Feasibility of brain age predictions from clinical T1-weighted MRIs.

An individual's brain predicted age minus chronological age (brain-PAD) obtained from MRIs could become a biomarker of disease in research studies. However, brain age reports from clinical MRIs are scant despite the rich clinical information hospitals provide. Since clinical MRI protocols are meant for specific clinical purposes, performance of brain age predictions on clinical data need to be tested. We explored the feasibility of using DeepBrainNet, a deep network previously trained on research-oriented MRIs, to predict the brain ages of 840 patients who visited 15 facilities of a health system in Florida. Anticipating a strong prediction bias in our clinical sample, we characterized it to propose a covariate model in group-level regressions of brain-PAD (recommended to avoid Type I, II errors), and tested its generalizability, a requirement for meaningful brain age predictions in new single clinical cases. The best bias-related covariate model was scanner-independent and linear in age, while the best method to estimate bias-free brain ages was the inverse of a scanner-independent and quadratic in brain age function. We demonstrated the feasibility to detect sex-related differences in brain-PAD using group-level regression accounting for the selected covariate model. These differences were preserved after bias correction. The Mean-Average Error (MAE) of the predictions in independent data was ∼8 years, 2-3 years greater than reports for research-oriented MRIs using DeepBrainNet, whereas an R2 (assuming no bias) was 0.33 and 0.76 for the uncorrected and corrected brain ages, respectively. DeepBrainNet on clinical populations seems feasible, but more accurate algorithms or transfer-learning retraining is needed.

Accelerated Brain Aging Mediates the Association Between Psychological Profiles and Clinical Pain in Knee Osteoarthritis.

Chronic pain is driven by factors across the biopsychosocial spectrum. Previously, we demonstrated that magnetic resonance images (MRI)-based brain-predicted age differences (brain-PAD: brain-predicted age minus chronological age) were significantly associated with pain severity in individuals with chronic knee pain. We also previously identified four distinct, replicable, multidimensional psychological profiles significantly associated with clinical pain. The brain aging-psychological characteristics interface in persons with chronic pain promises elucidating factors contributing to their poor health outcomes, yet this relationship is barely understood. That is why we examined the interplay between the psychological profiles in participants having chronic knee pain impacting function, brain-PAD, and clinical pain severity. Controlling for demographics and MRI scanner, we compared the brain-PAD among psychological profiles at baseline (n = 164) and over two years (n = 90). We also explored whether profile-related differences in pain severity were mediated by brain-PAD. Brain-PAD differed significantly between profiles (ANOVA's omnibus test, P = .039). Specifically, participants in the profile 3 group (high negative/low positive emotions) had an average brain-PAD ∼4 years higher than those in profile- (low somatic reactivity), with P = .047, Bonferroni-corrected, and than those in profile 2 (high coping), with P = .027, uncorrected. Repeated measures ANOVA revealed no significant change in profile-related brain-PAD differences over time, but there was a significant decrease in brain-PAD for profile 4 (high optimism/high positive affect), with P = .045. Moreover, profile-related differences in pain severity at baseline were partly explained by brain-PAD differences between profile 3 and 1, or 2; but brain-PAD did not significantly mediate the influence of variations in profiles on changes in pain severity over time. PERSPECTIVE: Accelerated brain aging could underlie the psychological-pain relationship, and psychological characteristics may predispose individuals with chronic knee pain to worse health outcomes via neuropsychological processes.

Toward MR protocol-agnostic, unbiased brain age predicted from clinical-grade MRIs.

The difference between the estimated brain age and the chronological age ('brain-PAD') could become a clinical biomarker. However, most brain age models were developed for research-grade high-resolution T1-weighted MRIs, limiting their applicability to clinical-grade MRIs from various protocols. We adopted a dual-transfer learning strategy to develop a model agnostic to modality, resolution, or slice orientation. We retrained a convolutional neural network (CNN) using 6281 clinical MRIs from 1559 patients, among 7 modalities and 8 scanner models. The CNN was trained to estimate brain age from synthetic research-grade magnetization-prepared rapid gradient-echo MRIs (MPRAGEs) generated by a 'super-resolution' method. The model failed with T2-weighted Gradient-Echo MRIs. The mean absolute error (MAE) was 5.86-8.59 years across the other modalities, still higher than for research-grade MRIs, but comparable between actual and synthetic MPRAGEs for some modalities. We modeled the "regression bias" in brain age, for its correction is crucial for providing unbiased summary statistics of brain age or for personalized brain age-based biomarkers. The bias model was generalizable as its correction eliminated any correlation between brain-PAD and chronological age in new samples. Brain-PAD was reliable across modalities. We demonstrate the feasibility of brain age predictions from arbitrary clinical-grade MRIs, thereby contributing to personalized medicine.

Brain-predicted age difference mediates the association between PROMIS sleep impairment, and self-reported pain measure in persons with knee pain.

Knee pain, the most common cause of musculoskeletal pain (MSK), constitutes a severe public health burden. Its neurobiological causes, however, remain poorly understood. Among many possible causes, it has been proposed that sleep problems could lead to an increase in chronic pain symptomatology, which may be driven by central nervous system changes. In fact, we previously found that brain cortical thickness mediated the relationship between sleep qualities and pain severity in older adults with MSK. We also demonstrated a significant difference in a machine-learning-derived brain-aging biomarker between participants with low-and high-impact knee pain. Considering this, we examined whether brain aging was associated with self-reported sleep and pain measures, and whether brain aging mediated the relationship between sleep problems and knee pain. Exploratory Spearman and Pearson partial correlations, controlling for age, sex, race and study site, showed a significant association of brain aging with sleep related impairment and self-reported pain measures. Moreover, mediation analysis showed that brain aging significantly mediated the effect of sleep related impairment on clinical pain and physical symptoms. Our findings extend our prior work demonstrating advanced brain aging among individuals with chronic pain and the mediating role of brain-aging on the association between sleep and pain severity. Future longitudinal studies are needed to further understand whether the brain can be a therapeutic target to reverse the possible effect of sleep problems on chronic pain.

Brain-predicted age difference estimated using DeepBrainNet is significantly associated with pain and function-a multi-institutional and multiscanner study.

ABSTRACT: Brain age predicted differences (brain-PAD: predicted brain age minus chronological age) have been reported to be significantly larger for individuals with chronic pain compared with those without. However, a debate remains after one article showed no significant differences. Using Gaussian Process Regression, an article provides evidence that these negative results might owe to the use of mixed samples by reporting a differential effect of chronic pain on brain-PAD across pain types. However, some remaining methodological issues regarding training sample size and sex-specific effects should be tackled before settling this controversy. Here, we explored differences in brain-PAD between musculoskeletal pain types and controls using a novel convolutional neural network for predicting brain-PADs, ie, DeepBrainNet. Based on a very large, multi-institutional, and heterogeneous training sample and requiring less magnetic resonance imaging preprocessing than other methods for brain age prediction, DeepBrainNet offers robust and reproducible brain-PADs, possibly highly sensitive to neuropathology. Controlling for scanner-related variability, we used a large sample (n = 660) with different scanners, ages (19-83 years), and musculoskeletal pain types (chronic low back [CBP] and osteoarthritis [OA] pain). Irrespective of sex, brain-PAD of OA pain participants was ∼3 to 4.7 years higher than that of CBP and controls, whereas brain-PAD did not significantly differ among controls and CBP. Moreover, brain-PAD was significantly related to multiple variables underlying the multidimensional pain experience. This comprehensive work adds evidence of pain type-specific effects of chronic pain on brain age. This could help in the clarification of the debate around possible relationships between brain aging mechanisms and pain.

Cross-Sectional Brain-Predicted Age Differences in Community-Dwelling Middle-Aged and Older Adults with High Impact Knee Pain.

Purpose: Knee OA-related pain varies in impact across individuals and may relate to central nervous system alterations like accelerated brain aging processes. We previously reported that older adults with chronic musculoskeletal pain had a significantly greater brain-predicted age, compared to pain-free controls, indicating an "older" appearing brain. Yet this association is not well understood. This cross-sectional study examines brain-predicted age differences associated with chronic knee osteoarthritis pain, in a larger, more demographically diverse sample with consideration for pain's impact. Patients and Methods: Participants (mean age = 57.8 ± 8.0 years) with/without knee OA-related pain were classified according to pain's impact on daily function (ie, impact): low-impact (n=111), and high-impact (n=60) pain, and pain-free controls (n=31). Participants completed demographic, pain, and psychosocial assessments, and T1-weighted magnetic resonance imaging. Brain-predicted age difference (brain-PAD) was compared across groups using analysis of covariance. Partial correlations examined associations of brain-PAD with pain and psychosocial variables. Results: Individuals with high-impact chronic knee pain had significantly "older" brains for their age compared to individuals with low-impact knee pain (p < 0.05). Brain-PAD was also significantly associated with clinical pain, negative affect, passive coping, and pain catastrophizing (p's<0.05). Conclusion: Our findings suggest that high impact chronic knee pain is associated with an older appearing brain on MRI. Future studies are needed to determine the impact of pain-related interference and pain management on somatosensory processing and brain aging biomarkers for high-risk populations and effective intervention strategies.

Experimental Pain Phenotype Profiles in Community-dwelling Older Adults.

OBJECTIVES: Pain sensitivity and the brain structure are critical in modulating pain and may contribute to the maintenance of pain in older adults. However, a paucity of evidence exists investigating the link between pain sensitivity and brain morphometry in older adults. The purpose of the study was to identify pain sensitivity profiles in healthy, community-dwelling older adults using a multimodal quantitative sensory testing protocol and to differentiate profiles based on brain morphometry. MATERIALS AND METHODS: This study was a secondary analysis of the Neuromodulatory Examination of Pain and Mobility Across the Lifespan (NEPAL) study. Participants completed demographic and psychological questionnaires, quantitative sensory testing, and a neuroimaging session. A Principal Component Analysis with Varimax rotation followed by hierarchical cluster analysis identified 4 pain sensitivity clusters (the "pain clusters"). RESULTS: Sixty-two older adults ranging from 60 to 94 years old without a specific pain condition (mean [SD] age=71.44 [6.69] y, 66.1% female) were analyzed. Four pain clusters were identified characterized by (1) thermal pain insensitivity; (2) high pinprick pain ratings and pressure pain insensitivity; (3) high thermal pain ratings and high temporal summation; and (4) thermal pain sensitivity, low thermal pain ratings, and low mechanical temporal summation. Sex differences were observed between pain clusters. Pain clusters 2 and 4 were distinguished by differences in the brain cortical volume in the parieto-occipital region. DISCUSSION: While sufficient evidence exists demonstrating pain sensitivity profiles in younger individuals and in those with chronic pain conditions, the finding that subgroups of experimental pain sensitivity also exist in healthy older adults is novel. Identifying these factors in older adults may help differentiate the underlying mechanisms contributing to pain and aging.

Psychological profiles in adults with knee OA-related pain: a replication study.

INTRODUCTION: Psychological factors have been associated with knee osteoarthritis pain severity and treatment outcomes, yet their combined contribution to phenotypic heterogeneity is poorly understood. In particular, empirically derived psychological profiles must be replicated before they can be targeted or considered for treatment studies. The objectives of this study were to (1) confirm previously identified psychological profiles using unsupervised clustering methods in persons with knee osteoarthritis pain, (2) determine the replicability of profiles using supervised machine learning in a different sample, and (3) examine associations with clinical pain, brain structure, and experimental pain. METHODS: Participants included two cohorts of individuals with knee osteoarthritis pain recruited as part of the multisite UPLOAD1 (n = 270, mean age = 56.8 ± 7.6, male = 37%) and UPLOAD2 (n = 164, mean age = 57.73 ± 7.8, male = 36%) studies. Similar psychological constructs (e.g. optimism, coping, somatization, affect, depression, and anxiety), sociodemographic and clinical characteristics, and somatosensory function were assessed across samples. UPLOAD2 participants also completed brain magnetic resonance imaging. Unsupervised hierarchical clustering analysis was first conducted in UPLOAD1 data to derive clusters, followed by supervised linear discriminative analysis to predict group membership in UPLOAD2 data. Associations among cluster membership and clinical variables were assessed, controlling for age, sex, education, ethnicity/race, study site, and number of pain sites. RESULTS: Four distinct profiles emerged in UPLOAD1 and were replicated in UPLOAD2. Identified psychological profiles were associated with psychological variables (ps < 0.001), and clinical outcomes (ps = 0.001-0.03), indicating good internal and external validation of the cluster solution. Significant associations between psychological profiles and somatosensory function and brain structure were also found. CONCLUSIONS: This study highlights the importance of considering the biopsychosocial model in knee osteoarthritis pain assessment and treatment.

Age Differences in Multimodal Quantitative Sensory Testing and Associations With Brain Volume.

BACKGROUND AND OBJECTIVES: Somatosensory function is critical for successful aging. Prior studies have shown declines in somatosensory function with age; however, this may be affected by testing site, modality, and biobehavioral factors. While somatosensory function declines are associated with peripheral nervous system degradation, little is known regarding correlates with the central nervous system and brain structure in particular. The objectives of this study were to examine age-related declines in somatosensory function using innocuous and noxious stimuli, across 2 anatomical testing sites, with considerations for affect and cognitive function, and associations between somatosensory function and brain structure in older adults. RESEARCH DESIGN AND METHODS: A cross-sectional analysis included 84 "younger" (n = 22, age range: 19-24 years) and "older" (n = 62, age range: 60-94 years) healthy adults who participated in the Neuromodulatory Examination of Pain and Mobility Across the Lifespan study. Participants were assessed on measures of somatosensory function (quantitative sensory testing), at 2 sites (metatarsal and thenar) using standardized procedures, and completed cognitive and psychological function measures and structural magnetic resonance imaging. RESULTS: Significant age × test site interaction effects were observed for warmth detection (p = .018, η p 2 = 0.10) and heat pain thresholds (p = .014, η p 2 = 0.12). Main age effects were observed for mechanical, vibratory, cold, and warmth detection thresholds (ps < .05), with older adults displaying a loss of sensory function. Significant associations between somatosensory function and brain gray matter structure emerged in the right occipital region, the right temporal region, and the left pericallosum. DISCUSSION AND IMPLICATIONS: Our findings indicate healthy older adults display alterations in sensory responses to innocuous and noxious stimuli compared to younger adults and, furthermore, these alterations are uniquely affected by anatomical site. These findings suggest a nonuniform decline in somatosensation in older adults, which may represent peripheral and central nervous system alterations part of aging processes.