Pub Date : 2018-10-01DOI: 10.1109/BIOCAS.2018.8584668
Ryan Weiss, J. Najem, Md. Sakib Hasan, Catherine D. Schuman, A. Belianinov, C. Collier, Stephen A. Sarles, G. Rose
The goal of neuromorphic computing is to recreate the computational power and efficiency of the human brain with circuitry. The ability of the brain to solve complex real time tasks, while consuming 20 W of power on average, is made possible through its connection density, adaptability, and parallel processing. Recreating these features using traditional electronics circuit elements is incredibly difficult, and therefore, soft-matter memristors made of biomolecules similar to those found in biological synapses and capable of emulating various synaptic features can be used as neuromorphic hardware. In this work, we introduce and experimentally demonstrate an electronic neuron circuit capable of interacting with ionic, soft-matter memristors. These memristors are proven to exhibit short-term plasticity, especially paired-pulse facilitation and depression found in presynaptic terminals - features that are not found in state-of-the-art solid-state memristors. We make use of these features for applications in online learning by developing a synapse-neuron circuit which implements spike-rate-dependent plasticity (SRDP) as a learning function.
{"title":"A Soft-Matter Biomolecular Memristor Synapse for Neuromorphic Systems","authors":"Ryan Weiss, J. Najem, Md. Sakib Hasan, Catherine D. Schuman, A. Belianinov, C. Collier, Stephen A. Sarles, G. Rose","doi":"10.1109/BIOCAS.2018.8584668","DOIUrl":"https://doi.org/10.1109/BIOCAS.2018.8584668","url":null,"abstract":"The goal of neuromorphic computing is to recreate the computational power and efficiency of the human brain with circuitry. The ability of the brain to solve complex real time tasks, while consuming 20 W of power on average, is made possible through its connection density, adaptability, and parallel processing. Recreating these features using traditional electronics circuit elements is incredibly difficult, and therefore, soft-matter memristors made of biomolecules similar to those found in biological synapses and capable of emulating various synaptic features can be used as neuromorphic hardware. In this work, we introduce and experimentally demonstrate an electronic neuron circuit capable of interacting with ionic, soft-matter memristors. These memristors are proven to exhibit short-term plasticity, especially paired-pulse facilitation and depression found in presynaptic terminals - features that are not found in state-of-the-art solid-state memristors. We make use of these features for applications in online learning by developing a synapse-neuron circuit which implements spike-rate-dependent plasticity (SRDP) as a learning function.","PeriodicalId":259162,"journal":{"name":"2018 IEEE Biomedical Circuits and Systems Conference (BioCAS)","volume":"12 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132036335","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2018-10-01DOI: 10.1109/BIOCAS.2018.8584695
G. Angotzi, Aziliz Lecomte, L. Giantomasi, L. Berdondini, M. Crepaldi, S. Rancati, D. D. Tonelli
Human-derived brain organoids were proposed for the generation of functional in vitro models and human brain tissues for drug-discovery, precision medicine and cell-based clinical therapies. However, their generation is currently subjected to a high variability which limits their routine exploitation. To achieve a quality-controlled production of brain organoids and to provide readout capabilities for assays development, we propose to realize active micro-scale devices that can be embedded into living 3D cell assemblies to provide in-tissue wireless sensing and monitoring of biosignals. Here, we evaluate a low-power solution that integrates into a 100μm×100μm area all circuits required for sensing and amplification of bioelectrical signals while providing RF wireless power delivery and data transmission. Circuit simulations on a 130nm RF-CMOS node demonstrate the feasibility of such solution with a 6.18µW of power consumption. Preliminary in vitro experiments with dummy Si micro-devices demonstrate their integration into 3D cell aggregates during cell culture.
{"title":"A µRadio CMOS Device for Real-Time In-Tissue Monitoring of Human Organoids","authors":"G. Angotzi, Aziliz Lecomte, L. Giantomasi, L. Berdondini, M. Crepaldi, S. Rancati, D. D. Tonelli","doi":"10.1109/BIOCAS.2018.8584695","DOIUrl":"https://doi.org/10.1109/BIOCAS.2018.8584695","url":null,"abstract":"Human-derived brain organoids were proposed for the generation of functional in vitro models and human brain tissues for drug-discovery, precision medicine and cell-based clinical therapies. However, their generation is currently subjected to a high variability which limits their routine exploitation. To achieve a quality-controlled production of brain organoids and to provide readout capabilities for assays development, we propose to realize active micro-scale devices that can be embedded into living 3D cell assemblies to provide in-tissue wireless sensing and monitoring of biosignals. Here, we evaluate a low-power solution that integrates into a 100μm×100μm area all circuits required for sensing and amplification of bioelectrical signals while providing RF wireless power delivery and data transmission. Circuit simulations on a 130nm RF-CMOS node demonstrate the feasibility of such solution with a 6.18µW of power consumption. Preliminary in vitro experiments with dummy Si micro-devices demonstrate their integration into 3D cell aggregates during cell culture.","PeriodicalId":259162,"journal":{"name":"2018 IEEE Biomedical Circuits and Systems Conference (BioCAS)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133832089","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2018-10-01DOI: 10.1109/BIOCAS.2018.8584757
Cynthia R. Steinhardt, Joseph L. Betthauser, Christopher L. Hunt, N. Thakor
Non-invasive recording of EMG signals from the arm of a typical subject or amputee has been popularized in control of a variety of devices, including upper limb prostheses. One of the most difficult challenges of using external recording devices, such as the Myo Armband, is the need to retrain a movement classifier due to the shift in positions and electrode location around the arm. Electrode shift causes distortion of the features to be extracted for classification and makes previous training unusable. For amputees, this means retraining movement classifiers several times per day. In this experiment, the Myo Armband is used to test the ability to predict the degree of electrode shift from the electrode sites used to originally train a classifier in order to correct by the detected shift and continue to use the same classifer, instead of training a new one. The Myo Armband was rotated around the arm of subjects with intact limbs as they performed six commonly used movements. The mean absolute value of each electrode was used to characterize the response at each electrode site. Shifts in orientation between one position and a new position were identified by minimizing the mean-squared error of their characteristic movement profiles. The correct shift was identified across subjects using only 0.25 s of data with over 90% accuracy using the “open” or “wrist supinate” grips. New movements at a shifted location were classified using the feature vectors of a previously collected training set and accounting for the shift; classification error averaged 95.7 ± 0.4%, indicating a possibility for real-time correction of electrode shift error.
{"title":"Registration of EMG Electrodes to Reduce Classification Errors due to Electrode Shift","authors":"Cynthia R. Steinhardt, Joseph L. Betthauser, Christopher L. Hunt, N. Thakor","doi":"10.1109/BIOCAS.2018.8584757","DOIUrl":"https://doi.org/10.1109/BIOCAS.2018.8584757","url":null,"abstract":"Non-invasive recording of EMG signals from the arm of a typical subject or amputee has been popularized in control of a variety of devices, including upper limb prostheses. One of the most difficult challenges of using external recording devices, such as the Myo Armband, is the need to retrain a movement classifier due to the shift in positions and electrode location around the arm. Electrode shift causes distortion of the features to be extracted for classification and makes previous training unusable. For amputees, this means retraining movement classifiers several times per day. In this experiment, the Myo Armband is used to test the ability to predict the degree of electrode shift from the electrode sites used to originally train a classifier in order to correct by the detected shift and continue to use the same classifer, instead of training a new one. The Myo Armband was rotated around the arm of subjects with intact limbs as they performed six commonly used movements. The mean absolute value of each electrode was used to characterize the response at each electrode site. Shifts in orientation between one position and a new position were identified by minimizing the mean-squared error of their characteristic movement profiles. The correct shift was identified across subjects using only 0.25 s of data with over 90% accuracy using the “open” or “wrist supinate” grips. New movements at a shifted location were classified using the feature vectors of a previously collected training set and accounting for the shift; classification error averaged 95.7 ± 0.4%, indicating a possibility for real-time correction of electrode shift error.","PeriodicalId":259162,"journal":{"name":"2018 IEEE Biomedical Circuits and Systems Conference (BioCAS)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132785565","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2018-10-01DOI: 10.1109/BIOCAS.2018.8584773
Parisa Forouzannezhad, Alireza Abbaspour, M. Cabrerizo, M. Adjouadi
Alzheimer‘s disease (AD) is a neurodegenerative disease which is progressive and can be described by amyloid deposition, and neuronal atrophy. In this study, a support vector machine (SVM) approach with radial basis function (RBF) has been proposed in order to detect the Alzheimer's disease in its early stage using multiple modalities, including positron emission tomography (PET), magnetic resonance imaging (MRI), and standard neuropsychological test scores. A total number of 896 participants from the Alzheimer's Disease Neuroimaging Initiative (ADNI) were considered in this study. The proposed approach is able to classify cognitively normal control (CN) group from early mild cognitive impairment (EMCI) with an accuracy of 81.1%. In addition, the accuracy of 91.9% for CN vs. late mild cognitive impairment and accuracy of 96.2% for CN vs. AD classifications have been achieved through the proposed model.
{"title":"Early Diagnosis of Mild Cognitive Impairment Using Random Forest Feature Selection","authors":"Parisa Forouzannezhad, Alireza Abbaspour, M. Cabrerizo, M. Adjouadi","doi":"10.1109/BIOCAS.2018.8584773","DOIUrl":"https://doi.org/10.1109/BIOCAS.2018.8584773","url":null,"abstract":"Alzheimer‘s disease (AD) is a neurodegenerative disease which is progressive and can be described by amyloid deposition, and neuronal atrophy. In this study, a support vector machine (SVM) approach with radial basis function (RBF) has been proposed in order to detect the Alzheimer's disease in its early stage using multiple modalities, including positron emission tomography (PET), magnetic resonance imaging (MRI), and standard neuropsychological test scores. A total number of 896 participants from the Alzheimer's Disease Neuroimaging Initiative (ADNI) were considered in this study. The proposed approach is able to classify cognitively normal control (CN) group from early mild cognitive impairment (EMCI) with an accuracy of 81.1%. In addition, the accuracy of 91.9% for CN vs. late mild cognitive impairment and accuracy of 96.2% for CN vs. AD classifications have been achieved through the proposed model.","PeriodicalId":259162,"journal":{"name":"2018 IEEE Biomedical Circuits and Systems Conference (BioCAS)","volume":"8 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116823832","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2018-10-01DOI: 10.1109/BIOCAS.2018.8584664
Dorian Haci, Yan Liu, K. Nikolic, D. Demarchi, T. Constandinou, P. Georgiou
This paper reports on the implementation and characterisation of a thermally controlled device for in vitro biomedical applications, based on standard Printed Circuit Board (PCB) technology. This is proposed as a low cost alternative to state-of-the-art microfluidic devices and Lab-on-Chip (LoC) platforms, which we refer to as the thermal Lab-on-PCB concept. In total, six different prototype boards have been manufactured to implement test mini-hotplate arrays. 3D mol-tiphysics software simulations show the thermal response of the modelled mini-hotplate boards to a current-controlled stimulus, highlighting their versatile heating capability. Comparing this with experimental results of the fabricated PCBs demonstrates the combined temperature sensing/heating feature of the mini-hotplate. This can provide a wider temperature range compared to that achieved in typical LoC devices. The thermal system is controllable by means of external off-the-shelf circuitry designed and implemented on a single-channel control board prototype.
{"title":"Thermally Controlled Lab-on-PCB for Biomedical Applications","authors":"Dorian Haci, Yan Liu, K. Nikolic, D. Demarchi, T. Constandinou, P. Georgiou","doi":"10.1109/BIOCAS.2018.8584664","DOIUrl":"https://doi.org/10.1109/BIOCAS.2018.8584664","url":null,"abstract":"This paper reports on the implementation and characterisation of a thermally controlled device for in vitro biomedical applications, based on standard Printed Circuit Board (PCB) technology. This is proposed as a low cost alternative to state-of-the-art microfluidic devices and Lab-on-Chip (LoC) platforms, which we refer to as the thermal Lab-on-PCB concept. In total, six different prototype boards have been manufactured to implement test mini-hotplate arrays. 3D mol-tiphysics software simulations show the thermal response of the modelled mini-hotplate boards to a current-controlled stimulus, highlighting their versatile heating capability. Comparing this with experimental results of the fabricated PCBs demonstrates the combined temperature sensing/heating feature of the mini-hotplate. This can provide a wider temperature range compared to that achieved in typical LoC devices. The thermal system is controllable by means of external off-the-shelf circuitry designed and implemented on a single-channel control board prototype.","PeriodicalId":259162,"journal":{"name":"2018 IEEE Biomedical Circuits and Systems Conference (BioCAS)","volume":"40 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114644401","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2018-10-01DOI: 10.1109/BIOCAS.2018.8584807
Daniel R. Mendat, A. Cassidy, Guido Zarrella, A. Andreou
Word2vec, like other ways of creating word embed-dings from a text corpus, has shown that interesting mathematical properties exist between the resulting word vectors. Word similarities as well as relationships can be discovered by determining which words are nearby in vector space and performing simple vector operations. In this work, IBM's TrueNorth Neurosynaptic System was used to implement massively-parallel word similarity computations using a large network of hardware spiking neurons. A 4-bit vector-matrix multiplication engine was implemented on TrueNorth in order to accommodate a word vector dictionary of 95,000 words trained on Wikipedia text, and it successfully performs word similarity searches using that dictionary while utilizing 3,991 cores out of the 4,096 available on TrueNorth and consuming less than 70 mW of power.
{"title":"Word2vec Word Similarities on IBM's TrueNorth Neurosynaptic System","authors":"Daniel R. Mendat, A. Cassidy, Guido Zarrella, A. Andreou","doi":"10.1109/BIOCAS.2018.8584807","DOIUrl":"https://doi.org/10.1109/BIOCAS.2018.8584807","url":null,"abstract":"Word2vec, like other ways of creating word embed-dings from a text corpus, has shown that interesting mathematical properties exist between the resulting word vectors. Word similarities as well as relationships can be discovered by determining which words are nearby in vector space and performing simple vector operations. In this work, IBM's TrueNorth Neurosynaptic System was used to implement massively-parallel word similarity computations using a large network of hardware spiking neurons. A 4-bit vector-matrix multiplication engine was implemented on TrueNorth in order to accommodate a word vector dictionary of 95,000 words trained on Wikipedia text, and it successfully performs word similarity searches using that dictionary while utilizing 3,991 cores out of the 4,096 available on TrueNorth and consuming less than 70 mW of power.","PeriodicalId":259162,"journal":{"name":"2018 IEEE Biomedical Circuits and Systems Conference (BioCAS)","volume":"96 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124180863","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2018-10-01DOI: 10.1109/BIOCAS.2018.8584838
N. Gurel, Donald Ward, Frank L. Hammond, O. Inan
This live demonstration presents a wearable system for neuro-vascular health assessment. The system is comprised of three key subcomponents: 1) a soft, fluid-driven thermal modulation pad, 2) a portable case containing electrical and mechanical hardware, and 3) off-board biosignal processing and power units. The soft thermal modulation pad (the only component that interfaces with the user) contains fluid channels, embedded temperature sensors, and a flexible protoboard, encased in graphite-based silicone. The pad induces fluid-based heating or cooling of the hand, and is connected to the portable case through inlets and outlets. The portable case contains electrical and mechanical actuation (temperature modulation, fluid flow), sensing, and control circuitry, none of which is in contact with the user. Components include Peltier tiles, temperature and flow rate sensors, fluid pump, reservoir, control circuitry, heat sinks, cooling fans, and a microcontroller. All components in the case are enclosed with a laser-cut acrylic sheet to shield them from outside world, except for the fluid reservoir (to be filled with water before use) and toggle switches. Separate from the portable case and the pad are a data acquisition system, photoplethysmography (PPG) sensor to be worn on the hand, a laptop for visualization of the data, and a power supply. To enable closed-loop temperature control, the current and voltage flowing through the Peltier tiles are controlled with a custom algorithm implemented on the microcontroller. Biosignals involved with thermoregulation are collected. PPG measures the changes in blood volume pulse at the collected location; the amplitude of the PPG signal reflects the change in dilation or constriction of the vasculature [1].
{"title":"Live Demonstration: A Soft Thermal Modulation System with Embedded Fluid Channels for Neuro-Vascular Assessment","authors":"N. Gurel, Donald Ward, Frank L. Hammond, O. Inan","doi":"10.1109/BIOCAS.2018.8584838","DOIUrl":"https://doi.org/10.1109/BIOCAS.2018.8584838","url":null,"abstract":"This live demonstration presents a wearable system for neuro-vascular health assessment. The system is comprised of three key subcomponents: 1) a soft, fluid-driven thermal modulation pad, 2) a portable case containing electrical and mechanical hardware, and 3) off-board biosignal processing and power units. The soft thermal modulation pad (the only component that interfaces with the user) contains fluid channels, embedded temperature sensors, and a flexible protoboard, encased in graphite-based silicone. The pad induces fluid-based heating or cooling of the hand, and is connected to the portable case through inlets and outlets. The portable case contains electrical and mechanical actuation (temperature modulation, fluid flow), sensing, and control circuitry, none of which is in contact with the user. Components include Peltier tiles, temperature and flow rate sensors, fluid pump, reservoir, control circuitry, heat sinks, cooling fans, and a microcontroller. All components in the case are enclosed with a laser-cut acrylic sheet to shield them from outside world, except for the fluid reservoir (to be filled with water before use) and toggle switches. Separate from the portable case and the pad are a data acquisition system, photoplethysmography (PPG) sensor to be worn on the hand, a laptop for visualization of the data, and a power supply. To enable closed-loop temperature control, the current and voltage flowing through the Peltier tiles are controlled with a custom algorithm implemented on the microcontroller. Biosignals involved with thermoregulation are collected. PPG measures the changes in blood volume pulse at the collected location; the amplitude of the PPG signal reflects the change in dilation or constriction of the vasculature [1].","PeriodicalId":259162,"journal":{"name":"2018 IEEE Biomedical Circuits and Systems Conference (BioCAS)","volume":"85 9 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127987832","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2018-10-01DOI: 10.1109/BIOCAS.2018.8584761
Federico Mazza, Yan Liu, N. Donaldson, T. Constandinou
Recent developments in the design of active implantable devices have achieved significant advances, for example, an increased number of recording channels, but too often practical clinical applications are restricted by device longevity. It is important however to complement efforts for increased functionality with translational work to develop implant technologies that are safe and reliable to be hosted inside the human body over long periods of time. This paper first examines techniques currently used to evaluate micro-package hermeticity and key challenges, highlighting the need for new, in situ instrumentation that can monitor the encapsulation status over time. Two novel circuits are then proposed to tackle the specific issue of moisture penetration inside a sub-mm, silicon-based package. They both share the use of metal tracks on the different layers of the CMOS stack to measure changes in impedance caused by moisture present in leak cracks or diffused into the oxide layers.
{"title":"Integrated Devices for Micro-Package Integrity Monitoring in mm-Scale Neural Implants","authors":"Federico Mazza, Yan Liu, N. Donaldson, T. Constandinou","doi":"10.1109/BIOCAS.2018.8584761","DOIUrl":"https://doi.org/10.1109/BIOCAS.2018.8584761","url":null,"abstract":"Recent developments in the design of active implantable devices have achieved significant advances, for example, an increased number of recording channels, but too often practical clinical applications are restricted by device longevity. It is important however to complement efforts for increased functionality with translational work to develop implant technologies that are safe and reliable to be hosted inside the human body over long periods of time. This paper first examines techniques currently used to evaluate micro-package hermeticity and key challenges, highlighting the need for new, in situ instrumentation that can monitor the encapsulation status over time. Two novel circuits are then proposed to tackle the specific issue of moisture penetration inside a sub-mm, silicon-based package. They both share the use of metal tracks on the different layers of the CMOS stack to measure changes in impedance caused by moisture present in leak cracks or diffused into the oxide layers.","PeriodicalId":259162,"journal":{"name":"2018 IEEE Biomedical Circuits and Systems Conference (BioCAS)","volume":"53 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129900019","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2018-10-01DOI: 10.1109/BIOCAS.2018.8584688
M. N. Sahadat, Nordine Sebkhi, Fanpeng Kong, Maysam Ghovanloo
People with motor disabilities affecting their four limbs (e.g. tetraplegia, ALS) can use their remaining abilities such as tongue, head, and eye motion to interact with devices, such as PC, smartphone, and wheelchair. Most of the existing assistive technologies (AT) rely on a single remaining ability, which is typically insufficient when performing complex computer tasks such as “drag and drop”, typing long sentences, or selecting multiple items. In this work, a multimodal AT is presented to leverage both tongue gestures and head motion, simultaneously, to interact with target devices at latency and accuracy of 10 ms and 95.9%, respectively, measured among 15 able-bodied participants. A wearable headset transmits commands wirelessly via Bluetooth Low Energy (BLE), utilizing a human interface device (HID) protocol for seamless interfacing with various applications running on PCs and smartphones without requiring a custom driver.
{"title":"Standalone Assistive System to Employ Multiple Remaining Abilities in People with Tetraplegia","authors":"M. N. Sahadat, Nordine Sebkhi, Fanpeng Kong, Maysam Ghovanloo","doi":"10.1109/BIOCAS.2018.8584688","DOIUrl":"https://doi.org/10.1109/BIOCAS.2018.8584688","url":null,"abstract":"People with motor disabilities affecting their four limbs (e.g. tetraplegia, ALS) can use their remaining abilities such as tongue, head, and eye motion to interact with devices, such as PC, smartphone, and wheelchair. Most of the existing assistive technologies (AT) rely on a single remaining ability, which is typically insufficient when performing complex computer tasks such as “drag and drop”, typing long sentences, or selecting multiple items. In this work, a multimodal AT is presented to leverage both tongue gestures and head motion, simultaneously, to interact with target devices at latency and accuracy of 10 ms and 95.9%, respectively, measured among 15 able-bodied participants. A wearable headset transmits commands wirelessly via Bluetooth Low Energy (BLE), utilizing a human interface device (HID) protocol for seamless interfacing with various applications running on PCs and smartphones without requiring a custom driver.","PeriodicalId":259162,"journal":{"name":"2018 IEEE Biomedical Circuits and Systems Conference (BioCAS)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130325122","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2018-10-01DOI: 10.1109/BIOCAS.2018.8584828
F. Khan, Usman Ashraf, Muhammad Awais Bin Altaf, Wala Saadeh
An electroencephalograph (EEG) based classification processor for the depth of Anesthesia (DoA) during the intraoperative procedure is presented. To enable a DoA to monitor the correct estimation across a range of patients, a novel feature extraction along with machine learning processor is utilized. The decisions are solely based on seven features extracted from EEG along with the EMG signal for motion artifacts rejection. To extract the features efficiently on hardware, a 128-point FFT is proposed that achieves an area reduction and energy/FFT-operation by 39% and 58%, respectively, compared to the conventional. A simple decision tree is used to perform a multiclass DoA classification. The system is synthesized using a 65nm process and experimental verification is done using FPGA based on the subset of patients from the University of Queensland Vital Signs. The proposed patient-specific DoA classification processor achieves a classification accuracy of 79%.
{"title":"A Patient-Specific Machine Learning based EEG Processor for Accurate Estimation of Depth of Anesthesia","authors":"F. Khan, Usman Ashraf, Muhammad Awais Bin Altaf, Wala Saadeh","doi":"10.1109/BIOCAS.2018.8584828","DOIUrl":"https://doi.org/10.1109/BIOCAS.2018.8584828","url":null,"abstract":"An electroencephalograph (EEG) based classification processor for the depth of Anesthesia (DoA) during the intraoperative procedure is presented. To enable a DoA to monitor the correct estimation across a range of patients, a novel feature extraction along with machine learning processor is utilized. The decisions are solely based on seven features extracted from EEG along with the EMG signal for motion artifacts rejection. To extract the features efficiently on hardware, a 128-point FFT is proposed that achieves an area reduction and energy/FFT-operation by 39% and 58%, respectively, compared to the conventional. A simple decision tree is used to perform a multiclass DoA classification. The system is synthesized using a 65nm process and experimental verification is done using FPGA based on the subset of patients from the University of Queensland Vital Signs. The proposed patient-specific DoA classification processor achieves a classification accuracy of 79%.","PeriodicalId":259162,"journal":{"name":"2018 IEEE Biomedical Circuits and Systems Conference (BioCAS)","volume":"8 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126378249","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：