Search

Abstract Background Constraint-induced movement therapy (CIMT) is a prominent neurorehabilitation approach for improving affected upper extremity motor function in children with unilateral cerebral palsy (UCP). However, the restraint of the less-affected upper extremity and intensive training protocol during CIMT may decrease children’s motivation and increase the therapist’s workload and family’s burden. A kinect-based CIMT program, aiming to mitigate the concerns of CIMT, has been developed. The preliminary results demonstrated that this program was child-friendly and feasible for improving upper extremity motor function. However, whether the kinect-based CIMT can achieve better or at least comparable effects to that of traditional CIMT (i.e., therapist-based CIMT) should be further investigated. Therefore, this study aimed to compare the effects of kinect-based CIMT with that of therapist-based CIMT on upper extremity and trunk motor control and on daily motor function in children with UCP. Methods Twenty-nine children with UCP were recruited and randomly allocated to kinect-based CIMT (n = 14) or therapist-based CIMT (n = 15). The intervention dosage was 2.25 h a day, 2 days a week for 8 weeks. Outcome measures, namely upper extremity and trunk motor control and daily motor function, were evaluated before and after 36-h interventions. Upper extremity and trunk motor control were assessed with unimanual reach-to-grasp kinematics, and daily motor function was evaluated with the Revised Pediatric Motor Activity Log. Between-group comparisons of effectiveness on all outcome measures were analyzed by analysis of covariance (α = 0.05). Results The two groups demonstrated similar improvements in upper extremity motor control and daily motor function. In addition, the kinect-based CIMT group demonstrated greater improvements in trunk motor control than the therapist-based CIMT group did (F(1,28) > 4.862, p

Background and Objective: Kidney ultrasound (US) imaging is a significant imaging modality for evaluating kidney health and is essential for diagnosis, treatment, surgical intervention planning, and follow-up assessments. Kidney US image segmentation consists of extracting useful objects or regions from the total image, which helps determine tissue organization and improve diagnosis. Thus, obtaining accurate kidney segmentation data is an important first step for precisely diagnosing kidney diseases. However, manual delineation of the kidney in US images is complex and tedious in clinical practice. To overcome these challenges, we developed a novel automatic method for US kidney segmentation. Methods: Our method comprises two cascaded steps for US kidney segmentation. The first step utilizes a coarse segmentation procedure based on a deep fusion learning network to roughly segment each input US kidney image. The second step utilizes a refinement procedure to fine-tune the result of the first step by combining an automatic searching polygon tracking method with a machine learning network. In the machine learning network, a suitable and explainable mathematical formula for kidney contours is denoted by basic parameters. Results: Our method is assessed using 1380 trans-abdominal US kidney images obtained from 115 patients. Based on comprehensive comparisons of different noise levels, our method achieves accurate and robust results for kidney segmentation. We use ablation experiments to assess the significance of each component of the method. Compared with state-of-the-art methods, the evaluation metrics of our method are significantly higher. The Dice similarity coefficient (DSC) of our method is 94.6 ± 3.4%, which is higher than those of recent deep learning and hybrid algorithms (89.4 ± 7.1% and 93.7 ± 3.8%, respectively). Conclusions: We develop a coarse-to-refined architecture for the accurate segmentation of US kidney images. It is important to precisely extract kidney contour features because segmentation errors can cause under-dosing of the target or over-dosing of neighboring normal tissues during US-guided brachytherapy. Hence, our method can be used to increase the rigor of kidney US segmentation.

Audio-based automatic speech recognition as a hearing aid is susceptible to background noise and overlapping speeches. Consequently, audio-visual speech recognition has been developed to complement the audio input with additional visual information. However, the huge improvement of neural networks in the visual task has resulted in a robust and reliable lip reading framework that can recognize speech from visual input alone. In this work, we propose a lip reading recognition model to predict daily Mandarin conversation and collect a new Daily Mandarin Conversation Lip Reading (DMCLR) dataset, consisting of 1,000 videos from 100 daily conversations spoken by ten speakers. Our model consists of a spatiotemporal convolution layer, a SE-ResNet-18 network, and a back-end module consisting of bi-directional gated recurrent unit (Bi-GRU), 1D convolution, and fully-connected layers. This model is able to reach 94.2% of accuracy in the DMCLR dataset. Such performance makes it possible for Mandarin lip reading applications to be practical in real life. Additionally, we are able to achieve 86.6% and 57.2% accuracy on Lip Reading in the Wild (LRW) and LRW-1000 (Mandarin), respectively. The results show that our method achieves state-of-the-art performance on these two challenging datasets.

Accurate lung segmentation in chest Computed Tomography (CT) scans is a challenging problem because of variations in lung volume shape, susceptibility to partial volume effects that affect thin antero-posterior junction lines, and lack of contrast between the lung and surrounding tissues. To address the need for a robust method for lung segmentation, we present a new method, called Pixel-based two-Scan Connected Component Labeling-Convex Hull-Closed Principal Curve method (PSCCL-CH-CPC), which automatically detects lung boundaries, and surpasses state-of-the-art performance. The proposed method has two main steps: 1) an image preprocessing step to extract coarse lung contours, and 2) a refinement step to refine the coarse segmentation result on the basis of the improved principal curve model and the machine learning model. Experimental results show that the proposed method has good performance, with a Dice Similarity Coefficient (DSC) as high as 98.21%. When compared with state-of-the-art methods, our proposed method achieved superior segmentation results, with an average DSC of 96.9%.

With the advance of deep neural networks (DNNs), artificial intelligence (AI) has been widely applied to various applications in our daily lives. These DNN-based models can be stored in portable storage disks or low-power Neural Compute Sticks. They can then be deployed in edge devices through the USB interface for AI-based applications, such as Automatic Diagnosis Systems or Smart Surveillance Systems, which provides solutions to incorporating AI into the Internet of Things (IoT). In this work, based on our observation and careful analysis, we propose a model-based deep encoding method built upon Huffman coding to compress a DNN model transmitted through the USB interface to edge devices. Based on the proposed lopsidedness estimation approach, we can exploit a modified Huffman coding method to increase the USB transmission efficiency for quantized DNN models while reducing the computational cost entailed by the coding process. We conducted experiments on several benchmarking DNN models compressed using three emerging quantization techniques, which indicates that our method can achieve a high compression ratio of 88.72%, with 93.76% of the stuffing bits saved on average.

Surface ElectroMyoGraphy (sEMG) is widely applied to a variety of applications. Managing the power consumption of battery-constrained sEMG Wireless Body Sensor Networks (WBSN) is an important topic. In this paper, we use fatigue assessments as a case study. We apply the concept of distributed computing to explore the impact of computation allocations on the client power consumption and the requirement of architecture specifications. Regarding the CPU clock rate, we propose a power-saving method based on the ping-pong buffer mechanism and evaluate all the crucial factors which affect the power consumption such as sEMG sample rates, algorithmic computational costs, wireless throughputs, and selection of wireless technologies. To sum up, we conduct a comprehensive analysis of all possible distributed computing architectures of WBSN to determine the lowest-power WBSN architecture. The results show that the implementation based on the lowest-power WBSN architecture has lower power consumption compared with other hardwares. The average current of the proposed architecture can be reduced by 81.7% compared with the previous work. Besides, the battery life is 4.48 times that of the previous work under the continuous wireless connection equipped with the same 300mAh lithium battery. Compared with the commercial device, the battery life is 1.6 times that of the commercial one.

Recently, surface electromyography (sEMG) has been used to detect running-related works. sEMG provides a non-invasive and real-time method that allows quantification of muscle energy. However, noises in sEMG signals are a serious issue to be considered as these will interrupt the analysis of muscular activity. Hence, this work aims at distinguishing between sEMG valid signals and noises during running exercise by taking advantage of the combination of 3D-CNN and LSTM, which we called 3D-LCNN. Furthermore, according to the possible cases that happen in the sEMG data-collection procedure, we proposed two data-augmentation approaches to expend our sEMG dataset, which are the simulation of the surface electrodes displacement on the skin and the muscle fatigue. Experiment results show that the classification accuracy of the proposed 3D-LCNN can achieve 90.52%. Additionally, this work provides excellent service-oriented architecture (SOA). The recognition process can be done after the subject placed the sEMG sensors and performed a trial. Therefore, the process can help clinicians or therapists to distinguish between sEMG valid signals and noises more efficiently.

Introduction: Cerebral palsy (CP) is the leading cause of childhood-onset physical disability. Children with CP often have impaired upper limb (UL) function. Constraint-induced therapy (CIT) is one of the most effective UL interventions for children with unilateral CP. However, concerns about CIT for children have been repeatedly raised due to frustration caused by restraint of the child’s less-affected UL and lack of motivation for the intensive protocol. Virtual reality (VR), which can mitigate the disadvantages of CIT, potentially can be used as an alternative mediator for implementing CIT. Therefore, we developed a VR-based CIT program for children with CP using the Kinect system.Aims: The feasibility of the Kinect-based CIT program was evaluated for children with unilateral CP using a two-phase study design.Materials and Methods: In phase 1, ten children with unilateral CP were recruited. To confirm the achievement of the motor training goals, maximal UL joint angles were evaluated during gameplay. To evaluate children’s perceptions of the game, a questionnaire was used. In phase 2, eight children with unilateral CP were recruited and received an 8 weeks Kinect-based CIT intervention. Performance scores of the game and outcomes of the box and block test (BBT) were recorded weekly.Results: In phase 1, results supported that the design of the program was CIT-specific and was motivational for children with unilateral CP. In phase 2, game performance and the BBT scores began showing stable improvements in the fifth week of intervention.Conclusion: It suggested the Kinect-based CIT program was beneficial to the motor function of the affected UL for children with unilateral CP. According to the results of this feasibility study, larger and controlled effectiveness studies of the Kinect-based CIT program can be conducted to further improve its clinical utility.Clinical Trial Registration: ClinicalTrials.gov, NCT02808195; Comparative effectiveness of a Kinect-based unilateral arm training system vs. CIT for children with CP

This article describes a system for analyzing acoustic data to assist in the diagnosis and classification of children’s speech sound disorders (SSDs) using a computer. The analysis concentrated on identifying and categorizing four distinct types of Chinese SSDs. The study collected and generated a speech corpus containing 2540 stopping, backing, final consonant deletion process (FCDP), and affrication samples from 90 children aged 3–6 years with normal or pathological articulatory features. Each recording was accompanied by a detailed diagnostic annotation by two speech–language pathologists (SLPs). Classification of the speech samples was accomplished using three well-established neural network models for image classification. The feature maps were created using three sets of MFCC (Mel-frequency cepstral coefficients) parameters extracted from speech sounds and aggregated into a three-dimensional data structure as model input. We employed six techniques for data augmentation to augment the available dataset while avoiding overfitting. The experiments examine the usability of four different categories of Chinese phrases and characters. Experiments with different data subsets demonstrate the system’s ability to accurately detect the analyzed pronunciation disorders. The best multi-class classification using a single Chinese phrase achieves an accuracy of 74.4 percent.

Music can generate a positive effect in runners’ performance and motivation. However, the practical implementation of music intervention during exercise is mostly absent from the literature. Therefore, this paper designs a playback sequence system for joggers by considering music emotion and physiological signals. This playback sequence is implemented by a music selection module that combines artificial intelligence techniques with physiological data and emotional music. In order to make the system operate for a long time, this paper improves the model and selection music module to achieve lower energy consumption. The proposed model obtains fewer FLOPs and parameters by using logarithm scaled Mel-spectrogram as input features. The accuracy, computational complexity, trainable parameters, and inference time are evaluated on the Bi-modal, 4Q emotion, and Soundtrack datasets. The experimental results show that the proposed model is better than that of Sarkar et al. and achieves competitive performance on Bi-modal (84.91%), 4Q emotion (92.04%), and Soundtrack (87.24%) datasets. More specifically, the proposed model reduces the computational complexity and inference time while maintaining the classification accuracy, compared to other models. Moreover, the size of the proposed model for network training is small, which can be applied to mobiles and other devices with limited computing resources. This study designed the overall playback sequence system by considering the relationship between music emotion and physiological situation during exercise. The playback sequence system can be adopted directly during exercise to improve users’ exercise efficiency.

In recent years, the surface electromyography (EMG) signal has received a lot of attention. EMG signals are used to analyze muscle activity or to evaluate a patient’s muscle status. However, commercial surface EMG systems are expensive and have high power consumption. Therefore, the purpose of this paper is to implement a surface EMG acquisition system that supports high sampling and ultra-low power consumption measurement. This work analyzes and optimizes each part of the EMG acquisition circuit and combines an MCU with BLE. Regarding the MCU power saving method, the system uses two different frequency MCU clock sources and we proposed a ping-pong buffer as the memory architecture to achieve the best power saving effect. The measured surface EMG signal samples can be forwarded immediately to the host for further processing and additional application. The results show that the average current of the proposed architecture can be reduced by 92.72% compared with commercial devices, and the battery life is 9.057 times longer. In addition, the correlation coefficients were up to 99.5%, which represents a high relative agreement between the commercial and the proposed system.

Hearing aids are increasingly essential for people with hearing loss. For this purpose, environmental noise estimation and classification are some of the required technologies. However, some noise classifiers utilize multiple audio features, which cause intense computation. In addition, such noise classifiers employ inputs of different time lengths, which may affect classification performance. Thus, this paper proposes a model architecture for noise classification, and performs experiments with three different audio segment time lengths. The proposed model attains fewer floating-point operations and parameters by utilizing the log-scaled mel-spectrogram as an input feature. The proposed models are evaluated with classification accuracy, computational complexity, trainable parameters, and inference time on the UrbanSound8k dataset and HANS dataset. The experimental results showed that the proposed model outperforms other models on two datasets. Furthermore, compared with other models, the proposed model reduces model complexity and inference time while maintaining classification accuracy. As a result, the proposed noise classification for hearing aids offers less computational complexity without compromising performance.

Object detection is an important aspect for autonomous driving vehicles (ADV), which may comprise of a machine learning model that detects a range of classes. As the deployment of ADV widens globally, the variety of objects to be detected may increase beyond the designated range of classes. Continual learning for object detection essentially ensure a robust adaptation of a model to detect additional classes on the fly. This study proposes a novel continual learning method for object detection that learns new object class(es) along with cumulative memory of classes from prior learning rounds to avoid any catastrophic forgetting. The results of PASCAL VOC 2007 have suggested that the proposed ER method obtains 4.3% of mAP drop compared against the all-classes learning, which is the lowest amongst other prior arts.

We study the foot plantar sensor placement by a deep reinforcement learning algorithm without using any prior knowledge of the foot anatomical area. To apply a reinforcement learning algorithm, we propose a sensor placement environment and reward system that aims to optimize fitting the center of pressure (COP) trajectory during the self-selected speed running task. In this environment, the agent considers placing eight sensors within a 7 × 20 grid coordinate system, and then the final pattern becomes the result of sensor placement. Our results show that this method (1) can generate a sensor placement, which has a low mean square error in fitting ground truth COP trajectory, and (2) robustly discovers the optimal sensor placement in a large number of combinations, which is more than 116 quadrillion. This method is also feasible for solving different tasks, regardless of the self-selected speed running task.

Maxillary sinuses are the most prevalent locations for paranasal infections on both children and adults. Common diagnostic material for this particular disease is through the screening of occipitomental-view skull radiography (SXR). With the growing cases on paranasal infections, expediting the diagnosis has become an important innovation aspect that could be addressed through the development of a computer-aided diagnosis system. As the preliminary stage of the development, an automatic segmentation over the maxillary sinuses is required to be developed. This study presents a computer-aided detection (CAD) module that segments maxillary sinuses from a plain SXR that has been preprocessed through the novel texture-based morphological analysis (ToMA). Later, the network model from the Transferable Fully Convolutional Network (T-FCN) performs pixel-wise segmentation of the maxillary sinuses. T-FCN is designed to be trained with multiple learning stages, which enables re-utilization of network weights to be adjusted based on newer dataset. According to the experiments, the proposed system achieved segmentation accuracy at 85.70%, with 50% faster learning time.

Automatically locating the lung regions effectively and efficiently in digital chest X-ray (CXR) images is important in computer-aided diagnosis. In this paper, we propose an adaptive pre-processing approach for segmenting the lung regions from CXR images using convolutional neural networks-based (CNN-based) architectures. It is comprised of three steps. First, a contrast enhancement method specifically designed for CXR images is adopted. Second, adaptive image binarization is applied to CXR images to separate the image foreground and background. Third, CNN-based architectures are trained on the binarized images for image segmentation. The experimental results show that the proposed pre-processing approach is applicable and effective to various CNN-based architectures and can achieve comparable segmentation accuracy to that of state-of-the-art methods while greatly expediting the model training by up to 20.74 % and reducing storage space for CRX image datasets by down to 94.6 % on average.

This paper aims to present an algorithm that specifically enhances maxillary sinuses using a novel contrast enhancement technique based on the adaptive morphological texture analysis for occipitomental view radiographs. First, the skull X-ray (SXR) is decomposed into rotational blocks (RBs). Second, each RB is rotated into various directions and processed using morphological kernels to obtain the dark and bright features. Third, a gradient-based block segmentation decomposes the interpolated feature maps into feature blocks (FBs). Finally, the histograms of FBs are equalized and overlaid locally to the input SXR. The performance of the proposed method was evaluated on an independent dataset, which comprises of 145 occipitomental view-based human SXR images. According to the experimental results, the proposed method is able to increase the diagnosis accuracy by 83.45% compared with the computed tomography modality as the gold standard. [ABSTRACT FROM AUTHOR]

Content-addressable memory (CAM) is frequently used in applications, such as lookup tables, databases, associative computing, and networking, that require high-speed searches due to its ability to improve application performance by using parallel comparison to reduce search time. Although the use of parallel comparison results in reduced search time, it also significantly increases power consumption. In this paper, we propose a Block-XOR approach to improve the efficiency of low power precomputation-based CAM (PB-CAM). Through mathematical analysis, we found that our approach can effectively reduce the number of comparison operations by 50% on average as compared with the ones-count approach for 32-bit-long inputs. In our experiment, we used Synopsys Nanosim to estimate the power consumption in TSMC 0.35-sum CMOS technology. Compared with the ones-count PB-CAM system, the experimental results show that our proposed approach can achieve on average 30% in power reduction and 32% in power performance reduction. The major contribution of this paper is that it presents theoretical and practical proofs to verify that our proposed Block-XOR PB-CAM system can achieve greater power reduction without the need for a special CAM cell design. This implies that our approach is more flexible and adaptive for general designs. [ABSTRACT FROM AUTHOR]

The use of deep-submicrometer (DSM) technology increases the capacitive coupling between adjacent wires leading to severe crosstalk noise, which causes power dissipation and may also lead to malfunction of a chip. In this paper, we present a technique that reduces crosstalk noise on instruction buses. While previous research focuses primarily on address buses, little work can be applied efficiently to instruction buses. This is due to the complex transition behavior of instruction streams. Based on instruction sequence profiling, we exploit an architecture that encodes pairs of bus wires and permute them in order to optimize power and noise. A close to optimal architecture configuration is obtained using a genetic algorithm. Unlike previous bus encoding approaches, crosstalk reduction can be balanced with delay and area overhead. Moreover, if delay (or area) is most critical, our architecture can be tailored to add nearly no overhead to the design. For our experiments, we used instruction bus traces obtained from 12 SPEC2000 benchmark programs. The results show that our approach can reduce crosstalk up to 50.79% and power consumption up to 55% on instruction buses. [ABSTRACT FROM AUTHOR]

In this paper, a novel contrast enhancement algorithm is proposed. The proposed approach enhances the contrast without losing the original histogram characteristics which is based on the Histogram Specification technique. It is expected to eliminate the annoying side effects effectively by using the differential information from the input histogram. The experimental results show that the proposed Dynamic Histogram Specification (DHS) algorithm not only keeps the original histogram shape features but also enhances the contrast effectively. Moreover, the DHS algorithm can be applied by simple hardware and processed in real-time system due to its simplicity. [ABSTRACT FROM AUTHOR]