A sensory-motor hand prosthesis with integrated thermal feedback

BACKGROUND

Recently, we reported the presence of phantom thermal sensations in amputees: thermal stimulation of specific spots on the residual arm elicited thermal sensations in their missing hands. Here, we exploit phantom thermal sensations via a standalone system integrated into a robotic prosthetic hand to provide real-time and natural temperature feedback.

METHODS

The subject (a male adult with unilateral transradial amputation) used the sensorized prosthesis to manipulate objects and distinguish their thermal properties. We tested his ability to discriminate between (1) hot, cold, and ambient temperature objects, (2) different materials (copper, glass, and plastic), and (3) artificial versus human hands. We also introduced the thermal box and block test (thermal BBT), a test to evaluate real-time temperature discrimination during standardized pick-and-place tasks.

FINDINGS

The subject performed all three discrimination tasks above chance level with similar accuracies as with his intact hand. Additionally, in all 15 sessions of the thermal BBT, he correctly placed more than half of the samples. Finally, the phantom thermal sensation was stable during the 13 recording sessions spread over 400 days.

CONCLUSION

Our study paves the way for more natural hand prostheses that restore the full palette of sensations.

FUNDING

This work was funded by the Bertarelli Foundation (including the Catalyst program); the Swiss National Science Foundation through the National Centre of Competence in Research (NCCR) Robotics; the European Union's Horizon 2020 research and innovation program; the Horizon Europe Research & Innovation Program; the Ministry of University and Research (MUR), National Recovery and Resilience Plan (NRRP); and the Tuscany Health Ecosystem.

Exploring the EMG transient: the muscular activation sequences used as novel time-domain features for hand gestures classification

Introduction

Muscular activation sequences have been shown to be suitable time-domain features for classification of motion gestures. However, their clinical application in myoelectric prosthesis control was never investigated so far. The aim of the paper is to evaluate the robustness of these features extracted from the EMG signal in transient state, on the forearm, for classifying common hand tasks.

Methods

The signal associated to four hand gestures and the rest condition were acquired from ten healthy people and two persons with trans-radial amputation. A feature extraction algorithm allowed for encoding the EMG signals into muscular activation sequences, which were used to train four commonly used classifiers, namely Linear Discriminant Analysis (LDA), Support Vector Machine (SVM), Non-linear Logistic Regression (NLR) and Artificial Neural Network (ANN). The offline performances were assessed with the entire sample of recruited people. The online performances were assessed with the amputee subjects. Moreover, a comparison of the proposed method with approaches based on the signal envelope in the transient state and in the steady state was conducted.

Results

The highest performance were obtained with the NLR classifier. Using the sequences, the offline classification accuracy was higher than 93% for healthy and amputee subjects and always higher than the approach with the signal envelope in transient state. As regards the comparison with the steady state, the performances obtained with the proposed method are slightly lower (<4%), but the classification occurred at least 200 ms earlier. In the online application, the motion completion rate reached up to 85% of the total classification attempts, with a motion selection time that never exceeded 218 ms.

Discussion

Muscular activation sequences are suitable alternatives to the time-domain features commonly used in classification problems belonging to the sole EMG transient state and could be potentially exploited in control strategies of myoelectric prosthesis hands.

A Workflow for Studying the Stump-Socket Interface in Persons with Transtibial Amputation through 3D Thermographic Mapping

The design and fitting of prosthetic sockets can significantly affect the acceptance of an artificial limb by persons with lower limb amputations. Clinical fitting is typically an iterative process, which requires patients' feedback and professional assessment. When feedback is unreliable due to the patient's physical or psychological conditions, quantitative measures can support decision-making. Specifically, monitoring the skin temperature of the residual limb can provide valuable information regarding unwanted mechanical stresses and reduced vascularization, which can lead to inflammation, skin sores and ulcerations. Multiple 2D images to examine a real-life 3D limb can be cumbersome and might only offer a partial assessment of critical areas. To overcome these issues, we developed a workflow for integrating thermographic information on the 3D scan of a residual limb, with intrinsic reconstruction quality measures. Specifically, workflow allows us to calculate a 3D thermal map of the skin of the stump at rest and after walking, and summarize this information with a single 3D differential map. The workflow was tested on a person with transtibial amputation, with a reconstruction accuracy lower than 3 mm, which is adequate for socket adaptation. We expect the workflow to improve socket acceptance and patients' quality of life.

Restoration of natural thermal sensation in upper-limb amputees

The use of hands for gathering rich sensory information is essential for proper interaction with the environment; therefore, the restoration of sensation is critical for reestablishing the sense of embodiment in hand amputees. Here, we show that a noninvasive wearable device can be used to provide thermal sensations on amputees' phantom hands. The device delivers thermal stimuli to specific regions of skin on their residual limb. These sensations were phenomenologically similar to those on the intact limbs and were stable over time. Using the device, the subjects could successfully exploit the thermal phantom hand maps to detect and discriminate different thermal stimuli. The use of a wearable device that provides thermal sensation can increase the sense of embodiment and improve life quality in hand amputees.

Primary stability of a press-fit cup in combination with impaction grafting in an acetabular defect model

The objectives of this study were to (a) assess primary stability of a press-fit cup in a simplified acetabular defect model, filled with compacted cancellous bone chips, and (b) to compare the results with primary stability of a press-fit cup combined with two different types of bone graft substitute in the same defect model. A previously developed acetabular test model made of polyurethane foam was used, in which a mainly medial contained defect was implemented. Three test groups (N = 6 each) were prepared: Cancellous bone chips (bone chips), tricalciumphosphate tetrapods + collagen matrix (tetrapods + coll), bioactive glass S53P4 + polyethylene glycol-glycerol matrix (b.a.glass + PEG). Each material was compacted into the acetabulum and a press-fit cup was implanted. The specimens were loaded dynamically in the direction of the maximum resultant force during level walking. Relative motion between cup and test model was assessed with an optical measurement system. At the last load step (3000 N), inducible displacement was highest for bone chips with median [25th percentile; 75th percentile] value of 113 [110; 114] µm and lowest for b.a.glass + PEG with 91 [89; 93] µm. Migration at this load step was highest for b.a.glass + PEG with 868 [845; 936] µm and lowest for tetrapods + coll with 491 [487; 497] µm. The results show a comparable behavior under load of tetrapods + coll and bone chips and suggest that tetrapods + coll could be an attractive alternative to bone chips. However, so far, this was found for one specific defect type and primary stability should be further investigated in additional/more severe defects.

Effect of cup medialization on primary stability of press-fit acetabular cups

BACKGROUND

Appropriate restoration of the native centre of rotation is of paramount importance in total hip arthroplasty. Reconstruction of the centre of rotation depends on reaming technique: conventional approaches require more cup medialization than anatomical preparations. To date, the influence of cup medialization on socket stability in cementless implants is still unknown.

METHODS

Ten cadaveric hemipelvises were sequentially reamed using anatomical technique (only subchondral bone removal with restoration of the native centre of rotation) and conventional preparation (reaming to the lamina and medializing the cup). A biomechanical test was performed on the reconstructions. Implant motions were measured with digital image correlation while a cyclic load of increasing magnitude was applied.

FINDINGS

No significant difference was measured between the two implantation techniques in terms of permanent cup migrations. The only significant difference was found for the cup inducible rotations, where the conventional technique was associated with larger rotations.

INTERPRETATION

Conventional reaming and cup medialization do not improve initial cup stability. Beyond the recently questioned concerns about medialization and hip biomechanics, this is another issue to bear in mind when reaming the acetabulum.

Primary Stability of Revision Acetabular Reconstructions Using an Innovative Bone Graft Substitute: A Comparative Biomechanical Study on Cadaveric Pelvises

Hip implant failure is mainly due to aseptic loosening of the cotyle and is typically accompanied by defects in the acetabular region. Revision surgery aims to repair such defects before implantation by means of reconstruction materials, whose morselized bone graft represents the gold standard. Due to the limited availability of bone tissue, synthetic substitutes are also used. The aim of this study was to evaluate if a synthetic fully resorbable tri-calcium phosphate-based substitute can provide adequate mechanical stability when employed to restore severe, contained defects, in comparison with morselized bone graft. Five cadaveric pelvises were adopted, one side was reconstructed with morselized bone graft and the other with the synthetic substitute, consisting of dense calcium phosphate granules within a collagen matrix. During the biomechanical test, cyclic load packages of increasing magnitude were applied to each specimen until failure. Bone/implant motions were measured through Digital Image Correlation and were expressed in terms of permanent and inducible translations and rotations. The reconstruction types exhibited a similar behavior, consisting of an initial settling trend followed by failure as bone fracture (i.e., no failure of the reconstruction material). When 2.2 Body Weight was applied, the permanent translations were not significantly different between the two reconstructions (p = 0.06–1.0) and were below 1.0 mm. Similarly, the inducible translations did not differ significantly (p = 0.06–1.0) and were below 0.160 mm. Rotations presented the same order of magnitude but were qualitatively different. Overall, the synthetic substitute provided adequate mechanical stability in comparison with morselized bone graft, thus representing a reliable alternative to treat severe, contained acetabular defects.

A method to assess primary stability of acetabular components in association with bone defects

The objectives of this study were to develop a simplified acetabular bone defect model based on a representative clinical case, derive four bone defect increments from the simplified defect to establish a step-wise testing procedure, and analyze the impact of bone defect and bone defect filling on primary stability of a press-fit cup in the smallest defined bone defect increment. The original bone defect was approximated with nine reaming procedures and by exclusion of specific procedures, four defect increments were derived. The smallest increment was used in an artificial acetabular test model to test primary stability of a press-fit cup in combination with bone graft substitute (BGS). A primary acetabular test model and a defect model without filling were used as reference. Load was applied in direction of level walking in sinusoidal waveform with an incrementally increasing maximum load (300 N/1000 cycles from 600 to 3000 N). Relative motions (inducible displacement, migration, and total motion) between cup and test model were assessed with an optical measurement system. Original and simplified bone defect volume showed a conformity of 99%. Maximum total motion in the primary setup at 600 N (45.7 ± 5.6 µm) was in a range comparable to tests in human donor specimens (36.0 ± 16.8 µm). Primary stability was reduced by the bone defect, but could mostly be reestablished by BGS-filling. The presented method could be used as platform to test and compare different treatment strategies for increasing bone defect severity in a standardized way.

Effect of different motor tasks on hip cup primary stability and on the strains in the periacetabular bone: An in vitro study

BACKGROUND

Excessive prosthesis/bone motions and the bone strains around the acetabulum may prevent osteointegration and lead to cup loosening. These two factors depend on post-operative joint loading. We investigated how Walking (which is often simulated) and Standing-Up from seated (possibly more critical) influence the cup primary stability and periacetabular strains.

METHODS

Twelve composite hemipelvises were used in two test campaigns. Simplified loading conditions were adopted to simulate Walking and Standing-Up. For each motor task, a single-direction force was applied in load packages of increasing amplitude. Stable and unstable uncemented cups were implanted. Digital image correlation was used to measure implant/bone motions (three-dimensional translations and rotations, both permanent and inducible), and the strain distribution around the acetabulum.

FINDINGS

When stable implants were tested, higher permanent cranial translations were found during Walking (however the resultant migrations were comparable with Standing-Up); higher rotations were found for Standing-Up. When unstable implants were tested, motions were 1-2 order of magnitude higher. Strains increased significantly from stable to unstable implants. The peak strains were in the superior aspect of the acetabulum during Walking and in the superior-posterior aspect of the acetabulum and at the bottom of the posterior column during Standing-Up.

INTERPRETATION

Different cup migration trends were caused by simulated Walking and Standing-Up, both similar to those observed clinically. The cup mobilization pattern depended on the different simulated motor tasks. Pre-clinical testing of new uncemented cups could include simulation of both motor tasks. Our study could also translate to indication of what tasks should be avoided.

Standardization of hemipelvis alignment for in vitro biomechanical testing

Although in vitro biomechanical tests are regularly performed, the definition of a suitable reference frame for hemipelvic specimens is still a challenge. The aims of the present study were to: (i) define a reference frame for the human hemipelvis suitable for in vitro applications, based on robust anatomical landmarks; (ii) identify the alignment of a hemipelvis based on the alignment of a whole pelvis (including right/left and male/female differences); (iii) identify the relative alignment of the proposed in vitro reference frame with respect to a reference frame commonly used in gait analysis; (iv) create an in vitro alignment procedure easy, robust and inexpensive; (v) quantify the intra-operator repeatability and inter-operator reproducibility of the procedure. A procedure to univocally identify the anatomical landmarks was created, exploiting the in vitro accessibility of the specimen's surface. Through the analysis on 53 CT scans (106 hemipelvises), the alignment of the hemipelvis based on the alignment of a whole pelvis was analyzed: differences between male/female and right/left hemipelvises were not statistically significant To overcome the uncertainty in the identification of the acetabular rim, a standard acetabular plane was defined. An alignment procedure was developed to implement such anatomical reference frame. The intra-operator repeatability and the inter-operator reproducibility were quantified with four operators, on male and female hemipelvises. The intra-operator repeatability was better than 1.5°. The inter-operator reproducibility was better than 2.0°. Alignment in the transverse plane was the most repeatable. The presented procedure to align hemipelvic specimens is sufficiently robust, standardized, and accessible. © 2017 Orthopaedic Research Society. Published by Wiley Periodicals, Inc. J Orthop Res 36:1645-1652, 2018.