Machine Learning in Neuro-Oncology, Epilepsy, Alzheimer's Disease, and Schizophrenia.

Applications of machine learning (ML) in translational medicine include therapeutic drug creation, diagnostic development, surgical planning, outcome prediction, and intraoperative assistance. Opportunities in the neurosciences are rich given advancement in our understanding of the brain, expanding indications for intervention, and diagnostic challenges often characterized by multiple clinical and environmental factors. We present a review of ML in neuro-oncology, epilepsy, Alzheimer's disease, and schizophrenia to highlight recent progression in these field, optimizing machine learning capabilities in their current forms. Supervised learning models appear to be the most commonly incorporated algorithm models for machine learning across the reviewed neuroscience disciplines with primary aim of diagnosis. Accuracy ranges are high from 63% to 99% across all algorithms investigated. Machine learning contributions to neurosurgery, neurology, psychiatry, and the clinical and basic science neurosciences may enhance current medical best practices while also broadening our understanding of dynamic neural networks and the brain.

Clinical Trial Designs for Neuromodulation in Chronic Spinal Cord Injury Using Epidural Stimulation.

STUDY DESIGN: This is a narrative review focused on specific challenges related to adequate controls that arise in neuromodulation clinical trials involving perceptible stimulation and physiological effects of stimulation activation. OBJECTIVES: 1) To present the strengths and limitations of available clinical trial research designs for the testing of epidural stimulation to improve recovery after spinal cord injury. 2) To describe how studies can control for the placebo effects that arise due to surgical implantation, the physical presence of the battery, generator, control interfaces, and rehabilitative activity aimed to promote use-dependent plasticity. 3) To mitigate Hawthorne effects that may occur in clinical trials with intensive supervised participation, including rehabilitation. MATERIALS AND METHODS: Focused literature review of neuromodulation clinical trials with integration to the specific context of epidural stimulation for persons with chronic spinal cord injury. CONCLUSIONS: Standard of care control groups fail to control for the multiple effects of knowledge of having undergone surgical procedures, having implanted stimulation systems, and being observed in a clinical trial. The irreducible effects that have been identified as "placebo" require sham controls or comparison groups in which both are implanted with potentially active devices and undergo similar rehabilitative training.

Epidural stimulation for cardiovascular function increases lower limb lean mass in individuals with chronic motor complete spinal cord injury.

NEW FINDINGS: What is the central question of this study? Spinal cord injury results in paralysis and deleterious neuromuscular and autonomic adaptations. Lumbosacral epidural stimulation can modulate motor and/or autonomic functions. Does long-term epidural stimulation for normalizing cardiovascular function affect leg muscle properties? What is the main finding and its importance? Leg lean mass increased after long-term epidural stimulation for cardiovascular function, which was applied in the sitting position and did not activate the leg muscles. Leg muscle strength and fatigue resistance, assessed in a subgroup of individuals, also increased. These adaptations might support interventions for motor recovery and warrant further mechanistic investigation. ABSTRACT: Chronic motor complete spinal cord injury (SCI) results in paralysis and deleterious neuromuscular and autonomic adaptations. Paralysed muscles demonstrate atrophy, loss of force and increased fatigability. Also, SCI-induced autonomic impairment results in persistently low resting blood pressure and heart rate, among other features. We previously reported that spinal cord epidural stimulation (scES) optimized for cardiovascular (CV) function (CV-scES), which is applied in sitting position and does not activate the leg muscles, can maintain systolic blood pressure within a normotensive range during quiet sitting and during orthostatic stress. In the present study, dual-energy X-ray absorptiometry collected from six individuals with chronic clinically motor complete SCI demonstrated that 88 ± 11 sessions of CV-scES (7 days week-1 ; 2 h day-1 in four individuals and 5 h day-1 in two individuals) over a period of ∼6 months significantly increased lower limb lean mass (by 0.67 ± 0.39 kg or 9.4 ± 8.1%; P < 0.001). Additionally, muscle strength and fatigability data elicited by neuromuscular electrical stimulation in three of these individuals demonstrated a general increase (57 ± 117%) in maximal torque output (between 2 and 44 N m in 14 of the 17 muscle groups tested overall) and torque-time integral during intermittent, fatiguing contractions (63 ± 71%; between 7 and 230% in 16 of the 17 muscle groups tested overall). In contrast, whole-body mass and composition did not change significantly. In conclusion, long-term use of CV-scES can have a significant impact on lower limb muscle properties after chronic motor complete SCI.

Effects of Respiratory Training on Heart Rate Variability and Baroreflex Sensitivity in Individuals With Chronic Spinal Cord Injury.

OBJECTIVE: To evaluate the effects of pressure threshold respiratory training (RT) on heart rate variability and baroreflex sensitivity in persons with chronic spinal cord injury (SCI). DESIGN: Before-after intervention case-controlled clinical study. SETTING: SCI research center and outpatient rehabilitation unit. PARTICIPANTS: Participants (N=44) consisted of persons with chronic SCI ranging from C2 to T11 who participated in RT (n=24), and untrained control subjects with chronic SCI ranging from C2 to T9 (n=20). INTERVENTIONS: A total of 21±2 RT sessions performed 5 days a week during a 4-week period using a combination of pressure threshold inspiratory and expiratory devices. MAIN OUTCOME MEASURES: Forced vital capacity (FVC), forced expiratory volume in 1 second (FEV1), and beat-to-beat arterial blood pressure and heart rate changes during the 5-second-long maximum expiratory pressure maneuver (5s MEP) and the sit-up orthostatic stress test, acquired before and after the RT program. RESULTS: In contrast to the untrained controls, individuals in the RT group experienced significantly increased FVC and FEV1 (both P<.01) in association with improved quality of sleep, cough, and speech. Sympathetically (phase II) and parasympathetically (phase IV) mediated baroreflex sensitivity both significantly (P<.05) increased during the 5s MEP. During the orthostatic stress test, improved autonomic control over heart rate was associated with significantly increased sympathetic and parasympathetic modulation (low- and high-frequency change: P<.01 and P<.05, respectively). CONCLUSIONS: Inspiratory-expiratory pressure threshold RT is a promising technique to positively affect both respiratory and cardiovascular dysregulation observed in persons with chronic SCI.

Increased Prevalence of Blood Pressure Instability Over Twenty-Four Hours in Chronic Spinal Cord Injury.

Spinal cord injury (SCI) leads to cardiovascular dysregulation, including persistent low blood pressure (BP), orthostatic hypotension, and autonomic dysreflexia, leading to daily BP instability that may not be adequately recognized. We compared mean systolic BP, diastolic BP, and heart rate from awake and asleep measurements over a 24-h period among persons with chronic SCI (n = 33; 30 cervical injuries and three upper thoracic injuries), ambulatory/non-injured (Ambulatory-NI; n = 13), and non-injured (NI) in a wheelchair (n = 9). Stability of awake BP was evaluated by deviation of systolic BP from 115 mmHg and percent of systolic BP measurement within and outside of 90-140 mmHg. Variability over 24 h was compared using coefficient of variation and average real variability. Awake hyper- and hypotensive events (change in systolic BP ±20 mmHg from the median) were compared to symptoms reported by the participants corresponding to BP events. Participants with SCI had a lower percentage of awake systolic BP measurements within 90-140 mmHg than Ambulatory-NI and a greater deviation below 115 mmHg. Coefficient of variation and successive differences of awake systolic and diastolic BP were greater in SCI than Ambulatory-NI. Finally, all SCI participants had hyper- and/or hypotensive events and 88% experienced the BP events asymptomatically. In conclusion, participants with SCI had significantly greater BP instability compared with NI, with many hyper- and hypotensive events occurring without symptoms. Clinical management of BP instability, regardless of symptoms, should be a priority after SCI to reduce the risk of cardiovascular disease and improve quality of life.

Beneficial Cardiac Structural and Functional Adaptations After Lumbosacral Spinal Cord Epidural Stimulation and Task-Specific Interventions: A Pilot Study.

Cardiac myocyte atrophy and the resulting decreases to the left ventricular mass and dimensions are well documented in spinal cord injury. Therapeutic interventions that increase preload can increase the chamber size and improve the diastolic filling ratios; however, there are no data describing cardiac adaptation to chronic afterload increases. Research from our center has demonstrated that spinal cord epidural stimulation (scES) can normalize arterial blood pressure, so we decided to investigate the effects of scES on cardiac function using echocardiography. Four individuals with chronic, motor-complete cervical spinal cord injury were implanted with a stimulator over the lumbosacral enlargement. We assessed the cardiac structure and function at the following time points: (a) prior to implantation; (b) after scES targeted to increase systolic blood pressure; (c) after the addition of scES targeted to facilitate voluntary (i.e., with intent) movement of the trunk and lower extremities; and (d) after the addition of scES targeted to facilitate independent, overground standing. We found significant improvements to the cardiac structure (left ventricular mass = 10 ± 2 g, p < 0.001; internal dimension during diastole = 0.1 ± 0.04 cm, p < 0.05; internal dimension during systole = 0.06 ± 0.03 cm, p < 0.05; interventricular septum dimension = 0.04 ± 0.02 cm, p < 0.05), systolic function (ejection fraction = 1 ± 0.4%, p < 0.05; velocity time integral = 2 ± 0.4 cm, p < 0.001; stroke volume = 4.4 ± 1.5 ml, p < 0.01), and diastolic function (mitral valve deceleration time = -32 ± 11 ms, p < 0.05; mitral valve deceleration slope = 50 ± 25 cm s-1, p < 0.05; isovolumic relaxation time = -6 ± 1.9 ms, p < 0.05) with each subsequent scES intervention. Despite the pilot nature of this study, statistically significant improvements to the cardiac structure, systolic function, and diastolic function demonstrate that scES combined with task-specific interventions led to beneficial cardiac remodeling, which can reverse atrophic changes that result from spinal cord injury. Long-term improvements to cardiac function have implications for increased quality of life and improved cardiovascular health in individuals with spinal cord injury, decreasing the risk of cardiovascular morbidity and mortality.

Systolic and diastolic function in chronic spinal cord injury.

Individuals with spinal cord injury develop cardiovascular disease more than age-matched, non-injured cohorts. However, progression of systolic and diastolic dysfunction into cardiovascular disease after spinal cord injury is not well described. We sought to investigate the relationship between systolic and diastolic function in chronic spinal cord injury to describe how biological sex, level, severity, and duration of injury correlate with structural changes in the left ventricle. Individuals with chronic spinal cord injury participated in this study (n = 70). Registered diagnostic cardiac sonographers used cardiac ultrasound to measure dimensions, mass, and systolic and diastolic function of the left ventricle. We found no significant relationship to severity or duration of injury with left ventricle measurements, systolic function outcome, or diastolic function outcome. Moreover, nearly all outcomes measured were within the American Society of Echocardiography-defined healthy range. Similar to non-injured individuals, when indexed by body surface area (BSA) left ventricle mass [-14 (5) g/m2, p < .01], end diastolic volume [-6 (3) mL/m2, p < .05], and end systolic volume [-4 (1) mL/m2, p < .01] were significantly decreased in women compared with men. Likewise, diastolic function outcomes significantly worsened with age: E-wave velocity [-5 (2), p < .01], E/A ratio [-0.23 (0.08), p < .01], and e' velocity [lateral: -1.5 (0.3) cm/s, p < .001; septal: -0.9 (0.2), p < .001] decreased with age while A-wave velocity [5 (1) cm/s, p < .001] and isovolumic relaxation time [6 (3) ms, p < .05] increased with age. Women demonstrated significantly decreased cardiac size and volumes compared with men, but there was no biological relationship to dysfunction. Moreover, individuals were within the range of ASE-defined healthy values with no evidence of systolic or diastolic function and no meaningful relationship to level, severity, or duration of injury. Decreases to left ventricular dimensions and mass seen in spinal cord injury may result from adaptation rather than maladaptive myocardial remodeling, and increased incidence of cardiovascular disease may be related to modifiable risk factors.

Heart rate and blood pressure response improve the prediction of orthostatic cardiovascular dysregulation in persons with chronic spinal cord injury.

Unstable blood pressure after spinal cord injury (SCI) is not routinely examined but rather predicted by level and completeness of injury (i.e., American Spinal Injury Association Impairment Scale AIS classification). Our aim was to investigate hemodynamic response to a sit-up test in a large cohort of individuals with chronic SCI to better understand cardiovascular function in this population. Continuous blood pressure and ECG were recorded from individuals with SCI (n = 159) and non-injured individuals (n = 48). We found orthostatic hypotension occurred within each level and AIS classification (n = 36). Moreover, 45 individuals with chronic SCI experienced a drop in blood pressure that did not meet the criteria for orthostatic hypotension, but was accompanied by dramatic increases in heart rate, reflecting orthostatic intolerance. A cluster analysis of hemodynamic response to a seated position identified eight distinct patterns of interaction between blood pressure and heart rate during orthostatic stress indicating varied autonomic responses. Algorithmic cluster analysis of heart rate and blood pressure is more sensitive to diagnosing orthostatic cardiovascular dysregulation. This indicates blood pressure instability cannot be predicted by level and completeness of SCI, and the consensus statement definition of orthostatic hypotension is insufficient to characterize the variability of blood pressure and heart rate responses during orthostatic stress. Both blood pressure and heart rate responses are needed to characterize autonomic function after SCI.

Prolonged Targeted Cardiovascular Epidural Stimulation Improves Immunological Molecular Profile: A Case Report in Chronic Severe Spinal Cord Injury.

In individuals with severe spinal cord injury (SCI), the autonomic nervous system (ANS) is affected leading to cardiovascular deficits, which include significant blood pressure instability, with the prevalence of systemic hypotension and orthostatic intolerance resulting in an increased risk of stroke. Additionally, persons with SCI rostral to thoracic vertebral level 5 (T5), where sympathetic nervous system fibers exit the spinal cord and innervate the immune system, have clinically significant systemic inflammation and increased infection risk. Our recent studies show that lumbosacral spinal cord epidural stimulation (scES), applied at the lumbosacral level using targeted configurations that promote cardiovascular stability (CV-scES), can safely and effectively normalize blood pressure in persons with chronic SCI. Herein we present a case report in a female (age 27 years) with chronic clinically motor complete cervical SCI demonstrating that 97-sessions of CV-scES, which increased systemic blood pressure, improved orthostatic tolerance in association with increased cerebral blood flow velocity in the middle cerebral artery, also promoted positive immunological changes in whole-blood gene expression. Specifically, there was evidence of the down-regulation of inflammatory pathways and the up-regulation of adaptative immune pathways. The findings of this case report suggest that the autonomic effects of epidural stimulation, targeted to promote cardiovascular homeostasis, also improves immune system function, which has a significant benefit to long-term cardiovascular and immunologic health in individuals with long-standing SCI. Clinical Trial Registration: www.ClinicalTrials.gov, identifier NCT02307565.

Epidural Spinal Cord Stimulation of Lumbosacral Networks Modulates Arterial Blood Pressure in Individuals With Spinal Cord Injury-Induced Cardiovascular Deficits.

Disruption of motor and autonomic pathways induced by spinal cord injury (SCI) often leads to persistent low arterial blood pressure and orthostatic intolerance. Spinal cord epidural stimulation (scES) has been shown to enable independent standing and voluntary movement in individuals with clinically motor complete SCI. In this study, we addressed whether scES configured to activate motor lumbosacral networks can also modulate arterial blood pressure by assessing continuous, beat-by-beat blood pressure and lower extremity electromyography during supine and standing in seven individuals with C5-T4 SCI. In three research participants with arterial hypotension, orthostatic intolerance, and low levels of circulating catecholamines (group 1), scES applied while supine and standing resulted in increased arterial blood pressure. In four research participants without evidence of arterial hypotension or orthostatic intolerance and normative circulating catecholamines (group 2), scES did not induce significant increases in arterial blood pressure. During scES, there were no significant differences in electromyographic (EMG) activity between group 1 and group 2. In group 1, during standing assisted by scES, blood pressure was maintained at 119/72 ± 7/14 mmHg (mean ± SD) compared with 70/45 ± 5/7 mmHg without scES. In group 2 there were no arterial blood pressure changes during standing with or without scES. These findings demonstrate that scES configured to facilitate motor function can acutely increase arterial blood pressure in individuals with SCI-induced cardiovascular deficits.

Baroreceptor reflex during forced expiratory maneuvers in individuals with chronic spinal cord injury.

Pulmonary and cardiovascular dysfunctions are leading causes of morbidity and mortality in patients with chronic Spinal Cord Injury (SCI). Impaired respiratory motor function and decreased Baroreflex Sensitivity (BS) are predictors for the development of cardiopulmonary disease. This observational case-controlled clinical study was undertaken to investigate if respiratory motor control deficits in individuals with SCI affect their ability to perform the Valsalva maneuver, and to determine if a sustained Maximum Expiratory Pressure (MEP) effort can serve as an acceptable maneuver for determination of the BS in the event that the Valsalva maneuver cannot be performed. The BS outcomes (ms/mmHg) were obtained using continuous beat-to-beat arterial blood pressure (BP) and heart rate (HR) recordings during Valsalva or MEP maneuvers in thirty nine individuals with chronic C3-T12 SCI. Twenty one participants (54%) reported signs of intolerance during the Valsalva maneuver and only 15 individuals (39%) were able to complete this task. Cervical level of injury was a significant risk factor (p=0.001) for failing to complete the Valsalva maneuver, and motor-complete injury was a significant risk factor for symptoms of intolerance (p=0.04). Twenty eight participants (72%) were able to perform the MEP maneuver; the other 11 participants failed to exceed the standard airway pressure threshold of 27cm H2O. Neither level nor completeness of injury were significant risk factors for failure of MEP maneuver. When the required airway pressure was sustained, there were no significant differences between BS outcomes obtained during Valsalva and MEP maneuvers. The results of this study indicate that individuals with high-level and motor-complete SCI are at increased risk of not completing the Valsalva maneuver and that baroreflex-mediated responses can be evaluated by using sustained MEP maneuver when the Valsalva maneuver cannot be performed.