Article
Chemistry, Multidisciplinary
Xiaoli Wang, Yan Piao, Yuanshang Jin, Jie Li, Zechuan Lin, Jie Cui, Tingfa Xu
Summary: Fourier ptychographic microscopy is a new computational imaging technology that allows for a large field of view and high resolution. This article proposes a Fourier ptychographic reconstruction method applied to a self-training physical model, incorporating the SwinIR network for the first time. Experimental results demonstrate that the proposed method achieves the best evaluation index values and produces the highest quality image reconstruction after model training.
APPLIED SCIENCES-BASEL
(2023)
Article
Nanoscience & Nanotechnology
Kyung Chul Lee, Kyungwon Lee, Jaewoo Jung, Se Hee Lee, Donghyun Kim, Seung Ah Lee
Summary: The research introduces a high-performance microscope device that can be mounted on a smartphone, utilizing the smartphone's camera, display screen, and processor for imaging and image reconstruction. The device achieves high resolution and a wide field of view, with potential for a variety of clinical applications.
Article
Biochemical Research Methods
Jizhou Zhang, Tingfa Xu, Jianan Li, Yuhan Zhang, Shenwang Jiang, Yiwen Chen, Jinhua Zhang
Summary: The article introduces a method for FPM reconstruction using physics-based neural network structures and channel attention modules, effectively addressing issues such as LED intensity correction and pupil function recovery, while demonstrating better performance in complex field reconstruction.
JOURNAL OF BIOPHOTONICS
(2022)
Article
Engineering, Electrical & Electronic
Ming Sun, Yutong Li, Guancheng Huang, Jiaxin Wang, Jiubin Tan, Shutian Liu, Bin Gao, Zhengjun Liu
Summary: SAS-FPM is a self-adapting search algorithm for FPM that improves data acquisition efficiency, verified through simulations and experiments to shorten the capture time by more than half compared to traditional FPM.
OPTICAL AND QUANTUM ELECTRONICS
(2021)
Article
Optics
Pavan Chandra Konda, Jonathan M. Taylor, Andrew R. Harvey
Summary: The novel microscopy concept, Multi-Aperture Fourier ptychographic microscopy (MA-FPM), introduces parallel detectors in microscopy to increase the space-bandwidth-time product. By using a synthetic aperture technique and FP algorithms to recover phase, high-resolution, wide field-of-view images are synthesized.
OPTICS AND LASERS IN ENGINEERING
(2021)
Article
Chemistry, Analytical
Xiaoli Wang, Yan Piao, Jinyang Yu, Jie Li, Haixin Sun, Yuanshang Jin, Limin Liu, Tingfa Xu
Summary: This paper presents a novel Fourier ptychographic microscopy imaging reconstruction method based on a deep multi-feature transfer network, which achieves anti-noise performance, high-resolution, and reduced image data. By extracting image features using transfer learning ResNet50, Xception, and DenseNet121 networks, and adopting cascaded feature fusion strategy, as well as using pre-upsampling reconstruction network, high-quality image reconstruction is achieved.
Article
Chemistry, Analytical
Lyes Bouchama, Bernadette Dorizzi, Jacques Klossa, Yaneck Gottesman
Summary: This paper introduces an algorithmic approach for Fourier Ptychographic Microscopy (FPM) image reconstruction using a physics-informed optimization deep neural network and statistical reconstruction learning. Simulation results demonstrate the conceptual benefits of the approach and show that high-quality images can be effectively reconstructed without resolution degradation. The learning step is also shown to be mandatory.
Article
Chemistry, Analytical
Jie Li, Jingzi Hao, Xiaoli Wang, Yongshan Wang, Yan Wang, Hao Wang, Xinbo Wang, Stefano Berretti, Jean-Baptiste Thomas, Baptiste Magnier, Khizar Hayat
Summary: In this paper, a hybrid attention network that combines spatial attention mechanisms with channel attention mechanisms into FPM reconstruction is introduced. The network can extract fine spatial features, reduce redundant features, and adaptively adjust hierarchical features to achieve the conversion of low-resolution complex amplitude images to high-resolution ones. The high-resolution images generated by this method can be applied to medical cell recognition, segmentation, classification, and other related studies, providing a better foundation for relevant research.
Article
Engineering, Electrical & Electronic
Vittorio Bianco, Mattia Delli Priscoli, Daniele Pirone, Gennaro Zanfardino, Pasquale Memmolo, Francesco Bardozzo, Lisa Miccio, Gioele Ciaparrone, Pietro Ferraro, Roberto Tagliaferri
Summary: Researchers use a generative adversarial network to reconstruct Fourier ptychographic images in real-time, even with severely misaligned setups. The network accurately retrieves important morphometric information relevant for neural tissue diagnosis, speeding up quantitative phase-contrast analysis of tissue samples.
IEEE JOURNAL OF SELECTED TOPICS IN QUANTUM ELECTRONICS
(2022)
Article
Optics
Peiwei Zhang, Jufeng Zhao, Binbin Lin, Xiaohui Wu, Guangmang Cui
Summary: This paper proposes an efficient strategy for hyperspectral image reconstruction based on the Fourier ptychographic microscopy (FPM) system. By optimizing the experimental setup and using a new algorithm, the paper achieves image reconstruction with high spatial resolution, wide field of view, and hyperspectral resolution.
Article
Engineering, Electrical & Electronic
Shuhe Zhang, Tos T. J. M. Berendschot, Jinhua Zhou
Summary: We propose a simple and efficient reconstruction algorithm, Elfpie, for Fourier ptychographic microscopy, which is robust to various system errors without the need for calibration or recovery. Elfpie utilizes a new image gradient-based data fidelity cost function regularized by global second-order total-variation regularization, and incorporates the AdaBelief optimizer with an adaptive learning rate.
Article
Optics
An Pan, Aiye Wang, Junfu Zheng, Yuting Gao, Caiwen Ma, Baoli Yao
Summary: Fourier ptychographic microscopy (FPM) is a computational imaging technique that offers high resolution, wide field-of-view, and quantitative phase recovery. However, the presence of an imperceptible artifact caused by edge effect degrades the precision of phase imaging in FPM. To address this issue, two opposite algorithms called discrete cosine transform (DCT) and periodic plus smooth image decomposition (PPSID) were proposed and discussed systematically. The PPSID-FPM algorithm significantly improves the accuracy of phase measurement and is comparable to the conventional FPM algorithm.
OPTICS AND LASERS IN ENGINEERING
(2023)
Article
Physics, Multidisciplinary
Yuting Gao, Jiurun Chen, Aiye Wang, An Pan, Caiwen Ma, Baoli Yao
Summary: The usage of full-color imaging in digital pathology produces significant results, but the conventional full-color digital pathology based on FPM is still time-consuming. The CFPM method significantly reduces reconstruction time while sacrificing only a small amount of precision by transferring color texture information.
SCIENCE CHINA-PHYSICS MECHANICS & ASTRONOMY
(2021)
Article
Optics
Yao Fan, Jiasong Sun, Yefeng Shu, Zuxin Zhang, Guoan Zheng, Wenjian Chen, Jin Zhang, Kun Gui, Kehui Wang, Qian Chen, Chao Zuo
Summary: Fourier ptychographic microscopy (FPM) is a computational high-throughput technique that achieves high-resolution and wide field-of-view (FOV) imaging. This paper proposes an efficient synthetic aperture scheme for FPM, called ESA-FPM, which utilizes both coherent and incoherent illuminations to maximize data utilization and achieve high-resolution reconstruction with few acquisitions. The experiment demonstrates that ESA-FPM achieves theoretical resolution with only 1.6% of the data of conventional FPM.
LASER & PHOTONICS REVIEWS
(2023)
Article
Optics
Youqiang Zhu, Minglu Sun, Peilin Wu, Quanquan Mu, Li Xuan, Dayu Li, Bin Wang
Summary: Fourier ptychographic microscopy, a super-resolution technique, overcomes the Space-Band Product limit by employing varied-illumination and phase retrieval algorithm. However, misalignment errors caused by inadequate installation accuracy of the LED array affect the calculation and result in artifacts. This paper proposes a Space based correction (SBC) method using Particle swarm optimization (PSO) algorithm, which is more stable and accurate compared to existing methods.
OPTICS COMMUNICATIONS
(2022)
Article
Computer Science, Artificial Intelligence
Yujie Wang, Chenggang Yan, Yutong Feng, Shaoyi Du, Qionghai Dai, Yue Gao
Summary: Partial point cloud registration transforms partial scans into a common coordinate system, which is crucial for generating complete 3D shapes. Traditional registration methods struggle with small point cloud overlaps, but the STORM method utilizes structure information to accurately detect overlap and generate precise partial correspondences, achieving superior performance.
IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE
(2023)
Article
Optics
Alex Matlock, Jiabei Zhu, Lei Tian
Summary: Recovering the 3D phase features of complex biological samples has traditionally involved sacrificing computational efficiency and processing time for physical model accuracy and reconstruction quality. However, this study introduces an approximant-guided deep learning framework that overcomes this challenge in a high-speed intensity diffraction tomography system. By training the network on natural image datasets using a physics model simulator-based learning strategy, complex 3D biological samples can be robustly reconstructed. This framework utilizes a lightweight 2D network structure with a multi-channel input to encode axial information, achieving highly efficient training and prediction.
Article
Optics
Yun Zhao, Hang Chen, Min Lin, Haiou Zhang, Tao Yan, Ruqi Huang, Xing Lin, Qionghai Dai
Summary: Increasing the layer number of on-chip photonic neural networks (PNNs) is important for improving model performance. To address the issue of larger chip area occupancy due to cascading hidden layers, the optical neural ordinary differential equations (ON-ODEs) architecture parameterizes the continuous dynamics of hidden layers with optical ODE solvers. The ON-ODEs demonstrate high accuracy in image classification tasks and improved model classification accuracy for diffraction-based all-optical linear hidden layer.
Article
Biochemical Research Methods
Jelena Platisa, Xin Ye, Allison M. Ahrens, Chang Liu, Ichun Anderson Chen, Ian G. Davison, Lei Tian, Vincent A. Pieribone, Jerry L. Chen
Summary: Monitoring spiking activity in large neuronal populations is crucial for understanding neural circuit function. Voltage imaging provides a new approach for this, but it faces challenges such as reduced fluorescence detection and limited imaging duration. This study developed improved voltage indicators, a high-speed two-photon microscope, and denoising software, enabling simultaneous high-speed deep-tissue imaging of more than 100 labeled neurons over 1 hour. This scalable approach offers a way to image voltage activity across increasing neuronal populations.
Article
Biochemical Research Methods
Zhi Lu, Yu Liu, Manchang Jin, Xin Luo, Huanjing Yue, Zian Wang, Siqing Zuo, Yunmin Zeng, Jiaqi Fan, Yanwei Pang, Jiamin Wu, Jingyu Yang, Qionghai Dai
Summary: Virtual-scanning light-field microscopy (VsLFM) uses a physics-based deep learning model to improve the quality and speed of LFM, reducing motion artifacts and enabling challenging demonstrations such as fast 3D voltage imaging in Drosophila. By constructing a 40 GB high-resolution scanning LFM dataset across different species, VsLFM exploits physical priors and bypasses hardware scanning to achieve ultrafast 3D imaging in various processes such as the beating heart in embryonic zebrafish and neutrophil migration in the mouse liver.
Article
Biochemical Research Methods
Yuanlong Zhang, Guoxun Zhang, Xiaofei Han, Jiamin Wu, Ziwei Li, Xinyang Li, Guihua Xiao, Hao Xie, Lu Fang, Qionghai Dai
Summary: Widefield microscopy allows optical access to large areas of mammalian brains and thousands of neurons, but signal deterioration due to tissue scattering and background contamination makes neuronal extraction challenging. DeepWonder, a deep-learning-based widefield neuron finder, effectively removes background contaminations and achieves high-fidelity neuronal extraction. It enhances signal-to-background ratio by 50-fold and extracts over 14,000 neurons in 17 hours, providing accurate and efficient neuronal segmentation.
Article
Optics
Hao Wang, Jiabei Zhu, Jangwoon Sung, Guorong Hu, Joseph Greene, Yunzhe LI, Seungbeom Park, Wookrae Kim, Myungjun Lee, Yusin Yang, Lei Tian
Summary: Topography measurement is crucial for surface characterization and inspection applications. This study presents a novel topography technique called Fourier ptychographic topography (FPT), which combines a computational microscope and a phase retrieval algorithm to achieve wide-field-of-view and high-resolution topography reconstruction with nanoscale accuracy. FPT has important implications for surface characterization, semiconductor metrology, and inspection applications.
Article
Optics
Jian Zhao, Lulu Jiang, Alex Matlock, Yihong Xu, Jiabei Zhu, Hongbo Zhu, Lei Tian, Benjamin Wolozin, Ji-Xin Cheng
Summary: Researchers developed a computational chemical microscope, FBS-IDT, which can extract molecular structure information of amyloid proteins in their native cellular environment. This technology enables label-free volumetric chemical imaging and 3D visualization of amyloid protein aggregates, as well as depth-resolved mid-infrared fingerprint spectroscopy for protein secondary structure analysis. It provides a new approach to study the relationship between neurodegenerative diseases and amyloid proteins.
LIGHT-SCIENCE & APPLICATIONS
(2023)
Article
Multidisciplinary Sciences
Yitong Chen, Tiankuang Zhou, Jiamin Wu, Hui Qiao, Xing Lin, Lu Fang, Qionghai Dai
Summary: With the explosive growth of global data, there is a growing need for high-throughput processing in image transmission systems. Existing methods rely on electronic circuits, limiting transmission throughput. In this study, an all-optical variational autoencoder called PED is proposed, which maps the physical system of image transmission into an optical generative neural network. This work establishes a large-scale high-throughput unsupervised optical computing framework that reduces system latency and transmission errors.
Article
Computer Science, Artificial Intelligence
Junhao Liang, Weisheng Zhang, Jianghui Yang, Meilong Wu, Qionghai Dai, Hongfang Yin, Ying Xiao, Lingjie Kong
Summary: Tissue biomarkers are important for cancer diagnosis, prognosis, and treatment planning, but there are few robust biomarkers with true value. A human-centric deep learning framework called PathFinder is presented to help pathologists discover new tissue biomarkers. Using sparse multi-class tissue spatial distribution information and attribution methods, PathFinder achieves localization, characterization, and verification of potential biomarkers with state-of-the-art prognostic performance. The framework has identified the spatial distribution of necrosis in liver cancer as a strong predictor of patient prognosis and proposed clinically independent indicators for practical prognosis.
NATURE MACHINE INTELLIGENCE
(2023)
Article
Computer Science, Artificial Intelligence
Bo Zhang, Jinli Suo, Qionghai Dai
Summary: Reliable object detection in dark environments is severely challenged by noise and uneven radiance. To address this, we propose illumination-aware spatio-temporal feature fusion modules for low-light video object detection under a TRansformer network structure. Extensive experiments validate the effectiveness of our approach and demonstrate that DVD-TR outperforms state-of-the-art video detectors on a large-scale multi-illuminance dark video benchmark.
INFORMATION FUSION
(2023)
Article
Nanoscience & Nanotechnology
Jianing Liu, Hao Wang, Yuyu Li, Lei Tian, Roberto Paiella
Summary: This paper introduces a new type of image sensor that can directly visualize phase objects without additional optical elements. It is particularly significant for applications involving space-constrained and portable setups, and is applicable to surface profiling and biomedical microscopy.
Article
Computer Science, Artificial Intelligence
Kaiming Dong, Yuchen Guo, Runzhao Yang, Yuxiao Cheng, Jinli Suo, Qionghai Dai
Summary: Video object detection has seen significant progress in recent decades, but existing detectors struggle with low lighting conditions and motion blur. To address this, we propose a method that multiplexes frame sequences into one snapshot and extracts motion cues for trajectory retrieval. By incorporating a programmable shutter and using a deep network called DECENT, we can effectively retrieve bounding boxes from blurred images of dynamic scenes. We generate quasi-real data for network learning, which allows for high generalization on real dark videos. This approach offers advantages of low bandwidth, low cost, compact setup, and high accuracy, and has been experimentally validated for night surveillance.
IEEE TRANSACTIONS ON IMAGE PROCESSING
(2023)
Article
Computer Science, Artificial Intelligence
Donglin Di, Changqing Zou, Yifan Feng, Haiyan Zhou, Rongrong Ji, Qionghai Dai, Yue Gao
Summary: This article proposes a multi-hypergraph based learning framework called HGSurvNet, which achieves an effective high-order global representation of WSIs for patient survival prediction. Extensive validation experiments demonstrate its superiority over state-of-the-art methods.
IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE
(2023)
Article
Computer Science, Artificial Intelligence
Zhihong Zhang, Yuxiao Cheng, Jinli Suo, Liheng Bian, Qionghai Dai
Summary: This article presents a novel non-blind deblurring method called INFWIDE, which can effectively tackle the challenges in low-light photography. By employing a two-branch architecture and fusion network, INFWIDE is able to remove noise, hallucinate saturated regions, and achieve high quality night photograph deblurring. The proposed approach demonstrates superior performance in both synthetic and real data experiments.
IEEE TRANSACTIONS ON IMAGE PROCESSING
(2023)
