Wu, Teresa

Abstract Background Post-traumatic headache (PTH) is a common acute and persistent symptom following mild traumatic brain injury (mTBI). Symptoms of cutaneous allodynia and presence of nociceptive sensitization might be associated with acute PTH and its persistence. The objectives of this study were to compare allodynia symptoms and cutaneous heat pain thresholds amongst males and females with acute PTH to healthy controls (HC) and determine if pain thresholds and allodynia symptoms are associated with PTH outcomes. Methods This prospective longitudinal study enrolled 139 adults with acute PTH attributed to mTBI as defined by the International Classification of Headache Disorders and 95 HC. All PTH participants completed a baseline research visit near PTH onset and a follow-up visit three to four months later. All PTH participants and a subset of HC completed the Allodynia Symptom Checklist (ASC-12) at each research visit. A different subset of the participants underwent quantitative sensory testing (QST) during baseline, 4-week, and 16-week research visits to quantify cutaneous heat pain thresholds at the forehead and forearms. Data from daily headache diaries were used to determine longitudinal PTH improvement versus non-improvement at three months. ASC-12 score and pain threshold comparisons were made between PTH and HC groups, PTH improved versus non-improved cohorts, and between PTH males and females. Results Participants with PTH had an average age of 42.6 years and 64.0% were female. HC had an average age of 40.0 years and 65.3% were female. At the first visit, PTH participant ASC-12 scores averaged 3.6 versus 0.1 amongst HC, p < 0.001. 44.8% of PTH participants had headache improvement at 3 months. ASC-12 scores were higher in the PTH non-improved versus improved group at baseline (4.0 versus 2.4, p = 0.038) and 3-month follow-up (3.4 versus 1.9, p = 0.012). ASC-12 scores were higher in females than males at baseline (4.7 versus 1.6, p < 0.001) and 3-months (3.9 versus 1.2, p < 0.001). Cutaneous heat pain thresholds at the forehead and forearm did not differ between any group. Conclusions PTH attributed to mTBI is associated with symptoms of cutaneous allodynia. Greater allodynia symptoms are present in females with PTH compared to males and may be associated with PTH non-improvement.

Abstract Objectives/Background Post‐traumatic headache (PTH) is a common symptom following mild traumatic brain injury (mTBI). Currently, there is no identified way to accurately predict if, when, and at what pace a person will have PTH improvement. In our prior studies, we focused on predicting headache improvement at 3 months post‐mTBI. However, that approach may overlook individual differences in how headaches evolve over time. This study aimed to identify individual subgroups based on their headache trajectories and to develop machine‐learning (ML) models for early prediction of headache evolution. Methods Participants with acute PTH completed a daily electronic headache diary over 3 months, recording their headache‐related symptoms. Tensor decomposition was utilized to extract latent factors underlying the time‐varying symptoms. We applied clustering techniques on the latent factors to identify patient subgroups with varying headache improvement trajectory. Next, we developed an ML method to classify each individual into a headache trajectory subgroup as early as possible within the 3‐month interval. Results Seventy‐three individuals with acute PTH (mean age = 44.8 years, SD = 14.0; 50 females/23 males) were enrolled between 0 and 59 days post‐mTBI. Data from 54 individuals were used as the training cohort for model training, and 19 individuals were used as the test cohort for model evaluation. Tensor decomposition extracted two latent factors: one factor representing the overall state of PTH severity and disability and the other representing the improvement state of these symptoms over 3 months. Clustering identified four patient subgroups with distinct headache evolution trajectories: (1) severe symptoms without improvement, (2) severe symptoms with mild improvement, (3) milder symptoms with substantial improvement, and (4) mildest symptoms with mild improvement. The proposed ML model achieved 0.80 cross‐validation accuracy in classifying individuals with PTH into subgroups for the training cohort and 0.84 accuracy for the test cohort. Notably, the model required only the first 2 weeks of headache data to accurately identify the subgroup with the mildest headaches, 3 additional weeks to identify the subgroup with the most severe headaches and no improvement in 3 months, and 2 additional weeks to distinguish the remaining subgroups. Conclusion This study identified subgroups of individuals with acute PTH with distinct headache improvement trajectories. The proposed ML method accurately classified individuals into these subgroups using the minimally necessary early headache data for each person, including detecting the subgroup with the mildest headaches at 2 weeks. This approach could offer an estimated forecast of headache burden over time and could assist clinicians with determining treatment needs and eligibility for PTH clinical trials.

Background Headache is a common symptom following mild traumatic brain injury (mTBI). Post-traumatic headache (PTH), a secondary headache disorder that develops after mTBI, often persists for months or years. To identify potential recovery mechanisms and prognostic biomarkers, the present study investigated whether longitudinal changes in pain-induced brain activation differ between healthy controls (HC) and PTH participants showing headache improvement and those without improvement. Methods Thirty-three participants who met International Classification of Headache Disorders, 3rd edition, criteria for acute PTH within 59 days post-mTBI and 33 HC participants were included with no significant differences in demographics. All participants underwent functional magnetic resonance imaging scans at baseline, four weeks, and 16 weeks post-enrollment using a thermal stimulation paradigm with noxious and non-painful heat stimuli. ‘Painful vs. Non-Painful Heat’ contrasts were generated using SPM12. PTH improvement was assessed at three months post-enrollment via electronic headache diaries. Two-sample t -tests compared the brain activation between HC and PTH at baseline. Linear mixed-effects models examined longitudinal changes for HC, PTH improvement and non-improvement groups across visits. Generalized linear models compared these groups within visits. Results Baseline analysis revealed several regions with significantly higher activation in acute PTH compared to HC, including bilateral postcentral gyrus, right superior temporal gyrus, right middle temporal gyrus, left inferior parietal gyrus, right superior parietal gyrus, left ventral striatum, left olfactory cortex, left gyrus rectus, and left middle occipital gyrus. Over time, the PTH improvement group demonstrated progressive normalization across all identified brain regions, whereas the non-improvement group showed only partial normalization in left ventral striatum, left olfactory cortex, and left gyrus rectus. Sustained elevated activation in specific regions distinguished PTH participants without headache improvement from those with headache improvement, suggesting potential biomarkers for persistent PTH. Conclusions Our findings demonstrate significantly altered pain-induced brain activations in participants with acute PTH compared to HC. Longitudinal analysis revealed distinct recovery trajectories: progressive normalization in the improvement group versus persistent alterations in the non-improvement group. These neuroimaging patterns may serve as biomarkers for identifying individuals at risk for persistent PTH, with implications for early intervention and personalized treatment approaches.

Introduction: We assess the risk of stroke in individuals with and without a migraine diagnosis, adjusting for demographic variables and stroke risk factors using participant information from the All of Us database. Method: Diagnoses of migraine, stroke and comorbidities were identified using ICD9 and ICD10 codes. A multivariable logistic regression analysis was performed to assess the association between migraine and risk of stroke adjusting for comorbidities including hypertension, atrial fibrillation, hyperlipidemia, diabetes, tobacco use, depression and demographics (age, sex at birth, race and ethnicity). To compare the prevalence of stroke between individuals with and without migraine, odds ratios using a 95% confidence interval (CI) were calculated. Results: Within the All of Us database, 31,444 individuals received a migraine diagnosis (female=25,374/81%, male=5,391/17%, other=679/2%; mean (std) age=54.9 (15.6)) and 379,283 did not have a migraine diagnosis (female=222,104/59%, male=149,182/39%, other=7,997/2%; mean (SD) age=55.9 (17.2)), see Figure 1 for detailed demographics. The migraine cohort had a greater proportion of women (81% vs 59%, p<0.001), a higher proportion of white individuals (61% vs. 55%, p<0.001) fewer African American individuals (15% vs. 19%, p<0.001) and a lower proportion of individuals who were 61 or older (39% vs 45%, p<0.001) compared to the non-migraine cohort. Compared to the non-migraine cohort, individuals with migraine had significantly higher rates of comorbidities, including depression (13% vs. 4%, p<0.001), diabetes (16% vs. 7%), tobacco use (23% vs. 9%, p<0.001), hyperlipidemia (48% vs. 22%, p<0.001), hypertension (57% vs. 27%, p<0.001), and atrial fibrillation (7% vs. 4%, p<0.001). A multivariate regression model adjusted for differences between cohort demographics and comorbidities found that individuals with migraine had 2.14 higher odds of overall stroke compared to the non-migraine cohort (OR 2.14, 95% CI [ 2.04, 2.24]. Discussion: After adjusting for cohort imbalances in demographics and comorbidities, individuals with migraine had a 2.14 times higher risk of stroke compared to those without a migraine diagnosis. Our analysis, using data from the All of Us database, confirms previous findings and suggests that while comorbidities are more prevalent in those with migraine, they do not fully account for the increased stroke risk.

Background The aim of this study was to examine the potential added value of including neuropsychiatric symptoms (NPS) in machine learning (ML) models, along with demographic features and Alzheimer's disease (AD) biomarkers, to predict decline or non-decline in global and domain-specific cognitive scores among community-dwelling older adults. Objective To evaluate the impact of adding NPS to AD biomarkers on ML model accuracy in predicting cognitive decline among older adults. Methods The study was conducted in the setting of the Mayo Clinic Study of Aging, including participants aged ≥ 50 years with information on demographics (i.e., age, sex, education), NPS (i.e., Neuropsychiatric Inventory Questionnaire; Beck Depression and Anxiety Inventories), at least one AD biomarker (i.e., plasma-, neuroimaging- and/or cerebrospinal fluid [CSF]-derived), and at least 2 repeated neuropsychological assessments. We trained and tested ML models using a stepwise feature addition approach to predict decline versus non-decline in global and domain-specific (i.e., memory, language, visuospatial, and attention/executive function) cognitive scores. Results ML models had better performance when NPS were included along with a) neuroimaging biomarkers for predicting decline in global cognition, as well as language and visuospatial skills; b) plasma-derived biomarkers for predicting decline in visuospatial skills; and c) CSF-derived biomarkers for predicting decline in attention/executive function, language, and memory. Conclusions NPS, added to ML models including demographic and AD biomarker data, improves prediction of downward trajectories in global and domain-specific cognitive scores among community-dwelling older adults, albeit effect sizes are small. These preliminary findings need to be confirmed by future cohort studies.

Alzheimer’s disease (AD) and Alzheimer’s Related Dementias (ADRD) are projected to affect 50 million people globally in the coming decades. Clinical research suggests that Mild Cognitive Impairment (MCI), a precursor to dementia, offers a critical window of opportunity for lifestyle interventions to delay or prevent the progression of AD/ADRD. Previous research indicates that lifestyle changes, including increased physical exercise, reduced caloric intake, and mentally stimulating activities, can reduce the risk of MCI. Early detection of MCI is challenging due to subtle and often unnoticed cognitive decline and is traditionally monitored through infrequent clinical tests. In this research, the Smart Driving System, a novel, unobtrusive, and economical technology to detect early stages of neurodegenerative diseases, is presented. The system comprises a multi-modal biosensing array (MMS) and AI algorithms, including driving performance and driver’s biometrics, offering insights into a driver’s cognitive function. This publication is the first work reported towards the ultimate goal of developing the Smart Driving Device and App, integrating it into vehicles, and validating its effectiveness in detecting MCI through comprehensive pilot studies.

Background: Prostate cancer is the second most common new cancer diagnosis in the United States. It is usually slow-growing, and when it is low-grade and confined to the prostate gland, it can be treated either conservatively (through active surveillance) or with surgery. However, if the cancer has spread beyond the prostate, such as to the lymph nodes, then that indicates a more aggressive cancer, and surgery may not be adequate. Methods: The challenge is that it is often difficult for radiologists reading prostate-specific imaging such as magnetic resonance images (MRIs) to differentiate malignant lymph nodes from non-malignant ones. An emerging field is the development of artificial intelligence (AI) models, including machine learning and deep learning, for medical imaging to assist in diagnostic tasks. Earlier research focused on implementing texture algorithms to extract imaging features used in classification models. More recently, researchers began studying the use of deep learning for both stand-alone feature extraction and end-to-end classification tasks. In order to tackle the challenges inherent in small datasets, this study was designed as a scalable hybrid framework utilizing pre-trained ResNet-18, a deep learning model, to extract features that were subsequently fed into a machine learning classifier to automatically identify malignant lymph nodes in patients with prostate cancer. For comparison, two texture algorithms were implemented, namely the gray-level co-occurrence matrix (GLCM) and Gabor. Results: Using an institutional prostate lymph node dataset (42 positives, 84 negatives), the proposed framework achieved an accuracy of 76.19%, a sensitivity of 79.76%, and a specificity of 69.05%. Using GLCM features, the classification achieved an accuracy of 61.90%, a sensitivity of 74.07%, and a specificity of 42.86%. Using Gabor features, the classification achieved an accuracy of 65.08%, a sensitivity of 73.47%, and a specificity of 52.50%. Conclusions: Our results demonstrate that a hybrid approach, i.e., using a pre-trainined deep learning model for feature extraction, followed by a machine learning classifier, is a viable solution. This hybrid approach is especially useful in medical-imaging-based applications with small datasets.

Abstract Most current algorithms for multivariate time series classification tend to overlook the correlations between time series of different variables. In this research, we propose a framework that leverages Eigen-entropy along with a cumulative moving window to derive time series signatures to support the classification task. These signatures are enumerations of correlations among different time series considering the temporal nature of the dataset. To manage dataset's dynamic nature, we employ preprocessing with dense multi scale entropy. Consequently, the proposed framework, Eigen-entropy-based Time Series Signatures, captures correlations among multivariate time series without losing its temporal and dynamic aspects. The efficacy of our algorithm is assessed using six binary datasets sourced from the University of East Anglia (UEA), in addition to a publicly available sepsis dataset and an institutional sepsis dataset from the Mayo Clinic. We employ recall as a metric for evaluation against the baseline algorithm, Dependent Dynamic Time Warping with 1 Nearest Neighbor. Our algorithm outperforms the baseline algorithm for UEA datasets in terms of recall with an average value of 10% across all datasets, and by 38% for an institutional sepsis dataset. For the public sepsis dataset, the performance of our algorithm is comparable to the baseline algorithm.