Influence of next-generation artificial intelligence on headache research, diagnosis and treatment: the junior editorial board members' vision - part 1

Migraine triggers, phases, and classification using machine learning models

Background

In many countries, patients with headache disorders such as migraine remain under-recognized and under-diagnosed. Patients affected by these disorders are often unaware of the seriousness of their conditions, as headaches are neither fatal nor contagious. In many cases, patients with migraine are often misdiagnosed as regular headaches.

Methods

In this article, we present a study on migraine, covering known triggers, different phases, classification of migraine into different types based on clinical studies, and the use of various machine learning algorithms such as logistic regression (LR), support vector machine (SVM), random forest (RF), and artificial neural network (ANN) to learn and classify different migraine types. This study will only consider using these methods for diagnostic purposes. Models based on these algorithms are then trained using the dataset, which includes a compilation of the types of migraine experienced by various patients. These models are then used to classify the types of migraines, and the results are analyzed.

Results

The results of the machine learning models trained on the dataset are verified for their performance. The results are further evaluated by selective sampling and tuning, and improved performance is observed. The precision and accuracy obtained by the support vector machine and artificial neural network are 91% compared to logistic regression (90%) and random forest (87%). These models are run with the dataset without optimal tuning across the entire dataset for different migraine types; which is further improved with selective sampling and optimal tuning. These results indicate that the discussed models are relatively good and can be used with high precision and accuracy for diagnosing different types of migraine.

Conclusion

Our study presents a realistic assessment of promising models that are dependable in aiding physicians. The study shows the performance of various models based on the classification metrics computed for each model. It is evident from the results that the artificial neural network (ANN) performs better, irrespective of the sampling techniques used. With these machine learning models, types of migraines can be classified with high accuracy and reliability, enabling physicians to make timely clinical diagnoses of patients.

Neuromodulation in Chronic Migraine: Evidence and Recommendations from the GRADE Framework

Chronic migraine (CM) affects approximately 2% of the general population and is defined by the persistence of migraine symptoms for at least 15 days per month for at least 3 months. CM is often refractory to common drug treatments and is associated with a significant burden in functions of daily life during ictal phases, productivity loss, and direct costs. Modulation of pain is considered pivotal to reduce its impact and to improve the quality of life among patients with CM. In recent years, neuromodulation in CM has received growing attention; however, there remains no consensus regarding the effectiveness and safety of these procedures. Previous invasive methods such as occipital nerve neurolysis and interruption of the trigeminal dorsal root are not indicated due to high rates of relapsing pain and frequent procedural complications. Although emerging neuromodulation methods, both noninvasive, such as vagus nerve stimulation (VNS), transcranial magnetic stimulation (TMS), remote electrical neuromodulation (REM), and invasive, such as deep brain stimulation (DBS), occipital nerve stimulation (ONS), and high-frequency 10-Hz spinal cord stimulation (HF-10 SNS) have demonstrated promising outcomes in early clinical trials, their use has yet to be integrated into routine clinical practice. In this review, study evidence and strength of recommendations are assessed by the Grading of Recommendations, Assessment, Development, and Evaluation (GRADE) system. Other conditions such as therapeutic risk/benefit, direct and indirect costs, use of resources, and patient/clinician preferences are also evaluated.

Unveiling new insights into migraine risk stratification using machine learning models of adjustable risk factors

BACKGROUND

Migraine ranks as the second-leading cause of global neurological disability, affecting approximately 1.1 billion individuals worldwide with severe quality-of-life impairments. Although adjustable risk factors-including environmental exposures, sleep disturbances, and dietary patterns-are increasingly implicated in pathogenesis of migraine, their causal roles remain insufficiently characterized, and the integration of multimodal evidence lags behind epidemiological needs.

METHODS

We developed a three-step analytical framework combining causal inference, predictive modeling, and burden projection to systematically evaluate modifiable factors associated with migraine. First, two-sample mendelian randomization (MR) assessed causality between five domains (metabolic profiles, body composition, cardiovascular markers, behavioral traits, and psychological states) and the risk of migraine. Second, we trained ensemble machine learning (ML) algorithms that incorporated these factors, with Shapley Additive exPlanations (SHAP) value analysis quantifying predictor importance. Finally, spatiotemporal burden mapping synthesized global incidence, prevalence, and disability-adjusted life years (DALYs) data to project region-specific risk and burden trajectories through 2050.

RESULTS

MR analyses identified significant causal associations between multiple adjustable factors (including overweight, obesity class 2, type 2 diabetes [T2DM], hip circumference [HC], body mass index [BMI], myocardial infarction, and feeling miserable) and the risk of migraine (P < 0.05, FDR-q < 0.05). The Random Forest (RF)-based model achieved excellent discrimination (Area under receiver operating characteristic curve [AUROC] = 0.927), identifying gender, age, HC, waist circumference [WC], BMI, and systolic blood pressure [SBP] as the predictors. Burden mapping projected a global decline in migraine incidence by 2050, yet persistently high prevalence and DALYs burdens underscored the urgency of timely interventions to maximize health gains.

CONCLUSIONS

Integrating causal inference, predictive modeling, and burden projection, this study establishes hierarchical evidence for adjustable migraine determinants and translates findings into scalable prevention frameworks. These findings bridge the gap between biological mechanisms, clinical practice, and public health policy, providing a tripartite framework that harmonizes causal inference, individualized risk prediction, and global burden mapping for migraine prevention.

Dropout Rate of Participants in Randomized Controlled Trials Using Different Exercise-Based Interventions in Patients with Migraine. A Systematic Review with Meta-Analysis

Background/Objectives: Exercise has gained attention as a potentially beneficial non-pharmacological intervention, but whether this type of intervention presents a higher dropout rate compared to other interventions is still unknown. This systematic review, with a meta-analysis of randomized controlled trials, aims to determine whether exercise or comparators present lower or higher attrition in patients with migraine. Methods: A search was conducted in PubMed, Scopus, Web of Science, and Cochrane Library until March 2025. The methodological quality was evaluated using the JBI scale for randomized trials. Proportion meta-analysis calculated the dropout rate. Results: Odds ratio meta-analysis under 1 indicated lower attrition in experimental participants. Subgroup meta-analyses sorted by type of exercise, control, and migraine were conducted to explore variability in results based on the mentioned moderators. The overall pooled dropout rate was 6.7%, 11.6% for the exercise groups, and 10.1% for the comparators. No statistical difference was found between groups of studies, type of migraine, type of exercise, and type of comparator (p ≥ 0.05). Only the odds ratio results for migraine with auras showed a lower pooled dropout rate in favor of control participants, OR = 1.18. Conclusions: Although there is no statistically significant difference, the meta-analysis of proportions shows a higher loss rate in exercise-based interventions. However, the high heterogeneity found in the included studies prevents us from drawing firm conclusions. Furthermore, adequate adherence to the CONSORT guidelines in reporting losses and their reasons could help design appropriate retention strategies for studies and interventions based on exercise in patients with migraines.

Hyper-DREAM, a Multimodal Digital Transformation Hypertension Management Platform Integrating Large Language Model and Digital Phenotyping: Multicenter Development and Initial Validation Study

Within the mHealth framework, systematic research that collects and analyzes patient data to establish comprehensive digital health archives for hypertensive patients, and leverages large language models (LLMs) to assist clinicians in health management and Blood Pressure (BP) control remains limited. In this study, our aims to describe the design, development and usability evaluation process of a management platform (Hyper-DREAM) for hypertension. Our multidisciplinary team employed an iterative design approach over the course of a year to develop the Hyper-DREAM platform. This platform's primary functionalities encompass multimodal data collection (personal hypertensive digital phenotype archive), multimodal interventions (BP measurement, medication assistance, behavior modification, and hypertension education) and multimodal interactions (clinician-patient engagement and BP Coach component). In August 2024, the mHealth App Usability Questionnaire (MAUQ) was conducted involving 51 hypertensive patients recruited from three distinct centers. In parallel, six clinicians engaged in management activities and contributed feedback via the Doctor's Software Satisfaction Questionnaire (DSSQ). Concurrently, a real-world comparative experiment was conducted to evaluate the usability of the BP Coach, ChatGPT-4o Mini, ChatGPT-4o and clinicians. The comparative experiment demonstrated that the BP Coach achieved significantly higher scores in utility (mean scores 4.05, SD 0.87) and completeness (mean scores 4.12, SD 0.78) when compared to ChatGPT-4o Mini, ChatGPT-4o, and clinicians. In terms of clarity, the BP Coach was slightly lower than clinicians (mean scores 4.03, SD 0.88). In addition, the BP Coach exhibited lower performance in conciseness (mean scores 3.00, SD 0.96). Clinicians reported a marked improvement in work efficiency (2.67 vs. 4.17, P < .001) and experienced faster and more effective patient interactions (3.0 vs. 4.17, P = .004). Furthermore, the Hyper-DREAM platform significantly decreased work intensity (2.5 vs. 3.5, P = .01) and minimized disruptions to daily routines (2.33 vs. 3.55, P = .004). The Hyper-DREAM platform demonstrated significantly greater overall satisfaction compared to the WeChat-based standard management (3.33 vs. 4.17, P = .01). Additionally, clinicians exhibited a markedly higher willingness to integrate the Hyper-DREAM platform into clinical practice (2.67 vs. 4.17, P < .001). Furthermore, patient management time decreased from 11.5 min (SD 1.87) with Wechat-based standard management to 7.5 min (SD 1.84, P = .01) with Hyper-DREAM. Hypertensive patients reported high satisfaction with the Hyper-DREAM platform, including ease of use (mean scores 1.60, SD 0.69), system information arrangement (mean scores 1.69, SD 0.71), and usefulness (mean scores 1.57, SD 0.58). In conclusion, our study presents Hyper-DREAM, a novel artificial intelligence-driven platform for hypertension management, designed to alleviate clinician workload and exhibiting significant promise for clinical application. The Hyper-DREAM platform is distinguished by its user-friendliness, high satisfaction rates, utility, and effective organization of information. Furthermore, the BP Coach component underscores the potential of LLMs in advancing mHealth approaches to hypertension management.

Influence of next-generation artificial intelligence on headache research, diagnosis and treatment: the junior editorial board members' vision - part 2

Part 2 explores the transformative potential of artificial intelligence (AI) in addressing the complexities of headache disorders through innovative approaches, including digital twin models, wearable healthcare technologies and biosensors, and AI-driven drug discovery. Digital twins, as dynamic digital representations of patients, offer opportunities for personalized headache management by integrating diverse datasets such as neuroimaging, multiomics, and wearable sensor data to advance headache research, optimize treatment, and enable virtual trials. In addition, AI-driven wearable devices equipped with next-generation biosensors combined with multi-agent chatbots could enable real-time physiological and biochemical monitoring, diagnosing, facilitating early headache attack forecasting and prevention, disease tracking, and personalized interventions. Furthermore, AI-driven advances in drug discovery leverage machine learning and generative AI to accelerate the identification of novel therapeutic targets and optimize treatment strategies for migraine and other headache disorders. Despite these advances, challenges such as data standardization, model explainability, and ethical considerations remain pivotal. Collaborative efforts between clinicians, biomedical and biotechnological engineers, AI scientists, legal representatives and bioethics experts are essential to overcoming these barriers and unlocking AI's full potential in transforming headache research and healthcare. This is a call to action in proposing novel frameworks for integrating AI-based technologies into headache care.

Machine learning classification meets migraine: recommendations for study evaluation

The integration of machine learning (ML) classification techniques into migraine research has offered new insights into the pathophysiology and classification of migraine types and subtypes. However, inconsistencies in study design, lack of methodological transparency, and the absence of external validation limit the impact and reproducibility of such studies. This paper presents a framework of six essential recommendations for evaluating ML-based classification in migraine research: (1) group homogenization by clinical phenotype, attack frequency, comorbidity, therapy, and demographics; (2) defining adequate sample size; (3) quality control of collected and preprocessed data; (4) transparent training, testing, and performance evaluation of ML models, including strategies for data splitting, overfitting control, and feature selection; (5) interpretability of results with clinical relevance; and (6) open data and code sharing to facilitate reproducibility. These recommendations aim to balance the trade-off between model generalization and precision while encouraging collaborative standardization across the ML and headache communities. Furthermore, this framework intends to stimulate discussion toward forming a consortium to establish definitive guidelines for ML-based classification research in migraine field.