Gait Rhythm Dynamics for Neuro-Degenerative Disease Classification via Persistence Landscape- Based Topological Representation

Topological Gait Analysis: A New Framework and Its Application to the Study of Human Gait

OBJECTIVE

This study introduces a physiologically driven topological gait analysis (TGA) framework to gain insights into pathological gait.

METHODS

A publicly available gait dataset consisting of four groups: healthy adults, people with Parkinson's disease (PD), Huntington's disease (HD), and amyotrophic lateral sclerosis (ALS) was used. The topological properties of the configuration space of three gait parameters were studied by approximating the underlying distribution through a Gaussian kernel-based density estimation technique. Thereafter, sublevel sets of the density estimate were analyzed using cubical persistence homology.

RESULTS

Three new features were constructed: 1. Probability density estimates (PDEs) that characterize the distribution of gait parameters over their configuration space. Healthy adults exhibited a unimodal distribution, while people with neurodegenerative disorders displayed a multi-modal distribution. 2. Persistence entropy plots that summarize changes in the PDEs and characterize the uncertainty in the underlying distribution. Gait of healthy adults was concentrated at higher entropy values as opposed to neurodegenerative gait. 3. A number that captures disease severity trends.

CONCLUSIONS

Topological features in PD and HD indicate a 'bias' to a certain set of gait configurations. This lack of exploration may reflect poor planning of the underlying topology, resulting in outward manifestations of impaired gait. The lower variegations in PDEs in ALS compared to PD and HD suggest that the planning of the topology of gait may occur at higher levels of the neural architecture.

SIGNIFICANCE

TGA offers characterization of gait at a hitherto uncharted level, potentially serving neuromotor markers for early diagnosis and personalized rehabilitation protocols.

A novel multi-task machine learning classifier for rare disease patterning using cardiac strain imaging data

To provide accurate predictions, current machine learning-based solutions require large, manually labeled training datasets. We implement persistent homology (PH), a topological tool for studying the pattern of data, to analyze echocardiography-based strain data and differentiate between rare diseases like constrictive pericarditis (CP) and restrictive cardiomyopathy (RCM). Patient population (retrospectively registered) included those presenting with heart failure due to CP (n = 51), RCM (n = 47), and patients without heart failure symptoms (n = 53). Longitudinal, radial, and circumferential strains/strain rates for left ventricular segments were processed into topological feature vectors using Machine learning PH workflow. In differentiating CP and RCM, the PH workflow model had a ROC AUC of 0.94 (Sensitivity = 92%, Specificity = 81%), compared with the GLS model AUC of 0.69 (Sensitivity = 65%, Specificity = 66%). In differentiating between all three conditions, the PH workflow model had an AUC of 0.83 (Sensitivity = 68%, Specificity = 84%), compared with the GLS model AUC of 0.68 (Sensitivity = 52% and Specificity = 76%). By employing persistent homology to differentiate the "pattern" of cardiac deformations, our machine-learning approach provides reasonable accuracy when evaluating small datasets and aids in understanding and visualizing patterns of cardiac imaging data in clinically challenging disease states.

Information-Based Similarity of Ordinal Pattern Sequences as a Novel Descriptor in Obstructive Sleep Apnea Screening Based on Wearable Photoplethysmography Bracelets

Obstructive sleep apnea (OSA) is a common respiratory disorder associated with autonomic nervous system (ANS) dysfunction, resulting in abnormal heart rate variability (HRV). Capable of acquiring heart rate (HR) information with more convenience, wearable photoplethysmography (PPG) bracelets are proven to be a potential surrogate for electrocardiogram (ECG)-based devices. Meanwhile, bracelet-type PPG has been heavily marketed and widely accepted. This study aims to investigate the algorithm that can identify OSA with wearable devices. The information-based similarity of ordinal pattern sequences (OP_IBS), which is a modified version of the information-based similarity (IBS), has been proposed as a novel index to detect OSA based on wearable PPG signals. A total of 92 PPG recordings (29 normal subjects, 39 mild-moderate OSA subjects and 24 severe OSA subjects) were included in this study. OP_IBS along with classical indices were calculated. For severe OSA detection, the accuracy of OP_IBS was 85.9%, much higher than that of the low-frequency power to high-frequency power ratio (70.7%). The combination of OP_IBS, IBS, CV and LF/HF can achieve 91.3% accuracy, 91.0% sensitivity and 91.5% specificity. The performance of OP_IBS is significantly improved compared with our previous study based on the same database with the IBS method. In the Physionet database, OP_IBS also performed exceptionally well with an accuracy of 91.7%. This research shows that the OP_IBS method can access the HR dynamics of OSA subjects and help diagnose OSA in clinical environments.

Algebraic topology-based machine learning using MRI predicts outcomes in primary sclerosing cholangitis

BACKGROUND

Primary sclerosing cholangitis (PSC) is a chronic cholestatic liver disease that can lead to cirrhosis and hepatic decompensation. However, predicting future outcomes in patients with PSC is challenging. Our aim was to extract magnetic resonance imaging (MRI) features that predict the development of hepatic decompensation by applying algebraic topology-based machine learning (ML).

METHODS

We conducted a retrospective multicenter study among adults with large duct PSC who underwent MRI. A topological data analysis-inspired nonlinear framework was used to predict the risk of hepatic decompensation, which was motivated by algebraic topology theory-based ML. The topological representations (persistence images) were employed as input for classification to predict who developed early hepatic decompensation within one year after their baseline MRI.

RESULTS

We reviewed 590 patients; 298 were excluded due to poor image quality or inadequate liver coverage, leaving 292 potentially eligible subjects, of which 169 subjects were included in the study. We trained our model using contrast-enhanced delayed phase T1-weighted images on a single center derivation cohort consisting of 54 patients (hepatic decompensation, n = 21; no hepatic decompensation, n = 33) and a multicenter independent validation cohort of 115 individuals (hepatic decompensation, n = 31; no hepatic decompensation, n = 84). When our model was applied in the independent validation cohort, it remained predictive of early hepatic decompensation (area under the receiver operating characteristic curve = 0.84).

CONCLUSIONS

Algebraic topology-based ML is a methodological approach that can predict outcomes in patients with PSC and has the potential for application in other chronic liver diseases.

Petri net transition times as training features for multiclass models to support the detection of neurodegenerative diseases

This paper proposes the transition times of Petri net models of human gait as training features for multiclass random forests (RFs) and classification trees (CTs). These models are designed to support screening for neurodegenerative diseases. The proposed Petri net describes gait in terms of nine cyclic phases and the timing of the nine events that mark the transition between phases. Since the transition times between strides vary, each is represented as a random variable characterized by its mean and standard deviation. These transition times are calculated using the PhysioNet database of vertical ground reaction forces (VGRFs) generated by feet-ground contact. This database comprises the VGRFs of four groups: amyotrophic lateral sclerosis, the control group, Huntington's disease, and Parkinson disease. The RF produced an overall classification accuracy of 91%, and the specificities and sensitivities for each class were between 80% and 100%. However, despite this high performance, the RF-generated models demonstrated lack of interpretability prompted the training of a CT using identical features. The obtained tree comprised only four features and required a maximum of three comparisons. However, this simplification dramatically reduced the overall accuracy from 90.6% to 62.3%. The proposed set features were compared with those included in PhysioNet database of VGRFs. In terms of both the RF and CT, more accurate models were established using our features than those of the PhysioNet.

Ventricular Fibrillation and Tachycardia Detection Using Features Derived from Topological Data Analysis

A rapid and accurate detection of ventricular arrhythmias is essential to take appropriate therapeutic actions when cardiac arrhythmias occur. Furthermore, the accurate discrimination between arrhythmias is also important, provided that the required shocking therapy would not be the same. In this work, the main novelty is the use of the mathematical method known as Topological Data Analysis (TDA) to generate new types of features which can contribute to the improvement of the detection and classification performance of cardiac arrhythmias such as Ventricular Fibrillation (VF) and Ventricular Tachycardia (VT). The electrocardiographic (ECG) signals used for this evaluation were obtained from the standard MIT-BIH and AHA databases. Two input data to the classify are evaluated: TDA features, and Persistence Diagram Image (PDI). Using the reduced TDA-obtained features, a high average accuracy near 99% was observed when discriminating four types of rhythms (98.68% to VF; 99.05% to VT; 98.76% to normal sinus; and 99.09% to Other rhythms) with specificity values higher than 97.16% in all cases. In addition, a higher accuracy of 99.51% was obtained when discriminating between shockable (VT/VF) and non-shockable rhythms (99.03% sensitivity and 99.67% specificity). These results show that the use of TDA-derived geometric features, combined in this case this the k-Nearest Neighbor (kNN) classifier, raises the classification performance above results in previous works. Considering that these results have been achieved without preselection of ECG episodes, it can be concluded that these features may be successfully introduced in Automated External Defibrillation (AED) and Implantable Cardioverter Defibrillation (ICD) therapies.

A topological data analysis-based method for gait signals with an application to the study of multiple sclerosis

In the past few years, light, affordable wearable inertial measurement units have been providing to clinicians and researchers the possibility to quantitatively study motor degeneracy by comparing gait trials from patients and/or healthy subjects. To do so, standard gait features can be used but they fail to detect subtle changes in several pathologies including multiple sclerosis. Multiple sclerosis is a demyelinating disease of the central nervous system whose symptoms include lower limb impairment, which is why gait trials are commonly used by clinicians for their patients’ follow-up. This article describes a method to compare pairs of gait signals, visualize the results and interpret them, based on topological data analysis techniques. Our method is non-parametric and requires no data other than gait signals acquired with inertial measurement units. We introduce tools from topological data analysis (sublevel sets, persistence barcodes) in a practical way to make it as accessible as possible in order to encourage its use by clinicians. We apply our method to study a cohort of patients suffering from progressive multiple sclerosis and healthy subjects. We show that it can help estimate the severity of the disease and also be used for longitudinal follow-up to detect an evolution of the disease or other phenomena such as asymmetry or outliers.

Sleep apnea screening based on Photoplethysmography data from wearable bracelets using an information-based similarity approach

BACKGROUND AND OBJECTIVE

Sleep apnea (SA) is a common sleep disorder in daily life and is also an aggravating factor for various diseases. Having the potential to replace traditional but complicated diagnostic equipment, portable medical devices are receiving increasing attention, and thus, the demand for supporting algorithms is growing. This study aims to identify SA with wearable devices.

METHODS

Static information-based similarity (sIBS) and dynamic information-based similarity (dIBS) were proposed to analyze short-term fluctuations in heart rate (HR) with wearable devices. This study included overnight photoplethysmography (PPG) signals from 92 subjects obtained from wearable bracelets.

RESULTS

The results showed that sIBS achieved the highest correlation coefficient with the apnea-hypopnea index (R=-0.653, p=0). dIBS showed a good balance in sensitivity and specificity (75.0% and 72.1%, respectively). Combining sIBS and dIBS with other classical time-frequency domain indices could simultaneously achieve good accuracy and balance (84.7% accuracy, 76.7% sensitivity and 89.6% specificity).

CONCLUSIONS

This research showed that both classic time-frequency domain indices and IBS indices changed significantly only in the severe SA group. This novel method could serve as an effective way to assess SA and provide new insight into its pathophysiology.

Computer-aided identification of degenerative neuromuscular diseases based on gait dynamics and ensemble decision tree classifiers

This study proposes a reliable computer-aided framework to identify gait fluctuations associated with a wide range of degenerative neuromuscular disease (DNDs) and health conditions. Investigated DNDs included amyotrophic lateral sclerosis (ALS), Parkinson's disease (PD), and Huntington's disease (HD). We further performed a statistical and classification comparison elucidating the discriminative capability of different gait signals, including vertical ground reaction force (VGRF), stride duration, stance duration, and swing duration. Feature representation of these gait signals was based on statistical amplitude quantification using the root mean square (RMS), variance, kurtosis, and skewness metrics. We investigated various decision tree (DT) based ensemble methods such as bagging, adaptive boosting (AdaBoost), random under-sampling boosting (RUSBoost), and random subspace to tackle the challenge of multi-class classification. Experimental results showed that AdaBoost ensembling provided a 6.49%, 0.78%, 2.31%, and 2.72% prediction rate improvement for the VGRF, stride, stance, and swing signals, respectively. The proposed approach achieved the highest classification accuracy of 99.17%, sensitivity of 98.23%, and specificity of 99.43%, using the VGRF-based features and the adaptive boosting classification model. This work demonstrates the effective capability of using simple gait fluctuation analysis and machine learning approaches to detect DNDs. Computer-aided analysis of gait fluctuations provides a promising advent to enhance clinical diagnosis of DNDs.