Biomarkers for Musculoskeletal Pain Conditions: Use of Brain Imaging and Machine Learning

Chronic neuropathic pain components in whiplash-associated disorders correlate with metabolite concentrations in the anterior cingulate and dorsolateral prefrontal cortex: a consensus-driven MRS re-examination

Introduction

Whiplash injury (WHI) is characterised by a forced neck flexion/extension, which frequently occurs after motor vehicle collisions. Previous studies characterising differences in brain metabolite concentrations and correlations with neuropathic pain (NP) components with chronic whiplash-associated disorders (WAD) have been demonstrated in affective pain-processing areas such as the anterior cingulate cortex (ACC). However, the detection of a difference in metabolite concentrations within these cortical areas with chronic WAD pain has been elusive. In this study, single-voxel magnetic resonance spectroscopy (MRS), following the latest MRSinMRS consensus group guidelines, was performed in the anterior cingulate cortex (ACC), left dorsolateral prefrontal cortex (DLPFC), and occipital cortex (OCC) to quantify differences in metabolite concentrations in individuals with chronic WAD with or without neuropathic pain (NP) components.

Materials and methods

Healthy individuals (n = 29) and participants with chronic WAD (n = 29) were screened with the Douleur Neuropathique 4 Questionnaire (DN4) and divided into groups without (WAD-noNP, n = 15) or with NP components (WAD-NP, n = 14). Metabolites were quantified with LCModel following a single session in a 3 T MRI scanner within the ACC, DLPFC, and OCC.

Results

Participants with WAD-NP presented moderate pain intensity and interference compared with the WAD-noNP group. Single-voxel MRS analysis demonstrated a higher glutamate concentration in the ACC and lower total choline (tCho) in the DLPFC in the WAD-NP versus WAD-noNP group, with no intergroup metabolite difference detected in the OCC. Best fit and stepwise multiple regression revealed that the normalised ACC glutamate/total creatine (tCr) (p = 0.01), DLPFC n-acetyl-aspartate (NAA)/tCr (p = 0.001), and DLPFC tCho/tCr levels (p = 0.02) predicted NP components in the WAD-NP group (ACC r 2 = 0.26, α = 0.81; DLPFC r 2 = 0.62, α = 0.98). The normalised Glu/tCr concentration was higher in the healthy than the WAD-noNP group within the ACC (p < 0.05), but not in the DLPFC or OCC. Neither sex nor age affected key normalised metabolite concentrations related to WAD-NP components when compared to the WAD-noNP group.

Discussion

This study demonstrates that elevated glutamate concentrations within the ACC are related to chronic WAD-NP components, while higher NAA and lower tCho metabolite levels suggest a role for increased neuronal-glial signalling and cell membrane dysfunction in individuals with chronic WAD-NP components.

Machine learning-based evaluation of spontaneous pain and analgesics from cellular calcium signals in the mouse primary somatosensory cortex using explainable features

Introduction

Pain that arises spontaneously is considered more clinically relevant than pain evoked by external stimuli. However, measuring spontaneous pain in animal models in preclinical studies is challenging due to methodological limitations. To address this issue, recently we developed a deep learning (DL) model to assess spontaneous pain using cellular calcium signals of the primary somatosensory cortex (S1) in awake head-fixed mice. However, DL operate like a "black box", where their decision-making process is not transparent and is difficult to understand, which is especially evident when our DL model classifies different states of pain based on cellular calcium signals. In this study, we introduce a novel machine learning (ML) model that utilizes features that were manually extracted from S1 calcium signals, including the dynamic changes in calcium levels and the cell-to-cell activity correlations.

Method

We focused on observing neural activity patterns in the primary somatosensory cortex (S1) of mice using two-photon calcium imaging after injecting a calcium indicator (GCaMP6s) into the S1 cortex neurons. We extracted features related to the ratio of up and down-regulated cells in calcium activity and the correlation level of activity between cells as input data for the ML model. The ML model was validated using a Leave-One-Subject-Out Cross-Validation approach to distinguish between non-pain, pain, and drug-induced analgesic states.

Results and discussion

The ML model was designed to classify data into three distinct categories: non-pain, pain, and drug-induced analgesic states. Its versatility was demonstrated by successfully classifying different states across various pain models, including inflammatory and neuropathic pain, as well as confirming its utility in identifying the analgesic effects of drugs like ketoprofen, morphine, and the efficacy of magnolin, a candidate analgesic compound. In conclusion, our ML model surpasses the limitations of previous DL approaches by leveraging manually extracted features. This not only clarifies the decision-making process of the ML model but also yields insights into neuronal activity patterns associated with pain, facilitating preclinical studies of analgesics with higher potential for clinical translation.

Recent developments and future avenues for human corticospinal neuroimaging

Non-invasive neuroimaging serves as a valuable tool for investigating the mechanisms within the central nervous system (CNS) related to somatosensory and motor processing, emotions, memory, cognition, and other functions. Despite the extensive use of brain imaging, spinal cord imaging has received relatively less attention, regardless of its potential to study peripheral communications with the brain and the descending corticospinal systems. To comprehensively understand the neural mechanisms underlying human sensory and motor functions, particularly in pathological conditions, simultaneous examination of neuronal activity in both the brain and spinal cord becomes imperative. Although technically demanding in terms of data acquisition and analysis, a growing but limited number of studies have successfully utilized specialized acquisition protocols for corticospinal imaging. These studies have effectively assessed sensorimotor, autonomic, and interneuronal signaling within the spinal cord, revealing interactions with cortical processes in the brain. In this mini-review, we aim to examine the expanding body of literature that employs cutting-edge corticospinal imaging to investigate the flow of sensorimotor information between the brain and spinal cord. Additionally, we will provide a concise overview of recent advancements in functional magnetic resonance imaging (fMRI) techniques. Furthermore, we will discuss potential future perspectives aimed at enhancing our comprehension of large-scale neuronal networks in the CNS and their disruptions in clinical disorders. This collective knowledge will aid in refining combined corticospinal fMRI methodologies, leading to the development of clinically relevant biomarkers for conditions affecting sensorimotor processing in the CNS.

Mapping the Design Space of Technology-Based Solutions for Better Chronic Pain Care: Introducing the Pain Tech Landscape

OBJECTIVES

Technology has substantial potential to transform and extend care for persons with chronic pain, a burdensome and costly condition. To catalyze the development of impactful applications of technology in this space, we developed the Pain Tech Landscape (PTL) model, which integrates pain care needs with characteristics of technological solutions.

METHODS

Our interdisciplinary group representing experts in pain and human factors research developed PTL through iterative discussions. To demonstrate one potential use of the model, we apply data generated from a narrative review of selected pain and technology journals (2000-2020) in the form of heat map overlays, to reveal where pain tech research attention has focused to date.

RESULTS

The PTL comprises three two-dimensional planes, with pain care needs on each x axis (measurement to management) and technology applications on the y axes according to a) user agency (user- to system-driven), b) usage time frame (temporary to lifelong), and c) collaboration (single-user to collaborative). Heat maps show that existing applications reside primarily in the "user-driven/management" quadrant (e.g., self-care apps). Examples of less developed areas include artificial intelligence and Internet of Things (i.e., Internet-linked household objects), and collaborative/social tools for pain management.

CONCLUSIONS

Collaborative development between the pain and tech fields in early developmental stages using the PTL as a common language could yield impactful solutions for chronic pain management. The PTL could also be used to track developments in the field over time. We encourage periodic reassessment and refinement of the PTL model, which can also be adapted to other chronic conditions.

Circuits and Biomarkers of the Central Nervous System Relating to Astronaut Performance: Summary Report for a NASA-Sponsored Technical Interchange Meeting

Biomarkers, ranging from molecules to behavior, can be used to identify thresholds beyond which performance of mission tasks may be compromised and could potentially trigger the activation of countermeasures. Identification of homologous brain regions and/or neural circuits related to operational performance may allow for translational studies between species. Three discussion groups were directed to use operationally relevant performance tasks as a driver when identifying biomarkers and brain regions or circuits for selected constructs. Here we summarize small-group discussions in tables of circuits and biomarkers categorized by (a) sensorimotor, (b) behavioral medicine and (c) integrated approaches (e.g., physiological responses). In total, hundreds of biomarkers have been identified and are summarized herein by the respective group leads. We hope the meeting proceedings become a rich resource for NASA's Human Research Program (HRP) and the community of researchers.

Understanding the role and adoption of artificial intelligence techniques in rheumatology research: An in-depth review of the literature

The major and upward trend in the number of published research related to rheumatic and musculoskeletal diseases, in which artificial intelligence plays a key role, has exhibited the interest of rheumatology researchers in using these techniques to answer their research questions. In this review, we analyse the original research articles that combine both worlds in a five- year period (2017-2021). In contrast to other published papers on the same topic, we first studied the review and recommendation articles that were published during that period, including up to October 2022, as well as the publication trends. Secondly, we review the published research articles and classify them into one of the following categories: disease identification and prediction, disease classification, patient stratification and disease subtype identification, disease progression and activity, treatment response, and predictors of outcomes. Thirdly, we provide a table with illustrative studies in which artificial intelligence techniques have played a central role in more than twenty rheumatic and musculoskeletal diseases. Finally, the findings of the research articles, in terms of disease and/or data science techniques employed, are highlighted in a discussion. Therefore, the present review aims to characterise how researchers are applying data science techniques in the rheumatology medical field. The most immediate conclusions that can be drawn from this work are: multiple and novel data science techniques have been used in a wide range of rheumatic and musculoskeletal diseases including rare diseases; the sample size and the data type used are heterogeneous, and new technical approaches are expected to arrive in the short-middle term.

Role of Internet of Things (IoT), Artificial Intelligence and Machine Learning in Musculoskeletal Pain: A Scoping Review

Artificial intelligence (AI), Internet of Things (IoT), and machine learning (ML) have considerably increased in numerous critical medical sectors and significantly impacted our daily lives. Digital health interventions support cost-effective, accessible, and preferred interventions that meet time and resource constraints for large patient populations. Musculoskeletal conditions significantly impact society, the economy, and people's life. Adults with chronic neck and back pain are frequently the victims, rendering them physically unable to move. They often experience discomfort, necessitating them to take over-the-counter medications or painkilling gels. Technologies driven by AI have been suggested as an alternative approach to improve adherence to exercise therapy, which in turn helps patients undertake exercises every day to relieve pain associated with the musculoskeletal system. Even though there are many computer-aided evaluations available for physiotherapy rehabilitation, current approaches to computer-aided performance and monitoring lack flexibility and robustness. A thorough literature search was conducted using key databases like PubMed and Google Scholar, as well as Medical Subject Headings (MeSH) terms and related keywords. This research aimed to determine if AI-operated digital health therapies that use cutting-edge IoT, brain imaging, and ML technologies are beneficial in lowering pain and enhancing functional impairment in patients with musculoskeletal diseases. The secondary goal was to ascertain whether solutions driven by machine learning or artificial intelligence can improve exercise compliance and be viewed as a lifestyle choice.

The reliability of two prospective cortical biomarkers for pain: EEG peak alpha frequency and TMS corticomotor excitability

BACKGROUND

Many pain biomarkers fail to move from discovery to clinical application, attributed to poor reliability and an inability to accurately classify at-risk individuals. Preliminary evidence has shown that high pain sensitivity is associated with slow peak alpha frequency (PAF), and depression of corticomotor excitability (CME), potentially due to impairments in ascending sensory and descending motor pathway signalling respectively NEW METHOD: The present study evaluated the reliability of PAF and CME responses during sustained pain. Specifically, we determined whether, over several days of pain, a) PAF remains stable and b) individuals show two stable and distinct CME responses: facilitation and depression. Participants were given an injection of nerve growth factor (NGF) into the right masseter muscle on Day 0 and Day 2, inducing sustained pain. Electroencephalography (EEG) to assess PAF and transcranial magnetic stimulation (TMS) to assess CME were recorded on Day 0, Day 2 and Day 5.

RESULTS

Using a weighted peak estimate, PAF reliability (n = 75) was in the excellent range even without standard pre-processing and ∼2 min recording length. Using a single peak estimate, PAF reliability was in the moderate-good range. For CME (n = 74), 80% of participants showed facilitation or depression of CME beyond an optimal cut-off point, with the stability of these changes in the good range.

COMPARISON WITH EXISTING METHODS

No study has assessed the reliability of PAF or feasibility of classifying individuals as facilitators/depressors, in response to sustained pain. PAF was reliable even in the presence of pain. The use of a weighted peak estimate for PAF is recommended, as excellent test-retest reliability can be obtained even when using minimal pre-processing and ∼2 min recording. We also showed that 80% of individuals exhibit either facilitation or depression of CME, with these changes being stable across sessions.

CONCLUSIONS

Our study provides support for the reliability of PAF and CME as prospective cortical biomarkers. As such, our paper adds important methodological advances to the rapidly growing field of pain biomarkers.

Value assessment of artificial intelligence in medical imaging: a scoping review

BACKGROUND

Artificial intelligence (AI) is seen as one of the major disrupting forces in the future healthcare system. However, the assessment of the value of these new technologies is still unclear, and no agreed international health technology assessment-based guideline exists. This study provides an overview of the available literature in the value assessment of AI in the field of medical imaging.

METHODS

We performed a systematic scoping review of published studies between January 2016 and September 2020 using 10 databases (Medline, Scopus, ProQuest, Google Scholar, and six related databases of grey literature). Information about the context (country, clinical area, and type of study) and mentioned domains with specific outcomes and items were extracted. An existing domain classification, from a European assessment framework, was used as a point of departure, and extracted data were grouped into domains and content analysis of data was performed covering predetermined themes.

RESULTS

Seventy-nine studies were included out of 5890 identified articles. An additional seven studies were identified by searching reference lists, and the analysis was performed on 86 included studies. Eleven domains were identified: (1) health problem and current use of technology, (2) technology aspects, (3) safety assessment, (4) clinical effectiveness, (5) economics, (6) ethical analysis, (7) organisational aspects, (8) patients and social aspects, (9) legal aspects, (10) development of AI algorithm, performance metrics and validation, and (11) other aspects. The frequency of mentioning a domain varied from 20 to 78% within the included papers. Only 15/86 studies were actual assessments of AI technologies. The majority of data were statements from reviews or papers voicing future needs or challenges of AI research, i.e. not actual outcomes of evaluations.

CONCLUSIONS

This review regarding value assessment of AI in medical imaging yielded 86 studies including 11 identified domains. The domain classification based on European assessment framework proved useful and current analysis added one new domain. Included studies had a broad range of essential domains about addressing AI technologies highlighting the importance of domains related to legal and ethical aspects.

Classification of primary dysmenorrhea by brain effective connectivity of the amygdala: a machine learning study

BACKGROUND

The amygdala plays a crucial role in the central pathogenesis mechanism of primary dysmenorrhea (PDM). However, the detailed pain modulation principles of the amygdala in PDM remain unclear. Here, we applied the Granger causality analysis (GCA) to investigate the directional effective connectivity (EC) alterations in the amygdala network of PDM patients.

METHODS

Thirty-seven patients with PDM and 38 healthy controls were enrolled in this study and underwent resting-state functional magnetic resonance imaging scans during the pain-free stage. GCA was employed to explore the amygdala-based EC network alteration in PDM. A multivariate pattern analysis (MVPA)-based machine learning approach was used to explore whether the altered amygdala EC could serve as an fMRI-based marker for classifying PDM and HC participants.

RESULTS

Compared to the healthy control group, patients with PDM showed significantly decreased EC from the amygdala to the right superior frontal gyrus (SFG), right superior parietal lobe/middle occipital gyrus, and left middle cingulate cortex, whereas increased EC was found from the amygdala to the bilateral medial orbitofrontal cortex. In addition, increased EC was found from the bilateral SFG to the amygdala, and decreased EC was found from the medial orbitofrontal cortex, caudate nucleus to the amygdala. The increased EC from the right SFG to the amygdala was associated with a plasma prostaglandin E2 level in PDM. The MVPA based on an altered amygdala EC pattern yielded a total accuracy of 86.84% for classifying the patients with PDM and HC.

CONCLUSION

Our study is the first to combine MVPA and EC to explore brain function alteration in PDM. The results could advance understanding of the neural theory of PDM in specifying the pain-free period.

Neuroimaging Assessment of Pain

Pain is an unpleasant sensory and emotional experience. Understanding the neural mechanisms of acute and chronic pain and the brain changes affecting pain factors is important for finding pain treatment methods. The emergence and progress of non-invasive neuroimaging technology can help us better understand pain at the neural level. Recent developments in identifying brain-based biomarkers of pain through advances in advanced imaging can provide some foundations for predicting and detecting pain. For example, a neurologic pain signature (involving brain regions that receive nociceptive afferents) and a stimulus intensity-independent pain signature (involving brain regions that do not show increased activity in proportion to noxious stimulus intensity) were developed based on multivariate modeling to identify processes related to the pain experience. However, an accurate and comprehensive review of common neuroimaging techniques for evaluating pain is lacking. This paper reviews the mechanism, clinical application, reliability, strengths, and limitations of common neuroimaging techniques for assessing pain to promote our further understanding of pain.

Validity and Diagnosis in Physical and Rehabilitation Medicine: Critical View and Future Perspectives

ABSTRACT

Obtaining a diagnosis is an essential and integral part of physical and rehabilitation medicine in practice and research. Standardized psychometric properties are required of any classifications, diagnostic criteria, and diagnostic rules used. Physicians and researchers, in physical and rehabilitation medicine, need to understand these properties to determine the accuracy and consistency of their diagnosis. Although chronic musculoskeletal pain disorders are among the highly prevalent disorders seen in physical and rehabilitation medicine, limitations regarding existing diagnostic criteria for chronic musculoskeletal pain disorders still exist. Hence, the quest for developing diagnostic tools for chronic musculoskeletal pain that align with the standard properties remains open. These are discussed with an example for existing diagnostic criteria for fibromyalgia. This article primarily aimed to provide an overview of standard psychometric properties. A secondary aim was to critically appraise the tools currently used to diagnose chronic musculoskeletal pain disorders. The challenges and limitations of existing diagnostic tools are discussed. Potential approaches on how to improve the conceptualization of the construct of musculoskeletal pain disorders are also discussed. Adopting a network perspective, for example, can better constitute the disease instead of a single known underlying etiology for persistent or recurrent pain symptoms.

Brain Imaging Biomarkers for Chronic Pain

The prevalence of chronic pain has reached epidemic levels. In addition to personal suffering chronic pain is associated with psychiatric and medical co-morbidities, notably substance misuse, and a huge a societal cost amounting to hundreds of billions of dollars annually in medical cost, lost wages, and productivity. Chronic pain does not have a cure or quantitative diagnostic or prognostic tools. In this manuscript we provide evidence that this situation is about to change. We first start by summarizing our current understanding of the role of the brain in the pathogenesis of chronic pain. We particularly focus on the concept of learning in the emergence of chronic pain, and the implication of the limbic brain circuitry and dopaminergic signaling, which underly emotional learning and decision making, in this process. Next, we summarize data from our labs and from other groups on the latest brain imaging findings in different chronic pain conditions focusing on results with significant potential for translation into clinical applications. The gaps in the study of chronic pain and brain imaging are highlighted in throughout the overview. Finally, we conclude by discussing the costs and benefits of using brain biomarkers of chronic pain and compare to other potential markers.

A classification algorithm to predict chronic pain using both regression and machine learning - A stepwise approach

This secondary data analysis study aimed to (1) investigate the use of two sense-based parameters (movement and sleep hours) as predictors of chronic pain when controlling for patient demographics and depression, and (2) identify a classification model with accuracy in predicting chronic pain. Data collected by Oregon Health & Science University between March 2018 and December 2019 under the Collaborative Aging Research Using Technology Initiative were analyzed in two stages. Data were collected by sensor technologies and questionnaires from older adults living independently or with a partner in the community. In Stage 1, regression models were employed to determine unique sensor-based behavioral predictors of pain. These sensor-based parameters were used to create a classification model to predict the weekly recalled pain intensity and interference level using a deep neural network model, a machine learning approach, in Stage 2. Daily step count was a unique predictor for both pain intensity (75% Accuracy, F1 = 0.58) and pain interference (82% Accuracy, F1 = 0.59). The developed classification model performed well in this dataset with acceptable accuracy scores. This study demonstrated that machine learning technique can be used to identify the relationship between patients' pain and the risk factors.

Screening the Influence of Biomarkers for Metabolic Syndrome in Occupational Population Based on the Lasso Algorithm

Aim: Metabolic syndrome (MS) screening is essential for the early detection of the occupational population. This study aimed to screen out biomarkers related to MS and establish a risk assessment and prediction model for the routine physical examination of an occupational population.

Methods: The least absolute shrinkage and selection operator (Lasso) regression algorithm of machine learning was used to screen biomarkers related to MS. Then, the accuracy of the logistic regression model was further verified based on the Lasso regression algorithm. The areas under the receiving operating characteristic curves were used to evaluate the selection accuracy of biomarkers in identifying MS subjects with risk. The screened biomarkers were used to establish a logistic regression model and calculate the odds ratio (OR) of the corresponding biomarkers. A nomogram risk prediction model was established based on the selected biomarkers, and the consistency index (C-index) and calibration curve were derived.

Results: A total of 2,844 occupational workers were included, and 10 biomarkers related to MS were screened. The number of non-MS cases was 2,189 and that of MS was 655. The area under the curve (AUC) value for non-Lasso and Lasso logistic regression was 0.652 and 0.907, respectively. The established risk assessment model revealed that the main risk biomarkers were absolute basophil count (OR: 3.38, CI:1.05–6.85), platelet packed volume (OR: 2.63, CI:2.31–3.79), leukocyte count (OR: 2.01, CI:1.79–2.19), red blood cell count (OR: 1.99, CI:1.80–2.71), and alanine aminotransferase level (OR: 1.53, CI:1.12–1.98). Furthermore, favorable results with C-indexes (0.840) and calibration curves closer to ideal curves indicated the accurate predictive ability of this nomogram.

Conclusions: The risk assessment model based on the Lasso logistic regression algorithm helped identify MS with high accuracy in physically examining an occupational population.

Functional and structural neuroplastic changes related to sensitization proxies in patients with Osteoarthritis: a systematic review

OBJECTIVE

Several reports in literature have identified sensitization as a possible basis for the enhanced pain reactions associated with Osteoarthritis (OA). The aim of this current systematic review is to summarize functional and structural brain changes associated with surrogate sensitization parameters assessed in patients with OA-related pain.

DESIGN

Systematic review.

SUBJECTS

Patients with OA related pain.

METHODS

A literature search was conducted systematically in MEDLINE, CINAHL, EMBASE databases for human studies up to December 2019. Articles were included if they assessed brain imaging and senzitisation parameters (quantitative sensory testing and questionnaires) in adults with OA related pain. Methodological quality was assessed using the Methodological Index for Non-Randomized Studies (MINORS) score.

RESULTS

Five studies reporting on 138 patients were included in this review. The MINORS scale yielded mean scores of 8.5/16 and 12.3/24, for the cohort and case-control studies respectively. Four low-quality studies suggest a greater pain matrix activation associated with clinical measures of sensitization in patients with OA, while another study underlined the presence of structural changes (reduced gray matter volume) in the cortical areas involved in the nociceptive processing possible also related to sensitization.

CONCLUSION

This review shows conflicting evidence for structural and functional neuroplastic brain changes related to sensitization proxies in patients with OA.

Characterizing Brain Tumor Regions Using Texture Analysis in Magnetic Resonance Imaging

Purpose

To extract texture features from magnetic resonance imaging (MRI) scans of patients with brain tumors and use them to train a classification model for supporting an early diagnosis.

Methods

Two groups of regions (control and tumor) were selected from MRI scans of 40 patients with meningioma or glioma. These regions were analyzed to obtain texture features. Statistical analysis was conducted using SPSS (version 20.0), including the Shapiro-Wilk test and Wilcoxon signed-rank test, which were used to test significant differences in each feature between the tumor and healthy regions. T-distributed stochastic neighbor embedding (t-SNE) was used to visualize the data distribution so as to avoid tumor selection bias. The Gini impurity index in random forests (RFs) was used to select the top five out of all features. Based on the five features, three classification models were built respectively with three machine learning classifiers: RF, support vector machine (SVM), and back propagation (BP) neural network.

Results

Sixteen of the 25 features were significantly different between the tumor and healthy areas. Through the Gini impurity index in RFs, standard deviation, first-order moment, variance, third-order absolute moment, and third-order central moment were selected to build the classification model. The classification model trained using the SVM classifier achieved the best performance, with sensitivity, specificity, and area under the curve of 94.04%, 92.3%, and 0.932, respectively.

Conclusion

Texture analysis with an SVM classifier can help differentiate between brain tumor and healthy areas with high speed and accuracy, which would facilitate its clinical application.

Implementation of the Uniformed Services University Pain Registry Biobank: A Military and Veteran Population Focused Biobank and Registry

OBJECTIVE

The objective of this overview is to discuss the development, implementation, data content, and structure of the Uniformed Services University Pain Registry Biobank. Additionally, procedures and policies for accessing samples for pain-related research purposes are detailed.

DESIGN

Cross-sectional overview.

SETTING

Multiple military treatment facilities.

SUBJECTS

Adult beneficiaries seeking care within the Military Health System.

METHODS

Participants complete a baseline battery of biopsychosocial survey measures, including PROMIS® measures, provide biologic samples (e.g. blood and saliva), and relevant health history, including medications and surgical history, is extracted from medical records. During the course of the next year, enrolled participants complete surveys and provide biologic samples at 3-months, 6-months, and 12-months. Thereafter, participants are contacted once annually to complete self-reported assessments and provide biologic samples.

RESULTS

In the first year alone 86 subjects have participated in the Uniformed Services University Pain Registry Biobank and provided 390 observations (e.g. biological samples and biopsychosocial patient-reported outcomes). The Uniformed Services University Pain Registry Biobank's integration of biological samples, patient-reported outcomes, and health record data over a longitudinal period across a diverse sample recruited from multiple military facilities addresses many of the limitations faced by other pain-related registries or biorepositories.

CONCLUSIONS

The Uniformed Services University Pain Registry Biobank will serve as a platform for conducting research closely aligned with the Federal Pain Research Strategy. The inclusion of active duty service members, beneficiaries, and civilians living with and without acute or chronic pain provides a unique data repository for all investigators interested in advancing pain science.

A calibrated deep learning ensemble for abnormality detection in musculoskeletal radiographs

Musculoskeletal disorders affect the locomotor system and are the leading contributor to disability worldwide. Patients suffer chronic pain and limitations in mobility, dexterity, and functional ability. Musculoskeletal (bone) X-ray is an essential tool in diagnosing the abnormalities. In recent years, deep learning algorithms have increasingly been applied in musculoskeletal radiology and have produced remarkable results. In our study, we introduce a new calibrated ensemble of deep learners for the task of identifying abnormal musculoskeletal radiographs. Our model leverages the strengths of three baseline deep neural networks (ConvNet, ResNet, and DenseNet), which are typically employed either directly or as the backbone architecture in the existing deep learning-based approaches in this domain. Experimental results based on the public MURA dataset demonstrate that our proposed model outperforms three individual models and a traditional ensemble learner, achieving an overall performance of (AUC: 0.93, Accuracy: 0.87, Precision: 0.93, Recall: 0.81, Cohen's kappa: 0.74). The model also outperforms expert radiologists in three out of the seven upper extremity anatomical regions with a leading performance of (AUC: 0.97, Accuracy: 0.93, Precision: 0.90, Recall:0.97, Cohen's kappa: 0.85) in the humerus region. We further apply the class activation map technique to highlight the areas essential to our model's decision-making process. Given that the best radiologist performance is between 0.73 and 0.78 in Cohen's kappa statistic, our study provides convincing results supporting the utility of a calibrated ensemble approach for assessing abnormalities in musculoskeletal X-rays.

Artificial intelligence for good health: a scoping review of the ethics literature

Background

Artificial intelligence (AI) has been described as the “fourth industrial revolution” with transformative and global implications, including in healthcare, public health, and global health. AI approaches hold promise for improving health systems worldwide, as well as individual and population health outcomes. While AI may have potential for advancing health equity within and between countries, we must consider the ethical implications of its deployment in order to mitigate its potential harms, particularly for the most vulnerable. This scoping review addresses the following question: What ethical issues have been identified in relation to AI in the field of health, including from a global health perspective?

Methods

Eight electronic databases were searched for peer reviewed and grey literature published before April 2018 using the concepts of health, ethics, and AI, and their related terms. Records were independently screened by two reviewers and were included if they reported on AI in relation to health and ethics and were written in the English language. Data was charted on a piloted data charting form, and a descriptive and thematic analysis was performed.

Results

Upon reviewing 12,722 articles, 103 met the predetermined inclusion criteria. The literature was primarily focused on the ethics of AI in health care, particularly on carer robots, diagnostics, and precision medicine, but was largely silent on ethics of AI in public and population health. The literature highlighted a number of common ethical concerns related to privacy, trust, accountability and responsibility, and bias. Largely missing from the literature was the ethics of AI in global health, particularly in the context of low- and middle-income countries (LMICs).

Conclusions

The ethical issues surrounding AI in the field of health are both vast and complex. While AI holds the potential to improve health and health systems, our analysis suggests that its introduction should be approached with cautious optimism. The dearth of literature on the ethics of AI within LMICs, as well as in public health, also points to a critical need for further research into the ethical implications of AI within both global and public health, to ensure that its development and implementation is ethical for everyone, everywhere.

An Ensemble of Psychological and Physical Health Indices Discriminates Between Individuals with Chronic Pain and Healthy Controls with High Reliability: A Machine Learning Study

INTRODUCTION

Chronic pain (CP) is a complex multidimensional experience severely affecting individuals' quality of life. Multiple cognitive, affective, emotional, and interpersonal factors play a major role in CP. Furthermore, the psychological, social, and physical circumstances leading to CP show high inter-individual variability, thus making it difficult to identify core syndrome characteristics. In a biopsychosocial perspective, we aim at identifying a pattern of psycho-physical impairments that can reliably discriminate between CP individuals and healthy controls (HC) with high accuracy and estimated generalizability using machine learning.

METHODS

A total of 118 CP and 86 HC were recruited. All individuals were administered several scales assessing quality of life, physical and mental health, personal functioning, anxiety, depression, beliefs about medical treatments, and cognitive ability. These features were trained to separate CP from HC using support vector classification and repeated nested cross-validation.

RESULTS

Our psycho-physical classifier was able to discriminate CP from HC with 86.5% balanced accuracy and significance (p = 0.0001). The most reliable features characterizing CP were anxiety and depression scores, and belief of harm from prolonged pharmacological treatments; for HP, the most reliable features were physical and occupational functioning, and vitality levels.

CONCLUSION

Our findings suggest that, using psychological and physical assessments, it is possible to classify CP from HC with high reliability and estimated generalizability via (i) a pattern of psychological symptoms and cognitive beliefs characteristic of CP, and (ii) a pattern of intact physical functioning characteristic of HC. We think that our algorithm enables novel insights into potential individualized targets for CP-related early intervention programs.

<p>Mitochondrial Imbalance as a New Approach to the Study of Fibromyalgia</p>

Background

Fibromyalgia (FM) is a common chronic pain disease, whose pathogenic mechanism still remains elusive. Oxidative stress markers and impaired bioenergetics homeostasis have been proposed as relevant events in the pathogenesis of the disease. Hence, the aim of the study is to analyse the potential biomarkers of mitochondrial imbalance in FM patients along with coenzyme Q10 (CoQ10) as a possible treatment.

Methods

The symptomatology of patients was recorded with an adaption of the Fibromyalgia Impact Questionnaire (FIQ). Mitochondrial imbalance was tested from blood extraction and serum isolation in 33 patients diagnosed with FM and 30 healthy controls. Western blot and HPLC techniques were performed to study the different parameters. Finally, bioinformatic analysis of machine learning was performed to predict possible associations of results.

Results

CoQ10 parameter did not show evidence to be a good marker of the disease, as the values are not significantly different between control and patient groups (Student’s t-test, CI 95%). For this reason, the focus of the study changed into the ratio between mitochondrial mass and autophagy levels. The bioinformatics analysis showed a possible association between this ratio and patients’ symptomatology. Finally, the effects of coenzyme Q10 as a potential treatment for the disease were different within patients, and its efficacy may be related to the initial mitochondrial status. However, there is no statistical significance due to limitations within the sample size.

Conclusion

Our study supports the hypothesis that an imbalance in mitochondrial homeostasis is involved in the FM pathogenesis. However, whether the increase in oxidative stress is the result of mitochondrial imbalance or the cause of this disease remains an open question. The measurement of this imbalance might be used as a preliminary biomarker for the diagnosis and follow-up of patients with FM, and even for the evaluation of the effects of the different antioxidants therapies.

Primer on machine learning: utilization of large data set analyses to individualize pain management

PURPOSE OF REVIEW

Pain researchers and clinicians increasingly encounter machine learning algorithms in both research methods and clinical practice. This review provides a summary of key machine learning principles, as well as applications to both structured and unstructured datasets.

RECENT FINDINGS

Aside from increasing use in the analysis of electronic health record data, machine and deep learning algorithms are now key tools in the analyses of neuroimaging and facial expression recognition data used in pain research.

SUMMARY

In the coming years, machine learning is likely to become a key component of evidence-based medicine, yet will require additional skills and perspectives for its successful and ethical use in research and clinical settings.

Sensor Technologies to Manage the Physiological Traits of Chronic Pain: A Review

Non-oncologic chronic pain is a common high-morbidity impairment worldwide and acknowledged as a condition with significant incidence on quality of life. Pain intensity is largely perceived as a subjective experience, what makes challenging its objective measurement. However, the physiological traces of pain make possible its correlation with vital signs, such as heart rate variability, skin conductance, electromyogram, etc., or health performance metrics derived from daily activity monitoring or facial expressions, which can be acquired with diverse sensor technologies and multisensory approaches. As the assessment and management of pain are essential issues for a wide range of clinical disorders and treatments, this paper reviews different sensor-based approaches applied to the objective evaluation of non-oncological chronic pain. The space of available technologies and resources aimed at pain assessment represent a diversified set of alternatives that can be exploited to address the multidimensional nature of pain.

Delineating conditions and subtypes in chronic pain using neuroimaging

Differentiating subtypes of chronic pain still remains a challenge—both from a subjective and objective point of view. Personalized medicine is the current goal of modern medical care and is limited by the subjective nature of patient self-reporting of symptoms and behavioral evaluation. Physiology-focused techniques such as genome and epigenetic analyses inform the delineation of pain groups; however, except under rare circumstances, they have diluted effects that again, share a common reliance on behavioral evaluation. The application of structural neuroimaging towards distinguishing pain subtypes is a growing field and may inform pain-group classification through the analysis of brain regions showing hypertrophic and atrophic changes in the presence of pain. Analytical techniques such as machine-learning classifiers have the capacity to process large volumes of data and delineate diagnostically relevant information from neuroimaging analysis. The issue of defining a “brain type” is an emerging field aimed at interpreting observed brain changes and delineating their clinical identity/significance. In this review, 2 chronic pain conditions (migraine and irritable bowel syndrome) with similar clinical phenotypes are compared in terms of their structural neuroimaging findings. Independent investigations are compared with findings from application of machine-learning algorithms. Findings are discussed in terms of differentiating patient subgroups using neuroimaging data in patients with chronic pain and how they may be applied towards defining a personalized pain signature that helps segregate patient subgroups (eg, migraine with and without aura, with or without nausea; irritable bowel syndrome vs other functional gastrointestinal disorders).

Neuroimaging-based biomarkers for pain: state of the field and current directions

Chronic pain is an endemic problem involving both peripheral and brain pathophysiology. Although biomarkers have revolutionized many areas of medicine, biomarkers for pain have remained controversial and relatively underdeveloped. With the realization that biomarkers can reveal pain-causing mechanisms of disease in brain circuits and in the periphery, this situation is poised to change. In particular, brain pathophysiology may be diagnosable with human brain imaging, particularly when imaging is combined with machine learning techniques designed to identify predictive measures embedded in complex data sets. In this review, we explicate the need for brain-based biomarkers for pain, some of their potential uses, and some of the most popular machine learning approaches that have been brought to bear. Then, we evaluate the current state of pain biomarkers developed with several commonly used methods, including structural magnetic resonance imaging, functional magnetic resonance imaging and electroencephalography. The field is in the early stages of biomarker development, but these complementary methodologies have already produced some encouraging predictive models that must be tested more extensively across laboratories and clinical populations.

Artificial intelligence and machine learning | applications in musculoskeletal physiotherapy

INTRODUCTION

Artificial intelligence (AI) is a field of mathematical engineering which has potential to enhance healthcare through new care delivery strategies, informed decision making and facilitation of patient engagement. Machine learning (ML) is a form of narrow artificial intelligence which can be used to automate decision making and make predictions based upon patient data.

PURPOSE

This review outlines key applications of supervised and unsupervised machine learning in musculoskeletal medicine; such as diagnostic imaging, patient measurement data, and clinical decision support. The current literature base is examined to identify areas where ML performs equal to or more accurately than human levels.

IMPLICATIONS

Potential is apparent for intelligent machines to enhance various areas of physiotherapy practice through automization of tasks which involve data analysis, classification and prediction. Changes to service provision through applications of ML, should encourage physiotherapists to increase their awareness of and experiences with emerging technologies. Data literacy should be a component of professional development plans to assist physiotherapists in the application of ML and the preparation of information technology systems to use these techniques.

Automated Assessment of Dynamic Knee Valgus and Risk of Knee Injury During the Single Leg Squat

Many clinical assessment protocols of the lower limb rely on the evaluation of functional movement tests such as the single leg squat (SLS), which are often assessed visually. Visual assessment is subjective and depends on the experience of the clinician. In this paper, an inertial measurement unit (IMU)-based method for automated assessment of squat quality is proposed to provide clinicians with a quantitative measure of SLS performance. A set of three IMUs was used to estimate the joint angles, velocities, and accelerations of the squatting leg. Statistical time domain features were generated from these measurements. The most informative features were used for classifier training. A data set of SLS performed by healthy participants was collected and labeled by three expert clinical raters using two different labeling criteria: "observed amount of knee valgus" and "overall risk of injury". The results showed that both flexion at the hip and knee, as well as hip and ankle internal rotation are discriminative features, and that participants with "poor" squats bend the hip and knee less than those with better squat performance. Furthermore, improved classification performance is achieved for females by training separate classifiers stratified by gender. Classification results showed excellent accuracy, 95.7 % for classifying squat quality as "poor" or "good" and 94.6% for differentiating between high and no risk of injury.

Multivariate pattern analysis utilizing structural or functional MRI-In individuals with musculoskeletal pain and healthy controls: A systematic review

OBJECTIVE

The purpose of this systematic review is to systematically review the evidence relating to findings generated by multivariate pattern analysis (MVPA) following structural or functional magnetic resonance imaging (fMRI) to determine if this analysis is able to: a) Discriminate between individuals with musculoskeletal pain and healthy controls, b) Predict pain perception in healthy individuals stimulated with a noxious stimulus compared to those stimulated with a non-noxious stimulus.

METHODS

MEDLINE, CINAHL, Embase, PEDro, Google Scholar, Cochrane library and Web of Science were systematically screened for relevant literature using different combinations of keywords regarding structural and functional MRI analysed with MVPA, both in individuals with musculoskeletal pain and healthy controls. Reference lists of included articles were hand-searched for additional literature. Eligible articles were assessed on risk of bias and reviewed by two independent researchers.

RESULTS

The search query returned 18 articles meeting the inclusion criteria. Methodological quality varied from poor to good. Seven studies investigated the ability of machine-learning algorithms to differentiate patient groups from healthy control participants. Overall, the review demonstrated that MVPA can discriminate between individuals with MSK pain and healthy controls with an overall accuracy ranging from 53% to 94%. Twelve studies utilized healthy control participants (using them as their own controls), during experimental pain paradigms aimed to investigate the ability of machine-learning to differentiate individuals stimulated with noxious stimuli from those stimulated with non-noxious stimuli, with 'pain' detection rates ranging from 60% to 94%. However, significant heterogeneity in patient conditions, study methodology and brain imaging techniques resulted in various findings that make study comparisons and formal conclusions challenging.

CONCLUSION

There is preliminary and emerging evidence that MVPA analyses of structural or functional MRI are able to discriminate between patients and healthy controls, and also discriminate between noxious and non-noxious stimulation. No prospective studies were found in this review to allow determination of the prognostic or diagnostic capabilities or treatment responsiveness of these analyses. Future studies would also benefit from combining various behavioural, genotype and phenotype data into analyses to assist with development of sensitive and specific signatures that could guide future individualized patient treatment options and evaluate how treatments exert their effects.