Oliver Ozioko, Prakash Karipoth, P. Escobedo, M. Ntagios, A. Pullanchiyodan, R. Dahiya
Herein, a novel tactile sensing device (SensAct) with a soft touch/pressure sensor seamlessly integrated on a flexible actuator is presented. The squishy touch sensor is developed with custom‐made graphite paste on a tiny permanent magnet, encapsulated in Sil‐Poxy, and the actuator (15 μ‐thick coil) is fabricated on polyimide by Lithographie Galvanoformung Abformung (LIGA) micromolding method. The actuator can operate in two modes (expansion and contraction/squeeze) and two states (vibration and nonvibration). The sensor was tested with up to 12 N applied forces and exhibited ≈70% average relative resistance variation (ΔR/Ro), ≈0.346 kPa−1 sensitivity, and ≈49 ms response time with excellent repeatability (≈12.7% coefficient of variation) at 5 N. During simultaneous sensing and actuation, the modulation of coil current, due to ΔR/Ro (≈14% at 2 N force) in the sensor, allows the close loop control (ΔI/Io ≈385%) of expansion/contraction (≈69.8 μm expansion in nonvibration state and ≈111.5 μm peak‐to‐peak in the vibration state). Finally, the soft sensor is embedded in the 3D‐printed fingertip of a robotic hand to demonstrate its use for pressure mapping along with remote vibrotactile stimulation using SensAct device. The self‐controllable actuation of SensAct could provide eSkin the ability to tune stiffness and the vibration states could be utilized for controlled haptic feedback.
{"title":"SensAct: The Soft and Squishy Tactile Sensor with Integrated Flexible Actuator","authors":"Oliver Ozioko, Prakash Karipoth, P. Escobedo, M. Ntagios, A. Pullanchiyodan, R. Dahiya","doi":"10.1002/aisy.201900145","DOIUrl":"https://doi.org/10.1002/aisy.201900145","url":null,"abstract":"Herein, a novel tactile sensing device (SensAct) with a soft touch/pressure sensor seamlessly integrated on a flexible actuator is presented. The squishy touch sensor is developed with custom‐made graphite paste on a tiny permanent magnet, encapsulated in Sil‐Poxy, and the actuator (15 μ‐thick coil) is fabricated on polyimide by Lithographie Galvanoformung Abformung (LIGA) micromolding method. The actuator can operate in two modes (expansion and contraction/squeeze) and two states (vibration and nonvibration). The sensor was tested with up to 12 N applied forces and exhibited ≈70% average relative resistance variation (ΔR/Ro), ≈0.346 kPa−1 sensitivity, and ≈49 ms response time with excellent repeatability (≈12.7% coefficient of variation) at 5 N. During simultaneous sensing and actuation, the modulation of coil current, due to ΔR/Ro (≈14% at 2 N force) in the sensor, allows the close loop control (ΔI/Io ≈385%) of expansion/contraction (≈69.8 μm expansion in nonvibration state and ≈111.5 μm peak‐to‐peak in the vibration state). Finally, the soft sensor is embedded in the 3D‐printed fingertip of a robotic hand to demonstrate its use for pressure mapping along with remote vibrotactile stimulation using SensAct device. The self‐controllable actuation of SensAct could provide eSkin the ability to tune stiffness and the vibration states could be utilized for controlled haptic feedback.","PeriodicalId":7187,"journal":{"name":"Advanced Intelligent Systems","volume":"7 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2021-01-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"73237203","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}
Wataru Akashi, H. Kambara, Yousuke Ogata, Y. Koike, L. Minati, N. Yoshimura
Recent advances in brain imaging technology have furthered our knowledge of the neural basis of auditory and speech processing, often via contributions from invasive brain signal recording and stimulation studies conducted intraoperatively. Herein, an approach for synthesizing vowel sounds straightforwardly from scalp‐recorded electroencephalography (EEG), a noninvasive neurophysiological recording method is demonstrated. Given cortical current signals derived from the EEG acquired while human participants listen to and recall (i.e., imagined) two vowels, /a/ and /i/, sound parameters are estimated by a convolutional neural network (CNN). The speech synthesized from the estimated parameters is sufficiently natural to achieve recognition rates >85% during a subsequent sound discrimination task. Notably, the CNN identifies the involvement of the brain areas mediating the “what” auditory stream, namely the superior, middle temporal, and Heschl's gyri, demonstrating the efficacy of the computational method in extracting auditory‐related information from neuroelectrical activity. Differences in cortical sound representation between listening versus recalling are further revealed, such that the fusiform, calcarine, and anterior cingulate gyri contributes during listening, whereas the inferior occipital gyrus is engaged during recollection. The proposed approach can expand the scope of EEG in decoding auditory perception that requires high spatial and temporal resolution.
{"title":"Vowel Sound Synthesis from Electroencephalography during Listening and Recalling","authors":"Wataru Akashi, H. Kambara, Yousuke Ogata, Y. Koike, L. Minati, N. Yoshimura","doi":"10.1002/aisy.202000164","DOIUrl":"https://doi.org/10.1002/aisy.202000164","url":null,"abstract":"Recent advances in brain imaging technology have furthered our knowledge of the neural basis of auditory and speech processing, often via contributions from invasive brain signal recording and stimulation studies conducted intraoperatively. Herein, an approach for synthesizing vowel sounds straightforwardly from scalp‐recorded electroencephalography (EEG), a noninvasive neurophysiological recording method is demonstrated. Given cortical current signals derived from the EEG acquired while human participants listen to and recall (i.e., imagined) two vowels, /a/ and /i/, sound parameters are estimated by a convolutional neural network (CNN). The speech synthesized from the estimated parameters is sufficiently natural to achieve recognition rates >85% during a subsequent sound discrimination task. Notably, the CNN identifies the involvement of the brain areas mediating the “what” auditory stream, namely the superior, middle temporal, and Heschl's gyri, demonstrating the efficacy of the computational method in extracting auditory‐related information from neuroelectrical activity. Differences in cortical sound representation between listening versus recalling are further revealed, such that the fusiform, calcarine, and anterior cingulate gyri contributes during listening, whereas the inferior occipital gyrus is engaged during recollection. The proposed approach can expand the scope of EEG in decoding auditory perception that requires high spatial and temporal resolution.","PeriodicalId":7187,"journal":{"name":"Advanced Intelligent Systems","volume":"27 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2021-01-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"74174372","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}
Xin Shu, S. Sansare, Di Jin, Xiang-Hui Zeng, K. Tong, Rishikesh Pandey, R. Zhou
Leukocyte differential test is a widely carried out clinical procedure for screening infectious diseases. Existing hematology analyzers require labor‐intensive work and a panel of expensive reagents. Herein, an artificial‐intelligence‐enabled reagent‐free imaging hematology analyzer (AIRFIHA) modality is reported that can accurately classify subpopulations of leukocytes with minimal sample preparation. AIRFIHA is realized through training a two‐step residual neural network using label‐free images of isolated leukocytes acquired from a custom‐built quantitative phase microscope. By leveraging the rich information contained in quantitative phase images, not only high accuracy is achieved in differentiating B and T lymphocytes, but also CD4 and CD8 T cells are classified, therefore outperforming the classification accuracy of most current hematology analyzers. The performance of AIRFIHA in a randomly selected test set is validated and is cross‐validated across all blood donors. Due to its easy operation, low cost, and accurate discerning capability of complex leukocyte subpopulations, AIRFIHA is clinically translatable and can also be deployed in resource‐limited settings, e.g., during pandemic situations for the rapid screening of infectious diseases.
{"title":"Artificial‐Intelligence‐Enabled Reagent‐Free Imaging Hematology Analyzer","authors":"Xin Shu, S. Sansare, Di Jin, Xiang-Hui Zeng, K. Tong, Rishikesh Pandey, R. Zhou","doi":"10.1002/aisy.202000277","DOIUrl":"https://doi.org/10.1002/aisy.202000277","url":null,"abstract":"Leukocyte differential test is a widely carried out clinical procedure for screening infectious diseases. Existing hematology analyzers require labor‐intensive work and a panel of expensive reagents. Herein, an artificial‐intelligence‐enabled reagent‐free imaging hematology analyzer (AIRFIHA) modality is reported that can accurately classify subpopulations of leukocytes with minimal sample preparation. AIRFIHA is realized through training a two‐step residual neural network using label‐free images of isolated leukocytes acquired from a custom‐built quantitative phase microscope. By leveraging the rich information contained in quantitative phase images, not only high accuracy is achieved in differentiating B and T lymphocytes, but also CD4 and CD8 T cells are classified, therefore outperforming the classification accuracy of most current hematology analyzers. The performance of AIRFIHA in a randomly selected test set is validated and is cross‐validated across all blood donors. Due to its easy operation, low cost, and accurate discerning capability of complex leukocyte subpopulations, AIRFIHA is clinically translatable and can also be deployed in resource‐limited settings, e.g., during pandemic situations for the rapid screening of infectious diseases.","PeriodicalId":7187,"journal":{"name":"Advanced Intelligent Systems","volume":"1 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2020-12-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"82302942","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}
Kameel Abdel-latif, Robert W. Epps, Fazel Bateni, Suyong Han, Kristofer G. Reyes, M. Abolhasani
Identifying the optimal formulation of emerging inorganic lead halide perovskite quantum dots (LHP QDs) with their vast colloidal synthesis universe and multiple synthesis/postsynthesis processing parameters is a challenging undertaking for material‐ and time‐intensive, batch synthesis strategies. Herein, a modular microfluidic synthesis strategy, integrated with an artificial intelligence (AI)‐guided decision‐making agent for intelligent navigation through the complex colloidal synthesis universe of LHP QDs with 10 individually controlled synthesis parameters and an accessible parameter space exceeding 2 × 107, is introduced. Utilizing the developed autonomous microfluidic experimentation strategy within a global learning framework, the optimal formulation of LHP QDs is rapidly identified through a two‐step colloidal synthesis and postsynthesis halide exchange reaction, for 10 different emission colors in less than 40 min per desired peak emission energy. Using two in‐series microfluidic reactors enables continuous bandgap engineering of LHP QDs via in‐line halide exchange reactions without the need for an intermediate washing step. Using an inert gas within a three‐phase flow format enables successful, self‐synchronized continuous delivery of halide salt precursor into moving droplets containing LHP QDs, resulting in accelerated closed‐loop formulation optimization and end‐to‐end continuous manufacturing of LHP QDs with desired optoelectronic properties.
{"title":"Self‐Driven Multistep Quantum Dot Synthesis Enabled by Autonomous Robotic Experimentation in Flow","authors":"Kameel Abdel-latif, Robert W. Epps, Fazel Bateni, Suyong Han, Kristofer G. Reyes, M. Abolhasani","doi":"10.1002/aisy.202000245","DOIUrl":"https://doi.org/10.1002/aisy.202000245","url":null,"abstract":"Identifying the optimal formulation of emerging inorganic lead halide perovskite quantum dots (LHP QDs) with their vast colloidal synthesis universe and multiple synthesis/postsynthesis processing parameters is a challenging undertaking for material‐ and time‐intensive, batch synthesis strategies. Herein, a modular microfluidic synthesis strategy, integrated with an artificial intelligence (AI)‐guided decision‐making agent for intelligent navigation through the complex colloidal synthesis universe of LHP QDs with 10 individually controlled synthesis parameters and an accessible parameter space exceeding 2 × 107, is introduced. Utilizing the developed autonomous microfluidic experimentation strategy within a global learning framework, the optimal formulation of LHP QDs is rapidly identified through a two‐step colloidal synthesis and postsynthesis halide exchange reaction, for 10 different emission colors in less than 40 min per desired peak emission energy. Using two in‐series microfluidic reactors enables continuous bandgap engineering of LHP QDs via in‐line halide exchange reactions without the need for an intermediate washing step. Using an inert gas within a three‐phase flow format enables successful, self‐synchronized continuous delivery of halide salt precursor into moving droplets containing LHP QDs, resulting in accelerated closed‐loop formulation optimization and end‐to‐end continuous manufacturing of LHP QDs with desired optoelectronic properties.","PeriodicalId":7187,"journal":{"name":"Advanced Intelligent Systems","volume":"61 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2020-12-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"74625601","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}
The increasing power consumption in the existing computation architecture presents grand challenges for the performance and reliability of very‐large‐scale integrated circuits. Inspired by the characteristics of the human brain for processing complicated tasks with low power, neuromorphic computing is intensively investigated for decreasing power consumption and enriching computation functions. Hardware implementation of neuromorphic computing with emerging devices substantially reduces power consumption down to a few mW cm−2, compared with the central processing unit based on conventional Si complementary metal–oxide semiconductor (CMOS) technologies (50–100 W cm−2). Herein, a brief introduction on the characteristics of neuromorphic computing is provided. Then, emerging devices for low‐power neuromorphic computing are overviewed, e.g., resistive random access memory with low power consumption (< pJ) per synaptic event. A few computation models for artificial neural networks (NNs), including spiking neural network (SNN) and deep neural network (DNN), which boost power efficiency by simplifying the computing procedure and minimizing memory access are discussed. A few examples for system‐level demonstration are described, such as mixed synchronous–asynchronous and reconfigurable convolution neuron network (CNN)–recurrent NN (RNN) for low‐power computing.
现有计算架构中不断增加的功耗对超大规模集成电路的性能和可靠性提出了巨大的挑战。受人脑处理低功耗复杂任务的特点的启发，神经形态计算在降低功耗和丰富计算功能方面得到了广泛的研究。与基于传统Si互补金属氧化物半导体(CMOS)技术(50-100 W cm - 2)的中央处理单元相比，新兴器件的神经形态计算硬件实现大大降低了功耗，功耗低至几mW cm - 2。本文简要介绍了神经形态计算的特点。然后，概述了用于低功耗神经形态计算的新兴器件，例如，每个突触事件低功耗(< pJ)的电阻式随机存取存储器。讨论了几种人工神经网络(NNs)的计算模型，包括尖峰神经网络(SNN)和深度神经网络(DNN)，它们通过简化计算过程和最小化内存访问来提高功率效率。本文描述了一些用于系统级演示的例子，例如用于低功耗计算的混合同步-异步和可重构卷积神经元网络(CNN) -循环神经网络(RNN)。
{"title":"Low‐Power Computing with Neuromorphic Engineering","authors":"Dingbang Liu, Hao Yu, Y. Chai","doi":"10.1002/aisy.202000150","DOIUrl":"https://doi.org/10.1002/aisy.202000150","url":null,"abstract":"The increasing power consumption in the existing computation architecture presents grand challenges for the performance and reliability of very‐large‐scale integrated circuits. Inspired by the characteristics of the human brain for processing complicated tasks with low power, neuromorphic computing is intensively investigated for decreasing power consumption and enriching computation functions. Hardware implementation of neuromorphic computing with emerging devices substantially reduces power consumption down to a few mW cm−2, compared with the central processing unit based on conventional Si complementary metal–oxide semiconductor (CMOS) technologies (50–100 W cm−2). Herein, a brief introduction on the characteristics of neuromorphic computing is provided. Then, emerging devices for low‐power neuromorphic computing are overviewed, e.g., resistive random access memory with low power consumption (< pJ) per synaptic event. A few computation models for artificial neural networks (NNs), including spiking neural network (SNN) and deep neural network (DNN), which boost power efficiency by simplifying the computing procedure and minimizing memory access are discussed. A few examples for system‐level demonstration are described, such as mixed synchronous–asynchronous and reconfigurable convolution neuron network (CNN)–recurrent NN (RNN) for low‐power computing.","PeriodicalId":7187,"journal":{"name":"Advanced Intelligent Systems","volume":"101 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2020-12-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"88412006","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}
Minimally invasive endovascular interventions have become the cornerstone of medical practice in the treatment of a variety of vascular diseases. Tools developed for these interventions have also opened new avenues for targeted delivery of therapeutics, such as chemotherapy or radiation therapy, using the vessels as highways into remote lesions. A more ambitious move toward an all‐endovascular approach to acute or chronic conditions, however, is fundamentally hindered by a variety of challenges, such as the complexity of the vascular anatomy, access to smaller vessels, the fragility of diseased vessels, emergency procedure requirements, prolonged exposure to ionizing X‐ray radiation, and patient‐specific factors including coagulopathy. These shortcomings necessitate new advances to the current practice. Smart soft‐body robots that fit the smallest vessels with high‐precision wireless control and autonomous capabilities have the potential to set the future standards of minimally invasive endovascular therapies. Herein, the current state of the small‐scale robotics from the viewpoint of endovascular applications is discussed, and their potential advantages to the existing tethered clinical devices are compared. Then, technical challenges and the clinical requirements toward realistic applications of small‐scale untethered robots inside the vasculature are discussed.
{"title":"Robotic Devices for Minimally Invasive Endovascular Interventions: A New Dawn for Interventional Radiology","authors":"S. Gunduz, H. Albadawi, R. Oklu","doi":"10.1002/aisy.202000181","DOIUrl":"https://doi.org/10.1002/aisy.202000181","url":null,"abstract":"Minimally invasive endovascular interventions have become the cornerstone of medical practice in the treatment of a variety of vascular diseases. Tools developed for these interventions have also opened new avenues for targeted delivery of therapeutics, such as chemotherapy or radiation therapy, using the vessels as highways into remote lesions. A more ambitious move toward an all‐endovascular approach to acute or chronic conditions, however, is fundamentally hindered by a variety of challenges, such as the complexity of the vascular anatomy, access to smaller vessels, the fragility of diseased vessels, emergency procedure requirements, prolonged exposure to ionizing X‐ray radiation, and patient‐specific factors including coagulopathy. These shortcomings necessitate new advances to the current practice. Smart soft‐body robots that fit the smallest vessels with high‐precision wireless control and autonomous capabilities have the potential to set the future standards of minimally invasive endovascular therapies. Herein, the current state of the small‐scale robotics from the viewpoint of endovascular applications is discussed, and their potential advantages to the existing tethered clinical devices are compared. Then, technical challenges and the clinical requirements toward realistic applications of small‐scale untethered robots inside the vasculature are discussed.","PeriodicalId":7187,"journal":{"name":"Advanced Intelligent Systems","volume":"3 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2020-11-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"79697482","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}
There is a lack of reliable prognostic biomarkers for hypoxic‐ischemic (HI) brain injury in preterm infants. Herein, spectrally detailed wavelet scalograms (WSs), derived from the 1024 Hz sampled electroencephalograms (EEG) of preterm fetal sheep after HI (n = 7), are infused into a high‐performance deep convolutional neural network (CNN) pattern classifier to identify high‐frequency spike transient biomarkers. The deep WS‐CNN pattern classifier identifies EEG spikes with remarkable accuracy of 99.81 = 0.15% (area under curve, AUC = 1.000), cross‐validated across 5010 EEG waveforms, during the first 6 h post‐HI (42 h total), an important clinical period for diagnosis of HI brain injury. Further, a feature‐fusion strategy is introduced to extract the spectrally dominant features of the raw EEG epochs to form robust 3D input matrix sets to be infused into the deep 2D‐CNNs for pattern classification. The results show that the proposed WS‐CNN approach is less sensitive to the potential morphological variations of spikes across all subjects compared to other deep CNNs and spectral‐fuzzy classifiers, allowing the user to flexibly choose an approach depending on their computational requirements. Collectively, the data provide a reliable framework that could help support well‐timed diagnosis of at‐risk neonates in clinical practice.
{"title":"Advanced Deep Learning Spectroscopy of Scalogram Infused CNN Classifiers for Robust Identification of Post‐Hypoxic Epileptiform EEG Spikes","authors":"H. Abbasi, A. Gunn, C. Unsworth, L. Bennet","doi":"10.1002/aisy.202000198","DOIUrl":"https://doi.org/10.1002/aisy.202000198","url":null,"abstract":"There is a lack of reliable prognostic biomarkers for hypoxic‐ischemic (HI) brain injury in preterm infants. Herein, spectrally detailed wavelet scalograms (WSs), derived from the 1024 Hz sampled electroencephalograms (EEG) of preterm fetal sheep after HI (n = 7), are infused into a high‐performance deep convolutional neural network (CNN) pattern classifier to identify high‐frequency spike transient biomarkers. The deep WS‐CNN pattern classifier identifies EEG spikes with remarkable accuracy of 99.81 = 0.15% (area under curve, AUC = 1.000), cross‐validated across 5010 EEG waveforms, during the first 6 h post‐HI (42 h total), an important clinical period for diagnosis of HI brain injury. Further, a feature‐fusion strategy is introduced to extract the spectrally dominant features of the raw EEG epochs to form robust 3D input matrix sets to be infused into the deep 2D‐CNNs for pattern classification. The results show that the proposed WS‐CNN approach is less sensitive to the potential morphological variations of spikes across all subjects compared to other deep CNNs and spectral‐fuzzy classifiers, allowing the user to flexibly choose an approach depending on their computational requirements. Collectively, the data provide a reliable framework that could help support well‐timed diagnosis of at‐risk neonates in clinical practice.","PeriodicalId":7187,"journal":{"name":"Advanced Intelligent Systems","volume":"7 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2020-11-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"77987668","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}
Leonardo Dominguez Rubio, M. Potomkin, R. Baker, Ayusman Sen, L. Berlyand, I. Aranson
Active particles consume energy stored in the environment and convert it into mechanical motion. Many potential applications of these systems involve their flowing, extrusion, and deposition through channels and nozzles, such as targeted drug delivery and out‐of‐equilibrium self‐assembly. However, understanding their fundamental interactions with flow and boundaries remain incomplete. Herein, experimental and theoretical studies of hydrogen peroxide (H2O2) powered self‐propelled gold–platinum nanorods in parallel channels and nozzles are conducted. The behaviors of active (self‐propelled) and passive rods are systematically compared. It is found that most active rods self‐align with the flow streamlines in areas with high shear and exhibit rheotaxis (swimming against the flow). In contrast, passive rods continue moving unaffected until the flow rate is very high, at which point they also start showing some alignment. The experimental results are rationalized by computational modeling delineating activity and rod‐flow interactions. The obtained results provide insight into the manipulation and control of active particle flow and extrusion in complex geometries.
{"title":"Self‐Propulsion and Shear Flow Align Active Particles in Nozzles and Channels","authors":"Leonardo Dominguez Rubio, M. Potomkin, R. Baker, Ayusman Sen, L. Berlyand, I. Aranson","doi":"10.1002/aisy.202000178","DOIUrl":"https://doi.org/10.1002/aisy.202000178","url":null,"abstract":"Active particles consume energy stored in the environment and convert it into mechanical motion. Many potential applications of these systems involve their flowing, extrusion, and deposition through channels and nozzles, such as targeted drug delivery and out‐of‐equilibrium self‐assembly. However, understanding their fundamental interactions with flow and boundaries remain incomplete. Herein, experimental and theoretical studies of hydrogen peroxide (H2O2) powered self‐propelled gold–platinum nanorods in parallel channels and nozzles are conducted. The behaviors of active (self‐propelled) and passive rods are systematically compared. It is found that most active rods self‐align with the flow streamlines in areas with high shear and exhibit rheotaxis (swimming against the flow). In contrast, passive rods continue moving unaffected until the flow rate is very high, at which point they also start showing some alignment. The experimental results are rationalized by computational modeling delineating activity and rod‐flow interactions. The obtained results provide insight into the manipulation and control of active particle flow and extrusion in complex geometries.","PeriodicalId":7187,"journal":{"name":"Advanced Intelligent Systems","volume":"11 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2020-11-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"77674263","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}
Chao-Chiun Liang, Hailin Jiang, Shaopeng Lin, Huashan Li, Biao Wang
Intelligent generation of time‐variant control series remains the critical challenge for acquiring the desired system evolution, due to the difficulties in perceiving temporal correlation and conducting appropriate feedback propagation. A machine learning (ML) algorithm named time‐series generative adversarial network (TSGAN) is developed to overcome the difficulties, by incorporating a long short‐term memory (LSTM) kernel for recognizing multirange temporal patterns beyond the Markovian approximation and an adversarial training mechanism for efficient optimization. A variety of time series are examined by temperature‐control experiments, and the results demonstrate an exceptional accuracy (>95%, 35% higher than prevalent ML methods) as well as strong transferability and stability of the TSGAN algorithm. The dependence of generation performance on underlying statistical mechanisms associated with different ML algorithms, including the deep neural network (DNN), hidden Markov model (HMM), LSTM, and TSGAN, is elucidated by analyzing the generation quality of characteristic temporal patterns. The capability of generating arbitrarily complex response series opens an opportunity for inverse design of time‐variant functionals as strenuously pursued in material science and modern technology.
{"title":"Intelligent Generation of Evolutionary Series in a Time‐Variant Physical System via Series Pattern Recognition","authors":"Chao-Chiun Liang, Hailin Jiang, Shaopeng Lin, Huashan Li, Biao Wang","doi":"10.1002/aisy.202000172","DOIUrl":"https://doi.org/10.1002/aisy.202000172","url":null,"abstract":"Intelligent generation of time‐variant control series remains the critical challenge for acquiring the desired system evolution, due to the difficulties in perceiving temporal correlation and conducting appropriate feedback propagation. A machine learning (ML) algorithm named time‐series generative adversarial network (TSGAN) is developed to overcome the difficulties, by incorporating a long short‐term memory (LSTM) kernel for recognizing multirange temporal patterns beyond the Markovian approximation and an adversarial training mechanism for efficient optimization. A variety of time series are examined by temperature‐control experiments, and the results demonstrate an exceptional accuracy (>95%, 35% higher than prevalent ML methods) as well as strong transferability and stability of the TSGAN algorithm. The dependence of generation performance on underlying statistical mechanisms associated with different ML algorithms, including the deep neural network (DNN), hidden Markov model (HMM), LSTM, and TSGAN, is elucidated by analyzing the generation quality of characteristic temporal patterns. The capability of generating arbitrarily complex response series opens an opportunity for inverse design of time‐variant functionals as strenuously pursued in material science and modern technology.","PeriodicalId":7187,"journal":{"name":"Advanced Intelligent Systems","volume":"112 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2020-11-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"79646364","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}
Gungun Lin, Yuan Liu, Guan Huang, Yinghui Chen, D. Makarov, Jun Lin, Z. Quan, D. Jin
Utilizing the magnetic interactions between microparticle building blocks allows creating long‐range ordered structures and constructing smart multifunctional systems at different scales. The elaborate control over the inter‐particle magnetic coupling interaction is entailed to unlock new magnetoactuation functionalities. Herein, dimer‐type microstructures consisting of a pair of magnetic emulsions with tailorable dimension and magnetic coupling strength are fabricated using a microfluidic emulsion‐templated assembly approach. The magnetite nanoparticles dispersed in vinylbenzene monomers are partitioned into a pair of emulsions with conserved volume, which are wrapped by an aqueous hydrogel shell and finally polymerized to form discrete structures. Tunable synchronous–asynchronous rotation over 60 dB is unlocked in magnetic dimers, which is shown to be dependent on the magnetic moments induced. This leads to a new class of magnetic actuators for the parallelized assay of distinctive virus DNAs and the dynamic optical evaluation of 3D cell cultures. The work suggests a new perspective to design smart multifunctional microstructures and devices by exploring their natural variance in magnetic coupling.
{"title":"3D Rotation‐Trackable and Differentiable Micromachines with Dimer‐Type Structures for Dynamic Bioanalysis","authors":"Gungun Lin, Yuan Liu, Guan Huang, Yinghui Chen, D. Makarov, Jun Lin, Z. Quan, D. Jin","doi":"10.1002/aisy.202000205","DOIUrl":"https://doi.org/10.1002/aisy.202000205","url":null,"abstract":"Utilizing the magnetic interactions between microparticle building blocks allows creating long‐range ordered structures and constructing smart multifunctional systems at different scales. The elaborate control over the inter‐particle magnetic coupling interaction is entailed to unlock new magnetoactuation functionalities. Herein, dimer‐type microstructures consisting of a pair of magnetic emulsions with tailorable dimension and magnetic coupling strength are fabricated using a microfluidic emulsion‐templated assembly approach. The magnetite nanoparticles dispersed in vinylbenzene monomers are partitioned into a pair of emulsions with conserved volume, which are wrapped by an aqueous hydrogel shell and finally polymerized to form discrete structures. Tunable synchronous–asynchronous rotation over 60 dB is unlocked in magnetic dimers, which is shown to be dependent on the magnetic moments induced. This leads to a new class of magnetic actuators for the parallelized assay of distinctive virus DNAs and the dynamic optical evaluation of 3D cell cultures. The work suggests a new perspective to design smart multifunctional microstructures and devices by exploring their natural variance in magnetic coupling.","PeriodicalId":7187,"journal":{"name":"Advanced Intelligent Systems","volume":"7 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2020-11-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"87190742","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：