Multiresolution semantic segmentation of biological structures in digital histopathology.

Purpose: Semantic segmentation in high-resolution, histopathology whole slide images (WSIs) is an important fundamental task in various pathology applications. Convolutional neural networks (CNN) are the state-of-the-art approach for image segmentation. A patch-based CNN approach is often employed because of the large size of WSIs; however, segmentation performance is sensitive to the field-of-view and resolution of the input patches, and balancing the trade-offs is challenging when there are drastic size variations in the segmented structures. We propose a multiresolution semantic segmentation approach, which is capable of addressing the threefold trade-off between field-of-view, computational efficiency, and spatial resolution in histopathology WSIs. Approach: We propose a two-stage multiresolution approach for semantic segmentation of histopathology WSIs of mouse lung tissue and human placenta. In the first stage, we use four different CNNs to extract the contextual information from input patches at four different resolutions. In the second stage, we use another CNN to aggregate the extracted information in the first stage and generate the final segmentation masks. Results: The proposed method reported 95.6%, 92.5%, and 97.1% in our single-class placenta dataset and 97.1%, 87.3%, and 83.3% in our multiclass lung dataset for pixel-wise accuracy, mean Dice similarity coefficient, and mean positive predictive value, respectively. Conclusions: The proposed multiresolution approach demonstrated high accuracy and consistency in the semantic segmentation of biological structures of different sizes in our single-class placenta and multiclass lung histopathology WSI datasets. Our study can potentially be used in automated analysis of biological structures, facilitating the clinical research in histopathology applications.

Investigating concurrent validity of inertial sensors to evaluate multiplanar spine movement.

Inertial measurement units (IMUs) offer a portable and inexpensive alternative to traditional optical motion capture systems, and have potential to support clinical diagnosis and treatment of low back pain; however, due to a lack of confidence regarding the validity of IMU-derived metrics, their uptake and acceptance remain a challenge. The objective of this work was to assess the concurrent validity of the Xsens DOT IMUs for tracking multiplanar spine movement, and to evaluate concurrent validity and reliability for estimating clinically relevant metrics relative to gold-standard optical motion capture equipment. Ten healthy controls performed spine range of motion (ROM) tasks, while data were simultaneously tracked from IMUs and optical marker clusters placed over the C7, T12, and S1 vertebrae. Root mean square error (RMSE), mean absolute error (MAE), and intraclass correlation coefficients (ICC2,1) were calculated to assess validity and reliability of absolute (abs; C7, T12, and S1 sensors) and relative joint (rel; intersegmental thoracic, lumbar, and total) motion. Overall RMSEabs = 1.33°, MAEabs = 0.74° ± 0.69, and ICC2,1,abs = 0.953 across all movements, sensors, and planes. Results were slightly better for uniplanar movements when evaluating the primary rotation axis (prim) absolute ROM (MAEabs,prim = 0.56° ± 0.49; ICC2,1,abs,prim = 0.999). Similarly, when evaluating relative intersegmental motion, overall RMSErel = 2.39°, MAErel = 1.10° ± 0.96, and ICC2,1,rel = 0.950, and relative primary rotation axis achieved MAErel,prim = 0.87° ± 0.77, and ICC2,1,rel,prim = 0.994. Findings from this study suggest that these IMUs can be considered valid for tracking multiplanar spine movement, and may be used to objectively assess spine movement and neuromuscular control in clinics.

A Systematic Review of Neurophysiology-Based Localization Techniques Used in Deep Brain Stimulation Surgery of the Subthalamic Nucleus.

OBJECTIVE: This systematic review is conducted to identify, compare, and analyze neurophysiological feature selection, extraction, and classification to provide a comprehensive reference on neurophysiology-based subthalamic nucleus (STN) localization. MATERIALS AND METHODS: The review was carried out using the methods and guidelines of the Kitchenham systematic review and provides an in-depth analysis on methods proposed on STN localization discussed in the literature between 2000 and 2021. Three research questions were formulated, and 115 publications were identified to answer the questions. RESULTS: The three research questions formulated are answered using the literature found on the respective topics. This review discussed the technologies used in past research, and the performance of the state-of-the-art techniques is also reviewed. CONCLUSION: This systematic review provides a comprehensive reference on neurophysiology-based STN localization by reviewing the research questions other new researchers may also have.

Automated detection of microscopic placental features indicative of maternal vascular malperfusion using machine learning.

INTRODUCTION: Hypertensive disorders of pregnancy (HDP) and fetal growth restriction (FGR) are common obstetrical complications, often with pathological features of maternal vascular malperfusion (MVM) in the placenta. Currently, clinical placental pathology methods involve a manual visual examination of histology sections, a practice that can be resource-intensive and demonstrates moderate-to-poor inter-pathologist agreement on diagnostic outcomes, dependant on the degree of pathologist sub-specialty training. METHODS: This study aims to apply machine learning (ML) feature extraction methods to classify digital images of placental histopathology specimens, collected from cases of HDP [pregnancy induced hypertension (PIH), preeclampsia (PE), PE + FGR], normotensive FGR, and healthy pregnancies, according to the presence or absence of MVM lesions. 159 digital images were captured from histological placental specimens, manually scored for MVM lesions (MVM- or MVM+) and used to develop a support vector machine (SVM) classifier model, using features extracted from pre-trained ResNet18. The model was trained with data augmentation and shuffling, with the performance assessed for patch-level and image-level classification through measurements of accuracy, precision, and recall using confusion matrices. RESULTS: The SVM model demonstrated accuracies of 70 % and 79 % for patch-level and image-level MVM classification, respectively, with poorest performance observed on images with borderline MVM presence, as determined through post hoc observation. DISCUSSION: The results are promising for the integration of ML methods into the placental histopathological examination process. Using this study as a proof-of-concept will lead our group and others to carry ML models further in placental histopathology.

Detection and Reconstruction of Poor-Quality Channels in High-Density EMG Array Measurements.

High-density electromyography (HD-EMG) arrays allow for the study of muscle activity in both time and space by recording electrical potentials produced by muscle contractions. HD-EMG array measurements are susceptible to noise and artifacts and frequently contain some poor-quality channels. This paper proposes an interpolation-based method for the detection and reconstruction of poor-quality channels in HD-EMG arrays. The proposed detection method identified artificially contaminated channels of HD-EMG for signal-to-noise ratio (SNR) levels 0 dB and lower with ≥99.9% precision and ≥97.6% recall. The interpolation-based detection method had the best overall performance compared with two other rule-based methods that used the root mean square (RMS) and normalized mutual information (NMI) to detect poor-quality channels in HD-EMG data. Unlike other detection methods, the interpolation-based method evaluated channel quality in a localized context in the HD-EMG array. For a single poor-quality channel with an SNR of 0 dB, the F1 scores for the interpolation-based, RMS, and NMI methods were 99.1%, 39.7%, and 75.9%, respectively. The interpolation-based method was also the most effective detection method for identifying poor channels in samples of real HD-EMG data. F1 scores for the detection of poor-quality channels in real data for the interpolation-based, RMS, and NMI methods were 96.4%, 64.5%, and 50.0%, respectively. Following the detection of poor-quality channels, 2D spline interpolation was used to successfully reconstruct these channels. Reconstruction of known target channels had a percent residual difference (PRD) of 15.5 ± 12.1%. The proposed interpolation-based method is an effective approach for the detection and reconstruction of poor-quality channels in HD-EMG.

A Review of Techniques for Surface Electromyography Signal Quality Analysis.

Electromyography (EMG) signals are instrumental in a variety of applications including prosthetic control, muscle health assessment, rehabilitation, and workplace monitoring. Signal contaminants including noise, interference, and artifacts can degrade the quality of the EMG signal, leading to misinterpretation; therefore it is important to ensure that collected EMG signals are of sufficient quality prior to further analysis. A literature search was conducted to identify current approaches for detecting, identifying, and quantifying contaminants within surface EMG signals. We identified two main strategies: 1) bottom-up approaches for identifying specific and well-characterized contaminants and 2) top-down approaches for detecting anomalous EMG signals or outlier channels in high-density EMG arrays. The best type(s) of approach are dependent on the circumstances of data collection including the environment, the susceptibility of the application to contaminants, and the resilience of the application to contaminants. Further research is needed for assessing EMG with multiple simultaneous contaminants, identifying ground-truths for clean EMG data, and developing user-friendly and autonomous methods for EMG signal quality analysis.

Signal Quality Assessment of Compressively Sensed Electrocardiogram.

OBJECTIVE: Develop a signal quality index (SQI) to determine the quality of compressively sensed electrocardiogram (ECG) by estimating the signal-to-noise ratio (SNR). METHODS: The SQI used random forests, with the ratio of the standard deviations of an ECG segment and a clean ECG and the Wasserstein metric between the amplitude distributions of an ECG segment and a clean ECG, as features. The SQI was tested using the Long-Term Atrial Fibrillation Database (LTAFDB) and the PhysioNet/CinC Challenge 2011 Database Set A (CinCDB). Clean ECG segments from the LTAFDB were corrupted using simulated motion artifact, with preset SNR between -12 dB and 12 dB. The CinCDB was used as-it-is. The databases were compressively sensed using three types of sensing matrices at three compression ratios (50%, 75%, and 95%). For LTAFDB, the RMSE and Spearman correlation between the SQI and the preset SNR were used for evaluation, while for CinCDB, accuracy and F1 score were used. RESULTS: The average RMSE was 3.18 dB and 3.47 dB in normal and abnormal ECG. The average Spearman correlation was 0.94 and 0.93 in normal and abnormal ECG, respectively. The average accuracy and F1 score were 0.90 and 0.88, respectively. CONCLUSION: The SQI determined the quality of compressively sensed ECG and generalized across different databases. There was no consequential effect on the SQI due to abnormal ECG or compression using different sensing matrices and compression ratios. SIGNIFICANCE: Without reconstruction, the SQI can inform which ECG should be analyzed to reduce false alarms due to contamination.

Fully automated estimation of the mean linear intercept in histopathology images of mouse lung tissue.

Purpose: The mean linear intercept (MLI) score is a common metric for quantification of injury in lung histopathology images. The automated estimation of the MLI score is a challenging task because it requires accurate segmentation of different biological components of the lung tissue. Therefore, the most widely used approaches for MLI quantification are based on manual/semi-automated assessment of lung histopathology images, which can be expensive and time-consuming. We describe a fully automated pipeline for MLI estimation, which is capable of producing results comparable to human raters. Approach: We use a convolutional neural network based on U-Net architecture to segment the diagnostically relevant tissue segments in the whole slide images (WSI) of the mouse lung tissue. The proposed method extracts multiple field-of-view (FOV) images from the tissue segments and screen the FOV images, rejecting images based on presence of certain biological structures (i.e., blood vessels and bronchi). We used color slicing and region growing for segmentation of different biological structures in each FOV image. Results: The proposed method was tested on ten WSIs from mice and compared against the scores provided by three human raters. In segmenting the relevant tissue segments, our method obtained a mean accuracy, Dice coefficient, and Hausdorff distance of 98.34%, 98.22%, and 109.68 µ m , respectively. Our proposed method yields a mean precision, recall, and F 1 -score of 93.37%, 83.47%, and 87.87%, respectively, in screening of FOV images. There was substantial agreement found between the proposed method and the manual scores (Fleiss Kappa score of 0.76). The mean difference between the calculated MLI score between the automated method and average rater's score was 2.33 ± 4.13 ( 4.25 % ± 5.67 % ). Conclusion: The proposed pipeline for automated calculation of the MLI score demonstrates high consistency and accuracy with human raters and can be a potential replacement for manual/semi-automated approaches in the field.

Simulating Motion Artifact Using an Autoregressive Model for Research in Biomedical Signal Quality Analysis.

Motion artifact contamination may adversely affect the interpretation of biological signals. The development of algorithms to detect, identify, quantify, and mitigate motion artifact is typically performed using a ground truth signal contaminated with previously recorded motion artifact, or simulated motion artifact. The diversity of available motion artifact recordings is limited, and the rationales for existing models of motion artifact are poorly described. In this paper we developed an autoregressive (AR) model of motion artifact based on data collected from 6 subjects walking at slow, medium, and fast paces. The AR model was evaluated for its ability to generate diverse data that replicated the properties of the experimental data. The simulated motion artifact data was successful at learning key time domain and frequency domain properties, including the mean, variance, and power spectrum of the data, but was ineffective for imitating the morphology and probability distribution of the motion artifact data (kurtosis % error of 100.9-103.6%). More sophisticated models of motion artifact may be necessary to develop simulations of motion artifact.

Transfer Learning for Detection of Atrial Fibrillation in Deterministic Compressive Sensed ECG.

Atrial Fibrillation (AF) is a cardiac condition resulting from uncoordinated contraction of the atria which may lead to an increase in the risk of heart attacks, strokes, and death. AF symptoms may go undetected and may require longterm monitoring of electrocardiogram (ECG) to be detected. Long-term ECG monitoring can generate a large amount of data which can increase power, storage, and the wireless transmission bandwidth of monitoring devices. Compressive Sensing (CS) is compression technique at the sampling stage which may save power, storage, and wireless bandwidth of monitoring devices. The reconstruction of compressive sensed ECG is a computationally expensive operation; therefore, detection of AF in compressive sensed ECG is warranted. This paper presents preliminary results of using deep learning to detect AF in deterministic compressive sensed ECG. MobileNetV2 convolutional neural network (CNN) was used in this paper. Transfer learning was utilized to leverage a pre-trained CNN with the final two layers retrained using 24 records from the Long-Term Atrial Fibrillation Database. The Short-Term Fourier Transform was used to generate spectrograms that were fed to the CNN. The CNN was tested on the MIT-BIH Atrial Fibrillation Database at the uncompressed, 50%, 75%, and 95% compressed ECG. The performance of the CNN was evaluated using weighted average precision (AP) and area under the curve (AUC) of the receiver operator curve (ROC). The CNN had AP of 0.80, 0.70, 0.70, and 0.57 at uncompressed, 50%, 75%, and 95% compression levels. The AUC was 0.87, 0.78, 0.79, and 0.75 at each compression level. The preliminary results show promise for using deep learning to detect AF in compressive sensed ECG.Clinical Relevance-This paper confirms that AF can be detected in compressive sensed ECG using deep learning, This will facilitate long-term ECG monitoring using wearable devices and will reduce adverse complications resulting from undiagnosed AF.

Decision Tree Learning Algorithm for Classifying Knee Injury Status Using Return-to-Activity Criteria.

Anterior cruciate ligament (ACL) injury rates in female adolescents are increasing. Irrespective of treatment options, approximately 1/3 will suffer secondary ACL injuries following their return to activity (RTA). Despite this, there are no evidence-informed RTA guidelines to aid clinicians in deciding when this should occur. The first step towards these guidelines is to identify relevant and feasible measures to assess the functional status of these patients. The purpose of this study was therefore to evaluate tests frequently used to assess functional capacity following surgery using a Reduced Error Pruning Tree (REPT). Thirty-six healthy and forty-two ACLinjured adolescent females performed a series of functional tasks. Motion analysis along with spatiotemporal measures were used to extract thirty clinically relevant variables. The REPT reduced these variables down to two limb symmetry measures (maximum anterior hop and maximum lateral hop), capable of classifying injury status between the healthy and ACL injured participants with a 69% sensitivity, 78% specificity and kappa statistic of 0.464. We, therefore, conclude that the REPT model was able to evaluate functional capacity as it relates to injury status in adolescent females. We also recommend considering these variables when developing RTA assessments and guidelines.Clinical Relevance- Our results indicate that spatiotemporal measures may differentiate ACL-injured and healthy female adolescents with moderate confidence using a REPT. The identified tests may reasonably be added to the clinical evaluation process when evaluating functional capacity and readiness to return to activity.

Concurrent validity of a wearable IMU for objective assessments of functional movement quality and control of the lumbar spine.

Inertial measurement units (IMUs) are being recognized in clinical and rehabilitation settings for their ability to assess movement-related disorders of the spine for better guidance of treatment-planning and tracking of recovery. This study evaluated the Mbientlab MetaMotionR IMUs, relative to Vicon motion capture equipment in measuring local dynamic stability of the spine (quantified using maximum finite-time Lyapunov exponent; λmax), lumbopelvic coordination (quantified using mean absolute relative phase; MARP), and intersegmental motor variability (quantified using deviation phase; DP) of lumbopelvic segments in 10 participants during 35 cycles of repetitive spine flexion-extension (FE). Intraclass correlations were strong between systems when using both the FE angle time-series and the sum of squares (SS) time-series to measure local dynamic stability (0.807 ≤ICC2,1λmax,FE ≤ 0.919; 0.738 ≤ ICC2,1λmax,SS ≤ 0.868), sagittal-plane lumbopelvic coordination (0.961 ≤ICC2,1MARP ≤ 0.963), and sagittal-plane lumbopelvic variability (0.961 ≤ICC2,1DP ≤ 0.963). It was concluded that the MetaMotionR IMUs can be reliably used for measuring features associated with spine movement quality and motor control during a repetitive FE task. Future work will assess the reliability of sensor placement, performance during multi-directional movements, and ability to discern clinical and healthy populations based on assessment of movement quality and control.

Realtime phase-amplitude coupling analysis of micro electrode recorded brain signals.

OBJECTIVE: To demonstrate a method to calculate phase amplitude coupling (PAC) quickly and robustly for realtime applications. METHODS: We designed and implemented a multirate PAC algorithm with efficient filter bank processing and efficient computation of PAC for many frequency-pair combinations. We tested the developed algorithm for computing PAC on simulated data and on intraoperative neural recording data obtained during deep brain stimulation (DBS) electrode implantation surgery. RESULTS: A combination of parallelized frequency-domain filtering and modulation index for PAC estimation provided robust results that could be calculated in real time on modest computing hardware. CONCLUSION: The standard methods for calculating PAC can be optimized for quick and robust performance. SIGNIFICANCE: These results demonstrated that PAC can be extracted in real time and is suitable for neurofeedback applications.

Postoperative real-time electrocardiography monitoring detects myocardial ischemia: a case report.

PURPOSE: This case report outlines the utility and challenges of remote continuous postoperative electrocardiography ECG) monitoring, which is routed through a secure smartphone to provide real-time detection and management of myocardial ischemia. CLINICAL FEATURES: A 42-yr-old male with previous myocardial infarction and angioplasty underwent a radical prostatectomy. At three hours and 45 min postoperatively, remote real-time ECG monitoring was initiated upon the patient's arrival on a regular surgical ward. Monitor alerts were routed to a study clinician's smartphone. About six hours postoperatively, alarms were received and horizontal ST segment depressions were observed. A 12-lead ECG validated the ST segment changes, prompting initiation of a metoprolol iv and a red blood cell transfusion. Approximately seven hours and 30 min postoperatively, the ST segments normalized. The patient was discharged on postoperative day 3 and followed for four years without any sequelae. CONCLUSION: This case report illustrates the use of remote ECG monitoring and clinician response in real time with the use of a smartphone. With each alert, a small ECG strip is transmitted to the smartphone for viewing. In our view, this technology and management system provides a possible means to interrupt myocardial ischemic cascades in real time and prevent postoperative myocardial infarction.

Signal Quality Analysis of Ambulatory Electrocardiograms to Gate False Myocardial Ischemia Alarms.

OBJECTIVE: The objective of this study is to propose and validate an alarm gating system for a myocardial ischemia monitoring system that uses ambulatory electrocardiogram. The PeriOperative ISchemic Evaluation study recommended the selective administration of ß blockers to patients at risk of cardiac events following noncardiac surgery. Patients at risk are identified by monitoring ST segment deviations in the electrocardiogram (ECG); however, patients are encouraged to ambulate to improve recovery, which deteriorates the signal quality of the ECG leading to false alarms. METHODS: The proposed alarm gating system computes a signal quality index (SQI) to quantify the ECG signal quality and rejects alarms with a low SQI. The system was validated by artificially contaminating ECG records with motion artifact records obtained from the long-term ST database and MIT-BIH noise stress test database, respectively. RESULTS: Without alarm gating, the myocardial ischemia monitoring system attained a Precision of 0.31 and a Recall of 0.78. The alarm gating improved the Precision to 0.58 with a reduction of Recall to 0.77. CONCLUSION: The proposed system successfully gated false alarms with future work exploring the misidentification of fiducial points by myocardial ischemia monitoring systems. SIGNIFICANCE: The reduction of false alarms due to the proposed system will decrease the incidence of the alarm fatigue condition typically found in clinicians. Alarm fatigue condition was rated as the top patient safety hazard from 2012 to 2015 by the Emergency Care Research Institute.

The Placental Distal Villous Hypoplasia Pattern: Interobserver Agreement and Automated Fractal Dimension as an Objective Metric.

The distal villous hypoplasia (DVH) pattern is a placental correlate of fetal growth restriction. Because the pattern seems to involve less complexity than do appropriately developed placental villi, we postulated that it may be associated with lower fractal dimension-a mathematical measure of complexity. Our study objectives were to evaluate interobserver agreement related to the DVH pattern among expert pathologists and to determine whether pathologist classification of DVH correlates with fractal dimension. A study set of 30 images of placental parenchyma at ×4 magnification was created by a single pathologist from a digital slide archive. The images were graded for the DVH pattern according to pre-specified definitions and included 10 images graded as "no DVH" (grade  =  0), 10 with mild to moderate DVH (grade  =  1), and 10 with severe DVH (grade  =  2). The images were randomly sorted and shown to a panel of 4 international experts who similarly graded the images for DVH. Weighted kappas were calculated. For each image, fractal dimension was calculated by the Box Counting method. The correlation coefficient between (1) the averaged DVH scores obtained by the 5 pathologists and (2) fractal dimension was calculated. The mean weighted kappa score among the observers was 0.59 (range: 0.42-0.70). The correlation coefficient between fractal dimension and the averaged DVH score was -0.915 (P < 0.001). Expert pathologists achieve fair to substantial agreement in grading DVH, indicating consensus on the definition of DVH. Distal villous hypoplasia correlates extremely well with fractal dimension and represents an objective measure for DVH.

A physical action potential generator: design, implementation and evaluation.

The objective was to develop a physical action potential generator (Paxon) with the ability to generate a stable, repeatable, programmable, and physiological-like action potential. The Paxon has an equivalent of 40 nodes of Ranvier that were mimicked using resin embedded gold wires (Ø = 20 µm). These nodes were software controlled and the action potentials were initiated by a start trigger. Clinically used Ag-AgCl electrodes were coupled to the Paxon for functional testing. The Paxon's action potential parameters were tunable using a second order mathematical equation to generate physiologically relevant output, which was accomplished by varying the number of nodes involved (1-40 in incremental steps of 1) and the node drive potential (0-2.8 V in 0.7 mV steps), while keeping a fixed inter-nodal timing and test electrode configuration. A system noise floor of 0.07 ± 0.01 µV was calculated over 50 runs. A differential test electrode recorded a peak positive amplitude of 1.5 ± 0.05 mV (gain of 40x) at time 196.4 ± 0.06 ms, including a post trigger delay. The Paxon's programmable action potential like signal has the possibility to be used as a validation test platform for medical surface electrodes and their attached systems.

Identification of contaminant type in surface electromyography (EMG) signals.

The ability to recognize various forms of contaminants in surface electromyography (EMG) signals and to ascertain the overall quality of such signals is important in many EMG-enabled rehabilitation systems. In this paper, new methods for the automatic identification of commonly occurring contaminant types in surface EMG signals are presented. Such methods are advantageous because the contaminant type is typically not known in advance. The presented approach uses support vector machines as the main classification system. Both simulated and real EMG signals are used to assess the performance of the methods. The contaminants considered include: 1) electrocardiogram interference; 2) motion artifact; 3) power line interference; 4) amplifier saturation; and 5) additive white Gaussian noise. Results show that the contaminants can readily be distinguished at lower signal to noise ratios, with a growing degree of confusion at higher signal to noise ratios, where their effects on signal quality are less significant.

Resolving the limb position effect in myoelectric pattern recognition.

Reported studies on pattern recognition of electromyograms (EMG) for the control of prosthetic devices traditionally focus on classification accuracy of signals recorded in a laboratory. The difference between the constrained nature in which such data are often collected and the unpredictable nature of prosthetic use is an example of the semantic gap between research findings and a viable clinical implementation. In this paper, we demonstrate that the variations in limb position associated with normal use can have a substantial impact on the robustness of EMG pattern recognition, as illustrated by an increase in average classification error from 3.8% to 18%. We propose to solve this problem by: 1) collecting EMG data and training the classifier in multiple limb positions and by 2) measuring the limb position with accelerometers. Applying these two methods to data from ten normally limbed subjects, we reduce the average classification error from 18% to 5.7% and 5.0%, respectively. Our study shows how sensor fusion (using EMG and accelerometers) may be an efficient method to mitigate the effect of limb position and improve classification accuracy.

A Lempel-Ziv complexity measure for muscle fatigue estimation.

This paper presents a Lempel-Ziv complexity measure for analysis of surface electromyography signals. The Lempel-Ziv measure provides a metric for the number of distinct deterministic patterns and the rate of their creation in signals. We propose a ternary Lempel-Ziv measure, improving upon the binary Lempel-Ziv measure, and making it more suited for the analysis of biological signals. The Lempel-Ziv measure is evaluated with a muscle fatigue experiment in which participants perform static, cyclic, and random contractions. Results show this complexity measure shows a greater correlation to a steadily increasing muscle fatigue level compared to the conventional median frequency. This measure is computationally easy to compute and does not require power spectrum estimation and signal stationarity assumptions.