Gaussian Processes for Hearing Threshold Estimation Using Auditory Brainstem Responses.

The Auditory Brainstem Response (ABR) plays an important role in diagnosing and managing hearing loss, but can be challenging and time-consuming to measure. Test times are especially long when multiple ABR measurements are needed, e.g., when estimating hearing threshold at a range of frequencies. While many detection methods have been developed to reduce ABR test times, the majority were designed to detect the ABR at a single stimulus level and do not consider correlations in ABR waveforms across levels. These correlations hold valuable information, and can be exploited for more efficient hearing threshold estimation. This was achieved in the current work using a Gaussian Process (GP), i.e., a Bayesian approach for non-linear regression. The function to estimate with the GP was the ABR's amplitude across stimulus levels, from which hearing threshold was ultimately inferred. Active learning rules were also designed to automatically adjust the stimulus level and efficiently locate hearing threshold. Simulation results show test time reductions of up to  âˆ¼ 50% for the GP compared to a sequentially applied Hotelling's T2 test, which does not consider correlations across ABR waveforms. A case study was also included to briefly assess the GP approach in ABR data from an adult volunteer.

Controlling test specificity for auditory evoked response detection using a frequency domain bootstrap.

BACKGROUND: Statistical detection methods are routinely used to automate auditory evoked response (AER) detection and assist clinicians with AER measurements. However, many of these methods are built around statistical assumptions that can be violated for AER data, potentially resulting in reduced or unpredictable test performances. This study explores a frequency domain bootstrap (FDB) and some FDB modifications to preserve test performance in serially correlated non-stationary data. METHOD: The FDB aims to generate many surrogate recordings, all with similar serial correlation as the original recording being analysed. Analysing the surrogates with the detection method then gives a distribution of values that can be used for inference. A potential limitation of the conventional FDB is the assumption of stationary data with a smooth power spectral density (PSD) function, which is addressed through two modifications. COMPARISONS WITH EXISTING METHODS: The FDB was compared to a conventional parametric approach and two modified FDB approaches that aim to account for heteroskedasticity and non-smooth PSD functions. Hotelling's T2(HT2) test applied to auditory brainstem responses was the test case. RESULTS: When using conventional HT2, false-positive rates deviated significantly from the nominal alpha-levels due to serial correlation. The false-positive rates of the modified FDB were consistently closer to the nominal alpha-levels, especially when data was strongly heteroskedastic or the underlying PSD function was not smooth due to e.g. power lines noise. CONCLUSION: The FDB and its modifications provide accurate, recording-dependent approximations of null distributions, and an improved control of false-positive rates relative to parametric inference for auditory brainstem response detection.

A group sequential test for ABR detection.

Objective: To detect the auditory brainstem response (ABR) automatically using an innovative sequentially applied Hotelling's T 2 test, with the overall goal of optimising test time whilst controlling the false-positive rate (FPR). Design: The stage-wise critical decision boundaries for accepting or rejecting the null hypothesis were found using a new approach called the Convolutional Group Sequential Test (CGST). Specificity, sensitivity, and test time were evaluated using simulations and subject recorded data. Study sample: Data consists of click-evoked ABR threshold series from 12 normal hearing adults, and recordings of EEG background activity from 17 normal hearing adults. Results: Reductions in mean test time of up to 40-45% were observed for the sequential test, relative to a conventional "single shot" test where the statistical test is applied to the data just once. To obtain these results, it will occasionally be necessary to run the test to a higher number of stimuli, i.e. the maximum test time needs to be increased. Conclusions: The CGST can be used to control the specificity of a sequentially applied ABR detection method. Doing so can reduce test time, relative to the "single shot" test, when considered across a cohort of test subjects.

Measuring depth of anaesthesia using changes in directional connectivity: a comparison with auditory middle latency response and estimated bispectral index during propofol anaesthesia.

General anaesthesia is associated with changes in connectivity between different regions of the brain, the assessment of which has the potential to provide a novel marker of anaesthetic effect. We propose an index that quantifies the strength and direction of information flow in electroencephalographic signals collected across the scalp, assess its performance in discriminating 'wakefulness' from 'anaesthesia', and compare it with estimated bispectral index and the auditory middle latency response. We used a step-wise slow induction of anaesthesia in 10 patients to assess graded changes in electroencephalographic directional connectivity at propofol effect-site concentrations of 2 µg.ml-1 , 3 µg.ml-1 and 4 µg.ml-1 . For each stable effect-site concentration, connectivity was estimated from multichannel electroencephalograms using directed coherence, together with middle latency response and estimated bispectral index. We used a linear support vector machine classifier to compare the performance of the different electroencephalographic features in discriminating wakefulness from anaesthesia. We found a significant reduction in the strength of long-range connectivity (interelectrode distance > 10 cm) (p < 0.008), and a reversal of information flow from markedly postero-frontal to fronto-posterior (p < 0.006) between wakefulness and a propofol effect-site concentration of 2 µg.ml-1 . This then remained relatively constant as effect-site concentration increased, consistent with a step change in directed coherence with anaesthesia. This contrasted with the gradual change with increasing anaesthetic dose observed for estimated bispectral index and middle latency response. Directed coherence performed best in discriminating wakefulness from anaesthesia with an accuracy of 95%, indicating the potential of this new method (on its own or combined with others) for monitoring adequacy of anaesthesia.

Objective measures for detecting the auditory brainstem response: comparisons of specificity, sensitivity and detection time.

OBJECTIVE: To evaluate and compare the specificity, sensitivity and detection time of various time-domain and multi-band frequency domain methods when detecting the auditory brainstem response (ABR). DESIGN: Simulations and subject recorded data were used to assess and compare the performance of the Hotelling's T2 test (applied in either time or frequency domain), two versions of the modified q-sample uniform scores test and both the Fsp and Fmp, which were evaluated using both conventional F-distributions with assumed degrees of freedom and a bootstrap approach. STUDY SAMPLE: Data consisted of click-evoked ABRs and recordings of EEG background activity from 12 to 17 normal hearing adults, respectively. RESULTS: An overall advantage in sensitivity and detection time was demonstrated for the Hotelling's T2 test. The false-positive rates (FPRs) of the Fsp and Fmp were also closer to the nominal alpha-level when evaluating statistical significance using the bootstrap approach, as opposed to using conventional F-distributions. The FPRs of the remaining methods were slightly higher than expected. CONCLUSIONS: In this work, Hotelling's T2 outperformed the alternative methods for automatically detecting ABRs. Its promise as a sensitive and efficient detection method should now be tested in a larger clinical study.

Directional connectivity in the EEG is able to discriminate wakefulness from NREM sleep.

A reliable measure of consciousness is of great interest for various clinical applications including sleep studies and the assessment of depth of anaesthesia. A number of measures of consciousness based on the EEG have been proposed in the literature and tested in studies of dreamless sleep, general anaesthesia and disorders of consciousness. However, reliability has remained a persistent challenge. Despite considerable theoretical and experimental effort, the neural mechanisms underlying consciousness remain unclear, but connectivity between brain regions is thought to be disrupted, impairing information flow. OBJECTIVE: The objective of the current work was to assess directional connectivity between brain regions using directed coherence and propose and assess an index that robustly reflects changes associated with non-REM sleep. APPROACH: We tested the performance on polysomnographic recordings from ten healthy subjects and compared directed coherence (and derived features) with more established measures calculated from EEG spectra. We compared the performance of the different indexes to discriminate the level of consciousness at group and individual level. MAIN RESULTS: At a group level all EEG measures could significantly discriminate NREM sleep from waking, but there was considerable individual variation. Across all individuals, normalized power, the strength of long-range connections and the direction of functional links strongly correlate with NREM sleep stages over the experimental timeline. At an individual level, of the EEG measures considered, the direction of functional links constitutes the most reliable index of the level of consciousness, highly correlating with the individual experimental time-line of sleep in all subjects. SIGNIFICANCE: Directed coherence provides a promising new means of assessing level of consciousness, firmly based on current physiological understanding of consciousness.

Elastase levels and activity are increased in dystrophic muscle and impair myoblast cell survival, proliferation and differentiation.

In Duchenne muscular dystrophy, progressive loss of muscle tissue is accompanied by fibrosis, chronic inflammation and reduced muscle regenerative capacity. Although much is known about the development of fibrosis and chronic inflammation in muscular dystrophy, less is known about how they are mechanistically linked to loss of muscle regenerative capacity. We have developed a proteomics method to discover dystrophy-associated changes in the muscle progenitor cell niche, which identified serine proteases, and especially neutrophil elastase, as candidates. We show that elastase activity is increased in dystrophic (mdx(4cv)) muscle and impairs myoblast survival in culture. While the effect of elastase on C2C12 cell survival correlates with the kinetics of elastase-mediated degradation of the substrate to which the cells adhere, the effect of elastase on satellite cell-derived primary myoblast growth and differentiation is substrate-independent and even more dramatic than the effect on C2C12 cells, suggesting a detrimental role for elastase on myogenesis in vivo. Additionally, elastase impairs differentiation of both primary and C2C12 myoblasts into myotubes. Our findings evidence the importance of neutrophil-mediated inflammation in muscular dystrophy and indicate elastase-mediated regulation of myoblast behaviour as a potential mechanism underlying loss of regenerative capacity in dystrophic muscle.

Brain morphometric correlates of metabolic variables in HIV: the CHARTER study.

Obesity and other metabolic variables are associated with abnormal brain structural volumes and cognitive dysfunction in HIV-uninfected populations. Since individuals with HIV infection on combined antiretroviral therapy (CART) often have systemic metabolic abnormalities and changes in brain morphology and function, we examined associations among brain volumes and metabolic factors in the multisite CNS HIV AntiRetroviral Therapy Effects Research (CHARTER) cohort, cross-sectional study of 222 HIV-infected individuals. Metabolic variables included body mass index (BMI), total blood cholesterol (C), low- and high-density lipoprotein C (LDL-C and HDL-C), blood pressure, random blood glucose, and diabetes. MRI measured volumes of cerebral white matter, abnormal white matter, cortical and subcortical gray matter, and ventricular and sulcal CSF. Multiple linear regression models allowed us to examine metabolic variables separately and in combination to predict each regional volume. Greater BMI was associated with smaller cortical gray and larger white matter volumes. Higher total cholesterol (C) levels were associated with smaller cortex volumes; higher LDL-C was associated with larger cerebral white matter volumes, while higher HDL-C levels were associated with larger sulci. Higher blood glucose levels and diabetes were associated with more abnormal white matter. Multiple atherogenic metabolic factors contribute to regional brain volumes in HIV-infected, CART-treated patients, reflecting associations similar to those found in HIV-uninfected individuals. These risk factors may accelerate cerebral atherosclerosis and consequent brain alterations and cognitive dysfunction.

Optimising the assessment of cerebral autoregulation from black box models.

Cerebral autoregulation (CA) mechanisms maintain blood flow approximately stable despite changes in arterial blood pressure. Mathematical models that characterise this system have been used extensively in the quantitative assessment of function/impairment of CA. Using spontaneous fluctuations in arterial blood pressure (ABP) as input and cerebral blood flow velocity (CBFV) as output, the autoregulatory mechanism can be modelled using linear and non-linear approaches, from which indexes can be extracted to provide an overall assessment of CA. Previous studies have considered a single--or at most a couple of measures, making it difficult to compare the performance of different CA parameters. We compare the performance of established autoregulatory parameters and propose novel measures. The key objective is to identify which model and index can best distinguish between normal and impaired CA. To this end 26 recordings of ABP and CBFV from normocapnia and hypercapnia (which temporarily impairs CA) in 13 healthy adults were analysed. In the absence of a 'gold' standard for the study of dynamic CA, lower inter- and intra-subject variability of the parameters in relation to the difference between normo- and hypercapnia were considered as criteria for identifying improved measures of CA. Significantly improved performance compared to some conventional approaches was achieved, with the simplest method emerging as probably the most promising for future studies.

Neurocognitive impairment in HIV-infected individuals with previous syphilis.

Cognitive impairment is common in HIV-infected individuals, as is syphilis. Treponema pallidum, the bacterium that causes syphilis, invades the central nervous system early in disease. We hypothesized that HIV-infected patients with a history of syphilis or neurosyphilis would have more cognitive impairment than HIV-infected individuals without these infections. Eighty-two of 1574 enrollees in CHARTER, a prospective, observational study, had reactive serum rapid plasma reagin (RPR) tests. They were matched to 84 controls with non-reactive RPR by age, gender, ethnicity and HIV risk factor. Participants underwent comprehensive neuropsychological (NP) evaluations. RPR results were confirmed and serum fluorescent treponemal antibody absorption (FTA-ABS) test reactivity determined at a central laboratory. Sera from 101 of 166 participants were FTA-ABS reactive, indicating past or current syphilis. Among the 136 individuals without confounding conditions, compared with patients who had never had syphilis, those with prior syphilis had a greater number of impaired NP test domains (1.90 SD [1.77] versus 1.25 [1.52], P = 0.03), a higher global deficit score (0.47 [0.46] versus 0.31 [0.33], P = 0.03), and more were impaired in the NP learning domain (36 [42.9%] of 84 versus 13 [25.0%] of 52, P = 0.04). These effects of prior syphilis remained after controlling for education and premorbid intelligence.

Apolipoprotein E4 genotype does not increase risk of HIV-associated neurocognitive disorders.

This is a cross-sectional, observational study to evaluate the hypothesis that HIV-seropositive (HIV+) apolipoprotein E4 (APOE4) carriers are at increased risk for HIV-associated neurocognitive disorders (HAND) compared to APOE4 noncarriers with HIV in the CNS HIV Antiretroviral Therapy Effects Research (CHARTER) Group sample. APOE genotype was determined in 466 CHARTER participants with varying disease stages and histories of antiretroviral treatment who did not have severe psychiatric or medical comorbid conditions that preclude diagnosis of HAND. HAND diagnoses were based on results of comprehensive neurobehavioral evaluation and use of current neuroAIDS diagnostic criteria. HAND status consists of two levels: neuropsychologically normal status (i.e., no HAND) and any HAND diagnosis (i.e., asymptomatic neurocognitive impairment, minor neurocognitive disorder, HIV-associated dementia). Logistic regression analyses revealed no association between APOE4 carrier status and HAND, and there were no interactions between APOE4 carrier status and ethnicity, age, substance use disorders, duration of infection, or nadir CD4. Results did not differ when analysis was restricted to symptomatic HAND, and no APOE4 gene dose-dependent relationship to HAND emerged. APOE4 status was not associated with concurrent HAND in this large, well-characterized sample. This does not preclude emergence of an association between APOE4 status and HAND as this population ages. Prospective, longitudinal studies are needed to examine APOE4 as a risk factor for neurocognitive decline, incident HAND at older ages, and potential associations with cerebrospinal fluid amyloid.

Defining neurocognitive impairment in HIV: deficit scores versus clinical ratings.

Because HIV-related neurocognitive impairment is usually mild and variable, clinical ratings (CR) and global deficit scores (GDS) are recommended for detecting HIV-associated neurocognitive disorders (HAND). The CR approach requires impairment in at least two ability domains while the GDS considers number and severity of impairments across all measures. We examined classification agreement and clinical correlates of the two methods. Neurocognitive functioning of 1574 HIV-infected participants was assessed via a comprehensive, seven-domain neuropsychological battery. Global neurocognitive impairment was defined for each participant independently by CR and GDS. Participants were classified into four categories (Dually-normal, [impaired by] CR-only, [impaired by] GDS-only, or Dually-impaired). There was 83% concordance between CR and GDS classifications; in total, 56% of participants were deemed impaired by CR and 41% were classified as impaired by GDS. Impairment by GDS virtually guaranteed CR impairment, but 16% of participants were additionally classified as impaired only by CR. As compared to Dually-normal participants, those classified as Dually and CR-only impaired were more likely to have AIDS, have more severe co-occurring conditions, have more severe depressive symptoms, be unemployed, and have more everyday functioning complaints (ps < .05). Impairment classifications of the two methods were in high agreement; however, more people were classified as impaired using the CR approach compared to the GDS approach. Those impaired according to CR-only showed fewer neurocognitive and functional deficits than the Dually-impaired participants, but more of these deficits than Dually-normal participants. The CR approach may be most appropriate for detecting more subtle forms of neurocognitive impairment. Clinicians and researchers should recognize the strengths and weaknesses of each method when evaluating neurocognitive complications in HIV.

Evidence-based guideline: Antiepileptic drug selection for people with HIV/AIDS: report of the Quality Standards Subcommittee of the American Academy of Neurology and the Ad Hoc Task Force of the Commission on Therapeutic Strategies of the International League Against Epilepsy.

OBJECTIVE: To develop guidelines for selection of antiepileptic drugs (AEDs) among people with HIV/AIDS. METHODS: The literature was systematically reviewed to assess the global burden of relevant comorbid entities, to determine the number of patients who potentially utilize AEDs and antiretroviral agents (ARVs), and to address AED-ARV interactions. RESULTS AND RECOMMENDATIONS: AED-ARV administration may be indicated in up to 55% of people taking ARVs. Patients receiving phenytoin may require a lopinavir/ritonavir dosage increase of ~50% to maintain unchanged serum concentrations (Level C). Patients receiving valproic acid may require a zidovudine dosage reduction to maintain unchanged serum zidovudine concentrations (Level C). Coadministration of valproic acid and efavirenz may not require efavirenz dosage adjustment (Level C). Patients receiving ritonavir/atazanavir may require a lamotrigine dosage increase of ∼50% to maintain unchanged lamotrigine serum concentrations (Level C). Coadministration of raltegravir/atazanavir and lamotrigine may not require lamotrigine dosage adjustment (Level C). Coadministration of raltegravir and midazolam may not require midazolam dosage adjustment (Level C). Patients may be counseled that it is unclear whether dosage adjustment is necessary when other AEDs and ARVs are combined (Level U). It may be important to avoid enzyme-inducing AEDs in people on ARV regimens that include protease inhibitors or nonnucleoside reverse transcriptase inhibitors, as pharmacokinetic interactions may result in virologic failure, which has clinical implications for disease progression and development of ARV resistance. If such regimens are required for seizure control, patients may be monitored through pharmacokinetic assessments to ensure efficacy of the ARV regimen (Level C).

Diagnosing symptomatic HIV-associated neurocognitive disorders: self-report versus performance-based assessment of everyday functioning.

Three types of HIV-associated neurocognitive disorders (HAND) exist that are distinguished by presence and severity of impairment in cognitive and everyday functioning. Although well-validated neurocognitive measures exist, determining impairment in everyday functioning remains a challenge. We aim to determine whether Self-Report measures of everyday functioning are as effective in characterizing HAND as Performance-Based measures. We assessed 674 HIV-infected participants with a comprehensive neurocognitive battery; 233 met criteria for a HAND diagnosis by having at least mild neurocognitive impairment. Functional decline was measured via Self-Report and Performance-Based measures. HAND diagnoses were determined according to published criteria using three approaches to assess functional decline: (1) Self-Report measures only, (2) Performance-Based measures only, and (3) Dual-method combining Self-Report and Performance-Based measures. The Dual-method classified the most symptomatic HAND, compared to either singular method. Singular method classifications were 76% concordant with each other. Participants classified as Performance-Based functionally impaired were more likely to be unemployed and more immunosuppressed, whereas those classified as Self-Report functionally impaired had more depressive symptoms. Multimodal methods of assessing everyday functioning facilitate detection of symptomatic HAND. Singular Performance-Based classifications were associated with objective functional and disease-related factors; reliance on Self-Report classifications may be biased by depressive symptoms.

HIV-associated neurocognitive disorders persist in the era of potent antiretroviral therapy: CHARTER Study.

OBJECTIVES: This is a cross-sectional, observational study to determine the frequency and associated features of HIV-associated neurocognitive disorders (HAND) in a large, diverse sample of infected individuals in the era of combination antiretroviral therapy (CART). METHODS: A total of 1,555 HIV-infected adults were recruited from 6 university clinics across the United States, with minimal exclusions. We used standardized neuromedical, psychiatric, and neuropsychological (NP) examinations, and recently published criteria for diagnosing HAND and classifying 3 levels of comorbidity (minimal to severe non-HIV risks for NP impairment). RESULTS: Fifty-two percent of the total sample had NP impairment, with higher rates in groups with greater comorbidity burden (40%, 59%, and 83%). Prevalence estimates for specific HAND diagnoses (excluding severely confounded cases) were 33% for asymptomatic neurocognitive impairment, 12% for mild neurocognitive disorder, and only 2% for HIV-associated dementia (HAD). Among participants with minimal comorbidities (n = 843), history of low nadir CD4 was a strong predictor of impairment, and the lowest impairment rate on CART occurred in the subset with suppressed plasma viral loads and nadir CD4 ≥200 cells/mm(3) (30% vs 47% in remaining subgroups). CONCLUSIONS: The most severe HAND diagnosis (HAD) was rare, but milder forms of impairment remained common, even among those receiving CART who had minimal comorbidities. Future studies should clarify whether early disease events (e.g., profound CD4 decline) may trigger chronic CNS changes, and whether early CART prevents or reverses these changes.

Nonlinear, multiple-input modeling of cerebral autoregulation using Volterra Kernel estimation.

Autoregulation refers to the automatic adjustment of blood flow to supply the required oxygen and glucose and remove waste, in proportion to the tissue's requirement at any instant of time. For the brain, cerebral autoregulation is an active process by which cerebral blood flow is controlled at an approximately steady level despite changes in the arterial blood pressure. Robust assessment of the cerebral autoregulation by a model that characterizes this system has been the goal of many studies, searching for techniques that can be used in clinical scenarios to detect potentially dangerous impairment of control. Multiple input, single output (MISO) models can be used to assess autoregulation, and system parameters can be estimated from spontaneous beat-to-beat variations in arterial blood pressure (ABP) and breath-by-breath end-tidal carbon dioxide (P(ETCO2)) as inputs, and cerebral blood flow velocity (CBFV) as the output. In this study a non-linear, multivariate approach, based on Volterra-type kernel estimation models is employed. The results are compared with linear models as well as nonlinear single-input single-output (SISO) models. The normalized mean squared error was used as the criteria of performance of each model in assessing cerebral autoregulation. Our simulation results indicate that for relatively short signals (around 300 sec), nonlinear, multiple-input models based on Volterra systems performed best, though the benefit varied considerably between subjects. When using a fixed model for all recordings, a linear SISO model with ABP as input provided the smallest average modeling error.

Association of self-reported painful symptoms with clinical and neurophysiologic signs in HIV-associated sensory neuropathy.

Sensory neuropathy (HIV-SN) is a common cause of pain in HIV-infected people. Establishing a diagnosis of HIV-SN is important, especially when contemplating opioid use in high-risk populations. However physical findings of HIV-SN may be subtle, and sensitive diagnostic tools require specialized expertise. We investigated the association between self-report of distal neuropathic pain and/or paresthesias (DNPP) and objective signs of HIV-SN. Data were obtained from the Central Nervous System HIV Antiretroviral Therapy Effects Research (CHARTER) study. Out of 237 participants, 101 (43%) reported DNPP. Signs of HIV-SN were measured by a modified Total Neuropathy Score (TNS), composed of six objective sensory subscores (pin sensibility, vibration sensibility, deep tendon reflexes, quantitative sensory testing for cooling and vibration, and sural sensory amplitude). Self-report of DNPP was associated with all six TNS items in univariate analysis and with four TNS items in multivariate analysis. The sensitivity and specificity of self-report of DNPP in detecting the presence of a sensory abnormality were 52% and 92%, respectively with a PPV of 96% and a NPV of 34%. Increasing intensity of pain measured on a visual analog scale was associated with increasing severity of sensory abnormality. In summary, our results suggest that HIV-infected patients reporting symptoms consistent with HIV-SN, such as tingling, pins and needles, or aching or stabbing pain in the distal lower extremities, usually have objective evidence of HIV-SN on neurologic examination or with neurophysiologic testing. This finding holds true regardless of demographic factors, depression or substance use history.