Combining electromyographic and electrical impedance data sets through machine learning: A study in D2-mdx and wild-type mice.

INTRODUCTION/AIMS: Needle impedance-electromyography (iEMG) assesses the active and passive electrical properties of muscles concurrently by using a novel needle with six electrodes, two for EMG and four for electrical impedance myography (EIM). Here, we assessed an approach for combining multifrequency EMG and EIM data via machine learning (ML) to discriminate D2-mdx muscular dystrophy and wild-type (WT) mouse skeletal muscle. METHODS: iEMG data were obtained from quadriceps of D2-mdx mice, a muscular dystrophy model, and WT animals. EIM data were collected with the animals under deep anesthesia and EMG data collected under light anesthesia, allowing for limited spontaneous movement. Fourier transformation was performed on the EMG data to provide power spectra that were sampled across the frequency range using three different approaches. Random forest-based, nested ML was applied to the EIM and EMG data sets separately and then together to assess healthy versus disease category classification using a nested cross-validation procedure. RESULTS: Data from 20 D2-mdx and 20 WT limbs were analyzed. EIM data fared better than EMG data in differentiating healthy from disease mice with 93.1% versus 75.6% accuracy, respectively. Combining EIM and EMG data sets yielded similar performance as EIM data alone with 92.2% accuracy. DISCUSSION: We have demonstrated an ML-based approach for combining EIM and EMG data obtained with an iEMG needle. While EIM-EMG in combination fared no better than EIM alone with this data set, the approach used here demonstrates a novel method of combining the two techniques to characterize the full electrical properties of skeletal muscle.

LASSI-L detects early cognitive changes in pre-motor manifest Huntington's disease: a replication and validation study.

Background and objectives: Cognitive decline is an important early sign in pre-motor manifest Huntington's disease (preHD) and is characterized by deficits across multiple domains including executive function, psychomotor processing speed, and memory retrieval. Prior work suggested that the Loewenstein-Acevedo Scale for Semantic Interference and Learning (LASSI-L)-a verbal learning task that simultaneously targets these domains - could capture early cognitive changes in preHD. The current study aimed to replicate, validate and further analyze the LASSI-L in preHD using larger datasets. Methods: LASSI-L was administered to 50 participants (25 preHD and 25 Healthy Controls) matched for age, education, and sex in a longitudinal study of disease progression and compared to performance on MMSE, Trail A & B, SCWT, SDMT, Semantic Fluency (Animals), and CVLT-II. Performance was then compared to a separate age-education matched-cohort of 25 preHD participants. Receiver operating curve (ROC) and practice effects (12 month interval) were investigated. Group comparisons were repeated using a preHD subgroup restricted to participants predicted to be far from diagnosis (Far subgroup), based on CAG-Age-Product scaled (CAPs) score. Construct validity was assessed through correlations with previously established measures of subcortical atrophy. Results: PreHD performance on all sections of the LASSI-L was significantly different from controls. The proactive semantic interference section (PSI) was sensitive (p = 0.0001, d = 1.548), similar across preHD datasets (p = 1.0), reliable on test-retest over 12 months (spearman rho = 0.88; p = <0.00001) and associated with an excellent area under ROC (AUROC) of 0.855. In the preHD Far subgroup comparison, PSI was the only cognitive assessment to survive FDR < 0.05 (p = 0.03). The number of intrusions on PSI was negatively correlated with caudate volume. Discussion: The LASSI-L is a sensitive, reliable, efficient tool for detecting cognitive decline in preHD. By using a unique verbal learning test paradigm that simultaneously targets executive function, processing speed and memory retrieval, the LASSI-L outperforms many other established tests and captures early signs of cognitive impairment. With further longitudinal validation, the LASSI-L could prove to be a useful biomarker for clinical research in preHD.

High CD4+:CD8+ ratios with herpes zoster infections in patients with multiple sclerosis on dimethyl fumarate.

BACKGROUND: Dimethyl fumarate (DMF) depletes CD8+ and CD4+ T cells, and cases of herpes zoster (HZ) in patients with multiple sclerosis (MS) on DMF have been documented. OBJECTIVES: To evaluate lymphocyte subsets in patients with MS who developed HZ on DMF (Tecfidera) compared to matched controls who did not develop HZ. METHODS: We used linear mixed-effects models to test for differences in white blood cell count, lymphocyte percentage, absolute lymphocyte count, CD3+ percentage, absolute CD3+ count, CD4+ percentage, absolute CD4+ count, CD8+ percentage, absolute CD8+ count, and CD4+:CD8+ ratio over time in HZ and non-HZ groups. RESULTS: Eighteen patients developed HZ while on DMF. The linear mixed-effects model for CD4+:CD8+ ratio showed a significant difference between the HZ and non-HZ groups (p = 0.033). CD4+:CD8+ ratio decreased over time in the HZ group and increased over time in the non-HZ group. CONCLUSION: Patients with MS who develop HZ while on DMF have high CD4+:CD8+ ratios, suggesting an imbalance of CD4+ and CD8+ cells that may put a patient at risk for developing HZ while on DMF. This result emphasizes the need for lymphocyte subset monitoring (including CD4+:CD8+ ratios) on DMF, as well as vaccination prior to DMF initiation.

Surface Electrical Impedance Myography Detects Skeletal Muscle Atrophy in Aged Wildtype Zebrafish and Aged gpr27 Knockout Zebrafish.

Throughout a vertebrate organism's lifespan, skeletal muscle mass and function progressively decline. This age-related condition is termed sarcopenia. In humans, sarcopenia is associated with risk of falling, cardiovascular disease, and all-cause mortality. As the world population ages, projected to reach 2 billion older adults worldwide in 2050, the economic burden on the healthcare system is also projected to increase considerably. Currently, there are no pharmacological treatments for sarcopenia, and given the long-term nature of aging studies, high-throughput chemical screens are impractical in mammalian models. Zebrafish is a promising, up-and-coming vertebrate model in the field of sarcopenia that could fill this gap. Here, we developed a surface electrical impedance myography (sEIM) platform to assess skeletal muscle health, quantitatively and noninvasively, in adult zebrafish (young, aged, and genetic mutant animals). In aged zebrafish (~85% lifespan) as compared to young zebrafish (~20% lifespan), sEIM parameters (2 kHz phase angle, 2 kHz reactance, and 2 kHz resistance) robustly detected muscle atrophy (p < 0.000001, q = 0.000002; p = 0.000004, q = 0.000006; p = 0.000867, q = 0.000683, respectively). Moreover, these same measurements exhibited strong correlations with an established morphometric parameter of muscle atrophy (myofiber cross-sectional area), as determined by histological-based morphometric analysis (r = 0.831, p = 2 × 10-12; r = 0.6959, p = 2 × 10-8; and r = 0.7220; p = 4 × 10-9, respectively). Finally, the genetic deletion of gpr27, an orphan G-protein coupled receptor (GPCR), exacerbated the atrophy of skeletal muscle in aged animals, as evidenced by both sEIM and histology. In conclusion, the data here show that surface EIM techniques can effectively discriminate between healthy young and sarcopenic aged muscle as well as the advanced atrophied muscle in the gpr27 KO animals. Moreover, these studies show how EIM values correlate with cell size across the animals, making it potentially possible to utilize sEIM as a "virtual biopsy" in zebrafish to noninvasively assess myofiber atrophy, a valuable measure for muscle and gerontology research.

Prevalence of neurocognitive disorder in Huntington's disease using the Enroll-HD dataset.

Background: Cognitive decline in Huntington's disease (HD) begins early in the disease course, however the reported prevalence and severity of cognitive impairment varies based on diagnostic approach. A Movement Disorders Society Task Force recently endorsed the use of standardized DSM-5-based criteria to diagnose neurocognitive disorder (NCD) in Huntington's disease. Objectives: To determine the prevalence and severity of cognitive impairment across different stages of HD by applying NCD criteria (mild and major) to participant data from the Enroll-HD database. Methods: Enroll-HD participants were triaged into either premanifest (preHD), manifest or control groups. PreHD was further dichotomized into preHD near or preHD far based on predicted time to diagnosis using the scaled CAG-age product score (CAPs). Embedded cognitive performance and functional independence measures were used to determine prevalence of NCD (mild and major) for all groups. Results: Prevalence of NCD-mild was 25.2%-38.4% for manifest HD, 22.8%-47.3% for preHD near, 11.5%-25.1% for preHD far, and 8.8%-19.1% for controls. Prevalence of NCD-major was 21.1%-57.7% for manifest HD, 0.5%-16.3% for preHD near, 0.0%-4.5% for preHD far, and 0.0%-3.0% for controls. Conclusion: The prevalence of NCD in HD is elevated in preHD and demonstrates a sharp rise prior to diagnosis. In manifest HD, the vast majority of participants meet criteria for NCD. These findings are important for optimizing clinical care and/or anticipating the need for supportive services.

Electrical Impedance Dermography Differentiates Squamous Cell Carcinoma In Situ from Inflamed Seborrheic Keratoses.

There are no currently available low-cost, noninvasive methods for discerning the depth of squamous cell carcinoma (SCC) invasion or distinguishing SCC from its benign mimics, such as inflamed seborrheic keratosis (SK). We studied 35 subjects with subsequently confirmed SCC or SK. Subjects underwent electrical impedance dermography measurements at six frequencies to assess the electrical properties of the lesion. Averaged greatest intrasession reproducibility values were 0.630 for invasive SCC at 128 kHz, 0.444 for SCC in situ at 16 kHz, and 0.460 for SK at 128 kHz. Electrical impedance dermography modeling revealed significant differences between SCC and inflamed SK in normal skin (P < 0.001) and also between invasive SCC and SCC in situ (P < 0.001), invasive SCC and inflamed SK (P < 0.001), and SCC in situ and inflamed SK (P < 0.001). A diagnostic algorithm classified SCC in situ from inflamed SK with an accuracy of 0.958, a sensitivity of 94.6%, and a specificity of 96.9%; it also classified SCC in situ from normal skin with an accuracy of 0.796, a sensitivity of 90.2%, and a specificity of 51.2%. This study provides preliminary data and a methodology that can be used in future studies to further advance the value of electrical impedance dermography and inform biopsy decision making in patients with lesions suspicious of SCC.

Electrical Impedance Myography Correlates with Functional Measures of Disease Progression in D2-mdx Mice and Boys with Duchenne Muscular Dystrophy.

BACKGROUND: Sensitive, objective, and longitudinal outcome measures applicable to both pre-clinical and clinical interventions are needed to assess muscle health in Duchenne muscular dystrophy (DMD). Electrical impedance myography (EIM) has the potential to non-invasively measure disease progression in mice and boys with DMD. OBJECTIVE: We sought to evaluate how electrical impedance values (i.e., phase, reactance, and resistance) correlate to established measures of disease in both D2-mdx and wild type (WT) mice and boys with and without DMD. METHODS: Histological, functional, and EIM data collected from previous studies of WT and D2-mdx mice at 6, 13, 21 and 43 weeks of age were reanalyzed. In parallel, previously collected functional outcome measures and EIM values were reanalyzed from boys with and without DMD at four different age groups from 2 to 14 years old. RESULTS: In mice, disease progression as detected by histological, functional, and EIM measures, was appreciable over this time period and grip strength best correlated to longitudinal phase and reactance impedance values. In boys, disease progression quantified through commonly utilized functional outcome measures was significant and longitudinal phase demonstrated the strongest correlation with functional outcome measures. CONCLUSION: Similar changes in EIM values, specifically in longitudinal reactance and phase, were found to show significant correlations to functional measures in both mice and boys. Thus, EIM demonstrates applicability in both pre-clinical and clinical settings and can be used as a safe, non-invasive, and longitudinal proxy biomarker to assess muscle health in DMD.

Electrical Impedance Myography in Dogs With Degenerative Myelopathy.

Canine degenerative myelopathy (DM) leads to disuse and neurogenic muscle atrophy. Currently there is a lack of non-invasive quantitative measures of muscle health in dogs with DM. Muscle pathology has been previously quantified in other disorders using the technique of electrical impedance myography (EIM) but it has not been reported for DM. The objective of this study was to compare EIM between DM-affected and similar aged healthy dogs as well as assess EIM changes over time in DM-affected dogs. Multifrequency EIM was performed on DM affected dogs at baseline and during disease progression and on age-matched healthy dogs. Muscles evaluated in the pelvic limbs included the craniotibialis, gastrocnemius, gracilis, sartorius, and biceps femoris. The 100 kHz phase angle was extracted from the full frequency set for analysis. Phase values were lower in DM dogs as compared to healthy controls. Specifically, phase of the gastrocnemius was lower on the left (θ = 7.69, 13.06; p =0.002) and right (θ= 6.11, 11.72; p = 0.001) in DM vs. control dogs, respectively. The mean phase value of all measured muscles was also lower on the left (θ = 9.24, 11.62; p = 0.012) and right (θ = 9.18, 11.72; p = 0.021). Other individual muscles measured did not reach statistical significance, although values were consistently lower in DM-affected dogs. With disease progression, downward trends in phase values were detected in DM-affected dogs when monitored serially over time. This study demonstrates that EIM 100 kHz phase values are sensitive to muscle pathology in DM and that phase values are decreased in dogs with DM. Measurements from the gastrocnemius muscle show the greatest differences from similar aged healthy dogs suggesting it may be the preferred muscle for future EIM studies.

Using machine learning algorithms to enhance the diagnostic performance of electrical impedance myography.

INTRODUCTION/AIMS: We assessed the classification performance of machine learning (ML) using multifrequency electrical impedance myography (EIM) values to improve upon diagnostic outcomes as compared to those based on a single EIM value. METHODS: EIM data was obtained from unilateral excised gastrocnemius in eighty diseased mice (26 D2-mdx, Duchenne muscular dystrophy model, 39 SOD1G93A ALS model, and 15 db/db, a model of obesity-induced muscle atrophy) and 33 wild-type (WT) animals. We assessed the classification performance of a ML random forest algorithm incorporating all the data (multifrequency resistance, reactance and phase values) comparing it to the 50 kHz phase value alone. RESULTS: ML outperformed the 50 kHz analysis as based on receiver-operating characteristic curves and measurement of the area under the curve (AUC). For example, comparing all diseases together versus WT from the test set outputs, the AUC was 0.52 for 50 kHz phase, but was 0.94 for the ML model. Similarly, when comparing ALS versus WT, the AUCs were 0.79 for 50 kHz phase and 0.99 for ML. DISCUSSION: Multifrequency EIM using ML improves upon classification compared to that achieved with a single-frequency value. ML approaches should be considered in all future basic and clinical diagnostic applications of EIM.

Electrical impedance myography in healthy dogs: Normative values, repeatability, and the impact of age.

Convenient tools to assess canine skeletal muscle health would be useful for a variety of applications, including standard veterinary assessments of dog fitness, as well as studies of muscle deterioration due to age or disease. One technology that can be applied conveniently to awake dogs with minimal restraint is electrical impedance myography (EIM). In EIM, a weak electrical current is applied via surface electrodes to a muscle of interest and consequent impedance characteristics of the muscle are obtained, providing insight into muscle condition and composition. In this study, we assessed a total of 73 dogs (42 males and 31 females), of varied neutering status and breed, ages 0.6 to 13.5 years. We identified age-dependent reference values for the 100 kHz phase value in three pelvic limb muscles, caudal sartorius, cranial tibial, and gastrocnemius. While phase values were generally higher in males than females, the difference did not reach significance. In general, values declined on average with age at about 0.5 degrees/year, but with the decline being most substantial in the oldest dogs. Limited reproducibility assessment of the technique suggested good repeatability with variation in values between measurements being under 5%. These results show that EIM has the potential for the assessment of canine muscle health and may find value in aging muscle research.

Relationships between in vivo surface and ex vivo electrical impedance myography measurements in three different neuromuscular disorder mouse models.

Electrical impedance myography (EIM) using surface techniques has shown promise as a means of diagnosing and tracking disorders affecting muscle and assessing treatment efficacy. However, the relationship between such surface-obtained impedance values and pure muscle impedance values has not been established. Here we studied three groups of diseased and wild-type (WT) animals, including a Duchenne muscular dystrophy model (the D2-mdx mouse), an amyotrophic lateral sclerosis (ALS) model (the SOD1 G93A mouse), and a model of fat-related atrophy (the db/db diabetic obese mouse), performing hind limb measurements using a standard surface array and ex vivo measurements on freshly excised gastrocnemius muscle. A total of 101 animals (23 D2-mdx, 43 ALS mice, 12 db/db mice, and corresponding 30 WT mice) were studied with EIM across a frequency range of 8 kHz to 1 MHz. For both D2-mdx and ALS models, moderate strength correlations (Spearman rho values generally ranging from 0.3-0.7, depending on the impedance parameter (i.e., resistance, reactance and phase) were obtained. In these groups of animals, there was an offset in frequency with impedance values obtained at higher surface frequencies correlating more strongly to impedance values obtained at lower ex vivo frequencies. For the db/db model, correlations were comparatively weaker and strongest at very high and very low frequencies. When combining impedance data from all three disease models together, moderate correlations persisted (with maximal Spearman rho values of 0.45). These data support that surface EIM data reflect ex vivo muscle tissue EIM values to a moderate degree across several different diseases, with the highest correlations occurring in the 10-200 kHz frequency range. Understanding these relationships will prove useful for future applications of the technique of EIM in the assessment of neuromuscular disorders.

Estimating myofiber cross-sectional area and connective tissue deposition with electrical impedance myography: A study in D2-mdx mice.

INTRODUCTION: Surface electrical impedance myography (sEIM) has the potential for providing information on muscle composition and structure noninvasively. We sought to evaluate its use to predict myofiber size and connective tissue deposition in the D2-mdx model of Duchenne muscular dystrophy (DMD). METHODS: We applied a prediction algorithm, the least absolute shrinkage and selection operator, to select specific EIM measurements obtained with surface and ex vivo EIM data from D2-mdx and wild-type (WT) mice (analyzed together or separately). We assessed myofiber cross-sectional area histologically and hydroxyproline (HP), a surrogate measure for connective tissue content, biochemically. RESULTS: Using WT and D2-mdx impedance values together in the algorithm, sEIM gave average root-mean-square errors (RMSEs) of 26.6% for CSA and 45.8% for HP, which translate into mean errors of ±363 µm2 for a mean CSA of 1365 µm2 and of ±1.44 µg HP/mg muscle for a mean HP content of 3.15 µg HP/mg muscle. Stronger predictions were obtained by analyzing sEIM data from D2-mdx animals alone (RMSEs of 15.3% for CSA and 34.1% for HP content). Predictions made using ex vivo EIM data from D2-mdx animals alone were nearly equivalent to those obtained with sEIM data (RMSE of 16.59% for CSA), and slightly more accurate for HP (RMSE of 26.7%). DISCUSSION: Surface EIM combined with a predictive algorithm can provide estimates of muscle pathology comparable to values obtained using ex vivo EIM, and can be used as a surrogate measure of disease severity and progression and response to therapy.

Predicting myofiber cross-sectional area and triglyceride content with electrical impedance myography: A study in db/db mice.

BACKGROUND: Electrical impedance myography (EIM) provides insight into muscle composition and structure. We sought to evaluate its use in a mouse obesity model characterized by myofiber atrophy. METHODS: We applied a prediction algorithm, ie, the least absolute shrinkage and selection operator (LASSO), to surface, needle array, and ex vivo EIM data from db/db and wild-type mice and assessed myofiber cross-sectional area (CSA) histologically and triglyceride (TG) content biochemically. RESULTS: EIM data from all three modalities provided acceptable predictions of myofiber CSA with average root mean square error (RMSE) of 15% in CSA (ie, ±209 µm2 for a mean CSA of 1439 µm2 ) and TG content with RMSE of 30% in TG content (ie, ±7.3 nmol TG/mg muscle for a mean TG content of 25.4 nmol TG/mg muscle). CONCLUSIONS: EIM combined with a predictive algorithm provides reasonable estimates of myofiber CSA and TG content without the need for biopsy.

Improved ALS clinical trials through frequent at-home self-assessment: a proof of concept study.

OBJECTIVE: To determine the potential for improving amyotrophic lateral sclerosis (ALS) clinical trials by having patients or caregivers perform frequent self-assessments at home. METHODS AND PARTICIPANTS: We enrolled ALS patients into a nonblinded, longitudinal 9-month study in which patients and caregivers obtained daily data using several different instruments, including a slow-vital capacity device, a hand grip dynamometer, an electrical impedance myography-based fitness device, an activity tracker, a speech app, and the ALS functional rating scale-revised. Questions as to acceptability were asked at two time points. RESULTS: A total of 113 individuals enrolled, with 61 (43 men, 18 women, mean age 60.1 ± 9.9 years) collecting a minimum of 7 days data and being included in the analysis. Daily measurements resulted in more accurate assessments of the slope of progression of the disease, resulting in smaller sample size estimates for a hypothetical clinical trial. For example, by performing daily slow-vital capacity measurements, calculated sample size was reduced to 182 subjects/study arm from 882/arm for monthly measurements. Similarly, performing the ALS functional rating scale weekly rather than monthly led to a calculated sample size of 73/arm as compared to 274/arm. Participants generally found the procedures acceptable and, for many, improved their sense of control of their disease. INTERPRETATION: Frequent at-home measurements using standard tools holds the prospect of tracking progression and reducing sample size requirements for clinical trials in ALS while also being acceptable to the patients. Future studies in this and other neurological disorders should consider adopting this approach to data collection.

A Serious Flaw in Nutrition Epidemiology: A Meta-Analysis Study.

Background Many researchers have studied the relationship between diet and health. Specifically, there are papers showing an association between the consumption of sugar sweetened beverages and Type 2 diabetes. Many meta-analyses use individual studies that do not attempt to adjust for multiple testing or multiple modeling. Hence the claims reported in a meta-analysis paper may be unreliable as the base papers do not ensure unbiased statistics. Objective Determine (i) the statistical reliability of 10 papers and (ii) indirectly the reliability of the meta-analysis study. Method We obtained copies of each of the 10 papers used in a metaanalysis paper and counted the numbers of outcomes, predictors, and covariates. We estimate the size of the potential analysis search space available to the authors of these papers; i. e. the number of comparisons and models available. The potential analysis search space is the number of outcomes times the number of predictors times 2 c , where c is the number of covariates. This formula was applied to information found in the abstracts (Space A) as well as the text (Space T) of each base paper. Results The median and range of the number of comparisons possible across the base papers are 6.5 and (2 12,288), respectively for Space A, and 196,608 and (3072-117,117,952), respectively for Space T. It is noted that the median of 6.5 for Space A may be misleading as each study has 60-165 foods that could be predictors. Conclusion Given that testing is at the 5% level and the number of comparisons is very large, nominal statistical significance is very weak support for a claim. The claims in these papers are not statistically supported and hence are unreliable so the meta-analysis paper is also unreliable.

Georgia's collaborative approach to expanding mosquito surveillance in response to Zika virus: a case study.

Zika virus (ZIKV) was declared an international public health emergency by the World Health Organization on February 1, 2016. Due to the known and estimated range of the ZIKV mosquito vectors, southern and central US states faced increased risk of ZIKV transmission. With the state of Georgia hosting the world's busiest international airport, a climate that supports the ZIKV vectors, and limited surveillance (13 counties) and response capacity, the Department of Public Health (DPH) was challenged to respond and prevent ZIKV transmission. This case study describes and evaluates the state's surveillance capacity before and after the declaration of ZIKV as a public health emergency. METHOD: We analyzed surveillance data from the DPH to compare the geographical distribution of counties conducting surveillance, total number, and overall percentage of mosquito species trapped in 2015 to 2016. Counties conducting surveillance before and after the identification of the ZIKV risk were mapped using ArcMap 10.4.1. Using SAS (version 9.2) (SAS Institute, Inc, Cary, NC), we performed the independent 2 sample t test to test for differences in prevalence in both years, and a χ² analysis to test for differences between numbers of species across the 13 counties. In addition, weighted frequency counts of mosquitoes were used to test (χ²) an association between major mosquito vector species and 7 urban counties. Lastly, using data from 2012-2016, a time-trend analysis was conducted to evaluate temporal trends in species prevalence. RESULTS: From 2015 to 2016, surveillance increased from 13 to 57 (338% increase) counties geographically dispersed across Georgia. A total of 76,052 mosquitoes were trapped and identified in 2015 compared to 144,731 (90.3% increase) in 2016. Significant differences between species (P<.001) and significant associations (P<.0001) between 7 urban counties and major mosquito vectors were found. Significant differences in prevalence were found between several species and year highlighting species-year temporal trends. CONCLUSIONS: The DPH collaborative response to ZIKV allowed a rapid increase in its surveillance footprint. Existing and new partnerships were developed with the military and local health departments to expand and share data. This additional surveillance data allowed DPH to make sound public health decisions regarding mosquito-borne disease risks and close gaps in data related to vector distribution.