Optimization for Gold Nanostructure-Based Surface Plasmon Biosensors Using a Microgenetic Algorithm

Fu, Po‐Han; Lo, Shang Lien; Tsai, Po-Cheng; Lee, Kuang-Li; Wei, Pei‐Kuen

doi:10.1021/acsphotonics.8b00136

Cited by 26 publications

(14 citation statements)

References 34 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Photonic devices optimized with the GA work in the long-wavelength range [39][40][41][42]. Moreover, the GA has been implemented for the optimization of a surfaceplasmon biosensor [20]. One can observe that powerful designs could be obtained for different applications using the GA optimization technique.…”

Section: Discussionmentioning

confidence: 99%

See 1 more Smart Citation

Genetic Algorithm-Driven Surface-Enhanced Raman Spectroscopy Substrate Optimization

et al. 2021

View full text Add to dashboard Cite

Surface-enhanced Raman spectroscopy (SERS) is a highly sensitive and molecule-specific detection technique that uses surface plasmon resonances to enhance Raman scattering from analytes. In SERS system design, the substrates must have minimal or no background at the incident laser wavelength and large Raman signal enhancement via plasmonic confinement and grating modes over large areas (i.e., squared millimeters). These requirements impose many competing design constraints that make exhaustive parametric computational optimization of SERS substrates prohibitively time consuming. Here, we demonstrate a genetic-algorithm (GA)-based optimization method for SERS substrates to achieve strong electric field localization over wide areas for reconfigurable and programmable photonic SERS sensors. We analyzed the GA parameters and tuned them for SERS substrate optimization in detail. We experimentally validated the model results by fabricating the predicted nanostructures using electron beam lithography. The experimental Raman spectrum signal enhancements of the optimized SERS substrates validated the model predictions and enabled the generation of a detailed Raman profile of methylene blue fluorescence dye. The GA and its optimization shown here could pave the way for photonic chips and components with arbitrary design constraints, wavelength bands, and performance targets.

show abstract

Section: Discussionmentioning

confidence: 99%

“…It is an effective technique for nonlinear problems with multiple local solutions. The GA is becoming more commonly used in optics for the optimization of polarization rotators [18], integrated optical devices [19], and biosensors [20].…”

Section: Introductionmentioning

confidence: 99%

Genetic Algorithm-Driven Surface-Enhanced Raman Spectroscopy Substrate Optimization

et al. 2021

View full text Add to dashboard Cite

show abstract

“…They discovered that refractive index resolution improved by six times more than a traditional one [24]. A gold nanostructure based SPR biosensor based on microgenetic algorithms was presented by Fu et al for enhancing detection sensitivity [25].…”

Section: Introductionmentioning

confidence: 99%

Sensitivity Improvement of a Surface Plasmon Resonance Sensor Based on Two-Dimensional Materials Hybrid Structure in Visible Region: A Theoretical Study

Lin

Chen

Lin

2020

Sensors

View full text Add to dashboard Cite

In this paper, we propose a surface plasmon resonance (SPR) sensor based on two-dimensional (2D) materials (graphene, MoS2, WS2 and WSe2) hybrid structure, and theoretically investigate its sensitivity improvement in the visible region. The thickness of metal (Au, Ag or Cu) and the layer number of each 2D material are optimized using genetic algorithms to obtain the highest sensitivity for a specific wavelength of incident light. Then, the sensitivities of proposed SPR sensors with different metal films at various wavelengths are compared. An Ag-based SPR sensor exhibits a higher sensitivity than an Au- or Cu-based one at most wavelengths in the visible region. In addition, the sensitivity of the proposed SPR sensor varies obviously with the wavelength of incident light, and shows a maximum value of 159, 194 or 155°/RIU for Au, Ag or Cu, respectively. It is demonstrated that the sensitivity of the SPR sensor based on 2D materials’ hybrid structure can be further improved by optimizing the wavelength of incident light.

show abstract

“…Hughes inspire us that the learning process of the ONNs can be converted to the inverse design problems that are solved by using gradient-based methods or gradient free methods [34,35]. Apart from gradient-based methods (such as the AVM), gradient free methods, for example genetic algorithms (GA) and particle swarm optimization (PSO), also can be applied in the inverse design of photonic devices [36][37][38][39]. Besides that, as an alternative approach to train ANNs, neuroevolution, which derives from the evolution process imitated the biological brain, is a typical gradient free method based on evolutionary algorithms [40].…”

mentioning

confidence: 99%

“…Two kinds of typical gradient free algorithms, the GA and PSO, are tried to validate the effectiveness of the training algorithms based on neuroevolution. The GA is a representative neuroevolution algorithm that is widely used in inverse design and performance optimization of photonic devices [37,38]. In this article, we use the GA to train the ONNs by optimizing for the phase shifters of optical mesh and hyper-parameters of nonlinear activation functions.…”

mentioning

confidence: 99%

Efficient training and design of photonic neural network through neuroevolution

Zhang¹,

Wang²,

Dan³

et al. 2019

Opt. Express

View full text Add to dashboard Cite

Recently, optical neural networks (ONNs) integrated in photonic chips has received extensive attention because they are expected to implement the same pattern recognition tasks in the electronic platforms with high efficiency and low power consumption. However, the current lack of various learning algorithms to train the ONNs obstructs their further development. In this article, we propose a novel learning strategy based on neuroevolution to design and train the ONNs. Two typical neuroevolution algorithms are used to determine the hyper-parameters of the ONNs and to optimize the weights (phase shifters) in the connections. In order to demonstrate the effectiveness of the training algorithms, the trained ONNs are applied in the classification tasks for iris plants dataset, wine recognition dataset and modulation formats recognition. The calculated results exhibit that the training algorithms based on neuroevolution are competitive with other traditional learning algorithms on both accuracy and stability. Compared with previous works, we introduce an efficient training method for the ONNs and demonstrate their broad application prospects in pattern recognition, reinforcement learning and so on. IntroductionArtificial neural networks (ANNs), deep learning [1] in particular, has attracted a great deal of research attentions for an impressively large number of applications, such as image processing [2], natural language processing [3], acoustical signal processing [4], time series processing [5], self-driving [6], games [7], robot [8] and so on. It should be noted that the training of the ANNs with deep hidden layers, especially for convolutional neural networks (CNNs) and recurrent neural networks (RNNs), for example AlexNet [9], VGGNet [10], GoogLeNet [11], ResNet [12] and long short-term memory [13], typically demands significant computational time and resources [1]. Thus, various electronic special-purpose platforms based on graphical processing units (GPUs) [14], field-programmable gate arrays (FPGAs) [15] and applicationspecific integrated circuits (ASICs) [16] were invented to accelerate the training and inference process of deep learning. On the other hand, in order to obtain general artificial intelligence, some brain-inspired chips including IBM TrueNorth [17], Intel Loihi [18], and SpiNNaker [19]were designed by imitating the structure of a brain. However, even both energy efficiency and speed were improved, the performances of the brain-inspired chips were difficult to compete with the state of the art of deep learning [20]. In the recent years, optical computing had been demonstrated as an effective alternative to traditionally electronic computing architectures and expected to alleviate the bandwidth bottlenecks and power consumption in electronics [21]. For example, new photonic approaches for spiking neuron and scalable network architecture based upon excitable lasers, broadcast-and-weight protocol and reservoir computing had been illustrated [22][23][24]. Despite ultrafast spiking response were achiev...

show abstract

Optimization for Gold Nanostructure-Based Surface Plasmon Biosensors Using a Microgenetic Algorithm

Cited by 26 publications

References 34 publications

Genetic Algorithm-Driven Surface-Enhanced Raman Spectroscopy Substrate Optimization

Genetic Algorithm-Driven Surface-Enhanced Raman Spectroscopy Substrate Optimization

Sensitivity Improvement of a Surface Plasmon Resonance Sensor Based on Two-Dimensional Materials Hybrid Structure in Visible Region: A Theoretical Study

Efficient training and design of photonic neural network through neuroevolution

Contact Info

Product

Resources

About