Hand sign recognition is gaining importance in human-human and in human-machine communication and interaction. Electrical Impedance Tomography (EIT) is thereby very interesting as it provides information on impedance changes in depth of the Section of the arm, which infers muscle contractions. This paper introduces an EIT imaging system for hand sign recognition and monitoring having a low complexity and including an electronic interface with 8 electrodes placed on the forearm, a Gauss-Newton image reconstruction algorithm, a robust CNN based hand sign classification and a virtual hand model for visualization. A database has been collected for EIT measurements in pole mode taken by eight subjects performing the American sign language numbers from 0 to 9. The overall imaging system is validated using a water tank system, where conductive objects can be changed in properties and positions. The correspondence between the reconstructed images and the expected muscle behavior for the hand signs is investigated. A robust Convolutional Neural Network (CNN) classification algorithm was implemented and optimized by implementing an Adam optimizer and conducting a dedicated study to avoid overfitting. The results obtained by CNN are compared to the results by a Support Vector Machine (SVM), and a Softmax classifier. They show a classification accuracy of 95.94%, 75.61%, and 62.9% respectively. In term of subject dependency, the system using the CNN model shows a higher performance, as the accuracy decreases only by 0.72% while increasing the number of subjects from one to eight. Finally, for visualization, a 3D virtual hand model is designed and controlled based on detected hand signs.

Bioimpedance spectroscopy (BIS) is an advanced measurement method for providing information on impedance changes at several frequencies by injecting a low current into a device under test and analyzing the response voltage. Several methods have been elaborated for BIS measurement, calculating impedance with a gain phase detector (GPD), IQ demodulation, and fast Fourier transform (FFT). Although the measurement method has a big influence on the measurement system performance, a systematical comparative study has not been performed yet. In this paper, we compare them based on simulations and experimental studies. To maintain similar conditions in the implementation of all methods, we use the same signal generator followed by a voltage-controlled current source (VCCS) as a signal generator. For performance analysis, three DUTs have been designed to imitate the typical behavior of biological tissues. A laboratory impedance analyzer is used as a reference. The comparison addresses magnitude measurement accuracy, phase measurement accuracy, signal processing, hardware complexity, and power consumption. The result shows that the FFT-based system excels with high accuracy for amplitude and phase measurement while providing the lowest hardware complexity, and power consumption, but it needs a much higher software complexity.

This paper presents the design and fabrication of Fil3D-Hand, a humanoid robotic hand with five fingers integrating nanocomposite filament strain sensors to monitor finger movements. The objective is to replicate the functionality of the human hand through an affordable design, with potential applications in humanoid robotics and prosthetic rehabilitation. Based on the investigation of the human hand’s anatomical structure and existing robotic hands, the mechanical design of Fil3D-Hand is elaborated. The kinematics analysis and finger workspace are formulated and illustrated using the Denavit-Hartenberg (D-H) convention and numerical simulations. The proposed design features 17 degrees of freedom, weighs 2.4 kg (including the forearm and 12 servo motors), and achieves a fabrication cost of less than $350. The Fil3D hand’s size is approximately 1.2 the size of the real human hand, allowing for enhanced versatility while maintaining a naturalistic appearance. This design strategy ensures a cost-effective, tendon-driven underactuated mechanism, utilizing servo motors for precise control of finger joints. It also integrates 14 filament strain sensors for accurate monitoring of finger positions. Experimental results demonstrate the performance of these sensors, revealing a highly linear correlation ( $$ R^2 = 0.9964 $$R 2 = 0.9964 ) between the finger’s proximal angles and the strain sensor resistance, guaranteeing precise real-time monitoring of finger movements. Evaluation of the Fil3D-Hand’s performance highlights its excellent grasping capabilities, showcasing its potential in practical applications.

A new version of a bioimpedance transducer is proposed for non-invasive monitoring the insulin bioavailability after subcutaneous administration. Both the firmware and the hardware were upgraded in order to improve safety, wearability and metrological performances of the device. As far as the firmware upgrade is concerned, a new block to manage a Bluetooth low energy 5.0 channel for data transmission was implemented. As far as the hardware upgrade is concerned, a shortening of the cables connecting the measurement system to the measurand was realized to minimize the inductive effect. Moreover, a feasibility of a new block to be added in the architecture of the bioimpedance meter was realized. In particular, the new block is able to identify and calibrate the residual cable effect on impedance measurements, by means of a builder of equivalent electrical circuits. After presenting the hardware, the firmware, and the software of the upgraded device, the metrological characterization during (i) laboratory (on passive electrical components), and (ii) in-vivo (on an human subject) experiments is reported. In laboratory tests, a mean percentage deterministic error of 0.20 % and 0.19 % was obtained for impedance magnitude and phase, respectively and an improvement of one order of magnitude was achieved respect to the previous version of the transducer. Regarding the in vivo-measurements, in addition to the impedance module trend already available with the old version of the transducer, it was possible to acquire a phase diagram in accordance with literature.

This study investigates the potential of impedimetric measurements providing deep muscle information for recognizing 36 American Sign Language (ASL) hand signs. Two measurement methods are considered together for the first time: electrical impedance myography (EIM) and electrical impedance tomography (EIT). EIM was measured along the anterior forearm, while 8-electrode EIT was recorded around the forearm below the elbow. Data were acquired from three volunteers, with each hand sign performed ten times. A correlation analysis was conducted to identify the relevant EIM frequencies to distinguish between hand signs. Among all evaluated algorithms, the random forest classifier achieves the highest classification performance. Classification based on the resistance and reactance at the selected EIM frequencies achieved 61.54 ± 0.85%, while classification based on EIT boundary voltages achieved 91.04% ± 0.46%. Combining the results from both classifiers into an EIM-EIT hybrid classifier improved the accuracy to 92.57 ± 0.41%, effectively reducing ambiguities between similar hand signs. Achieved results considerably outperform state-of-the-art works, which typically classify fewer hand signs or achieve lower accuracy.

This paper presents a framework for hand gesture recognition based on the information fusion from a body-attached multisensor technology comprising a smart glove, a smart band, and an inertial measurement unit (IMU). The smart glove is integrated with nanocomposite filament strain sensors developed from carbon nanotubes (CNT) dispersed in thermoplastic polyurethane (TPU) and extruded as filaments using a micro-compounder. The smart band is integrated with nanocomposite pressure sensors developed from CNTs dispersed in a silicone polymer, polydimethylsiloxane (PDMS), and deposited as a thin sheet that is cut as circular disks and coupled with underlying interdigital electrodes. The (IMU) comprising of a three-axis accelerometer, a three-axis gyroscope, and a three-axis magnetometer and is fabricated along with the sensor interface and single processing unit. The paper elaborates on the development of the nanocomposite sensors and the performance of these three sensor technologies by studying ten American Sign Language (ASL) gestures representing numbers 1 to 10 without the involvement of a sophisticated machine learning algorithm for gesture classification. The outcome of this investigation provides valuable insights into the performance of the individual sensor technology in comparison to their counterparts and sets guidelines for further development of such body-attached systems in terms of the type and selection of sensor technology and the required number of sensors. The hardware architecture of the developed framework can be directly implemented as a potential tool for human–machine interface-related activities.