods to compare them. ResultsAll but one of the latent trait models, representing FND as a single entity with heterogeneous phenotype, fit the data better than all the LCA models. Secondary analysis of the LCA models showed results compatible with the models capturing discretisation of continuous variation rather than true discrete categories. DiscussionOur results suggest that the symptom structure of FND is the result of a single pathophysiological process, either as a single entity, or a common pathway preceded by multiple causative processes where the common pathway is solely responsible for the phenotype of the condition.

Amyotrophic Lateral Sclerosis (Lou Gehrigs disease) is a progressive neurodegenerative disease affecting hundreds of thousands of people worldwide. It is characterized by the degeneration of the neurons in the brain and spinal cord of the patients, leading to a loss of control of muscles. Over time, without nerves to stimulate them muscles tend to atrophy. ALS may occur sporadically or run in families; many mutations have been identified for the latter. Treatment of ALS is mostly limited to three approved therapeutic agents: riluzole, edaravone, and tauroursidiol/ sodium phenylbutyrate. Among these, riluzole remains the most effective despite its early discovery. There are no conclusive meta-analysis comparing riluzole monotherapy to all possible co-therapies present. In this work we have attempted to address such a concern and observed that no adjunct therapy significantly improved the performance of riluzole. However, mitochondrial/ oxidative stress modulator and neuroimmune/ neuroexcitability modulator co-therapy exhibited positive trends. Surprisingly, trials were mainly confined to the USA and European countries, indicating unequal demographic representation in ASL research. We have concluded that large double blinded inter-continental RCTs to be carried out for better understanding of the scenario.

ObjectiveTo describe the design, feasibility, and baseline characteristics of the Migraine Impact on Neurocognitive Dynamics (MIND) study, a 30-day smartphone-based cohort for high-frequency assessment of cognition and symptoms in adults with migraine. BackgroundCognitive symptoms are an important component of migraine burden, but they are difficult to measure using single-visit testing or retrospective questionnaires. Repeated smartphone-based assessment may better capture real-world variability in cognition and symptoms. MethodsAdults meeting International Classification of Headache Disorders, 3rd edition, criteria for migraine were enrolled remotely and completed 30 days of once-daily ecological momentary assessments and mobile cognitive tasks delivered through the Mobile Monitoring of Cognitive Change platform. Baseline measures assessed demographics, migraine characteristics, disability, mood, stress, and treatment patterns. Feasibility was evaluated using enrollment, completion, and retention metrics. ResultsA total of 177 participants enrolled (mean age 38.8 {+/-} 11.9 years; 79.7% female), including 80/177 (45.2%) with chronic migraine. Across the 30-day protocol, 3688 daily assessments were completed, representing 70.8% of all possible study days, and 70.6% of participants completed at least 20 days of monitoring. Completion remained above 60% across study days. At baseline, chronic migraine was associated with greater burden than low-frequency and high-frequency episodic migraine, including higher MIDAS scores (98.6 vs. 38.7 and 70.3), more days with concentration difficulty (16.0 vs. 7.9 and 11.5), and more days with functional interference (18.5 vs. 7.6 and 13.0). ConclusionsThe MIND study demonstrates the feasibility of high-frequency smartphone-based assessment of cognition and symptoms in migraine and provides a methodological foundation for future analyses of within-person cognitive and symptom dynamics across the migraine cycle.

IntroductionChronic pain in fibromyalgia may be driven by abnormal ongoing activity in a subclass of C-fibre nociceptors known as Type1B or CMi nociceptors. As is common in C-nociceptor microneurography studies, the modest patient numbers in these prior studies generate large confidence intervals around the point estimate of the prevalence of this abnormal activity. This complicates the interpretation of the relative importance of this ongoing nociceptor activity as a pain generating mechanism in fibromyalgia. The study aims to improve precision via an adaptive Bayesian protocol that maximises the yield and quality of data collection whilst minimising patient burden. MethodsThe study employs an optimised microneurography protocol with an adaptive study design. The microneurography protocol incorporates early identification of CMi nociceptors via an abbreviated activity dependent slowing protocol to increase yields enabling efficient collection of the primary outcome data. The adaptive study design will use Bayesian principles to iteratively assess the predictive probability of futility, and terminate early if there is high confidence that the hypothesis is false. Furthermore, the study will employ questionnaires to explore links with pain in the area under study to the electrophysiology data. Finally, quantitative sensory testing will be used to investigate whether the irritable nociceptor phenotype is associated with abnormalities in CMi nociceptor physiology. Ethics & DisseminationThis study has received HRA REC approval in the UK. Participants will provide written informed consent, and may withdraw at any time without consequence. At the end of the study, the results will be disseminated through peer-reviewed publication, and the data made available via a data repository. Strengths & limitations of this studyBayesian predictive probability of futility to minimise patient burden in microneurography Microneurography for objective interrogation of the peripheral nervous system Optimised microneurography protocol to efficiently answer primary hypotheses Subjective elements of early termination criteria of the study assessed and co-developed with Patient and Public Inclusion and Engagement Group

BackgroundPeople with multiple sclerosis (pwMS) often experience impaired quality of life (QoL) despite receiving standard care. Digital therapeutics (DTx) may offer support, but prior trials yielded mixed results, possibly due to active controls and high baseline QoL. We therefore evaluated a DTx (levidex) as an adjunct to treatment as usual (TAU) in pwMS with impaired QoL. MethodsIn this pragmatic, online randomised controlled trial (LAMONT; NCT06090305), n = 470 pwMS with a score [&ge;]2 on the Hamburg Quality of Life Questionnaire in Multiple Sclerosis (HAQUAMS) were randomised to levidex + TAU or TAU alone. The primary endpoint was HAQUAMS total score at 6 months, analysed by intention-to-treat ANCOVA. ResultsCompared with TAU, levidex + TAU improved MS-specific QoL at 6 months (baseline-adjusted mean difference -0.10; 95% CI -0.18 to -0.03; p = 0.008; Cohens d = 0.26). Clinically relevant HAQUAMS improvement ([&ge;]0.22) occurred more often with levidex (39.5% vs 27.8%; number needed to treat = 9). Benefits also emerged for depressive symptoms and social/work functioning but not for anxiety. No serious adverse events occurred and user satisfaction was high. ConclusionsIn pwMS with impaired QoL, adding the scalable DTx levidex to TAU yields meaningful improvements in QoL and functioning.

Background: For early detection of Alzheimer's disease, it is essential to identify individuals showing cognitive performance consistent with the mild cognitive impairment (MCI) range during preliminary screening, ideally using methods that extend beyond conventional cognitive assessments. Non-invasive, easily accessible screening tools applicable in daily life are increasingly needed. Facial expressions, particularly during rest, may offer promising biomarkers for MCI level detection. This study aimed to identify specific facial features associated with MCI level during rest to inform development of facial expression-based screening tools. Methods: Participants were classified into an MCI level group and a healthy control (HC) group based on the Montreal Cognitive Assessment (MoCA) scores. Facial Action Units (AUs) were extracted from video recordings of resting-state facial expressions in 31 individuals with MCI level and 14 HC. Two statistical models were employed: a multilevel zero-inflated beta regression model for intensity of 17 AUs and a multilevel logistic regression model for presence or absence of 18 AUs. Results: In the zero-inflated beta regression, the AU relates to upper lip raiser showed a significant group effect (MCI level vs. HC; p <0.001), remaining significant after multiple comparison correction. The logistic regression revealed significant group differences for the AUs related to lip tightener (p <0.001) and lip suck (p <0.001), both remained significant after multiple comparison correction. Conclusions: Distinctive facial action patterns during rest were observed in individuals with MCI level. These findings highlight the potential of resting-state facial expressions as a basis for novel, unobtrusive screening tools for early MCI level detection.

Causal discovery algorithms are often leveraged for inferring causal relationships and recovering a causal model from data. However, causal discovery from data alone is limited by the structural constraints of the used dataset, the lack of causal logic, and the lack of external knowledge. Thus, data-driven causal discovery can only suggest possible causal relationships at best. To overcome these limitations, Large Language Models (LLMs) and knowledge systems, such as Retrieval-Augmented Generation (RAG), have been proposed as alternatives to data-driven causal discovery and as a method to augment causal discovery algorithms. Using an expert-defined causal graph of chronic lower back pain, we further propose knowledge graph based RAG systems, such as GraphRAG, as an improvement over RAG systems for augmenting causal discovery (F1 0.745), benchmarking its performance against augmenting causal discovery with an LLM (F1 0.636), augmenting causal discovery with RAG (F1 0.714), and causal discovery alone (F1 0.396). We also explore the impact of different prompting methods for causality, such as querying for the plausibility of causal relationships, the presence of statistical associations, and the existence of temporal causal relationships, as inspired by the methodology of the domain experts constructing our ground truth. Lastly, we discuss how applications of LLMs, RAG, and graph-based RAG systems can impact and accelerate the causal modeling of chronic lower back pain by bridging the gap between domain knowledge and data driven approaches to causal modeling. Graphical Abstract O_FIG O_LINKSMALLFIG WIDTH=200 HEIGHT=93 SRC="FIGDIR/small/26346255v1_ufig1.gif" ALT="Figure 1"> View larger version (31K): org.highwire.dtl.DTLVardef@f3387org.highwire.dtl.DTLVardef@2dforg.highwire.dtl.DTLVardef@bc839aorg.highwire.dtl.DTLVardef@63f6ea_HPS_FORMAT_FIGEXP M_FIG C_FIG

Background: Most trials of Parkinson's disease (PD) measure progression over a short to medium time-period using continuous rating scales that may be hard to interpret and less meaningful for patients. There is a lack of evidence connecting changes in these scales to changes in outcomes important to patients. Objectives: We present causal modelling to translate the causal, short-term disease-modifying treatment effects on functional rating scales to the 10-year risk of serious clinical progression milestones. Methods: We selected four important clinical milestones of disease progression from the Oxford Parkinson's Disease Centre "Discovery" cohort: dementia, any falls, frequent falls, and mortality. We proposed a causal framework for our research objectives so we could model the potential impact of a 30% reduction in disease progression slopes ("treatment effect") using the summation of parts I and II of the Movement Disorders Society Unified Parkinson's Disease Rating Scale (UPDRS12). This outcome was regressed on time to milestone using flexible parametric survival models. Marginal predictions of survival and survival difference at year 10 were then calculated for the Discovery cohort, and a counterfactual cohort applying the treatment effect to estimate the relative and absolute reductions for the four clinical milestones. Results: The model increase in risk for each unit change in the UPDRS12 were as follows: dementia hazard ratio (HR)=1.52 (95% Confidence Interval (CI) 1.36-1.70), any falls HR=1.37 (95% CI 1.29-1.46), frequent falls HR=1.68 (95% CI 1.49-1.89), mortality=1.29 (95% CI 1.17-1.42). These models led to marginal predictions of absolute reductions, when the progression was reduced by 30%, between 4.0% (mortality) and 7.5% (frequent falls) at 10 years follow up. Conclusions: We have demonstrated how a treatment effect in a trial specified in terms of a progression change of a rating scale can be contextualised into a long-term reduction in the probability of clinically relevant milestones. Whilst we have used PD as our exemplar, we believe this methodological approach is generalisable to other chronic progressive diseases where trials are often limited to a relatively short follow-up period and use some scalar measure of progression, but significant clinical milestones usually take longer to be observed. Keywords: Clinical trials; disease modifying therapies; causal estimation; prediction models

IntroductionThis clinical trial investigates the efficacy and safety of a personalized 15-day accelerated intermittent theta-burst stimulation (aiTBS) protocol, targeted at either the default mode network (DMN) or the fronto-parietal network (FPN), in individuals with mild Alzheimers disease (AD). Methods45 patients with mild AD were randomized 1:1:1 to receive 15 consecutive days of high-dose aiTBS (7200 pulses/day) targeting the DMN or FPN, or sham. The primary outcome was the change in ADAS-Cog after 15 days of treatment. ResultsBoth active aiTBS groups demonstrated significantly greater ADAS-Cog improvement than sham at the primary endpoint. Response rates for a clinically meaningful improvement ([&ge;]3-points on ADAS-Cog) were significantly higher in the active groups (DMN: 38%; FPN: 47%) than in the sham group (0%). The improvement in active groups was sustained at 3-month follow-up. DiscussionPersonalized aiTBS targeting the DMN or FPN produced clinically meaningful cognitive benefits in mild AD and was safe.

BackgroundMultiple sclerosis (MS) is a chronic neurodegenerative disease charac-terised by progressive neurological disability and heterogeneous symptom trajectories. Cur-rent clinical monitoring methods, including magnetic resonance imaging (MRI) and episodic neurological assessments, provide limited insight into subtle disease progression and real-world functional changes. Digital health technologies integrating multimodal biosignals and behavioural assessments may enable continuous monitoring and personalised rehabilitation for patients with MS. ObjectiveThis study aims to evaluate the clinical utility of the BodyMirror Clinical MS platform, a multimodal software-as-a-medical-device (SaMD) that combines wearable biosensors, neuroscience-based games, and machine learning algorithms to remotely monitor disease progression and deliver personalised neurorehabilitation for individuals with multiple sclerosis. MethodsThis study is a prospective, randomised, double-blind, controlled, multisite clinical trial enrolling 400 participants, including 300 individuals with multiple sclerosis and 100 healthy controls. MS participants will be randomly assigned (1:1) to either an adaptive neurorehabilitation intervention group or a control group receiving non-therapeutic digital activities matched for engagement and exposure. Participants will perform three 30-minute sessions per week over a 24-month period using the BodyMirror platform. The system integrates multiple biosignals, including electroencephalography (EEG), electromyography (EMG), inertial measurement unit (IMU) motion data, speech analysis, and behavioural performance metrics, to generate digital biomarkers of neurological function. The primary endpoint is change in Expanded Disability Status Scale (EDSS) score from baseline to 24 months. Secondary outcomes include changes in Multiple Sclerosis Functional Composite (MSFC), MRI brain volume, cognitive performance, patient-reported outcomes, adherence to digital rehabilitation, and health-economic outcomes. ConclusionsThis trial will provide the first large-scale clinical evaluation of a mul-timodal digital neurotechnology platform combining wearable biosensors and game-based neurorehabilitation for remote management of multiple sclerosis. If successful, BodyMirror Clinical MS may enable scalable remote monitoring, earlier detection of disease progres-sion, and personalised digital rehabilitation for individuals living with MS.

IntroductionSleep apnoea is common in older adults and a risk factor for cognitive decline and dementia but is rarely assessed in memory clinics. The Sleep Apnoea and Memory (SAM) study will assess the prevalence of sleep apnoea and identify optimal screening for sleep apnoea in memory clinics. MethodsSAM is a prospective observational multi-site study recruiting adults attending NHS memory clinics. Participants will undergo a single night of polygraphy using a home sleep apnoea test (WatchPAT 300) and complete questionnaires based on NICE guidance for sleep apnoea assessment. The primary outcome will be the prevalence of sleep apnoea. Secondary outcomes include determining sleep apnoea prevalence across different cognitive diagnoses, identifying which symptoms and risk factors which best predict sleep apnoea, and assessing feasibility of remote sleep apnoea screening. DiscussionThe SAM study will improve understanding of the extent of sleep apnoea in people attending memory clinics and inform design of an interventional trial for treating sleep apnoea in patients with cognitive impairment. Treating sleep apnoea in memory clinics may help to improve symptoms and/or prognosis for people experiencing memory problems.

ImportanceMigraine attacks often occur unpredictably, limiting the ability of individuals to initiate timely preventive or preemptive treatment. Short-term probabilistic forecasting of migraine risk could enable more targeted management strategies. ObjectiveTo externally validate the previously developed Headache Prediction Model (HAPRED-I), evaluate an updated continuously learning model (HAPRED-II), and assess the feasibility and short-term safety of delivering individualized probabilistic migraine forecasts directly to patients. Design, Setting, and ParticipantsProspective 8-week cohort study conducted remotely at two academic medical centers in the United States (Massachusetts General Hospital and Wake Forest Health Sciences) between 2015 and 2019. Adults with recurrent migraine or tension-type headache completed twice-daily electronic diaries. A total of 230 participants contributed 23,335 diary entries across 11,862 participant-days of observation. Main Outcomes and MeasuresOccurrence of a headache attack within 24 hours following each evening diary entry. Model performance was evaluated using discrimination (area under the receiver operating characteristic curve [AUC]) and calibration. ResultsExternal validation of HAPRED-I demonstrated modest discrimination (AUC, 0.59; 95% CI, 0.57-0.61) and poor calibration, with predicted probabilities consistently exceeding observed headache risk. In contrast, the continuously updating HAPRED-II model demonstrated progressive improvement in predictive performance as participant-specific data accumulated. Discrimination increased from an AUC of 0.59 (95% CI, 0.57-0.61) during the first 14 days to 0.66 (95% CI, 0.63-0.70) after the first month, accompanied by improved calibration across predicted risk levels. Over the study period, 6999 individualized forecasts were delivered directly to participants. No evidence suggested that receipt of forecasts was associated with increasing headache frequency or worsening predicted headache risk trajectories. Conclusions and RelevanceA static migraine forecasting model demonstrated limited transportability to new individuals. In contrast, models that continuously update within individuals may improve predictive accuracy over time and enable real-time delivery of personalized migraine risk forecasts. Further work incorporating richer physiologic and contextual predictors will likely be necessary before such systems can reliably guide clinical treatment decisions.

Traumatic brain injury (TBI) is a significant risk factor for the development of Alzheimers disease (AD) and related dementia. In both TBI and AD, inflammation plays a pivotal role. Recent studies have found that TBI triggers activation of NLRP3 inflammasome in the brain whereas the NLRP3 inflammasome plays an important role in pathogenesis of AD. To evaluate the influence of TBI on the immune response and the importance of the NLRP3 inflammasome in mediating this process we have examined the immune profiles in the brain using a novel transgenic AD mouse line with the NLRP3 gene deleted. Briefly, a group of 3xTg and 3xTg/NLRP3-/- mice received a moderate TBI in the form of lateral fluid percusive injury FPI or sham surgery at the age of 4 months old, an age before the onset of AD. Injury-induced changes in immune profiling were assessed using flow cytometry, real-time quantitative PCR, or Western blot at the acute and subacute stages following TBI. We found that TBI altered immune profiles in 3xTg mice and NLRP3 gene knock out counteracts the injury effect with a significant sex-related difference. More specifically, at the acute stage after TBI in both male and female mice, TBI induced a significant infiltration of neutrophils, macrophages and {gamma}{delta} T-cells in the brains of 3xTg mice, however, NLRP3 knock down significantly blocked this injury effect in males and less so in females. We also found that NLRP3 gene knock down counteracted TBI-enhanced inflammatory cytokine IL1-{beta}, TNF-, and IL-17f expression. The abnormal immune profiling was not significant at 7 days post-injury, however, changes in cerebral vascular associated proteins including GFAP, AQP4, CD31, Occludin were observed in related to injury, NLRP3 and sex. In conclusion, our study has confirmed that TBI significantly alters immune profiles and vascular integrity in the context of predisposition of AD and the NLRP3 inflammasome is important in mediating these TBI-induced changes. Sex-related differences warrant further evaluation to define the role of NLRP3-related immune response and vascular pathology as a result to TBI and the predisposition to AD development.