Delaram Jarchi

In healthcare applications, often it is not possible to record sufficient data as required for deep learning or data-driven classification and feature detection systems due to the patient condition, various clinical or experimental limitations, or time constraints. On the other hand, data imbalance invalidates many of the test results crucial for clinical approvals. Generating synthetic (artificial or dummy) data has become a potential solution to address this issue. Such data should possess adequate information, properties, and characteristics to mimic the real-world data recorded in natural circumstances. Several methods have been proposed for this purpose, and results often show that adding surrogates improves the decision-making accuracy. This article evaluates the most recent surrogate data generation and data synthesis methods to investigate the effects of the number of surrogates on improving the classification results. It is shown that the data analysis/classification results improve with an increasing number of surrogates, but this no longer continues after a certain number of surrogates. This achievement helps in deciding on the number of surrogates for each strategy, resulting in the alleviation of the computation cost.

Monitoring in-bed pose estimation based on the Internet of Medical Things (IoMT) and ambient technology has a significant impact on many applications such as sleep-related disorders including obstructive sleep apnea syndrome, assessment of sleep quality, and health risk of pressure ulcers. In this research, a new multimodal in-bed pose estimation has been proposed using a deep learning framework. The Simultaneously-collected multimodal Lying Pose (SLP) dataset has been used for performance evaluation of the proposed framework where two modalities including long wave infrared (LWIR) and depth images are used to train the proposed model. The main contribution of this research is the feature fusion network and the use of a generative model to generate RGB images having similar poses to other modalities (LWIR/depth). The inclusion of a generative model helps to improve the overall accuracy of the pose estimation algorithm. Moreover, the method can be generalized for situations to recover human pose both in home and hospital settings under various cover thickness levels. The proposed model is compared with other fusion-based models and shows an improved performance of 97.8% at PCKh@0.5. In addition, performance has been evaluated for different cover conditions, and under home and hospital environments which present improvements using our proposed model.

This paper presents a deep learning system applied for detecting anomalies from respiratory sound recordings. Our system initially performs audio feature extraction using Continuous Wavelet transformation. This transformation converts the respiratory sound input into a two-dimensional spectrogram where both spectral and temporal features are presented. Then, our proposed deep learning architecture inspired by the Inception-residual-based backbone performs the spatial-temporal focusing and multi-head attention mechanism to classify respiratory anomalies. In this work, we evaluate our proposed models on the benchmark SPRSound (The Open-Source SJTU Paediatric Respiratory Sound) database proposed by the IEEE BioCAS 2023 challenge. As regards the Score computed by an average between the average score and harmonic score, our robust system has achieved Top-1 performance with Scores of 0.810, 0.667, 0.744, and 0.608 in Tasks 1-1, 1-2, 2-1, and 2-2, respectively.

The application of machine learning-based tele-rehabilitation faces the challenge of limited availability of data. To overcome this challenge, data augmentation techniques are commonly employed to generate synthetic data that reflect the configurations of real data. One such promising data augmentation technique is the Generative Adversarial Network (GAN). However, GANs have been found to suffer from mode collapse, a common issue where the generated data fails to capture all the relevant information from the original dataset. In this paper, we aim to address the problem of mode collapse in GAN-based data augmentation techniques for post-stroke assessment. We applied the GAN to generate synthetic data for two post-stroke rehabilitation datasets and observed that the original GAN suffered from mode collapse, as expected. To address this issue, we propose a Time Series Siamese GAN (TS-SGAN) that incorporates a Siamese network and an additional discriminator. Our analysis, using the longest common sub-sequence (LCSS), demonstrates that TS-SGAN generates data uniformly for all elements of two testing datasets, in contrast to the original GAN. To further evaluate the effectiveness of TS-SGAN, we encode the generated dataset into images using Gramian Angular Field and classify them using ResNet-18. Our results show that TS-SGAN achieves a significant accuracy increase of classification accuracy (35.2%-42.07%) for both selected datasets. This represents a substantial improvement over the original GAN.

Objectives: Identification of cow lameness is important to farmers to improve and manage cattle health and welfare. No validated tools exist for automatic lameness detection. In this research, we aim to early detect the cow lameness by identifying the instantaneous fundamental gait harmonics from low frequency (16Hz) acceleration signals recorded using leg-worn sensors. Methods: A triaxial accelerometer has been worn on each cow leg. Synchrosqueezed wavelet transform (SSWT) has been applied to acceleration signals to generate the initial time-frequency spectrum related to the gait. This spectrum is given as an input to a designed deep neural network including time-frequency based long short-term memory (LSTM) to estimate instantaneous frequencies at each time point. An inverse SSWT (ISSWT) is then used to recover the gait harmonic and to estimate an enhanced spectrum. Results: Validation of instantaneous frequencies has been provided for each cow leg (combined signals from 23 cows) and the time-series cross validator across the three folds are provided. The average of mean squared errors in frequencies across 3 folds for each leg is obtained as 0.036, 0.033, 0.044 and 0.042 for left-front, right-front, right-back and left-back legs, respectively. Conclusion: Estimation of instantaneous gait frequencies is proved useful for identification of cow gait phases, lameness detection, accurate estimation of gait speed, coherency in movement among the legs and identification of non-gait episodes. Moreover, the proposed method can be used as a new frequency ridge estimation method exploiting SSWT for many other applications.

Accurately diagnosing sleep disorders is essential for clinical assessments and treatments. Polysomnography (PSG) has long been used for detection of various sleep disorders. In this research, electrocardiography (ECG) and electromayography (EMG) have been used for recognition of breathing and movement-related sleep disorders. Bio-signal processing has been performed by extracting EMG features exploiting entropy and statistical moments, in addition to developing an iterative pulse peak detection algorithm using synchrosqueezed wavelet transform (SSWT) for reliable extraction of heart rate and breathing-related features from ECG. A deep learning framework has been designed to incorporate EMG and ECG features. The framework has been used to classify four groups: healthy subjects, patients with obstructive sleep apnea (OSA), patients with restless leg syndrome (RLS) and patients with both OSA and RLS. The proposed deep learning framework produced a mean accuracy of 72% and weighted F1 score of 0.57 across subjects for our formulated four-class problem.