Search results
Found 23057 matches for
Are Time Series Foundation Models Ready for Vital Sign Forecasting in Healthcare?
The rise of foundation models, particularly large language models like ChatGPT, has revolutionized natural language processing and demonstrated remarkable generalization across numerous healthcare applications. Building on this success, foundation models for time series forecasting have emerged, offering new opportunities by leveraging pretraining on large-scale datasets. However, existing time series foundation models are pretrained with minimal clinical data, and their potentials for continuously recorded clinical time series, such as vital signs, remain largely under-explored. This motivates our endeavor to integrate time series foundation models with vital sign data to address critical clinical challenges, particularly in predicting patient deterioration. Through an extensive evaluation of various settings and configurations of these models, alongside comparisons with conventional forecasting models, we highlight the significant opportunities for improvement in developing clinically useful time series forecasting models. In a word, the “ChatGPT” moment for time series foundation models, in the typical clinical domain, is yet to come.
Decision cost hypersensitivity underlies Huntington's disease apathy.
The neuropsychiatric syndrome of apathy is now recognized to be a common and disabling condition in Huntington's disease. However, the mechanisms underlying it are poorly understood. One way to investigate apathy is to use a theoretical framework of normal motivated behaviour, to determine where breakdown has occurred in people with this behavioural disruption. A fundamental computation underlying motivated, goal-directed behaviour across species is weighing up the costs and rewards associated with actions. Here, we asked whether people with apathy are more sensitive to costs of actions (physical effort and time delay), less sensitive to rewarding outcomes, or both. Based on the unique anatomical substrates associated with Huntington's disease pathology, we hypothesized that a general hypersensitivity to costs would underpin Huntington's disease apathy. Genetically confirmed carriers of the expanded Huntingtin gene (premanifest to mild motor manifest disease, n = 53) were compared to healthy controls (n = 38). Participants performed a physical effort-based decision-making task (Apple Gathering Task) and a delay discounting task (Money Choice Questionnaire). Choice data was analysed using linear regression and drift diffusion models that also accounted for the time taken to make decisions. Apathetic people with Huntington's disease accepted fewer offers overall on the Apple Gathering Task, specifically driven by increased sensitivity to physical effort costs, and not explained by motor severity, mood, cognition or medication. Drift diffusion modelling provided further evidence of effort hypersensitivity, with apathy associated with a faster drift rate towards rejecting offers as a function of varying effort. Increased delay sensitivity was also associated with apathy, both when analysing raw choice and drift rate, where there was moderate evidence of Huntington's disease apathy drifting faster towards the immediately available (low-cost) option. Furthermore, the effort and delay sensitivity parameters from these tasks were positively correlated. The results demonstrate a clear mechanism for apathy in Huntington's disease, cost hypersensitivity, which manifests in both the effort and time costs associated with actions towards rewarding goals. This suggests that Huntington's disease pathology may cause a domain-general disruption of cost processing, which is distinct from apathy occurrence in other brain disorders and may require different therapeutic approaches.
Evaluating Traditional, Deep Learning and Subfield Methods for Automatically Segmenting the Hippocampus From MRI
ABSTRACTGiven the relationship between hippocampal atrophy and cognitive impairment in various pathological conditions, hippocampus segmentation from MRI is an important task in neuroimaging. Manual segmentation, though considered the gold standard, is time‐consuming and error‐prone, leading to the development of numerous automatic segmentation methods. However, no study has yet independently compared the performance of traditional, deep learning‐based and hippocampal subfield segmentation methods within a single investigation. We evaluated 10 automatic hippocampal segmentation methods (FreeSurfer, SynthSeg, FastSurfer, FIRST, e2dhipseg, Hippmapper, Hippodeep, FreeSurfer‐Subfields, HippUnfold and HSF) across 3 datasets with manually segmented hippocampus labels. Performance metrics included overlap with manual labels, correlations between manual and automatic volumes, volume similarity, diagnostic group differentiation and systematically located false positives and negatives. Most methods, especially deep learning‐based ones that were trained on manual labels, performed well on public datasets but showed more error and variability on clinical data. Many methods tended to over‐segment, particularly at the anterior hippocampus border, but were able to distinguish between healthy controls, MCI, and dementia patients based on hippocampal volume. Our findings highlight the challenges in hippocampal segmentation from MRI and the need for more publicly accessible datasets with manual labels across diverse ages and pathological conditions.
Investigating the impact of electroconvulsive therapy on brain networks and sleep: an observational study protocol.
INTRODUCTION: Electroconvulsive therapy (ECT) is a highly effective treatment for refractory depression, but it may also cause cognitive side effects. Despite decades of use, the mechanisms by which ECT exerts both its antidepressant and cognitive effects are still poorly understood, with the latter substantially limiting referral and adherence to therapy. ECT induces changes in correlated neural activity-functional connectivity-across various brain networks, which may underlie both its clinical efficacy and associated cognitive side effects. Electroencephalography (EEG) could address these knowledge gaps by identifying biomarkers that predict therapeutic outcomes or cognitive side effects. Such developments could ultimately improve patient selection and adherence. Such markers likely span large-scale functional brain networks or temporal dynamics of brain activity during sleep. We hypothesise that enhancement in slow wave sleep mediates the relationship between antidepressant effects and changes in functional connectivity throughout the course of ECT. METHODS AND ANALYSIS: Disruptions of Brain Networks and Sleep by Electroconvulsive Therapy (DNS-ECT) is an ongoing observational study investigating the impact of ECT on large-scale brain functional networks and their relationships to sleep slow waves, an EEG marker linked to synaptic plasticity. The novelty of this study stems from our focus on the assessment of EEG markers during sleep, wakefulness and ECT-induced seizures over the course of therapy. Graph-based network analyses of high-density EEG signals allow characterisation of functional networks locally in specific subnetworks and globally over large-scale functional networks. Longitudinal assessments of EEG alongside clinical and cognitive outcomes provide a unique opportunity to improve our understanding of the circuit mechanisms underlying the development of cognitive impairments and antidepressant effects incurred during ECT. ETHICS AND DISSEMINATION: Recruitment for this 5-year study started in March 2023. Dissemination plans include presentations at scientific conferences and peer-reviewed publications. This study has been registered with ClinicalTrials.gov registry under identifier. TRIAL REGISTRATION NUMBER: NCT05905705.
What is quality in long covid care? Lessons from a national quality improvement collaborative and multi-site ethnography
Abstract Background Long covid (post covid-19 condition) is a complex condition with diverse manifestations, uncertain prognosis and wide variation in current approaches to management. There have been calls for formal quality standards to reduce a so-called “postcode lottery” of care. The original aim of this study—to examine the nature of quality in long covid care and reduce unwarranted variation in services—evolved to focus on examining the reasons why standardizing care was so challenging in this condition. Methods In 2021–2023, we ran a quality improvement collaborative across 10 UK sites. The dataset reported here was mostly but not entirely qualitative. It included data on the origins and current context of each clinic, interviews with staff and patients, and ethnographic observations at 13 clinics (50 consultations) and 45 multidisciplinary team (MDT) meetings (244 patient cases). Data collection and analysis were informed by relevant lenses from clinical care (e.g. evidence-based guidelines), improvement science (e.g. quality improvement cycles) and philosophy of knowledge. Results Participating clinics made progress towards standardizing assessment and management in some topics; some variation remained but this could usually be explained. Clinics had different histories and path dependencies, occupied a different place in their healthcare ecosystem and served a varied caseload including a high proportion of patients with comorbidities. A key mechanism for achieving high-quality long covid care was when local MDTs deliberated on unusual, complex or challenging cases for which evidence-based guidelines provided no easy answers. In such cases, collective learning occurred through idiographic (case-based) reasoning, in which practitioners build lessons from the particular to the general. This contrasts with the nomothetic reasoning implicit in evidence-based guidelines, in which reasoning is assumed to go from the general (e.g. findings of clinical trials) to the particular (management of individual patients). Conclusion Not all variation in long covid services is unwarranted. Largely because long covid’s manifestations are so varied and comorbidities common, generic “evidence-based” standards require much individual adaptation. In this complex condition, quality improvement resources may be productively spent supporting MDTs to optimise their case-based learning through interdisciplinary discussion. Quality assessment of a long covid service should include review of a sample of individual cases to assess how guidelines have been interpreted and personalized to meet patients’ unique needs. Study registration NCT05057260, ISRCTN15022307.
Prevalence of orthostatic intolerance in long covid clinic patients and healthy volunteers: A multicenter study.
Orthostatic intolerance (OI), including postural orthostatic tachycardia syndrome (PoTS) and orthostatic hypotension (OH), are often reported in long covid, but published studies are small with inconsistent results. We sought to estimate the prevalence of objective OI in patients attending long covid clinics and healthy volunteers and associations with OI symptoms and comorbidities. Participants with a diagnosis of long covid were recruited from eight UK long covid clinics, and healthy volunteers from general population. All undertook standardized National Aeronautics and Space Administration Lean Test (NLT). Participants' history of typical OI symptoms (e.g., dizziness, palpitations) before and during the NLT were recorded. Two hundred seventy-seven long covid patients and 50 frequency-matched healthy volunteers were tested. Healthy volunteers had no history of OI symptoms or symptoms during NLT or PoTS, 10% had asymptomatic OH. One hundred thirty (47%) long covid patients had previous history of OI symptoms and 144 (52%) developed symptoms during the NLT. Forty-one (15%) had an abnormal NLT, 20 (7%) met criteria for PoTS, and 21 (8%) had OH. Of patients with an abnormal NLT, 45% had no prior symptoms of OI. Relaxing the diagnostic thresholds for PoTS from two consecutive abnormal readings to one abnormal reading during the NLT, resulted in 11% of long covid participants (an additional 4%) meeting criteria for PoTS, but not in healthy volunteers. More than half of long covid patients experienced OI symptoms during NLT and more than one in 10 patients met the criteria for either PoTS or OH, half of whom did not report previous typical OI symptoms. We therefore recommend all patients attending long covid clinics are offered an NLT and appropriate management commenced.
Thoracic Outlet Syndrome, United Kingdom: A Retrospective Review of Practice.
BACKGROUND: Thoracic outlet syndrome (TOS) is caused by compression of the neurovascular bundle at the thoracic outlet which often poses a diagnostic challenge. Patient management is often based on surgeon choice and experience. This study aims to describe practices relating to the diagnosis and management of TOS in the UK over a 1-year period. METHODS: This multicenter retrospective UK study included data from 16 vascular centers, analyzing surgical management and postoperative outcomes of patients treated for TOS in 2019. Outcomes were evaluated by TOS type: neurogenic (nTOS), venous (vTOS), or arterial TOS (aTOS). RESULTS: Data on 133 patients from 16 units were collected over a 1-year period. Most patients were female (87 of 133; 65%). Surgeries addressed nTOS (53 of 133; 40%), vTOS (48 of 133; 36%), and aTOS (32 of 133; 24%), with TOS type unspecified in 2 patients. Five imaging modalities were used for diagnosis. Surgical approaches included supraclavicular (90 of 133; 68%), transaxillary (23 of 133; 17%), infraclavicular (13 of 133; 10%), paraclavicular (6 of 133; 5%), and thoracoscopic (1 of 133; <1%). Pleural injury was the most reported complication (16 of 133; 12%). Most patients with pleural injury were managed conservatively, with only one-quarter requiring the insertion of a chest drain (4 of 16; 25%). Most patients (119 of 133; 89%) had symptom resolution, lower in nTOS compared to arterial and vTOS (P
The role of N-glycans in regulatory T cells in autoimmunity.
Regulatory T cells (Tregs) have a key role in the maintenance of immune tolerance and in the prevention of autoimmunity. Recent studies have shown an association between decreased Treg frequency and a deficient suppressive activity with the development of many autoimmune diseases. Although glycosylation, which consists in the addition of glycans to proteins and lipids on the cell surface, is recognized as a critical modification for T cell development and function, the relevance of glycans in Treg biology and activity, as well as their impact in the immunopathogenesis of autoimmune diseases, deserves more attention, as it is far from being fully understood. This review discusses the biological impact of N-glycans in Treg biology, highlighting their potential to uncover novel pathogenic mechanisms in autoimmunity and new targets for promising therapeutic approaches with clinical applications in autoimmune disease patients.
Augmenting rehabilitation robotics with spinal cord neuromodulation: A proof of concept.
Rehabilitation robotics aims to promote activity-dependent reorganization of the nervous system. However, people with paralysis cannot generate sufficient activity during robot-assisted rehabilitation and, consequently, do not benefit from these therapies. Here, we developed an implantable spinal cord neuroprosthesis operating in a closed loop to promote robust activity during walking and cycling assisted by robotic devices. This neuroprosthesis is device agnostic and designed for seamless implementation by nonexpert users. Preliminary evaluations in participants with paralysis showed that the neuroprosthesis enabled well-organized patterns of muscle activity during robot-assisted walking and cycling. A proof-of-concept study suggested that robot-assisted rehabilitation augmented by the neuroprosthesis promoted sustained neurological improvements. Moreover, the neuroprosthesis augmented recreational walking and cycling activities outdoors. Future clinical trials will have to confirm these findings in a broader population.
Isolation of the parent triplet titanocene via NHC stabilisation
We present the synthesis and characterization of the parent isolable monomeric triplet titanocene complex, stabilized by the N-heterocyclic carbene (NHC) IMe4. Investigated by SQUID magnetometry and quantum-chemical calculations, reactivity studies of the titanium precursor [Cp2Ti(btmsa)] (2) with the NHC IiPr2Me2 and the zirconocene complex [Cp2Zr(py)(btmsa)] (1) revealed divergent reactivity patterns, highlighting the role of steric and electronic effects in stabilization.