The state of health (SOH) of lithium-ion batteries (LIBs) is a critical parameter of the battery management system. Because of the complex internal electrochemical properties of LIBs and uncertain external working environment, it is difficult to achieve an accurate SOH determination. In this study, we proposed a novel SOH estimation method by using prior knowledge-based neural network (PKNN) and Markov chain for a single lithium-ion battery. Firstly, we extract multiple features to capture the battery aging process. Due to its effective fitting ability for complex nonlinear problems, the neural network with prior knowledge-based optimization strategy is adopted for the battery SOH prediction. The Markov chain, with the advantageous prediction performance for the long-term system, is established to modify the PKNN estimation results based on the prediction error. Experimental results show that the maximum estimation error of the state of health is reduced to less than 1.7% by adopting the proposed method. By comparing with the Group Method of Data Handling (GMDH) and back propagation neural network in conjunction with Levenberg-Marquardt algorithm (LMNN), the proposed estimation method obtains the highest SOH accuracy.

Quantitative assessment of parkinsonian tremor based on inertial sensors can provide reliable feedback on the effect of medication. In this regard, the features of parkinsonian tremor and its unique properties such as motor fluctuations and dyskinesia are taken into account. Least-square-estimation models are used to assess the severities of rest, postural, and action tremors. In addition, a time-frequency signal analysis algorithm for tremor state detection was also included in the tremor assessment method. This inertial sensor-based method was verified through comparison with an electromagnetic motion tracking system. Seven Parkinson’s disease (PD) patients were tested using this tremor assessment system. The measured tremor amplitudes correlated well with the judgments of a neurologist (r = 0.98). The systematic analysis of sensor-based tremor quantification and the corresponding experiments could be of great help in monitoring the severity of parkinsonian tremor.

Magneto-actuated soft robots have broad prospects for biomedical applications, such as cell manipulation, drug delivery, and noninvasive intervention. However, the single function and unchangeable stiffness restrict their practicality. Therefore, the work presented a functionally reconfigurable soft robot based on Gallium/NdFeB/PDMS composites. A heating-free reprogrammable magnetization technique was proposed to regulate the robot's response actions through re-patterning the magnetic domains. Moreover, a stiffness programming technique in terms of the liquid-metal thermotropic phase transition was employed to regulate the robot's stiffness. Four robot prototypes were constructed and tested. Experimental results showed that the robot performed variable responses and changeable stiffness, which satisfied different manipulation tasks. Therefore, the soft robot was endowed with variable functions and reusable characteristics. The proposed technique is expected to be applied to the batch manufacturing of intelligent soft machines.

The position and orientation of an object embedding with a permanent magnet can be acquired in real time via a magnetic tracking system. However, one characteristic of the magnetic tracking technique is its varied tracking accuracy along with the tracking distance. Hence, this paper tends to investigate the relationship between the tracking accuracy and the distance from the magnet to the sensor array by both simulations and experiments. Results show that the relationship is expressed as a cubic polynomial equation, and that the equation coefficients are related to the properties of the magnet and the signal-to-noise ratio of the sensor outputs. When the magnet is located at a distance from 36 to 96 mm above the sensor array, the system had the best performance, while average localization and orientation errors were 0.70 mm and 1.22°, respectively. Thus, this system achieved the best tracking accuracy compared with the state-of-the-art of magnetic tracking systems. This paper is helpful to the researchers who want to implement a magnetic tracking system or need to know clearly the valid tracking distance of a magnetic tracking system.

As the most characteristic feature of Parkinson’s disease (PD), bradykinesia (slowness of movement) affects all patients with Parkinson’s disease and interferes with their daily activities. This study introduces a wearable bradykinesia assessment system whose core component is composed of an inertial measurement unit. The system diagram and assessment task were defined in accordance with clinical requirements from neurologists. Based on hand grasping actions, calculations of hand grasping ranges and statistical methods of quantitatively assessing parkinsonian bradykinesia were presented. Seven control subjects and eight patients were tested with this system. Experimental results show that a calculated bradykinesia parameter (modified mean range, instead of mean and standard deviation of the grasp ranges) correlated well with the evaluations of a neurologist (Pearson’s correlation coefficient r = −0.83, p < 0.001). The bradykinesia assessment system was tested on both health subjects and PD patients. The results show that this system has greater correlation with the evaluations by neurologists than other parkinsonian bradykinesia assessment systems. The modified mean range was verified as the major bradykinesia parameter (key indicator). This study is helpful to those who want to use consumer-grade inertial sensors for quantitative assessment of motor symptoms during treatment.

Neurologists judge the severity of Parkinsonian motor symptoms according to clinical scales, and their judgments exist inconsistent because of differences in clinical experience. Correspondingly, inertial sensing-based wearable devices (ISWDs) produce objective and standardized quantifications. However, ISWDs indirectly quantify symptoms by parametric modeling of angular velocities and linear accelerations and trained by the judgments of several neurologists through supervised learning algorithms. Hence, the ISWD outputs are biased along with the scores provided by neurologists. To investigate the effectiveness ISWDs for Parkinsonian symptoms quantification, technical verification and clinical validation of both tremor and bradykinesia quantification methods were carried out. A total of 45 Parkinson's disease patients and 30 healthy controls performed the tremor and finger-tapping tasks, which were tracked simultaneously by an ISWD and a 6-axis high-precision electromagnetic tracking system (EMTS). The Unified Parkinson's Disease Rating Scale (UPDRS) prescribed parameters obtained from the EMTS, which directly provides linear and rotational displacements, were compared with the scores provided by both the ISWD and seven neurologists. EMTS-based parameters were regarded as the ground truth and were employed to train several common machine learning (ML) algorithms, i.e., support vector machine (SVM), k-nearest neighbors (KNN), and random forest (RF) algorithms. Inconsistency among the scores provided by the neurologists was proven. Besides, the quantification performance (sensitivity, specificity, and accuracy) of the ISWD employed with ML algorithms were better than that of the neurologists. Furthermore, EMTS can be utilized to both modify the quantification algorithms of ISWDs and improve the assessment skills of young neurologists.