Article
Multidisciplinary Sciences
Hui Xu, Nange Jin, Jen-Zen Chuang, Zhao Zhang, Xiaoyue Zhong, Zhijing Zhang, Ching-Hwa Sung, Christophe P. Ribelayga, Yingbin Fu
Summary: This study investigates the physiological roles of cone opsins in mice using a loss-of-function approach. It finds that mice with cone opsin deficiency are unable to form normal outer segments but can survive for an extended period of time. Although these mutant cones do not respond to light directly, they continue to mediate visual signaling by relaying the rod signals through rod-cone gap junctions.
PROCEEDINGS OF THE NATIONAL ACADEMY OF SCIENCES OF THE UNITED STATES OF AMERICA
(2022)
Article
Biotechnology & Applied Microbiology
Xikang Feng, Lingxi Chen, Yuhao Qing, Ruikang Li, Chaohui Li, Shuai Cheng Li
Summary: SCYN, a CNV segmentation method powered with dynamic programming, demonstrates precise segmentation in silico dataset and accurate copy number inferring on triple negative breast cancer scDNA data as well as newly emerged 10x Genomics CNV solution. It is significantly faster than the state of the art tool when dealing with datasets of approximately 2000 cells.
Article
Computer Science, Artificial Intelligence
Huafei Yu, Tinghua Ai, Min Yang, Lina Huang, Aji Gao
Summary: In this study, a deep learning method based on graph convolution neural network (GCNN) is proposed for the segmentation of parallel drainage pattern (SPDP). By constructing a dual drainage graph and defining drainage features, the segmentation task is accomplished. The experiment demonstrates that the proposed method outperforms other machine learning methods and GCNNs, providing a crucial reference for hydrology research.
EXPERT SYSTEMS WITH APPLICATIONS
(2023)
Article
Biochemical Research Methods
Eva Valterova, Jan D. Unterlauft, Mike Francke, Toralf Kirsten, Radim Kolar, Franziska G. Rauscher
Summary: This study presents a fully automated method for retinal analysis using a flood illuminated adaptive optics retinal camera (AO-FIO). The method includes steps such as image registration, photoreceptor detection, and density map generation. It enables comprehensive analysis and comparison with histological data and other published studies. The proposed method is suitable for large studies and allows for fully automated generation of AO-based photoreceptor density maps.
BIOMEDICAL OPTICS EXPRESS
(2023)
Article
Chemistry, Analytical
Maryam Khoshkhabar, Saeed Meshgini, Reza Afrouzian, Sebelan Danishvar
Summary: This paper presents a novel deep learning-based technique for accurately segmenting liver tumors and identifying liver organs. The proposed method achieves satisfactory results in terms of accuracy, Dice coefficient, mean IoU, sensitivity, precision, and recall based on the LiTS17 dataset. The technique is also evaluated in a noisy environment and demonstrates good adaptability and stability. The proposed model is expected to be used to assist radiologists and specialist doctors in the near future based on the positive results.
Article
Chemistry, Multidisciplinary
Agnese Simoni, Eleonora Barcali, Cosimo Lorenzetto, Eleonora Tiribilli, Vieri Rastrelli, Leonardo Manetti, Cosimo Nardi, Ernesto Iadanza, Leonardo Bocchi
Summary: The aim of this study was to provide a tool to aid medical doctors in planning endovascular surgery by quickly detecting stenotic vessels and quantifying the degree of stenosis. The use of skeletonization improved the visualization of vessels and the distance transform provided a linear representation of the diameter of critical vessels selected by the user. The system also estimated the exact distance between landmarks and the occlusion, which is crucial information for surgery planning. The proposed tool offers the advantage of examining chosen vessels in a linear representation free from tortuous vascular courses and vessel crossings.
APPLIED SCIENCES-BASEL
(2023)
Article
Computer Science, Information Systems
Yuanyuan Zeng, Yangfan Li, Xu Zhou, Jianye Yang, Wenjun Jiang, Kenli Li
Summary: This paper proposes a novel Edge-Cut Partitioning approach based on Game theory for dynamic graphs (ECPG) that can handle dynamic graphs and improve partitioning quality. Experimental results show that ECPG significantly outperforms existing algorithms in terms of performance and partitioning quality.
INFORMATION SCIENCES
(2022)
Article
Biochemical Research Methods
Hector Carrion, Mohammad Jafari, Michelle Dawn Bagood, Hsin-ya Yang, Roslyn Rivkah Isseroff, Marcella Gomez
Summary: A deep learning-based image analysis pipeline is developed to efficiently and accurately track the location and size metrics of wounds in experimental animals. The proposed system demonstrates high fidelity results on unseen data and minimal human intervention.
PLOS COMPUTATIONAL BIOLOGY
(2022)
Article
Computer Science, Information Systems
Zhaobin Wang, Xiong Gao, Runliang Wu, Jianfang Kang, Yaonan Zhang
Summary: This study proposes a fully automated image segmentation method based on FCN and Graph Cuts. It utilizes color histograms and mathematical morphological operations to generate seed regions and performs iterative segmentation using superpixel-level Graph Cuts, resulting in higher segmentation accuracy.
MULTIMEDIA SYSTEMS
(2022)
Article
Computer Science, Information Systems
L. Nahabedian, V Braberman, N. D'Ippolito, J. Kramer, S. Uchitel
Summary: The paper introduces a fully automated technique based on formal specifications and discrete event controller synthesis to produce correct-by-construction reconfiguration strategies, addressing transition requirements and process rules effectively.
INFORMATION SYSTEMS
(2022)
Article
Computer Science, Artificial Intelligence
Daniel Fernandez-Alvarez, Jose Emilio Labra-Gayo, Daniel Gayo-Avello
Summary: This paper introduces sheXer, a system that extracts shapes by mining the graph structure. sheXer is offered as a free Python library capable of producing both ShEx and SHACL content. Compared to other automatic shape extractors, sheXer includes some novel features such as shape inter-linkage and computation of big real-world datasets. The features and limitations of sheXer are analyzed with different experiments using the English chapter of DBpedia.
KNOWLEDGE-BASED SYSTEMS
(2022)
Article
Genetics & Heredity
Rebeca A. S. Amaral, Fabiana L. Motta, Olivia A. Zin, Mariana M. da Palma, Gabriela D. Rodrigues, Juliana M. F. Sallum
Summary: Achromatopsia (ACHM) is a congenital cone photoreceptor disorder characterized by reduced visual acuity, nystagmus, photophobia, and very poor or absent color vision. Most ACHM cases are caused by pathogenic variants in the CNGA3 and CNGB3 genes. This study provides a clinical and molecular overview of 42 Brazilian patients with ACHM related to biallelic pathogenic variants in the CNGA3 and CNGB3 genes, and identifies a novel variant in the CNGB3 gene.
Article
Radiology, Nuclear Medicine & Medical Imaging
Zahra Sedghi Gamechi, Andres M. Arias-Lorza, Zaigham Saghir, Daniel Bos, Marleen de Bruijne
Summary: The study proposes a fully automatic method based on an optimal surface graph-cut algorithm for accurate segmentation of the pulmonary arteries and aorta in noncontrast computed tomography (CT) scans. Evaluation results show that the method performs well in segmentation accuracy and diameter measurements.
Article
Engineering, Biomedical
Linbo Wang, Meng Li, Xianyong Fang, Michele Nappi, Shaohua Wan
Summary: The proposed method addresses the issue of distant repetitive patterns in object segmentation by introducing nonlocal affinity and a multi-scale superpixel bipartite graph, which is demonstrated to be effective through extensive experiments on multiple datasets.
BIOMEDICAL SIGNAL PROCESSING AND CONTROL
(2022)
Article
Computer Science, Software Engineering
Recep Sinan Tumen, Metin Sezgin
Summary: This article introduces a method that combines dynamic programming with a novel stroke ordering method for efficient segmentation and recognition in offline drawings. Through rigorous evaluation, it is demonstrated that the combined system outperforms or matches the state of the art in established databases and benchmarks.
IEEE COMPUTER GRAPHICS AND APPLICATIONS
(2022)
Article
Ophthalmology
Jessica Loo, Cindy X. Cai, John Choong, Emily Y. Chew, Martin Friedlander, Glenn J. Jaffe, Sina Farsiu
Summary: The aim of this study was to develop an automatic algorithm for segmenting retinal cavitations on OCT images of patients with MacTel2. The proposed method demonstrated high accuracy and efficiency, and could be a useful tool for clinicians to measure and assess changes in retinal cavitations over time.
BRITISH JOURNAL OF OPHTHALMOLOGY
(2022)
Article
Biochemical Research Methods
Ziyun Yang, Somayyeh Soltanian-Zadeh, Kengyeh K. Chu, Haoran Zhang, Lama Moussa, Ariel E. Watts, Nicholas J. Shaheen, Adam Wax, Sina Farsiu
Summary: A novel neural network model, Bicon-CE, was proposed for in vivo human esophageal OCT layer segmentation, outperforming other neural networks by combining pixel connectivity modeling and pixel-wise tissue classification. The model can reduce common topological prediction issues in tissues from healthy patients and from patients with Barrett's esophagus.
BIOMEDICAL OPTICS EXPRESS
(2021)
Article
Engineering, Electrical & Electronic
Ali Hasan, Joao M. Pereira, Sina Farsiu, Vahid Tarokh
Summary: The method proposed in this study learns latent stochastic differential equations (SDEs) from high dimensional time series data using self-supervised learning, recovering both the underlying SDE coefficients and original latent variables. Validation was conducted through simulated video processing tasks and real world datasets.
IEEE TRANSACTIONS ON SIGNAL PROCESSING
(2022)
Article
Optics
Bartlomiej Kowalski, Vyas Akondi, Alfredo Dubra
Summary: The accuracy and precision of optical scanners can be improved by recording scanner orientation in synchrony with light detection, followed by data resampling. This approach effectively corrects image warping, sampling jitter, and image distortion caused by angular speed fluctuations. It is applicable to both resonant and non-resonant scanners and has been demonstrated using synthetic and experimental data.
Article
Biochemical Research Methods
Justin Migacz, Oscar Otero-Marquez, Rebecca Zhou, Kara Rickford, Brian Murillo, Davis B. Zhou, Maria Castanos, Nripun Sredar, Alfredo Dubra, Richard B. Rosen, Toco Y. P. Chui
Summary: This study demonstrates the non-invasive visualization of the movement and morphological changes of vitreous cortex hyalocytes using advanced imaging technology. The findings suggest that these cells move in quick bursts and may have potential applications in the treatment of eye diseases.
BIOMEDICAL OPTICS EXPRESS
(2022)
Editorial Material
Biochemical Research Methods
Ruikang K. Wang, Sina Farsiu
Summary: The new Editor-in-Chief Ruikang (Ricky) Wang and Deputy Editor Sina Farsiu of Biomedical Optics Express share their introductory message as they start their editorial terms on 1 January 2022.
BIOMEDICAL OPTICS EXPRESS
(2022)
Article
Multidisciplinary Sciences
Ruobing Qian, Kevin C. Zhou, Jingkai Zhang, Christian Viehland, Al-Hafeez Dhalla, Joseph A. Izatt
Summary: This paper presents a high-speed FMCW-based 3D imaging system that combines grating beam steering with compressed time-frequency analysis for depth retrieval. The system achieves real-time densely sampled 3D imaging of moving objects with submillimeter localization accuracy.
NATURE COMMUNICATIONS
(2022)
Article
Biochemical Research Methods
Xiaojing Huang, Trevor Anderson, Alfredo Dubra
Summary: This study proposes methods for calculating retinal magnification factors (RMFs) that take into account individual ocular biometry. The results show minimal differences between the two methods compared to widely used approaches. The study also reveals a relationship between RMF changes and refractive error, as well as the axial separation between the entrance pupil and the exit pupil of the ophthalmoscope. Furthermore, the study finds weak correlation between surface radii and refractive error, while vitreous thickness shows a strong correlation. Lastly, the study suggests reporting individual ocular biometry data and detailed RMF calculation method descriptions in scientific publications to facilitate data comparison in retinal imaging biomarkers across studies.
BIOMEDICAL OPTICS EXPRESS
(2022)
Article
Optics
Kevin C. Zhou, Ryan P. McNabb, Ruobing Qian, Simone Degan, Al-Hafeez Dhalla, Sina Farsiu, Joseph A. Izatt
Summary: Optical coherence tomography (OCT) is a clinical diagnostic 3D imaging modality that has been successful in various fields. However, OCT has limitations that restrict its use as a 3D microscopy tool. In this paper, a computational extension of OCT called 3D optical coherence refraction tomography (OCRT) is introduced, which provides resolution-enhanced and noise-reduced 3D reconstructions.
Article
Robotics
William Edwards, Gao Tang, Yuan Tian, Mark Draelos, Joseph Izatt, Anthony Kuo, Kris Hauser
Summary: Deep anterior lamellar keratoplasty (DALK) is a technique for cornea transplantation that reduces patient morbidity. Robot microsurgery has been explored as a potential application for DALK due to its challenging nature for human surgeons. In this study, we have developed a data-driven autoregressive dynamic model and a model predictive controller to improve the accuracy of needle insertion during the surgery. Our experiments show that our controller significantly improves needle positioning accuracy compared to previous methods.
IEEE ROBOTICS AND AUTOMATION LETTERS
(2022)
Editorial Material
Biochemical Research Methods
Ruikang K. Wang, Sina Farsiu
Summary: The Editor-in-Chief and Deputy Editor of Biomedical Optics Express have introduced a new prize to honor the best paper published in the Journal between 2019 and 2021.
BIOMEDICAL OPTICS EXPRESS
(2023)
Article
Biochemical Research Methods
Somayyeh Soltanian-Zadeh, Zhuolin Liu, Yan Liu, Ayoub Lassoued, Catherine A. Cukras, Donald T. Miller, Daniel X. Hammer, Sina Farsiu
Summary: Objective quantification of photoreceptor cell morphology using AO-OCT is crucial for early and accurate diagnosis of retinal neurodegenerative diseases. We propose a comprehensive deep learning framework to automate the process of segmenting individual cone cells in AO-OCT scans, achieving human-level performance in assessing cone photoreceptors.
BIOMEDICAL OPTICS EXPRESS
(2023)
Article
Computer Science, Interdisciplinary Applications
Reza Rasti, Armin Biglari, Mohammad Rezapourian, Ziyun Yang, Sina Farsiu
Summary: In this study, a new convolutional neural network architecture called RetiFluidNet is proposed for multi-class retinal fluid segmentation. The model incorporates a self-adaptive dual-attention module, multiple self-adaptive attention-based skip connections, and a novel multi-scale deep self-supervision learning scheme for hierarchical representation learning. The model is optimized using a joint loss function comprising a weighted version of dice overlap and edge-preserved connectivity-based losses. Experimental results on three publicly available datasets demonstrate the effectiveness of the proposed model in retinal OCT fluid segmentation, outperforming existing state-of-the-art algorithms.
IEEE TRANSACTIONS ON MEDICAL IMAGING
(2023)
Article
Medicine, General & Internal
Amir Akhavanrezayat, Anadi Khatri, Neil Gregory L. Onghanseng, Muhammad Sohail Halim, Christopher Or, Nripun Sredar, Moataz Razeen, Murat Hasanreisoglu, Jonathan Regenold, Zheng Xian Thng, S. Saeed Mohammadi, Tanya Jain, Negin Yavari, Vahid Bazojoo, Ankur Sudhir Gupta, Azadeh Mobasserian, Cigdem Yasar, Ngoc Trong Tuong Than, Gunay Uludag Kirimli, Irmak Karaca, Yong-Un Shin, Woong-Sun Yoo, Hashem Ghoraba, Diana V. Do, Alfredo Dubra, Quan Dong Nguyen
Summary: This study utilizes multiple diagnostic modalities/tests to describe the longitudinal changes in patients with non-paraneoplastic autoimmune retinopathy (npAIR). The results show that wide-angle fundus photography, wide-angle fundus autofluorescence, and spectral-domain optical coherence tomography are commonly used methods to detect structural changes, while adaptive optics scanning laser ophthalmoscopy demonstrates better ability in capturing retinal microstructure changes during follow-ups. Functional changes can be detected through visual field, microperimetry, and electrophysiologic testing. Multimodal imaging/tests aid in the early detection of anomalous changes in npAIR patients.
Article
Computer Science, Artificial Intelligence
Yijun Bao, Somayyeh Soltanian-Zadeh, Sina Farsiu, Yiyang Gong
Summary: The Shallow U-Net Neuron Segmentation (SUNS) method can accurately and quickly segment active neurons from two-photon fluorescence imaging videos. This method is faster and more accurate than existing techniques, and can be used online for real-time neuroscience experiments.
NATURE MACHINE INTELLIGENCE
(2021)