Publications
Academic publications
2020 |
Ozioko, O; Karipoth, P; Escobedo, P; Ntagios, M; Pullanchiyodan, A; Dahiya, R SensAct: The Soft and Squishy Tactile Sensor with Integrated Flexible Actuator Journal Article Forthcoming Advanced Intelligent Systems, Forthcoming, ISSN: 2640-4567. @article{Ozioko2020, title = {SensAct: The Soft and Squishy Tactile Sensor with Integrated Flexible Actuator}, author = {O. Ozioko and P. Karipoth and P. Escobedo and M. Ntagios and A. Pullanchiyodan and R. Dahiya}, issn = {2640-4567}, year = {2020}, date = {2020-01-20}, journal = {Advanced Intelligent Systems}, keywords = {}, pubstate = {forthcoming}, tppubtype = {article} } |
2019 |
Rongala, Udaya B; Mazzoni, Alberto ; Spanne, Anton ; Jörntell, Henrik ; Oddo, Calogero M Cuneate spiking neural network learning to classify naturalistic texture stimuli under varying sensing conditions Journal Article Journal Article Neural Networks, 123 , pp. 273-287, 2019. @article{Rongala2019, title = {Cuneate spiking neural network learning to classify naturalistic texture stimuli under varying sensing conditions Journal Article}, author = {Rongala, Udaya B and Mazzoni, Alberto and Spanne, Anton and Jörntell, Henrik and Oddo, Calogero M }, url = {https://doi.org/10.6084/m9.figshare.11791581.v1 https://figshare.com/projects/Cuneate_Neuron_based_Spiking_Neural_Network_Rongal[...] doi:doi.org/10.1016/j.neunet.2019.11.020}, doi = {https://doi.org/10.1016/j.neunet.2019.11.020}, year = {2019}, date = {2019-12-09}, journal = {Neural Networks}, volume = {123}, pages = {273-287}, abstract = {We implemented a functional neuronal network that was able to learn and discriminate haptic features from biomimetic tactile sensor inputs using a two-layer spiking neuron model and homeostatic synaptic learning mechanism. The first order neuron model was used to emulate biological tactile afferents and the second order neuron model was used to emulate biological cuneate neurons. We have evaluated 10 naturalistic textures using a passive touch protocol, under varying sensing conditions. Tactile sensor data acquired with five textures under five sensing conditions were used for a synaptic learning process, to tune the synaptic weights between tactile afferents and cuneate neurons. Using post-learning synaptic weights, we evaluated the individual and population cuneate neuron responses by decoding across 10 stimuli, under varying sensing conditions. This resulted in a high decoding performance. We further validated the decoding performance across stimuli, irrespective of sensing velocities using a set of 25 cuneate neuron responses. This resulted in a median decoding performance of 96% across the set of cuneate neurons. Being able to learn and perform generalized discrimination across tactile stimuli, makes this functional spiking tactile system effective and suitable for further robotic applications.}, keywords = {}, pubstate = {published}, tppubtype = {article} } We implemented a functional neuronal network that was able to learn and discriminate haptic features from biomimetic tactile sensor inputs using a two-layer spiking neuron model and homeostatic synaptic learning mechanism. The first order neuron model was used to emulate biological tactile afferents and the second order neuron model was used to emulate biological cuneate neurons. We have evaluated 10 naturalistic textures using a passive touch protocol, under varying sensing conditions. Tactile sensor data acquired with five textures under five sensing conditions were used for a synaptic learning process, to tune the synaptic weights between tactile afferents and cuneate neurons. Using post-learning synaptic weights, we evaluated the individual and population cuneate neuron responses by decoding across 10 stimuli, under varying sensing conditions. This resulted in a high decoding performance. We further validated the decoding performance across stimuli, irrespective of sensing velocities using a set of 25 cuneate neuron responses. This resulted in a median decoding performance of 96% across the set of cuneate neurons. Being able to learn and perform generalized discrimination across tactile stimuli, makes this functional spiking tactile system effective and suitable for further robotic applications. |
Norrlid, Johanna; Enander, Jonas M D; Mogensen, Hannes; Jörntell, Henrik Structured internal cortical states deduced from fixed tactile input patterns Journal Article Forthcoming Forthcoming. @article{Norrlid2019, title = {Structured internal cortical states deduced from fixed tactile input patterns}, author = {Johanna Norrlid and Jonas M.D. Enander and Hannes Mogensen and Henrik Jörntell}, url = {https://www.biorxiv.org/content/10.1101/810770v1}, doi = {https://doi.org/10.1101/810770}, year = {2019}, date = {2019-10-18}, abstract = {The brain has a never-ending internal activity, whose spatiotemporal evolution impacts how we perceive external inputs and generate illusions. The spatiotemporal evolution depends on the neuronal network structure, which forms such a rich dynamic system that the internal interaction with external inputs has remained poorly understood. We used reproducible touch-related spatiotemporal inputs and recorded intracellularly from rat neocortical neurons to characterize this interaction at the circuitry level. Although repeated presentations of the same input generated variable responses, they tended to sort into a set of preferred response states, unique for each neuron. This suggests that sensory inputs combine with internal brain network dynamics to cause it to fall into one out of many possible local minima solutions with disparate instantiations in the subnetworks connected to each neuron}, keywords = {}, pubstate = {forthcoming}, tppubtype = {article} } The brain has a never-ending internal activity, whose spatiotemporal evolution impacts how we perceive external inputs and generate illusions. The spatiotemporal evolution depends on the neuronal network structure, which forms such a rich dynamic system that the internal interaction with external inputs has remained poorly understood. We used reproducible touch-related spatiotemporal inputs and recorded intracellularly from rat neocortical neurons to characterize this interaction at the circuitry level. Although repeated presentations of the same input generated variable responses, they tended to sort into a set of preferred response states, unique for each neuron. This suggests that sensory inputs combine with internal brain network dynamics to cause it to fall into one out of many possible local minima solutions with disparate instantiations in the subnetworks connected to each neuron |
Dahiya, Ravinder; Yogeswaran, Nivasan; Liu, Fengyuan; Manjakkal, Libu; Burdet, Etienne; Hayward, Vincent; Jörntell, Henrik Large-Area Soft e-Skin: TheChallenges BeyondSensor Designs Journal Article Proceedings Of The IEEE, 107 (10), pp. 2016-2033, 2019. @article{RAVINDERDAHIYA2019, title = {Large-Area Soft e-Skin: TheChallenges BeyondSensor Designs}, author = {Ravinder Dahiya and Nivasan Yogeswaran and Fengyuan Liu and Libu Manjakkal and Etienne Burdet and Vincent Hayward and Henrik Jörntell}, url = {https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=8858052 https://figshare.com/projects/Large-Area_Soft_e-Skin_The_Challenges_Beyond_Sensor_Designs/74109}, doi = {10.1109/JPROC.2019.2941366}, year = {2019}, date = {2019-10-03}, journal = {Proceedings Of The IEEE}, volume = {107}, number = {10}, pages = {2016-2033}, abstract = {Sensory feedback from touch is critical for manytasks carried out by robots and humans, such as graspingobjects or identifying materials. Electronic skin (e-skin) is acrucial technology for these purposes. Artificial tactile skin thatcan play the roles of human skin remains a distant possibilitybecause of hard issues in resilience, manufacturing, mechan-ics, sensorics, electronics, energetics, information processing,and transport. Taken together, these issues make it difficultto bestow robots, or prosthetic devices, with effective tactileskins. Nonetheless, progress over the past few years in relationwith the above issues has been encouraging, and we haveachieved close to providing some of the abilities of biologicalskin with the advent of deformable sensors and flexible elec-tronics. The naive imitation of skin morphology and sensingan impoverished set of mechanical and thermal quantities arenot sufficient. There is a need to find more efficient ways toextract tactile information from mechanical contact than thosepreviously available. Renewed interest in neuromorphic tactileskin is expected to bring some fresh ideas in this field. This article reviews these new developments, particularly related tothe handling of tactile data, energy autonomy, and large-areamanufacturing. The challenges in relation with these advancesfor tactile sensing and haptics in robotics and prosthetics arediscussed along with potential solutions}, keywords = {}, pubstate = {published}, tppubtype = {article} } Sensory feedback from touch is critical for manytasks carried out by robots and humans, such as graspingobjects or identifying materials. Electronic skin (e-skin) is acrucial technology for these purposes. Artificial tactile skin thatcan play the roles of human skin remains a distant possibilitybecause of hard issues in resilience, manufacturing, mechan-ics, sensorics, electronics, energetics, information processing,and transport. Taken together, these issues make it difficultto bestow robots, or prosthetic devices, with effective tactileskins. Nonetheless, progress over the past few years in relationwith the above issues has been encouraging, and we haveachieved close to providing some of the abilities of biologicalskin with the advent of deformable sensors and flexible elec-tronics. The naive imitation of skin morphology and sensingan impoverished set of mechanical and thermal quantities arenot sufficient. There is a need to find more efficient ways toextract tactile information from mechanical contact than thosepreviously available. Renewed interest in neuromorphic tactileskin is expected to bring some fresh ideas in this field. This article reviews these new developments, particularly related tothe handling of tactile data, energy autonomy, and large-areamanufacturing. The challenges in relation with these advancesfor tactile sensing and haptics in robotics and prosthetics arediscussed along with potential solutions |
Mogensen, Hannes; Norrlid, Johanna; Enander, Jonas M D; Wahlbom, Anders; Jörntell, Henrik Absence of Repetitive Correlation Patterns Between Pairs of Adjacent Neocortical Neurons in vivo Journal Article Front. Neural Circuits, 13 (48), 2019. @article{Mogensen2019, title = {Absence of Repetitive Correlation Patterns Between Pairs of Adjacent Neocortical Neurons in vivo}, author = {Hannes Mogensen and Johanna Norrlid and Jonas M. D. Enander and Anders Wahlbom and Henrik Jörntell}, url = {https://www.frontiersin.org/articles/10.3389/fncir.2019.00048/full https://figshare.com/projects/Paired_Neurons/73455 }, doi = {10.3389/fncir.2019.00048}, year = {2019}, date = {2019-07-19}, journal = {Front. Neural Circuits}, volume = {13}, number = {48}, abstract = {Neuroanatomy suggests that adjacent neocortical neurons share a similar set of afferent synaptic inputs, as opposed to neurons localized to different areas of the neocortex. In the present study, we made simultaneous single-electrode patch clamp recordings from two or three adjacent neurons in the primary somatosensory cortex (S1) of the ketamine- xylazine anesthetized rat in vivo to study the correlation patterns in their spike firing during both spontaneous and sensory-evoked activity. One difference with previous studies of pairwise neuronal spike firing correlations was that here we identified several different quantifiable parameters in the correlation patterns by which different pairs could be compared. The questions asked were if the correlation patterns between adjacent pairs were similar and if there was a relationship between the degree of similarity and the layer location of the pairs. In contrast, our results show that for putative pyramidal neurons within layer III and within layer V, each pair of neurons is to some extent unique in terms of their spiking correlation patterns. Interestingly, our results also indicated that these correlation patterns did not substantially alter between spontaneous and evoked activity. Our findings are compatible with the view that the synaptic input connectivity to each neocortical neuron is at least in some aspects unique. A possible interpretation is that plasticity mechanisms, which could either be initiating or be supported by transcriptomic differences, tend to differentiate rather than harmonize the synaptic weight distributions between adjacent neurons of the same type}, keywords = {}, pubstate = {published}, tppubtype = {article} } Neuroanatomy suggests that adjacent neocortical neurons share a similar set of afferent synaptic inputs, as opposed to neurons localized to different areas of the neocortex. In the present study, we made simultaneous single-electrode patch clamp recordings from two or three adjacent neurons in the primary somatosensory cortex (S1) of the ketamine- xylazine anesthetized rat in vivo to study the correlation patterns in their spike firing during both spontaneous and sensory-evoked activity. One difference with previous studies of pairwise neuronal spike firing correlations was that here we identified several different quantifiable parameters in the correlation patterns by which different pairs could be compared. The questions asked were if the correlation patterns between adjacent pairs were similar and if there was a relationship between the degree of similarity and the layer location of the pairs. In contrast, our results show that for putative pyramidal neurons within layer III and within layer V, each pair of neurons is to some extent unique in terms of their spiking correlation patterns. Interestingly, our results also indicated that these correlation patterns did not substantially alter between spontaneous and evoked activity. Our findings are compatible with the view that the synaptic input connectivity to each neocortical neuron is at least in some aspects unique. A possible interpretation is that plasticity mechanisms, which could either be initiating or be supported by transcriptomic differences, tend to differentiate rather than harmonize the synaptic weight distributions between adjacent neurons of the same type |
Wahlbom, Anders; Enander, Jonas M D; Bengtsson, Fredrik; Jörntell, Henrik Focal neocortical lesions impair distant neuronal information processing Journal Article The Journal of Physiology, 597 (16), pp. 4357-4371, 2019. @article{Wahlbom2019, title = {Focal neocortical lesions impair distant neuronal information processing}, author = {Anders Wahlbom and Jonas M. D. Enander and Fredrik Bengtsson and Henrik Jörntell}, url = {https://physoc.onlinelibrary.wiley.com/doi/full/10.1113/JP277717 https://figshare.com/projects/Stroke_paper/73458 }, doi = {https://doi.org/10.1113/JP277717}, year = {2019}, date = {2019-07-09}, journal = {The Journal of Physiology}, volume = {597}, number = {16}, pages = {4357-4371}, abstract = {Recent clinical studies report a surprisingly weak relationship between the location of cortical brain lesions and the resulting functional deficits. From a neuroscience point of view, such findings raise questions as to what extent functional localization applies in the neocortex and to what extent the functions of different regions depend on the integrity of others. Here we provide an in‐depth analysis of the changes in the function of the neocortical neuronal networks after distant focal stroke‐like lesions in the anaesthetized rat. Using a recently introduced high resolution analysis of neuronal information processing, consisting of pre‐set spatiotemporal patterns of tactile afferent activation against which the neuronal decoding performance can be quantified, we found that stroke‐like lesions in distant parts of the cortex significantly degraded the decoding performance of individual neocortical neurons in the primary somatosensory cortex (decoding performance decreased from 30.9% to 24.2% for n = 22 neurons, Wilcoxon signed rank test, P = 0.028). This degrading effect was not due to changes in the firing frequency of the neuron (Wilcoxon signed rank test, P = 0.499) and was stronger the higher the decoding performance of the neuron, indicating a specific impact on the information processing capacity in the cortex. These findings suggest that even primary sensory processing depends on widely distributed cortical networks and could explain observations of focal stroke lesions affecting a large range of functions.}, keywords = {}, pubstate = {published}, tppubtype = {article} } Recent clinical studies report a surprisingly weak relationship between the location of cortical brain lesions and the resulting functional deficits. From a neuroscience point of view, such findings raise questions as to what extent functional localization applies in the neocortex and to what extent the functions of different regions depend on the integrity of others. Here we provide an in‐depth analysis of the changes in the function of the neocortical neuronal networks after distant focal stroke‐like lesions in the anaesthetized rat. Using a recently introduced high resolution analysis of neuronal information processing, consisting of pre‐set spatiotemporal patterns of tactile afferent activation against which the neuronal decoding performance can be quantified, we found that stroke‐like lesions in distant parts of the cortex significantly degraded the decoding performance of individual neocortical neurons in the primary somatosensory cortex (decoding performance decreased from 30.9% to 24.2% for n = 22 neurons, Wilcoxon signed rank test, P = 0.028). This degrading effect was not due to changes in the firing frequency of the neuron (Wilcoxon signed rank test, P = 0.499) and was stronger the higher the decoding performance of the neuron, indicating a specific impact on the information processing capacity in the cortex. These findings suggest that even primary sensory processing depends on widely distributed cortical networks and could explain observations of focal stroke lesions affecting a large range of functions. |
Enander, Jonas M D; Spanne, Anton; Mazzoni, Alberto; Bengtsson, Fredrik; Oddo, Calogero Maria; Jörntell, Henrik Ubiquitous Neocortical Decoding of Tactile Input Patterns Journal Article Frontiers in Cellular Neuroscience, 13 , pp. 140, 2019, ISSN: 1662-5102. @article{Enander2019b, title = {Ubiquitous Neocortical Decoding of Tactile Input Patterns}, author = {Jonas M.D. Enander and Anton Spanne and Alberto Mazzoni and Fredrik Bengtsson and Calogero Maria Oddo and Henrik Jörntell}, editor = { Front. Cell. Neurosci., 12 April 2019 | https://doi.org/10.3389/fncel.2019.00140 }, url = {https://www.frontiersin.org/article/10.3389/fncel.2019.00140 https://figshare.com/s/297c1c4e8f5b4c20c037 https://doi.org/10.6084/m9.figshare.11771157.v1}, doi = {10.3389/fncel.2019.00140}, issn = {1662-5102}, year = {2019}, date = {2019-04-19}, journal = {Frontiers in Cellular Neuroscience}, volume = {13}, pages = {140}, abstract = {Whereas functional localization historically has been a key concept in neuroscience, direct neuronal recordings show that input of a particular modality can be recorded well outside its primary receiving areas in the neocortex. Here we wanted to explore if such spatially unbounded inputs potentially contain any information about the quality of the input received. We utilized a recently introduced approach to study the neuronal decoding capacity at a high resolution by delivering a set of electrical, highly reproducible spatiotemporal tactile afferent activation patterns to the skin of the contralateral second digit of the forepaw of the anesthetized rat. Surprisingly, we found that neurons in all areas recorded from, across all cortical depths tested, could decode the tactile input patterns, including neurons of the primary visual cortex. Within both somatosensory and visual cortical areas, the combined decoding accuracy of a population of neurons was higher than for the best performing single neuron within the respective area. Such cooperative decoding indicates that not only did individual neurons decode the input, they did so by generating responses with different temporal profiles compared to other neurons, which suggests that each neuron could have unique contributions to the tactile information processing. These findings suggest that tactile processing in principle could be globally distributed in the neocortex, possibly for comparison with internal expectations and disambiguation processes relying on other modalities.}, keywords = {}, pubstate = {published}, tppubtype = {article} } Whereas functional localization historically has been a key concept in neuroscience, direct neuronal recordings show that input of a particular modality can be recorded well outside its primary receiving areas in the neocortex. Here we wanted to explore if such spatially unbounded inputs potentially contain any information about the quality of the input received. We utilized a recently introduced approach to study the neuronal decoding capacity at a high resolution by delivering a set of electrical, highly reproducible spatiotemporal tactile afferent activation patterns to the skin of the contralateral second digit of the forepaw of the anesthetized rat. Surprisingly, we found that neurons in all areas recorded from, across all cortical depths tested, could decode the tactile input patterns, including neurons of the primary visual cortex. Within both somatosensory and visual cortical areas, the combined decoding accuracy of a population of neurons was higher than for the best performing single neuron within the respective area. Such cooperative decoding indicates that not only did individual neurons decode the input, they did so by generating responses with different temporal profiles compared to other neurons, which suggests that each neuron could have unique contributions to the tactile information processing. These findings suggest that tactile processing in principle could be globally distributed in the neocortex, possibly for comparison with internal expectations and disambiguation processes relying on other modalities. |
Enander, Jonas M D; Jörntell, Henrik Somatosensory Cortical Neurons Decode Tactile Input Patterns and Location from Both Dominant and Non-dominant Digits Journal Article Cell Reports, 26 (13), pp. 3551-3560, 2019. @article{Enander2019CellReports, title = {Somatosensory Cortical Neurons Decode Tactile Input Patterns and Location from Both Dominant and Non-dominant Digits}, author = {Jonas M.D. Enander and Henrik Jörntell}, url = {https://doi.org/10.6084/m9.figshare.11762370.v1 https://doi.org/10.6084/m9.figshare.7687604.v1}, doi = {10.1016/j.celrep.2019.02.099}, year = {2019}, date = {2019-03-26}, journal = {Cell Reports}, volume = {26}, number = {13}, pages = {3551-3560}, keywords = {}, pubstate = {published}, tppubtype = {article} } |