Classification of low back pain from dynamic motion characteristics using an artificial neural network

Artificial intelligence: a new cutting-edge tool in spine surgery

The purpose of this narrative review was to comprehensively elaborate the various components of artificial intelligence (AI), their applications in spine surgery, practical concerns, and future directions. Over the years, spine surgery has been continuously transformed in various aspects, including diagnostic strategies, surgical approaches, procedures, and instrumentation, to provide better-quality patient care. Surgeons have also augmented their surgical expertise with rapidly growing technological advancements. AI is an advancing field that has the potential to revolutionize many aspects of spine surgery. We performed a comprehensive narrative review of the various aspects of AI and machine learning in spine surgery. To elaborate on the current role of AI in spine surgery, a review of the literature was performed using PubMed and Google Scholar databases for articles published in English in the last 20 years. The initial search using the keywords "artificial intelligence" AND "spine," "machine learning" AND "spine," and "deep learning" AND "spine" extracted a total of 78, 60, and 37 articles and 11,500, 4,610, and 2,270 articles on PubMed and Google Scholar. After the initial screening and exclusion of unrelated articles, duplicates, and non-English articles, 405 articles were identified. After the second stage of screening, 93 articles were included in the review. Studies have shown that AI can be used to analyze patient data and provide personalized treatment recommendations in spine care. It also provides valuable insights for planning surgeries and assisting with precise surgical maneuvers and decisionmaking during the procedures. As more data become available and with further advancements, AI is likely to improve patient outcomes.

Biomechanical Phenotyping of Chronic Low Back Pain: Protocol for BACPAC

OBJECTIVE

Biomechanics represents the common final output through which all biopsychosocial constructs of back pain must pass, making it a rich target for phenotyping. To exploit this feature, several sites within the NIH Back Pain Consortium (BACPAC) have developed biomechanics measurement and phenotyping tools. The overall aims of this article were to: 1) provide a narrative review of biomechanics as a phenotyping tool; 2) describe the diverse array of tools and outcome measures that exist within BACPAC; and 3) highlight how leveraging these technologies with the other data collected within BACPAC could elucidate the relationship between biomechanics and other metrics used to characterize low back pain (LBP).

METHODS

The narrative review highlights how biomechanical outcomes can discriminate between those with and without LBP, as well as among levels of severity of LBP. It also addresses how biomechanical outcomes track with functional improvements in LBP. Additionally, we present the clinical use case for biomechanical outcome measures that can be met via emerging technologies.

RESULTS

To answer the need for measuring biomechanical performance, our "Results" section describes the spectrum of technologies that have been developed and are being used within BACPAC.

CONCLUSION AND FUTURE DIRECTIONS

The outcome measures collected by these technologies will be an integral part of longitudinal and cross-sectional studies conducted in BACPAC. Linking these measures with other biopsychosocial data collected within BACPAC increases our potential to use biomechanics as a tool for understanding the mechanisms of LBP, phenotyping unique LBP subgroups, and matching these individuals with an appropriate treatment paradigm.

Artificial Intelligence and Computer Aided Diagnosis in Chronic Low Back Pain: A Systematic Review

Low Back Pain (LBP) is currently the first cause of disability in the world, with a significant socioeconomic burden. Diagnosis and treatment of LBP often involve a multidisciplinary, individualized approach consisting of several outcome measures and imaging data along with emerging technologies. The increased amount of data generated in this process has led to the development of methods related to artificial intelligence (AI), and to computer-aided diagnosis (CAD) in particular, which aim to assist and improve the diagnosis and treatment of LBP. In this manuscript, we have systematically reviewed the available literature on the use of CAD in the diagnosis and treatment of chronic LBP. A systematic research of PubMed, Scopus, and Web of Science electronic databases was performed. The search strategy was set as the combinations of the following keywords: “Artificial Intelligence”, “Machine Learning”, “Deep Learning”, “Neural Network”, “Computer Aided Diagnosis”, “Low Back Pain”, “Lumbar”, “Intervertebral Disc Degeneration”, “Spine Surgery”, etc. The search returned a total of 1536 articles. After duplication removal and evaluation of the abstracts, 1386 were excluded, whereas 93 papers were excluded after full-text examination, taking the number of eligible articles to 57. The main applications of CAD in LBP included classification and regression. Classification is used to identify or categorize a disease, whereas regression is used to produce a numerical output as a quantitative evaluation of some measure. The best performing systems were developed to diagnose degenerative changes of the spine from imaging data, with average accuracy rates >80%. However, notable outcomes were also reported for CAD tools executing different tasks including analysis of clinical, biomechanical, electrophysiological, and functional imaging data. Further studies are needed to better define the role of CAD in LBP care.

Artificial intelligence to improve back pain outcomes and lessons learnt from clinical classification approaches: three systematic reviews

Artificial intelligence and machine learning (AI/ML) could enhance the ability to detect patterns of clinical characteristics in low-back pain (LBP) and guide treatment. We conducted three systematic reviews to address the following aims: (a) review the status of AI/ML research in LBP, (b) compare its status to that of two established LBP classification systems (STarT Back, McKenzie). AI/ML in LBP is in its infancy: 45 of 48 studies assessed sample sizes <1000 people, 19 of 48 studies used ≤5 parameters in models, 13 of 48 studies applied multiple models and attained high accuracy, 25 of 48 studies assessed the binary classification of LBP versus no-LBP only. Beyond the 48 studies using AI/ML for LBP classification, no studies examined use of AI/ML in prognosis prediction of specific sub-groups, and AI/ML techniques are yet to be implemented in guiding LBP treatment. In contrast, the STarT Back tool has been assessed for internal consistency, test-retest reliability, validity, pain and disability prognosis, and influence on pain and disability treatment outcomes. McKenzie has been assessed for inter- and intra-tester reliability, prognosis, and impact on pain and disability outcomes relative to other treatments. For AI/ML methods to contribute to the refinement of LBP (sub-)classification and guide treatment allocation, large data sets containing known and exploratory clinical features should be examined. There is also a need to establish reliability, validity, and prognostic capacity of AI/ML techniques in LBP as well as its ability to inform treatment allocation for improved patient outcomes and/or reduced healthcare costs.

A Review on the Use of Artificial Intelligence in Spinal Diseases

Artificial neural networks (ANNs) have been used in a wide variety of real-world applications and it emerges as a promising field across various branches of medicine. This review aims to identify the role of ANNs in spinal diseases. Literature were searched from electronic databases of Scopus and Medline from 1993 to 2020 with English publications reported on the application of ANNs in spinal diseases. The search strategy was set as the combinations of the following keywords: “artificial neural networks,” “spine,” “back pain,” “prognosis,” “grading,” “classification,” “prediction,” “segmentation,” “biomechanics,” “deep learning,” and “imaging.” The main findings of the included studies were summarized, with an emphasis on the recent advances in spinal diseases and its application in the diagnostic and prognostic procedures. According to the search strategy, a set of 3,653 articles were retrieved from Medline and Scopus databases. After careful evaluation of the abstracts, the full texts of 89 eligible papers were further examined, of which 79 articles satisfied the inclusion criteria of this review. Our review indicates several applications of ANNs in the management of spinal diseases including (1) diagnosis and assessment of spinal disease progression in the patients with low back pain, perioperative complications, and readmission rate following spine surgery; (2) enhancement of the clinically relevant information extracted from radiographic images to predict Pfirrmann grades, Modic changes, and spinal stenosis grades on magnetic resonance images automatically; (3) prediction of outcomes in lumbar spinal stenosis, lumbar disc herniation and patient-reported outcomes in lumbar fusion surgery, and preoperative planning and intraoperative assistance; and (4) its application in the biomechanical assessment of spinal diseases. The evidence suggests that ANNs can be successfully used for optimizing the diagnosis, prognosis and outcome prediction in spinal diseases. Therefore, incorporation of ANNs into spine clinical practice may improve clinical decision making.

Effectiveness of lumbar supports in low back functionality and disability in assembly-line workers

Low back pain (LBP) is a common problem in manufacturing workers. Several strategies have been proposed in order to reduce the pain and/or improve functionality. Among them, lumbar supports are a common solution prescribed for lumbar pain relief. Most of the studies in the literature only consider subjective sensations of the workers for evaluation assessment. This study applies biomechanical tests (a flexion-relaxation test and a functional movement evaluation test) to analyse the effectiveness of flexible lumbar supports in functionality and disability versus placebo intervention, consisting of kinesiotape placed on the low back without any stress. 28 workers participated in the study, randomised in control and intervention groups with a two months' intervention. None of the biomechanical tests showed statistical differences in between-groups pre-post changes. No benefits of wearing a flexible lumbar support during the workday have been found in these assembly-line workers versus placebo intervention.

Prediction Effects of Personal, Psychosocial, and Occupational Risk Factors on Low Back Pain Severity Using Artificial Neural Networks Approach in Industrial Workers

OBJECTIVES

This study aimed to provide an empirical model of predicting low back pain (LBP) by considering the occupational, personal, and psychological risk factor interactions in workers population employed in industrial units using an artificial neural networks approach.

METHODS

A total of 92 workers with LBP as the case group and 68 healthy workers as a control group were selected in various industrial units with similar occupational conditions. The demographic information and personal, occupational, and psychosocial factors of the participants were collected via interview, related questionnaires, consultation with occupational medicine, and also the Rapid Entire Body Assessment worksheet and National Aeronautics and Space Administration Task Load Index software. Then, 16 risk factors for LBP were used as input variables to develop the prediction model. Networks with various multilayered structures were developed using MATLAB.

RESULTS

The developed neural networks with 1 hidden layer and 26 neurons had the least error of classification in both training and testing phases. The mean of classification accuracy of the developed neural networks for the testing and training phase data were about 88% and 96%, respectively. In addition, the mean of classification accuracy of both training and testing data was 92%, indicating much better results compared with other methods.

CONCLUSION

It appears that the prediction model using the neural network approach is more accurate compared with other applied methods. Because occupational LBP is usually untreatable, the results of prediction may be suitable for developing preventive strategies and corrective interventions.

Reliability and measurement error of frontal and horizontal 3D spinal motion parameters in 219 patients with chronic low back pain

Background

In order for measurements to be clinically useful, data on psychometric conditions such as reliability should be available in the population for which the measurements are intended to be used. This study comprises a test-retest design separated by 7 to 14 days, and evaluates the intra and interrater reliability of regional frontal and horizontal spinal motion in 219 chronic LBP patients using the CA6000 Spine Motion Analyzer. In addition, it compares these results on the frontal and horizontal plane with previously published results on the sagittal plane. 219 individuals with chronic mechanical LBP, classified as either Quebec Task Force group 1, 2, 3 or 4 were included, and kinematics of the lumbar spine were sampled during standardized spinal lateral flexion and rotation motion using a 6-df instrumented spatial linkage system. Test-retest reliability and measurement error were evaluated using intraclass correlation coefficients ICC_(1,1) and Bland-Altman limits of agreement (LOAs).

Results

The reliability analysis based on the whole study sample showed ICC_(1,1) coefficients varying between 0.68 and 0.73 for the frontal plane and 0.33 and 0.49 for the horizontal plane. Relatively wide LOAs were observed for all parameters. Reliability measures in patient subgroups ICC_(1,1) ranged between 0.55 and 0.81 for the frontal plane and 0.28 and 0.69 for the horizontal plane. Greater ICC_(1,1) coefficients and smaller LOA were observed when patients were examined by the same examiner, had a stable pain level between tests, and were male. ROM measurements were more reliable in patients with a BMI higher than 30, and measurements on patients with LBP and leg pain showed higher reliability and smaller measurement error in all parameters except for the jerk index.

Conclusion

Frontal plane measurements obtained using the CA6000 Spine Motion Analyzer are sufficiently reliable to be used for group comparisons but not individual comparisons. Measurements in the horizontal plane can be used for neither group nor individual comparisons.

Artificial neural networks in neurosurgery

Artificial neural networks (ANNs) effectively analyze non-linear data sets. The aimed was A review of the relevant published articles that focused on the application of ANNs as a tool for assisting clinical decision-making in neurosurgery. A literature review of all full publications in English biomedical journals (1993-2013) was undertaken. The strategy included a combination of key words 'artificial neural networks', 'prognostic', 'brain', 'tumor tracking', 'head', 'tumor', 'spine', 'classification' and 'back pain' in the title and abstract of the manuscripts using the PubMed search engine. The major findings are summarized, with a focus on the application of ANNs for diagnostic and prognostic purposes. Finally, the future of ANNs in neurosurgery is explored. A total of 1093 citations were identified and screened. In all, 57 citations were found to be relevant. Of these, 50 articles were eligible for inclusion in this review. The synthesis of the data showed several applications of ANN in neurosurgery, including: (1) diagnosis and assessment of disease progression in low back pain, brain tumours and primary epilepsy; (2) enhancing clinically relevant information extraction from radiographic images, intracranial pressure processing, low back pain and real-time tumour tracking; (3) outcome prediction in epilepsy, brain metastases, lumbar spinal stenosis, lumbar disc herniation, childhood hydrocephalus, trauma mortality, and the occurrence of symptomatic cerebral vasospasm in patients with aneurysmal subarachnoid haemorrhage; (4) the use in the biomechanical assessments of spinal disease. ANNs can be effectively employed for diagnosis, prognosis and outcome prediction in neurosurgery.

Lumbar motion changes in chronic low back pain patients: a secondary analysis of data from a randomized clinical trial

BACKGROUND CONTEXT

Several therapies have been used in the treatment of chronic low back pain (LBP), including various exercise strategies and spinal manipulative therapy (SMT). A common belief is that spinal motion changes in particular ways in direct response to specific interventions, such as exercise or spinal manipulation.

PURPOSE

The purpose of this study was to assess changes in lumbar region motion for more than 12 weeks by evaluating four motion parameters in the sagittal plane and two in the horizontal plane in LBP patients treated with either exercise therapy or spinal manipulation.

STUDY DESIGN/SETTING

Secondary analysis of a subset of participants from a randomized clinical trial.

PATIENT SAMPLE

One hundred ninety-nine study participants with LBP of more than 6 weeks' duration who had spinal motion measures obtained before and after the period of intervention.

OUTCOME MEASURES

Lumbar region spinal kinematics sampled using a six-degree-of-freedom instrumented spatial linkage system.

METHODS

Trained therapists collected regional lumbar spinal motion data at baseline and 12 weeks of follow-up. The lumbar region spinal motion data were analyzed as a total cohort and relative to treatment modality (high dose, supervised low-tech trunk exercise, SMT, and a short course of home exercise and self-care advice). The study was supported by grants from Health Resources and Services Administration, Danish Agency for Science Technology and Innovation, Danish Chiropractors Research Foundation, and the University of Southern Denmark. No conflicts of interest reported.

RESULTS

For the cohort as a whole, lumbar region motion parameters were altered over the 12-week period, except for the jerk index parameter. The group receiving spinal manipulation changed significantly in all, and the exercise groups in half, the motion parameters included in the analysis. The spinal manipulation group changed to a smoother motion pattern (reduced jerk index), whereas the exercise groups did not.

CONCLUSION

This study provides evidence that spinal motion changes can occur in chronic LBP patients over a 12-week period and that these changes are associated with the type of treatment.

Developing functional workspace for the movement of trunk circumduction in healthy young subjects: a reliability study

Background

The lumbar range of motion has traditionally been used to assess disability in patients with low back disorders. Controversy exists about how movement ranges in static positions or in a single straight plane is related to the functional status of the patients. The trunk circumduction, as the result of neuromuscular coordination, is the integrated movements from three dimensions. The functional workspace stands for the volume of movement configuration from the trunk circumduction and represents all possible positions in three dimensions. By using single quantitative value, the functional workspace substitutes the complicated joint linear or angular motions. The aim of this study is to develop the functional workspace of the trunk circumduction (FWTC) considering possible functional positions in three dimensional planes. The reliability of the trunk circumduction is examined.

Methods

Test-retest reliability was performed with 18 healthy young subjects. A three-dimensional (3-D) Motion Analysis System was used to record the trunk circumduction. The FWTC was defined and calculated based on the volume of the cone that was formed as the resultant scanned area of markers, multiplied by the length of the body segment. The statistical analysis of correlation was performed to describe the relation of maximal displacements of trunk circumduction and straight planes: sagittal and coronal.

Results

The results of this study indicate that the movement of trunk circumduction measured by motion analysis instruments is a reliable tool. The ICC value is 0.90-0.96, and the means and standard deviations of the normalized workspace are: C7 0.425 (0.1162); L1 0.843 (0.2965); and knee 0.014 (0.0106). Little correlations between the maximal displacement of trunk circumduction and that of straight planes are shown and therefore suggest different movement patterns exist.

Conclusions

This study demonstrates high statistical reliability for the FWTC, which is important for the potential development as the functional assessment technique. The FWTC provides a single integrated value to represent angular and linear measurements of different joints and planes. Future study is expected to carry out the FWTC to evaluate the amount of workspace for the functional status of patients with low back injuries or patients with spinal surgery.

Pain-related fear is associated with avoidance of spinal motion during recovery from low back pain

STUDY DESIGN

A longitudinal assessment of the association between pain-related fear and joint motions in 36 participants with subacute low back pain.

OBJECTIVES

To determine how the psychologic construct of pain-related fear maps to motor behavior in standardized reaching tasks.

SUMMARY OF BACKGROUND DATA

Pain-related fear is a significant predictor of avoidance behavior and occupational disability in individuals with low back pain. However, it is not currently known how pain-related fear maps to motor behavior.

METHODS

Participants with an episode of subacute LBP were tested at 3, 6, and 12 weeks following pain onset. Participants performed reaching tasks at comfortable and fast-paced movement speeds to 3 targets (high, middle, low) located in a midsagittal plane. Three-dimensional joint motions of the thoracic spine, lumbar spine, and hip were recorded using an electromagnetic tracking device. Group differences in joint excursions were assessed using 2 groups (high pain-related fear, low pain-related fear) x 3 times (3, 6, 12 weeks) x 2 movement speeds (comfortable, fast paced) x 3 target heights (high, middle, low) MANOVAs.

RESULTS

Individuals with high pain-related fear had smaller excursions of the lumbar spine for reaches to all targets at 3 and 6 weeks, but not at 12 weeks following pain onset.

CONCLUSION

Individuals with high pain-related fear adopt alternative movement strategies and avoid motion of the lumbar spine when performing a common reaching movement. Identifying how pain-related fear maps to actual motor behavior (i.e., alternative movement strategies) is a crucial first step in determining how pain-related fear and motor behavior interact to promote or delay recovery from acute low back pain.

Reliability of the Assessment of Lumbar Range of Motion and Maximal Isometric Strength

OBJECTIVE

To examine the interobserver reliability and intrasubject variability of the assessment of lumbar range of motion (ROM) and maximal isometric strength in asymptomatic subjects by using commercially available equipment.

DESIGN

A cross-sectional repeated-measures design.

SETTING

Ambulatory care in a university hospital.

PARTICIPANTS

Convenience sample of 61 asymptomatic healthy subjects aged 20 to 55 years.

INTERVENTIONS

Not applicable.

MAIN OUTCOME MEASURES

Six movements of the lumbar spine were assessed with commercially available equipment. Both the ROM and the maximal isometric strength for flexion, extension, lateroflexion, and rotation of the lumbar spine were assessed by 2 investigators who were blinded to the outcome of the assessment performed by their colleague.

RESULTS

The intraclass correlation coefficient (ICC) was above .95 for all the strength measurements. For the assessment of the ROM of the lumbar spine, the ICC varied between .77 and .94. There was a significant intrasubject variability for 8 of 12 measurements.

CONCLUSIONS

The interobserver reliability is excellent for the measurement of the maximal isometric strength and good for the assessment of the ROM of the lumbar spine. There is a significant intrasubject variability, which requires the use of the mean or the best value of different trials.

Applications of artificial neural nets in clinical biomechanics

The purpose of this article is to provide an overview of current applications of artificial neural networks in the area of clinical biomechanics. The body of literature on artificial neural networks grew intractably vast during the last 15 years. Conventional statistical models may present certain limitations that can be overcome by neural networks. Artificial neural networks in general are introduced, some limitations, and some proven benefits are discussed.

Classification of neck movement patterns related to whiplash-associated disorders using neural networks

This paper presents a new method for classification of neck movement patterns related to Whiplash-associated disorders (WAD) using a resilient backpropagation neural network (BPNN). WAD are a common diagnosis after neck trauma, typically caused by rear-end car accidents. Since physical injuries seldom are found with present imaging techniques, the diagnosis can be difficult to make. The active range of the neck is often visually inspected in patients with neck pain, but this is a subjective measure, and a more objective decision support system, that gives a reliable and more detailed analysis of neck movement pattern, is needed. The objective of this study was to evaluate the predictive ability of a BPNN, using neck movement variables as input. Three-dimensional (3-D) neck movement data from 59 subjects with WAD and 56 control subjects were collected with a ProReflex system. Rotation angle and angle velocity were calculated using the instantaneous helical axis method and motion variables were extracted. A principal component analysis was performed in order to reduce data and improve the BPNN performance. BPNNs with six hidden nodes had a predictivity of 0.89, a sensitivity of 0.90 and a specificity of 0.88, which are very promising results. This shows that neck movement analysis combined with a neural network could build the basis of a decision support system for classifying suspected WAD, even though further evaluation of the method is needed.

Biomechanical assessments of lumbar spinal function. how low back pain sufferers differ from normals. implications for outcome measures research. part i: kinematic assessments of lumbar function

OBJECTIVE

To review new and advanced biomechanical assessment techniques for the lumbar spine and illustrate the differences in lumbar function in patients with low back pain and asymtomatic subjects.

DATA SOURCES

The biomedical literature was searched for research and reviews on spinal kinematic differences between low back pain subjects and healthy controls. A data search for articles indexed on MEDLINE until April 2002 was performed.

RESULTS

Kinematic measurements of lumbar function were categorized into 3 areas where low back patients may differ from normals: (1) end range of motion during simple movements; (2) higher order kinematics (displacement, velocity, and acceleration) during complex movement tasks; and (3) spinal proprioception. The assessment of higher order kinematics during complex movement tasks is the most highly researched and the most successful in describing differences between the populations. The use of simple end range of motion appears questionable, while assessing spinal proprioception is the least researched, yet shows potential in highlighting differences between low back sufferers and asymptomatics.

CONCLUSION

Current kinematic biomechanical assessment techniques are capable of identifying functional differences between low back pain populations and controls. The use and validity of the majority of these techniques as outcome measures are currently unknown, yet may be valuable in generating functional diagnoses, evaluating the mechanisms of current therapies, and prescribing specific rehabilitation programs.

A pilot study for a randomized clinical trial assessing chiropractic care, medical care, and self-care education for acute and subacute neck pain patients

OBJECTIVE

To conduct a pilot study in preparation for a full-scale randomized clinical trial assessing conservative treatments for acute and subacute neck pain. Study design Prospective, randomized pilot study.

SETTING

Primary contact chiropractic and medical clinics.

PATIENTS

Ages 21 to 65 with current episode of neck pain less than 12 weeks in duration. Outcome measures Patient self-report questionnaires and cervical spine motion were assessed at baseline and 3 and 12 weeks post-randomization.

INTERVENTIONS

Chiropractic spinal manipulation, prescription medications, and self-care education.

RESULTS

Recruitment took place over a 1-month period. Twenty-eight patients were randomized to treatment, and 1 patient (medical care group) refused their treatment assignment and was lost to further follow-up. Twenty-three patients were either "very satisfied" or "completely satisfied" with the care they received in the study. More than half the patients reported 75% or 100% improvement (n = 17). No between-group comparisons were planned or performed due to the small sample size.

CONCLUSION

Recruitment of patients appears feasible for a full-scale randomized clinical trial evaluating chiropractic spinal manipulation, medical care, and self-care education for acute and subacute neck pain. Patient and provider compliance with study protocols was excellent, and the pilot study allowed us to further develop and optimize our data collection processes. Although pilot studies such as these require substantial time, money, and effort, they provide valuable information for future research efforts.

Relationship between pain and vertebral motion in chronic low-back pain subjects

OBJECTIVES

To investigate the relationship between intervertebral motion, intravertebral deformation and pain in chronic low-back pain patients.

DESIGN

This study measured vertebral motion of the lumbar spine and associated pain in a select group of chronic low-back pain patients as they performed a standard battery of motions in all planes.

BACKGROUND

Numerous studies have demonstrated that individuals with low-back pain have impaired spinal motion, yet few studies have examined the specific relationship between pain and motion parameters. Although it is accepted that the pain in mechanical low-back patients is due to specific spinal motions, no studies have related specific motions to pain measures.

METHODS

Percutaneous intra-pedicle screws were placed into the right and left L4 (or L5) and S1 segments of nine chronic low-back pain patients. The external fixator frame was removed following the clinical external fixation test. The 3D locations of the pedicle screws and the level of pain were recorded as the subjects performed a battery of motions. The relationship between the pain and motion parameters was assessed using linear discriminant analysis and neural network models.

RESULTS

The neural network model showed a strong relationship between observed and predicted pain (R(2)=0.997). The discriminant analysis showed a weak relationship (R(2)=0.5).

CONCLUSIONS

Vertebral motion parameters are strongly predictive of pain in this select group of chronic low-back pain patients. The nature of the relationship is nonlinear and involves interactions; neural networks are able to effectively describe these relationships.

RELEVANCE

Specific patterns of intervertebral motion and intravertebral deformation result in pain in chronic low-back pain patients. This substantiates the mechanical back pain aetiology.

Pelvis-thorax coordination in the transverse plane during walking in persons with nonspecific low back pain

STUDY DESIGN

Transverse pelvis and thorax rotations were studied during walking in 39 patients with nonspecific low back pain and 19 healthy participants.

OBJECTIVES

To gain insight into the consequences of low back pain for gait and to identify clinically useful measures for characterizing the quality of walking in patients with low back pain.

SUMMARY OF BACKGROUND DATA

Gait studies in patients with low back pain have reported a decrease in walking velocity. In normal gait, in-phase pelvis-thorax coordination (synchronicity) evolves toward antiphase coordination (counterrotation) as walking velocity increases. This study examined the effect of walking velocity on pelvis and thorax rotations in patients with low back pain.

METHODS

Amplitudes of pelvis and thorax rotations were calculated, and spectral analyses were performed. Pelvis-thorax coordination was characterized in terms of relative Fourier phase, and coupling strength was assessed by means of cross-spectral analysis.

RESULTS

In comparison with healthy participants, relative Fourier phase was significantly smaller in low back pain patients for walking velocities of 3.8 km/h and higher, whereas coupling strength was significantly higher for velocities from 1.4 to 3.0 km/h. No significant group differences were found in amplitude or spectral content of individual pelvis and thorax rotations.

CONCLUSION

In comparison with healthy participants, the gait of patients with low back pain was characterized by a more rigid, less flexible pelvis-thorax coordination in the absence of significant differences in the kinematics of the component rotations. This result suggests that coordination measures are more adequate in assessing quality of walking in patients with low back pain than are kinematic measures pertaining to the individual segment rotations, and that conservative therapy should use methods aimed at improving intersegmental coordination.

Artificial neural network assessment of substitutive pharmacological treatments in hospitalised intravenous drug users

Artificial neural networks (ANNs) provide better solutions than linear discriminant analysis (LDA) to problems of classification and estimation involving a large number of non-homogeneous (categorical and metric) variables. In this study, we compared the ability of traditional LDA and a feed-forward back-propagation (FF-BP) ANN with self-momentum to predict pharmacological treatments received by intravenous drug users (IDUs) hospitalised for coexisting medical illness. When medical staff considered detoxification appropriate they usually suggested methadone (MET) and (or) benzodiazepines (BDZ). Given four different treatment options (MET, BDZ, MET+BDZ, no treatment) as dependent variables and 38 independent variables, the FF-BP ANN provided the best prediction of the consultant's decision (overall accuracy: 62.7%). It achieved the highest level of predictive accuracy for the BDZ option (90.5%), the lowest for no treatment (29.6), often misclassifying no treatment as BDZ. The LDA yielded a lower mean accuracy (50.3%). When the untreated group was excluded, ANN improved its absolute recognition rate by only 1.2% and the BDZ group remained the best predicted. In contrast, LDA improved its absolute recognition rate from 50.3 to 58.9%, maximum 65.7% for the BDZ group. In conclusion, the FF-BP ANN was more accurate than the statistical model (discriminant analysis) in predicting the pharmacological treatment of IDUs.

Symmetry and linearity of trunk function in subjects with non-specific low back pain

Objective. To investigate the effect of a standard rehabilitation program on the symmetry of trunk kinematics in subjects with non-specific low back pain.Design. Assessing lumbar spine kinematics in the cardinal planes using dynamometry.Background. Previous evaluations of trunk symmetry focussed more on anatomical rather than functional symmetry. Correlation of functional symmetry with low back pain was occasionally performed, but only for base line evaluations. To the best of the author's knowledge, there is no study examining the effect of exercise on the functional symmetry of the trunk, especially in non-specific low back pain subjects.Methods. Two groups of young male subjects whose working conditions incurred long daily hours of sitting and standing participated in the study. Muscles of the lumbar spine were initially evaluated in the cardinal planes using dynamometry. The same parameters (maximum isometric torque, dynamic torque, angular velocity and range of motion) were then repeatedly measured throughout a standardized strength protocol lasting for 12 sessions over a four-week period.Results. As pain gradually disappeared over a one-month period of rehabilitation, certain factors of the trunk kinematics exhibited convergence towards perfect symmetry while others showed oscillations. Moreover, global right-left data for certain trunk kinematics exhibited near-perfect linear relationship.Conclusion. Functional symmetry of the trunk in the coronal and transverse planes can be used to assess the progress of rehabilitation programs of non-specific low back pain subjects. RelevanceRehabilitation programs focus on the progress of directly measured trunk kinematics, which do not always exhibit monotonic behavior. This paper alludes to the importance of tracking symmetry of trunk kinematics as it may help clinicians modify the strengthening protocols in order to achieve more rapid relief from back pain.

The effects of motion on trunk biomechanics

OBJECTIVE

To review the literature that evaluates the influence of trunk motion on trunk strength and structural loading.

BACKGROUND

In recent years, trunk dynamics have been identified as potential risk factors for developing low-back disorders. Consequently, a better understanding of the underlying mechanisms involved in trunk motion is needed.

METHODS

This review summarizes the results of 53 studies that have evaluated trunk motion and its impact on several biomechanical outcome measures. The biomechanical measures consisted of trunk strength, intra-abdominal pressure, muscle activity, imposed trunk moments, and spinal loads. Each of these biomechanical measures was discussed in relation to the existing knowledge within each plane of motion (extension, flexion, lateral flexion, twisting, and asymmetric extension).

RESULTS

Trunk strength was drastically reduced as dynamic motion increased, and males were impacted more than females. Intra-abdominal pressure seemed to only be affected by trunk dynamics at high levels of force. Trunk moments were found to increase monotonically with increased trunk motion. Both agonistic and antagonistic muscle activities were greater as dynamic characteristics increased. As a result, the three-dimensional spinal loads increase significantly for dynamic exertions as compared to isometric conditions.

CONCLUSIONS

Trunk motion has a dramatic affect on the muscle coactivity, which seems to be the underlying source for the decrease strength capability as well as the increased muscle force, IAP, and spinal loads. This review suggests that the ability of the individual to perform a task "safely" might be significantly compromised by the muscle coactivity that accompanies dynamic exertions. It is also important to consider various workplace and individual factors when attempting to reduce the impact of trunk motions during dynamic exertions. Relevance This review provides insight as to why trunk motions are important risk factors to consider when attempting to control low-back disorders in the workplace. It is apparent that trunk motion increases the risk of low-back disorders. To better control low-back disorders in industry, more comprehensive knowledge about the impact of trunk motion is needed. A better understanding of muscle coactivity may ultimately lead to reducing the risk associated with dynamic exertions.

Disability measurement in persons with back pain: a validity study of spinal range of motion and velocity

OBJECTIVE

To evaluate the criterion validity and responsiveness to change of spine kinematic variables to assess disability in patients with low back pain.

DESIGN

Blinded comparison between spine kinematic variables, Oswestry disability questionnaire scores, and work status.

SETTING

Multidisciplinary occupational rehabilitation clinic of a university hospital.

PATIENTS

Population-based cohort of 111 patients with subacute work-related back pain who were absent from regular work for more than 4 weeks because of back pain.

INTERVENTIONS

This study was part of a population-based randomized clinical trial. Patients were randomized to 4 different methods of management: usual care, rehabilitation, ergonomics, or rehabilitation and ergonomics.

MAIN OUTCOME MEASURES

Oswestry disability questionnaire, kinematic analysis of the spine during flexion and extension of the trunk, and work status were collected at weeks 4, 12, 24, and 52 after the back accident.

RESULTS

Kinematic variables were poorly to moderately related to work status and Oswestry questionnaire scores. Kinematic variables were also unresponsive to change in work status and Oswestry questionnaire scores over time.

CONCLUSION

Spine kinematics during flexion and extension of the trunk do not appear to be a valid measure of disability in patients with subacute and chronic back pain.

Ruptured abdominal aortic aneurysm: a novel method of outcome prediction using neural network technology

BACKGROUND

reported survival following emergency surgery for ruptured abdominal aortic aneurysm (RAAA) varies widely between institutions. This is largely attributable to differences in case mix. The aim of this study was to identify and evaluate a set of prognostic variables that would accurately predict outcome for individual patients from perioperative indices.

METHODS

perioperative factors associated with subsequent mortality at our institution were identified by retrospective review of 102 consecutive operations for RAAA over a 7-year period (January 1990 to January 1997). Logistic regression analysis was used to select the most significant variables associated with subsequent mortality. These were used to construct, train, and validate a neural network designed to predict survival from surgery in individual cases on a prospective basis.

RESULTS

the 30-day mortality rate was 53%. Multivariate analysis identified four highly significant independent predictors of mortality; preoperative hypotension, intraperitoneal rupture, preoperative coagulopathy, and preoperative cardiac arrest. Using these inputs, the neural network correctly predicted outcome in 82.5% of individual cases.

CONCLUSION

a neural network based on just four perioperative variables can accurately predict outcome of RAAA. Prognostic variables should be reported in studies as a measure of the effect of case mix on survival data. Neural networks have potential to aid decision-making relating to outcome for individual cases.