Computer-aided diagnosis of diabetic subjects by heart rate variability signals using discrete wavelet transform method

Acharya, U. Rajendra; Vidya, K. Sudarshan; Ghista, Dhanjoo N.; Lim, Wei Jie Eugene; Molinari, Filippo; Sankaranarayanan, Meena

doi:10.1016/j.knosys.2015.02.005

Cited by 97 publications

(16 citation statements)

References 44 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The Wilcoxon method evaluates the difference between the two correlative samples and is suitable for analyzing two different assessment sets derived from the same data. Bhattacharyya determines divergence between statistical populations by probability distributions; the features are ranked by their capacity to discriminate the training data (Acharya et al, 2015c).…”

Section: Feature Rankingmentioning

confidence: 99%

Early Detection of Sudden Cardiac Death by Using Ensemble Empirical Mode Decomposition-Based Entropy and Classical Linear Features From Heart Rate Variability Signals

Shi

Geng

et al. 2020

Front. Physiol.

View full text Add to dashboard Cite

Sudden cardiac death (SCD), which can deprive a person of life within minutes, is a destructive heart abnormality. Thus, providing early warning information for patients at risk of SCD, especially those outside hospitals, is essential. In this study, we investigated the performances of ensemble empirical mode decomposition (EEMD)-based entropy features on SCD identification. EEMD-based entropy features were obtained by using the following technology: (1) EEMD was performed on HRV beats to decompose them into intrinsic mode functions (IMFs), (2) five entropy parameters, namely Rényi entropy (RenEn), fuzzy entropy (FuEn), dispersion Entropy (DisEn), improved multiscale permutation entropy (IMPE), and Renyi distribution entropy(RdisEn), were computed from the first four IMFs obtained, which were named EEMD-based entropy features. Additionally, an automated scheme combining EEMD-based entropy and classical linear (time and frequency domains) features was proposed with the intention of detecting SCD early by analyzing 14 min (at seven successive intervals of 2 min) heart rate variability (HRV) in signals from a normal population and subjects at risk of SCD. Firstly, EEMDbased entropy and classical linear measurements were extracted from HRV beats, and then the integrated measurements were ranked by various methodologies, i.e., t-test, entropy, receiver-operating characteristics (ROC), Wilcoxon, and Bhattacharyya. Finally, these ranked features were fed into a k-Nearest Neighbor algorithm for classification. Compared with several state-of-the-art methods, the proposed scheme firstly predicted subjects at risk of SCD up to 14 min earlier with an accuracy of 96.1%, a sensitivity of 97.5%, and a specificity of 94.4% 14 min before SCD onset. The simulation results exhibited that EEMD-based entropy estimators showed significant difference between SCD patients and normal individuals and outperformed the classical linear estimators

show abstract

Section: Feature Rankingmentioning

confidence: 99%

Early Detection of Sudden Cardiac Death by Using Ensemble Empirical Mode Decomposition-Based Entropy and Classical Linear Features From Heart Rate Variability Signals

Shi

Geng

et al. 2020

Front. Physiol.

View full text Add to dashboard Cite

show abstract

“…However, due to the nonstationary nature of HRV, valuable information embedded within the signal might not be completely extracted by these conventional methods. Therefore, more advanced analysis such as Time Frequency Distribution (TFD) and Wavelet transform (WT) are proposed [1,18,28,84,125]. However, in studies related to adverse childhood experience and stress reactivity, very few used HRV to measure the stress response and only conventional frequency domain method was used [27,87,129].…”

Section: Stress Classificationmentioning

confidence: 99%

“…A Wavelet analysis provides both time and frequency localizations and the resultant wavelet coefficients can be used as features in classifiers. To decompose the HRV signal, wavelet family ψ a,b , a basis function generated by dilatations and translations of a unique admissible mother wavelet ψ(t) is defined as, (1) where a denotes the scale and b the location [43,63].…”

Section: Wavelet Transform Analysismentioning

confidence: 99%

“…To decompose a signal, the choice of the mother wavelet must be similar to the signal being analysed. In this study, a discrete wavelet transform using the eight-order Daubechies mother wavelet (db8) up to five levels was applied to the HRV to create a five-level wavelet decomposition of the signal [1,43]. The Daubechies wavelet was selected due to the similarity between the wavelet function and the HRV signal [58].…”

Section: Wavelet Transform Analysismentioning

confidence: 99%

See 1 more Smart Citation

Fusion of heart rate variability and salivary cortisol for stress response identification based on adverse childhood experience

Aimie-Salleh

Malarvili

Whittaker

2019

Med Biol Eng Comput

View full text Add to dashboard Cite

Adverse childhood experiences has been suggested to cause changes in physiological processes and can determine the magnitude of the stress response which might have a significant impact on health later in life. To detect the stress response, biomarkers that represent both Autonomic Nervous System (ANS) and Hypothalamic-Pituitary-Adrenal (HPA) axis is proposed. Among the available biomarkers, Heart Rate Variability (HRV) has been proven as a powerful biomarker that represents ANS. Meanwhile, salivary cortisol has been suggested as a biomarker that reflects the HPA axis. Even though many studies used multiple biomarkers to measure the stress response, the results for each biomarker were analysed separately. Therefore, the objective of this study is to propose a fusion of ANS and HPA axis biomarkers in order to classify the stress response based on adverse childhood experience. Electrocardiograph, blood pressure (BP), pulse rate (PR) and salivary cortisol (SCort) measures were collected from 23 healthy participants, 12 participants had adverse childhood experience while the remaining 11 acted as the no adversity control group. HRV was then computed from the ECG and the HRV features were extracted. Next, the selected HRV features were combined with the other biomarkers using Euclidean distance (e d) and serial fusion, and the performance of the fused features were compared using Support Vector Machine. From the result, HRV-SCort using Euclidean distance achieved the most satisfactory performance with 80.0% accuracy, 83.3% sensitivity and 78.3% specificity. Furthermore, the performance of the stress response classification of the fused biomarker, HRV-SCort, outperformed that of the single biomarkers: HRV (61% Accuracy), Cort (59.4% Accuracy), BP (78.3% accuracy) and PR (53.3% accuracy). From this study, it was proven that the fused biomarkers that represent both ANS and HPA (HRV-SCort) able to demonstrate a better classification performance in discriminating the stress response. Furthermore, a new approach for classification of stress response using Euclidean distance and SVM named as e d-SVM was proven to be an effective method for the HRV-SCort in classifying the stress response from PASAT. The robustness of this method is crucial in contributing to the effectiveness of the stress response measures and could further be used as an indicator for future health.

show abstract

“…Recent studies found that classification is the most widely implemented ML task in the medical sector and solutions using the AdaBoost algorithm [22] form a significant subset of the available research. Clinical applications include the diagnosis of Alzheimer's disease, diabetes, hypertension and various cancers [23,24,25,26]. There are also non-clinical assessments of self-reported mental health, and subhealth status.…”

Section: Introductionmentioning

confidence: 99%

Ada-WHIPS: Explaining AdaBoost Classification with Applications in the Health Sciences

Hatwell

Gaber

Azad

2020

Preprint

View full text Add to dashboard Cite

Background Computer Aided Diagnostics (CAD) can support medical practitioners to make critical decisions about their patients' disease conditions. Practitioners require access to the chain of reasoning behind CAD to build trust in the CAD advice and to supplement their own expertise. Yet, CAD systems might be based on black box machine learning (ML) models and high dimensional data sources (electronic health records, MRI scans, cardiotocograms, etc). These foundations make interpretation and explanation of the CAD advice very challenging. This challenge is recognised throughout the machine learning research community. eXplainable Artificial Intelligence (XAI) is emerging as one of the most important research areas of recent years because it addresses the interpretability and trust concerns of critical decision makers, including those in clinical and medical practice. Methods In this work, we focus on AdaBoost, a black box ML model that has been widely adopted in the CAD literature. We address the challenge -- to explain AdaBoost classification -- with a novel algorithm that extracts simple, logical rules from AdaBoost models. Our algorithm, Adaptive-Weighted High Importance Path Snippets (Ada-WHIPS), makes use of AdaBoost's adaptive classifier weights. Using a novel formulation, Ada-WHIPS uniquely redistributes the weights among individual decision nodes of the internal decision trees (DT) of the AdaBoost model. Then, a simple heuristic search of the weighted nodes finds a single rule that dominated the model's decision. We compare the explanations generated by our novel approach with the state of the art in an experimental study. We evaluate the derived explanations with simple statistical tests of well-known quality measures, precision and coverage, and a novel measure stability that is better suited to the XAI setting.Results Experiments on 9 CAD-related data sets showed that Ada-WHIPS explanations consistently generalise better (mean coverage 15%-68%) than the state of the art while remaining competitive for speciﬁcity (mean precision 80%-99%). A very small trade-oﬀ in speciﬁcity is shown to guard againstover-ﬁtting which is a known problem in the state of the art methods.Conclusions The experimental results demonstrate the beneﬁts of using our novel algorithm for explaining CAD AdaBoost classiﬁers widely found in the literature. Our tightly coupled, AdaBoost-speciﬁc approach outperforms model-agnostic explanation methods and should be considered by practitioners looking for an XAI solution for this class of models.

show abstract

Computer-aided diagnosis of diabetic subjects by heart rate variability signals using discrete wavelet transform method

Cited by 97 publications

References 44 publications

Early Detection of Sudden Cardiac Death by Using Ensemble Empirical Mode Decomposition-Based Entropy and Classical Linear Features From Heart Rate Variability Signals

Early Detection of Sudden Cardiac Death by Using Ensemble Empirical Mode Decomposition-Based Entropy and Classical Linear Features From Heart Rate Variability Signals

Fusion of heart rate variability and salivary cortisol for stress response identification based on adverse childhood experience

Ada-WHIPS: Explaining AdaBoost Classification with Applications in the Health Sciences

Contact Info

Product

Resources

About