Using hierarchical unsupervised learning to integrate and reduce multi-level and multi-paraspinal muscle MRI data in relation to low back pain.

PURPOSE: The paraspinal muscles (PSM) are a key feature potentially related to low back pain (LBP), and their structure and composition can be quantified using MRI. Most commonly, quantifying PSM measures across individual muscles and individual spinal levels renders numerous separate metrics that are analyzed in isolation. However, comprehensive multivariate approaches would be more appropriate for analyzing the PSM within an individual. To establish and test these methods, we hypothesized that multivariate summaries of PSM MRI measures would associate with the presence of LBP symptoms (i.e., pain intensity). METHODS: We applied hierarchical multiple factor analysis (hMFA), an unsupervised integrative method, to clinical PSM MRI data from unique cohort datasets including a longitudinal cohort of astronauts with pre- and post-spaceflight data and a cohort of chronic LBP subjects and asymptomatic controls. Three specific use cases were investigated: (1) predicting longitudinal changes in pain using combinations of baseline PSM measures; (2) integrating baseline and post-spaceflight MRI to assess longitudinal change in PSM and how it relates to pain; and (3) integrating PSM quality and adjacent spinal pathology between LBP patients and controls. RESULTS: Overall, we found distinct complex relationships with pain intensity between particular muscles and spinal levels. Subjects with high asymmetry between left and right lean muscle composition and differences between spinal segments PSM quality and structure are more likely to increase in pain reported outcome after prolonged time in microgravity. Moreover, changes in PSM quality and structure between pre and post-spaceflight relate to increase in pain after prolonged microgravity. Finally, we show how unsupervised hMFA recapitulates previous research on the association of CEP damage and LBP diagnostic. CONCLUSION: Our analysis considers the spine as a multi-segmental unit as opposed to a series of discrete and isolated spine segments. Integrative and multivariate approaches can be used to distill large and complex imaging datasets thereby improving the clinical utility of MRI-based biomarkers, and providing metrics for further analytical goals, including phenotyping.

Biomechanical changes in the lumbar spine following spaceflight and factors associated with postspaceflight disc herniation.

BACKGROUND CONTEXT: For chronic low back pain, the causal mechanisms between pathological features from imaging and patient symptoms are unclear. For instance, disc herniations can often be present without symptoms. There remains a need for improved knowledge of the pathophysiological mechanisms that explore spinal tissue damage and clinical manifestations of pain and disability. Spaceflight and astronaut health provides a rare opportunity to study potential low back pain mechanisms longitudinally. Spaceflight disrupts diurnal loading on the spine and several lines of evidence indicate that astronauts are at a heightened risk for low back pain and disc herniation following spaceflight. PURPOSE: To examine the relationship between prolonged exposure to microgravity and the elevated incidence of postflight disc herniation, we conducted a longitudinal study to track the spinal health of twelve NASA astronauts before and after approximately 6 months in space. We hypothesize that the incidence of postflight disc herniation and low back complaints associates with spaceflight-included muscle atrophy and pre-existing spinal pathology. STUDY DESIGN: This is a prospective longitudinal study. PATIENT SAMPLE: Our sample included a cohort of twelve astronaut crewmembers. OUTCOME MEASURES: From 3T MRI, we quantified disc water content (ms), disc degeneration (Pfirrmann grade), vertebral endplate irregularities, facet arthropathy and/ fluid, high intensity zones, disc herniation, multifidus total cross-sectional area (cm2), multifidus lean muscle cross-sectional area (cm2), and muscle quality/composition (%). From quantitative fluoroscopy we quantified, maximum flexion-extension ROM (°), maximum lateral bending ROM (°), and maximum translation (%). Lastly, patient outcomes and clinical notes were used for identifying postflight symptoms associated with disc herniations from 3T MRI. METHODS: Advanced imaging data from 3T MRI were collected at three separate time points in relation to spending six months in space: (1) within a year before launch ("pre-flight"), (2) within a week after return to Earth ("post-flight"), and (3) between 1 and 2 months after return to Earth ("recovery"). Fluoroscopy of segmental kinematics was collected at preflight and postflight timepoints. We assessed the effect of spaceflight and postflight recovery on longitudinal changes in spinal structure and function, as well as differences between crew members who did and did not present a symptomatic disc herniation following spaceflight. RESULTS: Half of our astronauts (n=6) experienced new symptoms associated with a new or previously asymptomatic lumbar disc protrusion or extrusion following spaceflight. We observed decreased multifidus muscle quality following spaceflight in the lower lumbar spine, with a reduced percentage of lean muscle at L4L5 (-6.2%, p=.009) and L5S1 (-7.0%, p=.006) associated with the incidence of new disc herniation. Additionally, we observed reduced lumbar segment flexion-extension ROM for L2L3 (-17.2%, p=.006) and L3L4 (-20.5%, p=.02) following spaceflight, and furthermore that reduced ROM among the upper three lumbar segments (-24.1%, p=.01) associated with the incidence of disc herniation. Existing endplate pathology was most prevalent in the upper lumbar spine and associated with reduced segmental ROM (-20.5%, p=.02). CONCLUSIONS: In conclusion from a 10-year study investigating the effects of spaceflight on the lumbar spine and risk for disc herniation, we found the incidence of lumbar disc herniation following spaceflight associates with compromised multifidus muscle quality and spinal segment kinematics, as well as pre-existing spinal endplate irregularities. These findings suggest differential effects of spinal stiffness and muscle loss in the upper versus lower lumbar spine regions that may specifically provoke risk for symptomatic disc herniation in the lower lumbar spine following spaceflight. Results from this study provide a unique longitudinal assessment of mechanisms and possible risk factors for developing disc herniations and related low back pain. Furthermore, these findings will help inform physiologic countermeasures to maintain spinal health in astronauts during long-duration missions in space.

A Monte Carlo study of accelerator head scatter.

The production of off-focus x rays in the head of a 6 MV linac has been investigated using the EGS4 Monte Carlo code. The purpose of the study was to identify the sources of off-focus radiation and the relative contribution for each source. Even though a particular energy and linac were modeled, the broad conclusions are expected to be general since the effects of head scatter are similar for most conventional head designs, regardless of manufacturer, energy, and model. The head components that were modeled include the exit window of the accelerating structure, target, beam stopper, flattening filter, monitor chamber, primary and secondary collimators, and air. Monoenergetic 6 MeV electrons were followed through the exit window, target, and beam stopper until all energy was expended. Primary- and higher-order x rays produced throughout the head were followed until they were either absorbed or passed through a plane at the isocenter. Sites of off-focus radiation were found to be diffusely distributed throughout the head, with the most intense sources being the primary collimator, flattening filter, and beam stopper. Data analysis shows that the collimator effect is determined primarily by the volume of the extended head-scatter source that is exposed to the point of measurement through the collimating system. The results of this study provide a rationale for developing extended source models to calculate the collimator factor for fields defined by arbitrary collimation. An additional advantage is an improvement in the agreement between measured and calculated isodose distributions.

Sources of electron contamination for the Clinac-35 25-MV photon beam.

A detailed Monte Carlo approach has been employed to investigate the sources of electron contamination for the 25-MV photon beam generated by Varian's Clinac-35. Three sources of contamination were examined: (a) the flattening filter and beam monitor chamber, (b) the fixed primary collimators downstream from the monitor chamber and the adjustable photon jaws, and (c) the air volume separating the treatment head from the observation point. Five source-to-surface distances (SSDs) were considered for a single field size, 28 cm in diameter at 80 cm SSD. It was found that for small SSDs (80-100 cm), the dominant sources of electron contamination were the flattening filter and the beam monitor chamber which accounted for 70% of the unwanted electrons. Thirteen percent of the remaining electrons originated in the downstream primary collimators and the photon jaws, and 17% were produced in air. At larger SSDs, the fraction of unwanted electrons originating in air increased. At 400 cm SSD, 61% of the contaminating electrons present in the beam were produced in air, 34% originated in the flattening filter and beam monitor chamber, and 5% were due to interactions in the fixed collimators downstream from the monitor chamber and the adjustable photon jaws. These calculated results are substantiated by recent experiments.

Investigation of buildup dose from electron contamination of clinical photon beams.

The contribution made by contaminating electrons present in a clinical photon beam to the buildup dose in a polystyrene phantom has been calculated and compared to measurements. A Monte Carlo technique was employed. The calculation was divided into two parts. First, the accelerator treatment head was simulated in detail using the EGS-PEGS electromagnetic shower code. Then, information obtained from these calculations was used to compute dose curves in a polystyrene phantom. Two cases were considered, one in which both electrons and photons were incident upon the phantom, and another in which electrons were eliminated from the incident beam. Results of these calculations agree with recent experimental findings. A decrease in buildup dose as well as a shift in dmax was observed when electrons were eliminated from the beam.