Human arm joints reconstruction algorithm in rehabilitation therapies assisted by end-effector robotic devices.

BACKGROUND: End-effector robots are commonly used in robot-assisted neuro-rehabilitation therapies for upper limbs where the patient's hand can be easily attached to a splint. Nevertheless, they are not able to estimate and control the kinematic configuration of the upper limb during the therapy. However, the Range of Motion (ROM) together with the clinical assessment scales offers a comprehensive assessment to the therapist. Our aim is to present a robust and stable kinematic reconstruction algorithm to accurately measure the upper limb joints using only an accelerometer placed onto the upper arm. METHODS: The proposed algorithm is based on the inverse of the augmented Jaciobian as the algorithm (Papaleo, et al., Med Biol Eng Comput 53(9):815-28, 2015). However, the estimation of the elbow joint location is performed through the computation of the rotation measured by the accelerometer during the arm movement, making the algorithm more robust against shoulder movements. Furthermore, we present a method to compute the initial configuration of the upper limb necessary to start the integration method, a protocol to manually measure the upper arm and forearm lengths, and a shoulder position estimation. An optoelectronic system was used to test the accuracy of the proposed algorithm whilst healthy subjects were performing upper limb movements holding the end effector of the seven Degrees of Freedom (DoF) robot. In addition, the previous and the proposed algorithms were studied during a neuro-rehabilitation therapy assisted by the 'PUPArm' planar robot with three post-stroke patients. RESULTS: The proposed algorithm reports a Root Mean Square Error (RMSE) of 2.13cm in the elbow joint location and 1.89cm in the wrist joint location with high correlation. These errors lead to a RMSE about 3.5 degrees (mean of the seven joints) with high correlation in all the joints with respect to the real upper limb acquired through the optoelectronic system. Then, the estimation of the upper limb joints through both algorithms reveal an instability on the previous when shoulder movement appear due to the inevitable trunk compensation in post-stroke patients. CONCLUSIONS: The proposed algorithm is able to accurately estimate the human upper limb joints during a neuro-rehabilitation therapy assisted by end-effector robots. In addition, the implemented protocol can be followed in a clinical environment without optoelectronic systems using only one accelerometer attached in the upper arm. Thus, the ROM can be perfectly determined and could become an objective assessment parameter for a comprehensive assessment.

A Motion Analysis Protocol for Kinematic Assessment of Poly-Articulated Prosthetic Hands With Cosmetic Gloves.

To provide upper-limb amputees with devices that best fit their needs and to test innovative solutions, it is necessary to quantitatively appraise a device performance with rigorous measurement methods. The aim of this work was to define an optimal motion analysis protocol, suitable for optoelectronic systems, to measure the kinematics of poly-articulated hands even when covered by a cosmetic glove. This is a fundamental aspect, because gloves can decrease device speed and range of motion and, ultimately, patients' acceptance of the artificial limb. In this work, different mathematical models of the joints and marker-sets for motion analysis were conceived. A regression model to choose a reduced marker-set for studying the hand performance with different cosmetic glove models was developed. The proposed approaches for finger motion analysis were experimentally tested on the index finger of the i-Limb, a commercial myoelectric poly-articulated prosthetic hand, but the results can be easily extended to the whole hand and to other poly-articulated prosthetic hands. The methods proposed for the performance analysis of prosthetic hands points out that the cosmetic gloves imply a reduction of the finger flexion/extension (F/E) angles and of the motion velocity. This draws attention to the need for performing independent cyclic tests on commercial products with various cosmetic solutions to better guide component selection.

Estimation of Human Arm Joints Using Two Wireless Sensors in Robotic Rehabilitation Tasks.

This paper presents a novel kinematic reconstruction of the human arm chain with five degrees of freedom and the estimation of the shoulder location during rehabilitation therapy assisted by end-effector robotic devices. This algorithm is based on the pseudoinverse of the Jacobian through the acceleration of the upper arm, measured using an accelerometer, and the orientation of the shoulder, estimated with a magnetic angular rate and gravity (MARG) device. The results show a high accuracy in terms of arm joints and shoulder movement with respect to the real arm measured through an optoelectronic system. Furthermore, the range of motion (ROM) of 50 healthy subjects is studied from two different trials, one trying to avoid shoulder movements and the second one forcing them. Moreover, the shoulder movement in the second trial is also estimated accurately. Besides the fact that the posture of the patient can be corrected during the exercise, the therapist could use the presented algorithm as an objective assessment tool. In conclusion, the joints' estimation enables a better adjustment of the therapy, taking into account the needs of the patient, and consequently, the arm motion improves faster.

Decoding lower-limb kinematic parameters during pedaling tasks using deep learning approaches and EEG.

Stroke is a neurological condition that usually results in the loss of voluntary control of body movements, making it difficult for individuals to perform activities of daily living (ADLs). Brain-computer interfaces (BCIs) integrated into robotic systems, such as motorized mini exercise bikes (MMEBs), have been demonstrated to be suitable for restoring gait-related functions. However, kinematic estimation of continuous motion in BCI systems based on electroencephalography (EEG) remains a challenge for the scientific community. This study proposes a comparative analysis to evaluate two artificial neural network (ANN)-based decoders to estimate three lower-limb kinematic parameters: x- and y-axis position of the ankle and knee joint angle during pedaling tasks. Long short-term memory (LSTM) was used as a recurrent neural network (RNN), which reached Pearson correlation coefficient (PCC) scores close to 0.58 by reconstructing kinematic parameters from the EEG features on the delta band using a time window of 250 ms. These estimates were evaluated through kinematic variance analysis, where our proposed algorithm showed promising results for identifying pedaling and rest periods, which could increase the usability of classification tasks. Additionally, negative linear correlations were found between pedaling speed and decoder performance, thereby indicating that kinematic parameters between slower speeds may be easier to estimate. The results allow concluding that the use of deep learning (DL)-based methods is feasible for the estimation of lower-limb kinematic parameters during pedaling tasks using EEG signals. This study opens new possibilities for implementing controllers most robust for MMEBs and BCIs based on continuous decoding, which may allow for maximizing the degrees of freedom and personalized rehabilitation.

Realistic Reference for Evaluation of Vehicle Safety Focusing on Pedestrian Head Protection Observed From Kinematic Reconstruction of Real-World Collisions.

Head-to-vehicle contact boundary condition and criteria and corresponding thresholds of head injuries are crucial in evaluation of vehicle safety performance for pedestrian protection, which need a constantly updated understanding of pedestrian head kinematic response and injury risk in real-world collisions. Thus, the purpose of the current study is to investigate the characteristics of pedestrian head-to-vehicle contact boundary condition and pedestrian AIS3+ (Abbreviated Injury Scale) head injury risk as functions of kinematic-based criteria, including HIC (Head Injury Criterion), HIP (Head Impact Power), GAMBIT (Generalized Acceleration Model for Brain Injury Threshold), RIC (Rotational Injury Criterion), and BrIC (Brain Injury Criteria), in real-world collisions. To achieve this, 57 vehicle-to-pedestrian collision cases were employed, and a multi-body modeling approach was applied to reconstruct pedestrian kinematics in these real-world collisions. The results show that head-to-windscreen contacts are dominant in pedestrian collisions of the analysis sample and that head WAD (Wrap Around Distance) floats from 1.5 to 2.3 m, with a mean value of 1.84 m; 80% of cases have a head linear contact velocity below 45 km/h or an angular contact velocity less than 40 rad/s; pedestrian head linear contact velocity is on average 83 ± 23% of the vehicle impact velocity, while the head angular contact velocity (in rad/s) is on average 75 ± 25% of the vehicle impact velocity in km/h; 77% of cases have a head contact time in the range 50-140 ms, and negative and positive linear correlations are observed for the relationships between pedestrian head contact time and WAD/height ratio and vehicle impact velocity, respectively; 70% of cases have a head contact angle floating from 40° to 70°, with an average value of 53°; the pedestrian head contact angles on windscreens (average = 48°) are significantly lower than those on bonnets (average = 60°); the predicted thresholds of HIC, HIP, GAMBIT, RIC, BrIC2011, and BrIC2013 for a 50% probability of AIS3+ head injury risk are 1,300, 60 kW, 0.74, 1,470 × 104, 0.56, and 0.57, respectively. The findings of the current work could provide realistic reference for evaluation of vehicle safety performance focusing on pedestrian protection.