Sleep staging is a critical process in the study and evaluation of sleep architecture, with its accuracy directly influencing the effectiveness of sleep disorder diagnosis. In temporal sleep staging, a common challenge arises from class imbalance—particularly, the significantly smaller number of N1 stage samples compared to other stages. Existing approaches have primarily focused on oversampling and expanding N1 stage data to address this issue. However, such methods may disrupt the temporal structure of the original signals and potentially lead to overfitting. To overcome these limitations, this paper proposes a Multi-Modal Spatio-Temporal Network for Imbalanced Sleep Staging (MSTN-ISNet), designed to perform data augmentation and enhancement for underrepresented sleep stages, thereby effectively alleviating the problem of insufficient training data for minority classes. First, we introduce a spatio-temporal statistics-driven time-frequency data augmentation module to enhance the representation of minority class samples. Next, we develop a deep interactive cross-modal feature encoding module to extract multi-modal spatio-temporal dynamic features from polysomnography (PSG) signals. Finally, a Markov chain entropy-weighted intelligent decision module is employed to calculate state transition probabilities among sleep stages, capturing the dynamic transitions between stages. Furthermore, to address the issue of class confusion, we implement strategies that reward minority classes, calculate inter-stage similarity, and utilize a weighted cross-entropy loss function to enhance the classification performance for minority stages. We evaluate the model on three benchmark datasets. Experimental results demonstrate classification accuracies of 90.8

Background: Prevention of communicable diseases is crucial for global health. Most studies focus on socioeconomic factors, while the impacts of natural environmental variables (e.g., altitude) on their epidemiology and quantitative relationship with disease burden remain under-researched. Using GBD data and China Health Statistical Yearbook, this study quantified the altitude-disease burden association and its heterogeneity. Methods: Data were from GBD 2023 and China Health Statistical Yearbook 2023, including 40 communicable disease subcategories (excluding maternal, neonatal and nutritional disorders). Incidence and DALYs were primary measures; GAM was used for analysis (5% significance threshold). Altitude (0–3000 m) and age were stratified into 5 and 17 five-year groups, respectively. Temporal trends (1990–2023) and sex-age disparities were evaluated; 2022 provincial notifiable Category A/B disease incidence (31 units) validated GBD estimates. Findings: Thirty diseases showed significant altitude-related burden gradients: 13 (leprosy, lower respiratory infections, tuberculosis, encephalitis, etc.) had >5% increased burden with higher altitude, while 17 (upper respiratory infections, COVID-19, etc.) had decreased burden. Validation confirmed consistent trends for 8 of 11 overlapping diseases. Males had higher burden than females at low altitudes (disparity attenuating at high altitudes), with greater differences after 35 years. Neglected tropical diseases and malaria peaked at 1000–1500 m. Age-altitude interactions were prominent: children under 5 had the heaviest burden at 0–500 m; HIV/AIDS DALYs peaked at 1500–2000 m (males 35–39, females 50–54); COVID-19 DALYs surged with age at 2000–3000 m (highest in males ≥80 and females ≥75). Interpretation: Altitude is an independent variable significantly associated with communicable disease burden; most diseases have increased burden with higher altitude, others decreased or no clear pattern. Sex-age-altitude interactions exist: male-predominant burden at low altitudes diminishes at high altitudes, and males have higher burden across ages (after 35). These findings support precision prevention, stratified resource allocation and targeted interventions based on altitude, sex and age.

Background Transcranial electrical stimulation (tES) is a promising noninvasive neuromodulation technique for home-based sleep intervention, yet traditional long-duration, high-current protocols cause discomfort and low compliance.Methods This study explored the regulatory effects and optimization potential of intermittent short-duration tES on sleep quality. Twenty-two healthy adults participated in three nocturnal sessions design: Sham, low-current stimulation (Lc-Stim), and high-current stimulation (Hc-Stim) conditions. Data were collected via polysomnography (PSG), and sleep stages were scored according to the American Academy of Sleep Medicine (AASM) criteria.Results The results show that intermittent short-duration tES did not significantly increase total sleep time or sleep efficiency (SE) but did significantly reduce sleep onset latency (SOL) in both the Lc-Stim and Hc-Stim groups. Moreover, the intervention facilitated the N2–N3 transition and attenuated sleep fragmentation. At the micro structure level, SO-spindle coupling density was higher in the Hc-Stim group than in Sham. Compared with conventional long-duration and Hc-Stim, scalp discomfort was significantly reduced in both stimulation groups. Additionally, stimulation effects exhibited interindividual variability, with Hc-Stim demonstrating an intensity-dependent threshold effect on micro structure characteristics.Conclusion The short-duration intermittent tES improves sleep quality safely and acceptably by reducing SOL, increasing deep sleep proportion, and enhancing sleep continuity.

Sleep staging is a crucial link between brain function monitoring and regulation. Current deep learning-based methods have unlocked a new paradigm for feature learning in automatic sleep staging modeling. However, most deep learning models still lack robustness to the domain-invariant features of sleep signals. Making them unable to effectively address individual differences and data imbalance in the automatic sleep staging task. In this paper, we propose a delayed adversarial optimization neural network approach that can improve the learning of domain-invariant features in the automatic sleep staging task. Specifically, we introduce the simplex equiangular tight frame (ETF) of neural collapse to depict the domain-invariant features of sleep signals. Constraining the imbalanced staging features to align with the ETF of balanced data learning. Then, to simulate the data property of individual differences among different subjects, we generate adversarial minority samples by perturbing the loss of the staging model. In addition, an ETF joint optimization strategy is established to fine-tune the staging model, achieving the domain-invariant feature gain. We validated the proposed method on MASS and Sleep_EDF datasets under a cross-subject test experiment. The test results showed that our method has significant advantages in addressing the problem of individual differences.

Cooperation and competition are fundamental to human social interaction. While recent hyperscanning studies have linked stronger interbrain synchrony (IBS) to successful cooperation, most have focused on dyadic interactions, leaving the underlying neural mechanisms of group-level social behavior largely unknown. Here, we employed EEG hyperscanning to investigate interbrain neural dynamics of triadic cooperative and competitive interactions. Distinct interbrain network patterns emerged in the delta and beta bands, with cooperation showing enhanced frontal-parietal IBS and more efficient network properties. Non-parametric cluster-based permutation tests further identified significant regional differences in a left-lateralized frontal-temporal-parietal cluster in both bands. Crucially, increased delta-band frontal-parietal IBS was closely associated with better group-level cooperative performance. Moreover, classification and prediction models based on delta-band interbrain metrics successfully distinguished interaction types and predicted cooperative outcomes. These findings uncover interbrain neurocognitive traits that reflect specific social behavioral contexts, highlighting the pivotal role of frontal-parietal synchrony and delta-band modulations in supporting group cooperation. Together, our results advance the understanding of the neural basis of triadic social interaction and underscore the potential of interbrain network signatures as biomarkers for decoding and predicting complex social behaviors.

Purpose This study addresses the data bias arising from different EEG devices in multicenter studies and proposes an EEG data standardization method based on amplitude-frequency response correction to mitigate the impact of device-specific performance variations on the acquired signals.Methods The effectiveness and applicability of the amplitude-frequency response correction algorithm were evaluated through simulation, the amplitude-frequency response data of 41 devices were collected and analyzed, and the effectiveness of the algorithm was verified by processing actual ERP data acquired using these devices.Results The amplitude-frequency response analysis reveals that, after eliminating interference from noise, bad lead, and other factors, responses across channels within the same device exhibit improved consistency. Accordingly, a correction template is generated for each device. Building on this template, the standardized amplitude-frequency response correction algorithm was enhanced. Experimental validation with real data confirmed that our method attenuates the influence of amplitude-frequency response discrepancies in the EEG acquisition system on the collected signals.Conclusion By systematically analyzing the differences in the amplitude-frequency responses of various EEG devices, we developed a set of standardization methods based on their performance parameters and provided a tool to support the standardization of experimental data across EEG acquisition systems in multi-center EEG studies.

The persistent global burden of herpes simplex virus type 2 (HSV-2) requires region-specific therapeutic strategies, yet the limited number of characterized clinical isolates from China has hindered accurate evaluation of local viral evolution and vaccine efficacy. To address this gap, we isolated and comprehensively characterized HSV-2/KM-1, a novel clinical strain obtained from a genital herpes patient in Kunming, China. Whole-genome sequencing showed 99.92% nucleotide identity with contemporary U.S. strains (MH790606, PP099973), which was markedly higher than its homology to China's reference strain HJ12 (128 amino acid differences). This unexpected phylogeographic similarity challenges existing models of geographically restricted HSV-2 evolution and suggests global viral gene flow facilitated by human mobility. Cell tropism analyses revealed accelerated replication in human foreskin fibroblasts (HFF-1), with early viral protein expression at 12 hpi, and high-titer production (7.0 log10 TCID50/mL) in Vero cells at a low MOI (0.001). Comparative genomics identified 27 amino acid substitutions in virulence determinants (ICP4, UL52, UL36, etc.) compared with the U.S strain (PP099973), 14 of which altered residue polarity, potentially influencing viral-host interactions, as suggested by previous studies on these proteins' roles. As a phylogenetically U.S.-linked clinical isolate from China, HSV-2/KM-1 helps to fill a gap in regional pathogen resources and provides a critical tool for assessing globally circulating strains and developing targeted interventions.

Background:Immune-mediated inflammatory disease (IMID) and cancer share underlying mechanisms. We aimed to comprehensively evaluate the associations between IMIDs and cancers from global, population and genetic perspectives. Methods:A triangulation framework was employed to assess the association between IMIDs and cancers, using the Global Burden of Disease Study (2012-2021) to analyse six IMIDs and 33 cancers. The UK Biobank (UKBB) prospective cohort was subsequently used to validate these associations, with hazard ratios (HRs) and 95% confidence intervals (CIs) estimated by Cox proportional hazards models. Causal inference based on genetic instruments was performed in the FinnGen and UKBB to assess the potential causal effects between IMIDs and cancers. Results:IMIDs were positively associated with the occurrence of cancers from a global perspective. Moreover, 170 specific IMID-cancer pairs revealed statistically significant associations. A total of 20 pairs of specific IMID-cancer associations were further confirmed in the UKBB cohort. Among these, the five most pronounced associations included atopic dermatitis with Hodgkin lymphoma (HR = 12.56, 95% CI: 1.76-89.59), with ovarian cancer (HR = 5.65, 95% CI: 1.41-22.65) and with non-Hodgkin lymphoma (HR = 5.11, 95% CI: 1.91-13.63); rheumatoid arthritis with Hodgkin lymphoma (HR = 3.85, 95% CI: 1.11-13.32); and psoriasis with Hodgkin lymphoma (HR = 3.43, 95% CI: 1.69-6.96). Additionally, a positive causal association between rheumatoid arthritis and Hodgkin lymphoma (inverse variance weighted OR = 1.31, 95% CI: 1.10-1.57) was observed. Conclusions:This study provides comprehensive evidence of the relationships between IMIDs and cancers from global, population and genetic perspectives and identifies 20 pairs of specific IMID-cancer associations, thereby contributing to advancements in cancer prevention and control.

Cognitive function is critical for overall health, with vitamin D’s impact under extensive investigation. This review explores the association between vitamin D and cognitive health, its neuroprotective mechanisms, and the therapeutic potential of supplementation in cognitive decline. Observational studies link low vitamin D levels to increased cognitive deterioration risk, particularly in Alzheimer’s disease, vascular dementia, Parkinson’s disease, and schizophrenia. Clinical trial results on vitamin D supplementation’s cognitive benefits are inconclusive. Vitamin D’s neuroprotective effects are complex, influencing cognitive abilities by interacting with neuronal and glial cells, modulating immune responses, and regulating key molecular pathways. Challenges remain in clinical applications, including determining optimal vitamin D levels, effective supplementation forms and doses, and identifying responsive populations. The review advocates for robust clinical trials to address these gaps, facilitating informed use of vitamin D in cognitive health. Future research should focus on the optimal timing, duration, and target groups for supplementation to enhance cognitive outcomes and reduce risks.

Electroencephalography (EEG) is an important electrical signal for recording physiological activities of the brain. Usually, its high temporal resolution can be susceptible to contamination by artefacts such as electrooculogram (EOG) and electromyogram (EMG), which affects the subsequent signal processing and analysis. Therefore, EEG artefact removal is particularly important and can lay the foundation for brain-computer interface (BCI) applications. However, most of the existing studies suffer from insufficient ability to extract global and local coarse-grained and fine-grained features interactively, insufficient effective constraints on latent features, and excessive computational complexity. To address these problems, this study designs a convolutional attention-based adaptive separation network (ASNet) for EEG artefact removal. Here, a separator is introduced based on the U-net architecture to constrain the potential features so as to adaptively separate the required discriminative features. In addition, this study improves on the disadvantages of the Transformer itself by designing a novel convolutional attention module to improve the coarse-grained and finegrained feature interaction learning capability while reducing the number of parameters and computational effort. The experimental results show that ASNet achieves impressive artefact removal performance on fully synthetic datasets, semi-synthetic datasets, and real datasets, and has low computational complexity, which is a significant advantage over existing state-of-the-art methods. The codes are available at https://github.com/ qwertwjq/ASNet/tree/main.

Recent neuroscience research has shed light on heart-brain interactions during diverse information processes across perception, affective, and cognitive domains. It remains unclear how the heartbeat-related interoceptive pathway affects the neural responses of somatosensory information processing. In this study, we combined EEG, ECG, and DTI to examine the effect of heart-brain interaction on cortical somatosensory processing and investigated both the cardiac phase and heart rate effects on somatosensory-evoked high-frequency oscillations (HFOs). First, we examined the somatosensory cortex activity in terms of HFOs along the cardiac cycle and observed an attenuated HFO response in the systole phase. Moreover, voluntary hyperventilation (VH) was adopted as the approach to interoceptive exposure, and a significant HFO decrease was observed after VH in both the systole and diastole phases. Then, we constructed robust fusion models to demonstrate the combined effects of cardiac activity, brain structure, and somatosensory stimulation input on the HFO response. The results showed that there was an important predictive effect of heart rate on somatosensory neural oscillations. These findings revealed the essential regulatory effect of dynamic cardiac activity on brain response during somatosensory information processing, and they may provide a better understanding of the mechanisms underlying the heart-brain interaction by integrating interoceptive and exteroceptive signals.

BACKGROUND:Acute hypoxia exposure leads to a high incidence of acute mountain sickness (AMS) and changes in body composition, while the relationship between body composition and AMS remains unclear. We designed this study to detect the body composition and discern its relationship with AMS. METHODS:Eighty-one subjects were transported from the plain (300 m) to the plateau (3680 m). The body weight and body composition were measured in the plain and at the plateau. The occurrence of AMS was investigated by using the Lake Louise Scoring (LLS) system for six consecutive days. Then, the relationship between body composition and AMS was further analyzed. RESULTS:The body weight, fat-free mass (FFM), total body water (TBW), intracellular water (ICW), extracellular water (ECW) and segmental lean mass decreased significantly with the prolonged stay at the plateau. Compared with those in the plain, FM increased significantly at the plateau, and the mineral increased significantly on the third day while decreasing significantly on the sixth day. The daily incidence of AMS during the first 6 days at the plateau was 23.46 %, 7.41 %, 2.47 %, 2.47 %, 1.23 %, and 2.47 %, respectively. Correlation analysis showed that the decline of FFM, TBW, and ECW was positively correlated with the LLS score. Among the AMS relevant symptoms, only fatigue was positively correlated with the decline of FFM, TBW, ICW, and ECW. CONCLUSIONS:Significant changes of body composition were observed in the early stage after ascent to plateau. The decline of FFM, TBW, and ECW might be related to the severity of AMS, and fatigue was positively correlated with the decline of FFM, TBW, ICW and ECW.

The sleep structure of healthy adults varies across continuous nights but exhibits periodicity and regularity. To explore the similarities and differences in sleep electroencephalogram structures on an intra- and inter-individual basis across nights, we present an open-access, continuous, multi-night sleep database. This database contains multi-modal sleep monitoring data from 20 healthy participants over three consecutive nights. Each recording contains eight channels, including electroencephalogram, electrooculogram, and electromyogram signals, as well as manually labelled sleep staging labels by three experts from the different sleep centres. To the best of our knowledge, this is the first continuous multi-night and multi-expert annotated open-access sleep database from healthy participants. It provides valuable opportunities to investigate the physiological mechanisms of sleep structure continuity, advancing sleep features recognition and prediction algorithms, and designing individualised templates for closed-loop sleep stimulation. This database aims to promote individualisation and precision in sleep research, and provide better protection for human sleep research.

Background: Sleep staging is essential for monitoring an individual’s sleep process. Recently, automatic sleep staging algorithms have gained popularity due to their ability to compensate for the limitations of traditional manual recognition. However, current studies fail to investigate the interplay between temporal, channel, and frequency in multimodal data. Moreover, neuroscience research has shown that integrating common and distinct feature representations across multimodal data can reveal latent brain features across different sleep stages. Methods: To address these issues, we propose a hybrid common-private domain sleep staging network with deep learning, called HybridDomainSleepNet. This network explores the correlation between multimodal data and multidomain to improve the representation of different sleep stages. The HybridDomainSleepNet is able to learn the common and private features of individuals in different sleep stages and capture the long-term and short-term dependencies during sleep. Meanwhile, we propose a multidomain attentional framework to explore the relevance of multimodal data from different sleep stages through a three-branch structure to capture brain signals that interact across domains. Results: The proposed HybridDomainSleepNet is validated on the public dataset MASS-SS3 and DRM-Pat, and the homemade dataset BP-SleepX. The experimental results demonstrate that the classification accuracy reaches 88.9%, 83.9% and 88.0%, respectively, which outperforms the existing competitive methods. The visualization results show that the model can accurately characterizes the brain patterns in different sleep stages. Conclusions: This study offers an opportunity to improve the performance of current sleep staging models and provides a stronger foundation for further research into the mechanisms of the sleep process.

Depressed mood has been proposed to possibly possess a unique mode of defocused attention. However, this argument needs to be supported by experimental evidence based on attentional performance. The present study used a perceptual load paradigm, combining factors of perceptual load, distractor-target compatibility, and eccentricity, to investigate the degree of attentional distraction in depressed mood. In addition, the mode of attentional distraction associated with depressed mood was explored with the time-frequency features of electroencephalography (EEG). The behavioral results showed that the high depressed mood (HD) group had significantly higher attentional distraction than the low depressed mood (LD) group. EEG results showed that 1) the beta power (especially beta-2, 18-30 Hz) of the two groups differed in the medio-late part of the attentional distraction, with significantly lower power in the HD group than in the LD group; 2) the results of the correlation between beta-2 power and depression scores revealed a significant negative correlation. These results imply that beta-2 is a potential marker that may be sensitive to depressed mood during attentional processing, which was further supported by the classification results of the support vector machine (SVM) with 80.65% accuracy between the HD and LD groups.

Background: Both hypoxia exposure and physical exercise before ascending have been proved to promote high altitude acclimatization, whether the combination of these two methods can bring about a better effect remains uncertain. Therefore, we designed this study to evaluate the effect of hypoxic preacclimatization combining intermittent hypoxia exposure (IHE) and physical exercise on the tolerance to acute hypoxia and screen the optimal preacclimatization scheme among the lowlanders.Methods: A total of 120 Han Chinese young men were enrolled and randomly assigned into four groups, including the control group and three experimental groups with hypoxic preacclimatization of 5-day rest, 5-day exercise, and 3-day exercise in a hypobaric chamber, respectively. Main physical parameters for hypoxia acclimatization, AMS incidence, physical and mental capacity were measured for each participant in the hypobaric chamber simulated to the altitude of 4500 m in the effect evaluation stage. The effect was compared between different schemes.Results: During the effect evaluation stage, SpO2 of the 5-day rest group and 5-day exercise group was significantly higher than that of the control group (p = 0.001 and p = 0.006, respectively). The participants with 5-day rest had significantly lower HR than the controls (p = 0.018). No significant differences of AMS incidence were found among the four groups, while the proportion of AMS headache symptom (moderate and severe vs. mild) was significantly lower in the 3-day exercise group than that in the control group (p = 0.002). The 5-day exercise group had significantly higher VO2max, than the other three groups (p = 0.033, p < 0.001, and p = 0.023, respectively). The 5-day exercise group also had significantly higher digital symbol and pursuit aiming test scores, while shorter color selection reaction time than the control group (p = 0.005, p = 0.005, and p = 0.004, respectively).Conclusion: Hypoxic preacclimatization combining IHE with physical exercise appears to be efficient in promoting the tolerance to acute hypoxia. Hypoxia duration and physical exercise of moderate intensity are helpful for improvement of SpO2 and HR, relief of AMS headache symptoms, and enhancement of mental and physical operation capacity.

Brain-computer interface (BCI) has been developed for decades to directly establish communication between human brain and external devices. Current BCI has been widely exploited in the emerging macroscopical field such as mind-controlled mechanical devices and robots; it is an open challenge to manipulate microscopic devices or nanoparticles through mind control. Here, we demonstrate a synthetic mind-controlled optical manipulation (MCOM) platform that integrates BCI with a spatial light modulator (SLM) to generate dynamic light fields (e.g., vortex beams). Through human mind instructions, the MCOM perceives the mind signal and translates the information into a series of optical holograms for complex particle manipulation. This work is the first implementation of the combination of the intelligence of the human brain with the manipulative power of optics, merging the two noncontact techniques into a real noncontact optical manipulation in the microscopic and nanoscopic world, which may find applications in soft-matter physics, biomedical engineering, and nanotechnology.

Time perception is one fundamental ability for human survival and everyday life, and it holds potential as a marker for psychopathology assessment and early intervention. Neural oscillations, particularly alpha oscillations, have been proposed as a key mechanism underlying the temporal organization of perception. However, there is limited causal evidence regarding how brain oscillations regulate timing processes and whether non-invasive brain stimulation can modulate the precision of timing by manipulating endogenous alpha rhythm. In this study, we investigated the causal relationship between alpha oscillations and time perception by employing tACS over occipital cortex, whereas participants performed a self-initiated sub-second timing task. tACS was delivered with different stimulation protocols at the individual alpha frequency (IAF), as well as the slightly slower and faster frequencies (IAF±2 Hz). Concurrently, EEG signals were recorded pre- and post-tACS during the tasks to provide direct electrophysiological evidence at various stages of the timing process and elucidate neural responses to tACS effects. The results demonstrated that IAF can predict individual timing precision. tACS with lower frequency (IAF-2Hz) significantly enhanced timing precision compared to tACS with IAF and IAF+2Hz which had no effect on timing behavior. Furthermore, slower tACS not only specifically modulated phase changes in alpha activity prior to timing initiation but also altered patterns of event-related desynchronization (ERD) in alpha activity during the timing process. These findings provide causal evidence for an association between alpha-dependent neural activity and internal time processing mechanisms, and suggested the potential applications aimed at enhancing the time perception in psychopathology clinics.

This article describes our initial work toward a general-purpose platform for non-invasive neurotechnology research. This platform consists of a multi-modal wireless recording device, an associated software API, and full integration into BCI2000 software. The device is placed on the forehead and features two electroencephalographic (EEG) sensors, an inertial movement sensor (IMU), a photoplethysmogram (PPG) sensor, a microphone, and vibration-based feedback. Herein, we demonstrate different technical characteristics of our platform and its use in the context of sleep monitoring/modulation, simultaneous and synchronized recordings from different hardware, and evoked potentials. With further development and widespread dissemination, our platform could become an important tool for research into new non-invasive neurotechnology protocols in humans.### Competing Interest StatementThe authors have declared no competing interest.

Objective: To compare the ability of the Chinese AMS Score (CAS) to detect acute mountain sickness (AMS) using the 2018 version of the Lake Louise Score (LLS) as reference. Methods: After flying from Chengdu (altitude: 500 m) to Lhasa (3,658 m), 2,486 young men completed a questionnaire. The questionnaire contained LLS and CAS items. An LLS >= 3 and/or a CAS >= cutoff were used as the criteria for AMS. Hierarchical cluster analysis and two-step cluster analysis were used to investigate relationships between the symptoms. Results: AMS incidence rates were 33.8% (n = 840) with the LLS and 59.3% (n = 1,473) with the CAS (chi(2) = 872.5, p < 0.001). The LLS and CAS had a linear relationship (orthogonal regression, Pearson r = 0.91, p < 0.001). With the LLS as the standard, the CAS had high diagnostic accuracy (area under the curve = 0.95, 95% confidence interval: 0.94-0.96). However, with the CAS, 25.5% (n = 633) more participants were labeled as having AMS than with the LLS (false positives). Two clusters were identified: one with headache only (419 participants, 66.2%) and one without headache but with other symptoms (214 participants, 33.8%). Reducing the weight of headache in the CAS allowed to align CAS and LLS. Conclusion: In comparison to the LLS, the CAS has a sensitivity close to 100% but lacks specificity given the high rate of false positives. The different weight of headaches may be the main reason for the discrepancy.