Search

change of blood glucose (BG) level stimulates the autonomic nervous system leading to variation in both human’s electrocardiogram (ECG) and photoplethysmogram (PPG). In this article, we aimed to construct a novel multimodal framework based on ECG and PPG signal fusion to establish a universal BG monitoring model. This is proposed as a spatiotemporal decision fusion strategy that uses weight-based Choquet integral for BG monitoring. Specifically, the multimodal framework performs three-level fusion. First, ECG and PPG signals are collected and coupled into different pools. Second, the temporal statistical features and spatial morphological features in the ECG and PPG signals are extracted through numerical analysis and residual networks, respectively. Furthermore, the suitable temporal statistical features are determined with three feature selection techniques, and the spatial morphological features are compressed by deep neural networks (DNNs). Lastly, weight-based Choquet integral multimodel fusion is integrated for coupling different BG monitoring algorithms based on the temporal statistical features and spatial morphological features. To verify the feasibility of the model, a total of 103 days of ECG and PPG signals encompassing 21 participants were collected in this article. The BG levels of participants ranged between 2.2 and 21.8 mmol/L. The results obtained show that the proposed model has excellent BG monitoring performance with a root-mean-square error (RMSE) of 1.49 mmol/L, mean absolute relative difference (MARD) of 13.42%, and Zone A + B of 99.49% in tenfold cross-validation. Therefore, we conclude that the proposed fusion approach for BG monitoring has potentials in practical applications of diabetes management.

The development of procedural competencies by interventional cardiology fellows for minimally invasive treatment of cardiac diseases depends on their effective training and evaluation. Learning tool manipulation for safe robotic PCIs requires expert supervision and use of high-fidelity systems, however, with limited proficiency for real-time hand motion recognition. Therefore, this study proposes a deep learning-based model for identifying operators’ actions. The proposed model is based on the convolutional neural network (CNN) that consists of 1-D convolutional layers for automatic feature map extraction, downsampling, and fully connected layers (FCLs) for inference. The developed models were evaluated using a multimodal dataset collected from sensory glove, electromagnetic (EM), and surface electromyography (sEMG) sensors and real-time angiograms in the course of in vivo catheterization trials performed by nine subjects (two experts and seven novices) using a custom-built robotic catheter system (RCS). The results indicate that the model achieves a 92%–96% accuracy in identifying five actions across four clusters compared with a recognition performance of 83% when recognizing all six actions. Furthermore, we compared the proposed model performance with the existing studies, and the analyses show that our model has a 2%–3% higher accuracy for five-action recognition. Therefore, the proposed model could be employed for real-time hand motion recognition in robot-assisted percutaneous coronary intervention (R-PCI) trials.

Robot-assisted intravascular interventions have evolved as unique treatments approach for cardiovascular diseases. However, the technology currently has low potentials for catheterization skill evaluation, slow learning curve, and inability to transfer experience gained from manual interventions. This study proposes a new weighting-based deep ensemble model for recognizing interventionalists' hand motions in manual and robotic intravascular catheterization. The model has a module of neural layers for extracting features in electromyography data, and an ensemble of machine learning methods for classifying interventionalists' hand gestures as one of the six hand motions used during catheterization. A soft-weighting technique is applied to guide the contributions of each base learners. The model is validated with electromyography data recorded during in-vitro and in-vivo trials and labeled as many-to-one sequences. Results obtained show the proposed model could achieve 97.52% and 47.80% recognition performances on test samples in the in-vitro and in-vivo data, respectively. For the latter, transfer learning was applied to update weights from the in-vitro data, and the retrained model was used for recognizing the hand motions in the in-vivo data. The weighting-based ensemble was evaluated against the base learners and the results obtained shows it has a more stable performance across the six hand motion classes. Also, the proposed model was compared with four existing methods used for hand motion recognition in intravascular catheterization. The results obtained show our model has the best recognition performances for both the in-vitro and in-vivo catheterization datasets. This study is developed toward increasing interventionalists' skills in robot-assisted catheterization.

In this study, a 1-D distal force sensor design is proposed for minimally invasive surgery (MIS) based on the fiber Bragg grating (FBG) sensing principle. The sensor is made with a miniaturized force-sensitive flexure within which an optical fiber is embedded. The former includes a spiral elastomer utilized for high axial force sensitivity, while the fiber includes dual FBGs that are tightly suspended around the distal parts of the flexure and along elastomer’s centerline, respectively. The theoretical model of the flexure was derived based on its components and a design approach for decoupling strain and temperature cross effects during sensor usage. In addition, design parameters of the flexure were analyzed on physics-based model optimization using a fmincon function, while the static and dynamic properties were analyzed in-silico using the finite-element method. Further experiments were also carried out to analyze the sensor’s performance, and the data obtained was used for calibration study. The latter was done by training a feed-forward neural network designed for predicting the force versus wavelength-shift relationship. The experimental results show the designed force sensor can sense force values in a range of 1 N with an average relative error less than 2% of full scale. Ex-vivo tissue palpation study was also done on a pig liver to verify the effectiveness and applicability of the proposed sensor.

E-Commerce has become very popular these days because it is convenient, reliable, and fast to use. In spite of these advantages, online buyers often experience difficulty in searching for products on the web, while online businesses are often overwhelmed by the rich data they have collected and find it difficult to promote products appropriate to specific customers. This paper proposes a hybrid recommender system that uses fuzzy logic to intelligently mine the requirements of each specific customer, together with some previous users' opinions about the product, to recommend a list of optimal products to meet users' needs. Experimental results of the proposed system with different brands of laptops prove its effectiveness.

Flexible pressure sensors have attracted increasing attention because they can mimic human skin to sense external pressure; however, for mimicking human skin, the sensing of a pressure point is far from sufficient. To realize fully biomimetic skins, it is crucial for flexible sensors to have high resolution and high sensitivity. We conducted simulations and experiments to determine the relationship between the sensor sensitivity and physical parameters, such as the effective relative permittivity and air ratio of the dielectric layer. According to the results, a micropillar–poly(vinylidene fluoride) (PVDF) dielectric layer was designed to achieve high sensitivity (0.43 kPa–1) in the low-pressure regime (<1 kPa). An 8 × 8 pixel sensor matrix was prepared based on a micropillar–PVDF (MP) film and electrode array (MPEA) to detect the pressure distribution with high resolution (13 dpi). Each pixel could reflect the point of applied pressure through an obvious change in the relative capacitance; moreover, objects with various geometries could be mapped by the pixels of the flexible sensor. A counterweight, a plastic flag, and pine leaves were placed on the flexible sensor, and the shapes were successfully mapped; in particular, the mapping of the ∼0.005 g ultra-lightweight pine leaves with a length of 7 mm and a width of 0.6 mm shows the high sensitivity and high resolution of our flexible pressure sensor.

This paper studies the reconstruction of the spatial-dependent-heat transfer coefficient γ (x) and- heat flux q (x) , using the Levenberg–Marquardt (L–M) connected with the surrogate functional method and the modified conjugate gradient method (MCGM) (problem I). In addition to studying the reconstruction of γ (x) and space–time-dependent heat flux q (x , t) using the MCGM (problem II). The numerical results obtained by the L–M and MCGM demonstrate the efficiency, accuracy, and robustness of the proposed methods. The obtained results are compared with those obtained from an inverse approach connected with COMSOL Multiphysics under the same conditions. Finally, a conclusion and some remarks are presented. [ABSTRACT FROM AUTHOR]