The somatosensory system has a primary role in daily life, since it provides important information for the human interaction with the surrounding environment. The loss of this information in hand amputees using passive prostheses can highly worsen manipulation activities. Thus, the integration of somatosensory feedback in closed-loop prostheses represents a significant advancement toward the restoration of natural functions of prosthetic hands. The potential benefits are multiple, including: the possibility to enhance the overall quality of life by providing amputees with a pleasant experience; the substantial enhancement of users’ self-esteem and independence; the improvement of performance, in terms of finer and more precise movements, which are useful for tasks needing high dexterity such as grasping small objects or handling delicate instruments. To develop closed-loop prostheses embedding somatosensory functions, several investigations on the human natural functions and on devices capable of artificially replicating them are needed. To this aim, this work was devoted to design and develop a modular and versatile mechatronic testbed to conduct heterogeneous somatosensory studies in humans, prostheses and in-vitro sensing. In literature, multiple tests, based on different devices and methods, have been performed not only on humans but also on animals and ex-vivo models. Depending on the nature of the sample under analysis and on the scientific aims of interest, several solutions for experimental stimulation and for investigations on sensation or pain have been adopted. A deep analysis of the available devices and methods has been performed, also analyzing the representative values adopted during literature experiments. Based on the analysis of their main features and on literature studies, the most suitable solution for humans, rodents, and ex-vivo models and investigation aims (sensation and pain) was pointed out. This analysis revealed a lack of a single device that includes both mechanical and thermal stimulators, suitable for use in both in-vivo and ex-vivo models. It is therefore evident that the development of the mechatronic testbed, which is the objective of this thesis, was necessary to address the aforementioned gap in the literature. The results obtained from the analysis of the state of the art were integrated with the ones coming from the commercially available solutions. Based on the final need of the novel mechatronic testbed to be applied in somatosensory studies on in-vivo and in-vitro models, the technical specifications were defined. The main specifications to be taken into account were: i) a positioning system with at least 2 degrees of freedom and a minimum displacement in each direction of 100 mm, ii) a mechanical stimulator able to apply at least a purely normal force and controllable in force (optionally in position), iii) a thermal stimulator based on a Peltier element (thus able to perform both hot and cold stimuli within a single device) with a temperature range of 0 – 50 °C. Commercial components have been selected and adjusted/assembled to meet the technical specification requirements previously defined. The final developed setup allows for the accurate delivery and measurement of mechanical and thermal stimuli of interest with high spatial resolution (0.5 μ m). Compared with state-of-the-art solutions, it allows automated stimulation, overcoming the disadvantages associated with the manual application of the stimuli. Moreover, the mechanical and thermal stimulators included in the setup are highly modular, allowing the use of multiple ending tips (i.e. diameter of 0.5 – 3 mm for the mechanical and of 2 – 3 mm for the thermal) and the application of a wide range of force/ temperature (0.3 – 5 N/ 5 – 60 °C). Thanks to its high versatility, the testbed was employed in two different scenarios: i) identification of the human somatosensory mechanical/thermal sensation and pain thresholds to be used as feedback in hand prostheses, ii) assessment of the performance of thermal sensors in realistic conditions and selection of the most suitable one for the integration in hand prostheses. The developed testbed was first employed to identify the mechanical/thermal innocuous and painful thresholds, as well as the human ability to distinguish the nature of a painful stimulus, on both the hand and the forearm of 12 healthy volunteers. The obtained mechanical thresholds showed no statistically significant difference when comparing the two stimulated anatomical sites. Furthermore, a statistically significant difference was found between mechanical innocuous and painful thresholds on both sites. When compared with the results in the literature, a difference between the presented continuously increasing mechanical stimulus and the discrete one provided by state-of-the-art equipment was found. The identified mechanical painful thresholds were found to be consistent with literature results, also confirming the dependency of the values from the dimension of the stimulation surface. A statistically significant difference was found when comparing the two stimulated sites only for the cold sensation and the hot pain. Moreover, when the sensation threshold was compared with the painful one, a statistically significant difference was found only for the cold stimulation on both the anatomical sites. In analyzing the ability to distinguish the nature of painful stimuli, the mechanical stimulus presented the highest recognition rate and the cold stimulus was never confused with the mechanical one. A systematic and rigorous comparison of the performance of commercial thermal sensors when integrated into a prosthetic fingertip was performed. Four thermal sensors were selected and a testing protocol for their assessment was devised based on the developed mechatronic testbed. The sensors were inserted into a 3D-printed fingertip and different object temperatures were considered. Moreover, a silicone layer was added to simulate the presence of a cosmetic glove and various object-to-sensor inclination angles were tested to reproduce a realistic contact. The optimal sensor for the integration into a prosthetic fingertip was identified through a comparison of the obtained results. An infrared sensor was selected, as it exhibited an average rise time less than 0.01 s with silicone and less than 0.04 ± 0.20 s without silicone and a mean error of 3.3 ± 5.1 °C and 4.0 ± 5.4 °C with and without silicone, respectively. All the tested sensors underwent a degradation of performance when the fingertip was covered with silicone and positioned at the highest inclination. These observations emphasize the significance of using the proposed assessment method to evaluate the performance of thermal sensors in a realistic environment.
Testbed for Accurate Mechanical and Thermal Stimulation and its Application in Somatosensory Studies / Marika Sperduti , 2024 Nov 06. 36. ciclo
Testbed for Accurate Mechanical and Thermal Stimulation and its Application in Somatosensory Studies
SPERDUTI, MARIKA
2024-11-06
Abstract
The somatosensory system has a primary role in daily life, since it provides important information for the human interaction with the surrounding environment. The loss of this information in hand amputees using passive prostheses can highly worsen manipulation activities. Thus, the integration of somatosensory feedback in closed-loop prostheses represents a significant advancement toward the restoration of natural functions of prosthetic hands. The potential benefits are multiple, including: the possibility to enhance the overall quality of life by providing amputees with a pleasant experience; the substantial enhancement of users’ self-esteem and independence; the improvement of performance, in terms of finer and more precise movements, which are useful for tasks needing high dexterity such as grasping small objects or handling delicate instruments. To develop closed-loop prostheses embedding somatosensory functions, several investigations on the human natural functions and on devices capable of artificially replicating them are needed. To this aim, this work was devoted to design and develop a modular and versatile mechatronic testbed to conduct heterogeneous somatosensory studies in humans, prostheses and in-vitro sensing. In literature, multiple tests, based on different devices and methods, have been performed not only on humans but also on animals and ex-vivo models. Depending on the nature of the sample under analysis and on the scientific aims of interest, several solutions for experimental stimulation and for investigations on sensation or pain have been adopted. A deep analysis of the available devices and methods has been performed, also analyzing the representative values adopted during literature experiments. Based on the analysis of their main features and on literature studies, the most suitable solution for humans, rodents, and ex-vivo models and investigation aims (sensation and pain) was pointed out. This analysis revealed a lack of a single device that includes both mechanical and thermal stimulators, suitable for use in both in-vivo and ex-vivo models. It is therefore evident that the development of the mechatronic testbed, which is the objective of this thesis, was necessary to address the aforementioned gap in the literature. The results obtained from the analysis of the state of the art were integrated with the ones coming from the commercially available solutions. Based on the final need of the novel mechatronic testbed to be applied in somatosensory studies on in-vivo and in-vitro models, the technical specifications were defined. The main specifications to be taken into account were: i) a positioning system with at least 2 degrees of freedom and a minimum displacement in each direction of 100 mm, ii) a mechanical stimulator able to apply at least a purely normal force and controllable in force (optionally in position), iii) a thermal stimulator based on a Peltier element (thus able to perform both hot and cold stimuli within a single device) with a temperature range of 0 – 50 °C. Commercial components have been selected and adjusted/assembled to meet the technical specification requirements previously defined. The final developed setup allows for the accurate delivery and measurement of mechanical and thermal stimuli of interest with high spatial resolution (0.5 μ m). Compared with state-of-the-art solutions, it allows automated stimulation, overcoming the disadvantages associated with the manual application of the stimuli. Moreover, the mechanical and thermal stimulators included in the setup are highly modular, allowing the use of multiple ending tips (i.e. diameter of 0.5 – 3 mm for the mechanical and of 2 – 3 mm for the thermal) and the application of a wide range of force/ temperature (0.3 – 5 N/ 5 – 60 °C). Thanks to its high versatility, the testbed was employed in two different scenarios: i) identification of the human somatosensory mechanical/thermal sensation and pain thresholds to be used as feedback in hand prostheses, ii) assessment of the performance of thermal sensors in realistic conditions and selection of the most suitable one for the integration in hand prostheses. The developed testbed was first employed to identify the mechanical/thermal innocuous and painful thresholds, as well as the human ability to distinguish the nature of a painful stimulus, on both the hand and the forearm of 12 healthy volunteers. The obtained mechanical thresholds showed no statistically significant difference when comparing the two stimulated anatomical sites. Furthermore, a statistically significant difference was found between mechanical innocuous and painful thresholds on both sites. When compared with the results in the literature, a difference between the presented continuously increasing mechanical stimulus and the discrete one provided by state-of-the-art equipment was found. The identified mechanical painful thresholds were found to be consistent with literature results, also confirming the dependency of the values from the dimension of the stimulation surface. A statistically significant difference was found when comparing the two stimulated sites only for the cold sensation and the hot pain. Moreover, when the sensation threshold was compared with the painful one, a statistically significant difference was found only for the cold stimulation on both the anatomical sites. In analyzing the ability to distinguish the nature of painful stimuli, the mechanical stimulus presented the highest recognition rate and the cold stimulus was never confused with the mechanical one. A systematic and rigorous comparison of the performance of commercial thermal sensors when integrated into a prosthetic fingertip was performed. Four thermal sensors were selected and a testing protocol for their assessment was devised based on the developed mechatronic testbed. The sensors were inserted into a 3D-printed fingertip and different object temperatures were considered. Moreover, a silicone layer was added to simulate the presence of a cosmetic glove and various object-to-sensor inclination angles were tested to reproduce a realistic contact. The optimal sensor for the integration into a prosthetic fingertip was identified through a comparison of the obtained results. An infrared sensor was selected, as it exhibited an average rise time less than 0.01 s with silicone and less than 0.04 ± 0.20 s without silicone and a mean error of 3.3 ± 5.1 °C and 4.0 ± 5.4 °C with and without silicone, respectively. All the tested sensors underwent a degradation of performance when the fingertip was covered with silicone and positioned at the highest inclination. These observations emphasize the significance of using the proposed assessment method to evaluate the performance of thermal sensors in a realistic environment.File | Dimensione | Formato | |
---|---|---|---|
PhD_Sperduti_Marika.pdf
embargo fino al 06/11/2026
Licenza:
Creative commons
Dimensione
11.01 MB
Formato
Adobe PDF
|
11.01 MB | Adobe PDF | Visualizza/Apri Richiedi una copia |
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.