scholarly journals DoGNet: A deep architecture for synapse detection in multiplexed fluorescence images

2019 ◽  
Vol 15 (5) ◽  
pp. e1007012 ◽  
Author(s):  
Victor Kulikov ◽  
Syuan-Ming Guo ◽  
Matthew Stone ◽  
Allen Goodman ◽  
Anne Carpenter ◽  
...  
Keyword(s):  
Deep Architecture ◽  
Fluorescence Images ◽  
Synapse Detection

LambdaNet: A Fully Convolutional Architecture for Directional Change Detection

2020 ◽  
Vol 2020 (8) ◽  
pp. 114-1-114-7
Author(s):  
Bryan Blakeslee ◽  
Andreas Savakis
Keyword(s):  
Change Detection ◽  
Satellite Imagery ◽  
Classification Scheme ◽  
Deep Architecture ◽  
Binary Process ◽  
Directional Change ◽  
Class Labels ◽  
Image Pairs

Change detection in image pairs has traditionally been a binary process, reporting either “Change” or “No Change.” In this paper, we present LambdaNet, a novel deep architecture for performing pixel-level directional change detection based on a four class classification scheme. LambdaNet successfully incorporates the notion of “directional change” and identifies differences between two images as “Additive Change” when a new object appears, “Subtractive Change” when an object is removed, “Exchange” when different objects are present in the same location, and “No Change.” To obtain pixel annotated change maps for training, we generated directional change class labels for the Change Detection 2014 dataset. Our tests illustrate that LambdaNet would be suitable for situations where the type of change is unstructured, such as change detection scenarios in satellite imagery.

scholarly journals RLC-GNN: An Improved Deep Architecture for Spatial-Based Graph Neural Network with Application to Fraud Detection

2021 ◽  
Vol 11 (12) ◽  
pp. 5656
Author(s):  
Yufan Zeng ◽  
Jiashan Tang
Keyword(s):  
Numerical Experiments ◽  
State Of The Art ◽  
Single Layer ◽  
Fraud Detection ◽  
Layer By Layer ◽  
Residual Structure ◽  
Detection Algorithms ◽  
Deep Architecture ◽  
Graph Neural Networks ◽  
Node Embeddings

Graph neural networks (GNNs) have been very successful at solving fraud detection tasks. The GNN-based detection algorithms learn node embeddings by aggregating neighboring information. Recently, CAmouflage-REsistant GNN (CARE-GNN) is proposed, and this algorithm achieves state-of-the-art results on fraud detection tasks by dealing with relation camouflages and feature camouflages. However, stacking multiple layers in a traditional way defined by hop leads to a rapid performance drop. As the single-layer CARE-GNN cannot extract more information to fix the potential mistakes, the performance heavily relies on the only one layer. In order to avoid the case of single-layer learning, in this paper, we consider a multi-layer architecture which can form a complementary relationship with residual structure. We propose an improved algorithm named Residual Layered CARE-GNN (RLC-GNN). The new algorithm learns layer by layer progressively and corrects mistakes continuously. We choose three metrics—recall, AUC, and F1-score—to evaluate proposed algorithm. Numerical experiments are conducted. We obtain up to 5.66%, 7.72%, and 9.09% improvements in recall, AUC, and F1-score, respectively, on Yelp dataset. Moreover, we also obtain up to 3.66%, 4.27%, and 3.25% improvements in the same three metrics on the Amazon dataset.

scholarly journals The Design and Preclinical Evaluation of a Single-Label Bimodal Nanobody Tracer for Image-Guided Surgery

Biomolecules ◽  
2021 ◽  
Vol 11 (3) ◽  
pp. 360
Author(s):  
Pieterjan Debie ◽  
Noemi B. Declerck ◽  
Danny van Willigen ◽  
Celine M. Huygen ◽  
Bieke De Sloovere ◽  
...  
Keyword(s):  
Single Molecule ◽  
Gamma Ray ◽  
Nuclear Imaging ◽  
Primary Amines ◽  
Resection Margins ◽  
Fluorescent Light ◽  
Single Structure ◽  
Fluorescence Images

Intraoperative guidance using targeted fluorescent tracers can potentially provide surgeons with real-time feedback on the presence of tumor tissue in resection margins. To overcome the limited depth penetration of fluorescent light, combining fluorescence with SPECT/CT imaging and/or gamma-ray tracing has been proposed. Here, we describe the design and preclinical validation of a novel bimodal nanobody-tracer, labeled using a “multifunctional single attachment point” (MSAP) label, integrating a Cy5 fluorophore and a diethylenetriaminepentaacetic acid (DTPA) chelator into a single structure. After conjugation of the bimodal MSAP to primary amines of the anti-HER2 nanobody 2Rs15d and 111In-labeling of DTPA, the tracer’s characteristics were evaluated in vitro. Subsequently, its biodistribution and tumor targeting were assessed by SPECT/CT and fluorescence imaging over 24 h. Finally, the tracer’s ability to identify small, disseminated tumor lesions was investigated in mice bearing HER2-overexpressing SKOV3.IP1 peritoneal lesions. [111In]In-MSAP.2Rs15d retained its affinity following conjugation and remained stable for 24 h. In vivo SPECT/CT and fluorescence images showed specific uptake in HER2-overexpressing tumors with low background. High tumor-to-muscle ratios were obtained at 1h p.i. and remained 19-fold on SPECT/CT and 3-fold on fluorescence images over 24 h. In the intraperitoneally disseminated model, the tracer allowed detection of larger lesions via nuclear imaging, while fluorescence enabled accurate removal of submillimeter lesions. Bimodal nuclear/fluorescent nanobody-tracers can thus be conveniently designed by conjugation of a single-molecule MSAP-reagent carrying a fluorophore and chelator for radioactive labeling. Such tracers hold promise for clinical applications.

A Data Fusion Method For The Delayering Of X-Ray Fluorescence Images Of Painted Works Of Art

2021 ◽  
Author(s):  
L. D. Fiske ◽  
A. K. Katsaggelos ◽  
M. C. G. Aalders ◽  
M. Alfeld ◽  
M. Walton ◽  
...  
Keyword(s):  
Data Fusion ◽  
Fusion Method ◽  
X Ray ◽  
Works Of Art ◽  
Fluorescence Images

scholarly journals Real-time intraoperative glioma diagnosis using fluorescence imaging and deep convolutional neural networks

2021 ◽  
Author(s):  
Biluo Shen ◽  
Zhe Zhang ◽  
Xiaojing Shi ◽  
Caiguang Cao ◽  
Zeyu Zhang ◽  
...  
Keyword(s):  
Neural Networks ◽  
Real Time ◽  
Fluorescence Imaging ◽  
Convolutional Neural Networks ◽  
Frozen Section ◽  
Fluorescent Imaging ◽  
Pathological Diagnosis ◽  
Pathological Examination ◽  
Deep Convolutional Neural Networks ◽  
Fluorescence Images

Abstract Purpose Surgery is the predominant treatment modality of human glioma but suffers difficulty on clearly identifying tumor boundaries in clinic. Conventional practice involves neurosurgeon’s visual evaluation and intraoperative histological examination of dissected tissues using frozen section, which is time-consuming and complex. The aim of this study was to develop fluorescent imaging coupled with artificial intelligence technique to quickly and accurately determine glioma in real-time during surgery. Methods Glioma patients (N = 23) were enrolled and injected with indocyanine green for fluorescence image–guided surgery. Tissue samples (N = 1874) were harvested from surgery of these patients, and the second near-infrared window (NIR-II, 1000–1700 nm) fluorescence images were obtained. Deep convolutional neural networks (CNNs) combined with NIR-II fluorescence imaging (named as FL-CNN) were explored to automatically provide pathological diagnosis of glioma in situ in real-time during patient surgery. The pathological examination results were used as the gold standard. Results The developed FL-CNN achieved the area under the curve (AUC) of 0.945. Comparing to neurosurgeons’ judgment, with the same level of specificity >80%, FL-CNN achieved a much higher sensitivity (93.8% versus 82.0%, P < 0.001) with zero time overhead. Further experiments demonstrated that FL-CNN corrected >70% of the errors made by neurosurgeons. FL-CNN was also able to rapidly predict grade and Ki-67 level (AUC 0.810 and 0.625) of tumor specimens intraoperatively. Conclusion Our study demonstrates that deep CNNs are better at capturing important information from fluorescence images than surgeons’ evaluation during patient surgery. FL-CNN is highly promising to provide pathological diagnosis intraoperatively and assist neurosurgeons to obtain maximum resection safely. Trial registration ChiCTR ChiCTR2000029402. Registered 29 January 2020, retrospectively registered

Video summarization via spatio-temporal deep architecture

2019 ◽  
Vol 332 ◽  
pp. 224-235 ◽  
Author(s):  
Sheng-hua Zhong ◽  
Jiaxin Wu ◽  
Jianmin Jiang
Keyword(s):  
Video Summarization ◽  
Deep Architecture ◽  
Spatio Temporal

Diagnostic imaging of the larynx: autofluorescence of laryngeal tumours using the helium-cadmium laser

1995 ◽  
Vol 109 (2) ◽  
pp. 108-110 ◽  
Author(s):  
Meredydd Lloyd Harries ◽  
Stephen Lam ◽  
Calum MacAulay ◽  
Jianan Qu ◽  
Branko Palcic
Keyword(s):  
Pilot Study ◽  
Diagnostic Imaging ◽  
Vocal Folds ◽  
Tumour Tissue ◽  
Laryngeal Tumours ◽  
Bronchial Tissue ◽  
Fluorescence Images ◽  
Detection And Localization

AbstractThe use of tissue autofluorescence for the detection and localization of cancer of the larynx is described. In this pilot study, eight patients with probable carcinoma of the vocal folds underwent laryngoscopy in which the tissue autofluorescence spectra of normal and pathologically confirmed tumour tissue were acquiredin vivo. Fluorescence images of the suspect areas were also acquired using the LIFE system (Xillix Technologies Corp.). The results suggest that the autofluorescence properties of laryngeal tissue, under 442 nm illumination, are similar to those of bronchial tissue and that the LIFE system has the potential to increase the accuracy of staging of cancer of the larynx and also to allow earlier diagnosis of tumours and theirrecurrence

Sign in / Sign up

Export Citation Format

Share Document