scholarly journals DeepSpot: a deep neural network for RNA spot enhancement in smFISH microscopy images

2021 ◽  
Author(s):  
Emmanuel Bouilhol ◽  
Edgar Lefevre ◽  
Benjamin Dartigues ◽  
Robyn Brackin ◽  
Anca F Savulescu ◽  
...  
Keyword(s):  
Deep Learning ◽  
Single Molecule ◽  
Parameter Tuning ◽  
Small Object ◽  
Learning Approaches ◽  
Wound Healing Assay ◽  
Detection Thresholds ◽  
Spot Detection ◽  
Small Object Detection ◽  
Microscopy Images

Detection of RNA spots in single molecule FISH microscopy images remains a difficult task especially when applied to large volumes of data. The small size of RNA spots combined with high noise level of images often requires a manual adaptation of the spot detection thresholds for each image. In this work we introduce DeepSpot, a Deep Learning based tool specifically designed to enhance RNA spots which enables spot detection without need to resort to image per image parameter tuning. We show how our method can enable the downstream accurate detection of spots. The architecture of DeepSpot is inspired by small object detection approaches. It incorporates dilated convolutions into a module specifically designed for the Context Aggregation for Small Object (CASO) and uses Residual Convolutions to propagate this information along the network. This enables DeepSpot to enhance all RNA spots to the same intensity and thus circumvents the need for parameter tuning. We evaluated how easily spots can be detected in images enhanced by our method, by training DeepSpot on 20 simulated and 1 experimental datasets, and have shown that more than 97% accuracy is achieved. Moreover, comparison with alternative deep learning approaches for mRNA spot detection (deepBlink) indicated that DeepSpot allows more precise mRNA detection. In addition, we generated smFISH images from mouse fibroblasts in a wound healing assay to evaluate whether DeepSpot enhancement can enable seamless mRNA spot detection and thus streamline studies of localized mRNA expression in cells.

scholarly journals RS-FISH: Precise, interactive and scalable smFISH spot detection using Radial Symmetry

2021 ◽  
Author(s):  
Ella Bahry ◽  
Laura Breimann ◽  
Leo Epstein ◽  
Klim Kolyvanov ◽  
Kyle I. S. Harrington ◽  
...  
Keyword(s):  
Single Molecule ◽  
Parameter Tuning ◽  
Radial Symmetry ◽  
Three Dimensions ◽  
Powerful Method ◽  
Accurate Identification ◽  
Large Sets ◽  
Rna Fish ◽  
Spot Detection ◽  
Microscopy Images

AbstractStudying transcription using single-molecule RNA-FISH (smFISH) is a powerful method to gain insights into gene regulation on a single cell basis, which relies on accurate identification of sub-resolution fluorescent spots in microscopy images. Here we present Radial Symmetry-FISH (RS-FISH), which can robustly and quickly detect even close smFISH spots in two and three dimensions with high precision, allows interactive parameter tuning, and can easily be applied to large sets of images.Availability and implementationRS-FISH is an open-source implementation written in Java/ImgLib2 and provided as a macro-scriptable Fiji plugin. Source code, tutorial, documentation, and example images are available at: https://github.com/PreibischLab/RadialSymmetryLocalization

scholarly journals A Survey of Small Object Detection Based on Deep Learning

2021 ◽  
Author(s):  
Zhang Zhenghua ◽  
Jiang Ling ◽  
Hong Qingqing
Keyword(s):  
Deep Learning ◽  
Object Detection ◽  
Small Object ◽  
Small Object Detection

A novel two-stage deep learning-based small-object detection using hyperspectral images

2019 ◽  
Vol 26 (6) ◽  
pp. 597-606 ◽  
Author(s):  
Lu Yan ◽  
Masahiro Yamaguchi ◽  
Naoki Noro ◽  
Yohei Takara ◽  
Fuminori Ando
Keyword(s):  
Deep Learning ◽  
Object Detection ◽  
Hyperspectral Images ◽  
Small Object ◽  
Two Stage ◽  
Small Object Detection

scholarly journals Deep Learning Approaches for Head and Operculum Segmentation in Zebrafish Microscopy Images

2021 ◽  
pp. 154-164
Author(s):  
Navdeep Kumar ◽  
Alessio Carletti ◽  
Paulo J. Gavaia ◽  
Marc Muller ◽  
M. Leonor Cancela ◽  
...  
Keyword(s):  
Deep Learning ◽  
Learning Approaches ◽  
Microscopy Images

scholarly journals An Evaluation of Deep Learning Methods for Small Object Detection

2020 ◽  
Vol 2020 ◽  
pp. 1-18 ◽  
Author(s):  
Nhat-Duy Nguyen ◽  
Tien Do ◽  
Thanh Duc Ngo ◽  
Duy-Dinh Le
Keyword(s):  
Deep Learning ◽  
Object Detection ◽  
State Of The Art ◽  
Rapid Development ◽  
Empirical Evaluation ◽  
Grid Cell ◽  
Small Object ◽  
Feature Maps ◽  
Comparative Results ◽  
Small Object Detection

Small object detection is an interesting topic in computer vision. With the rapid development in deep learning, it has drawn attention of several researchers with innovations in approaches to join a race. These innovations proposed comprise region proposals, divided grid cell, multiscale feature maps, and new loss function. As a result, performance of object detection has recently had significant improvements. However, most of the state-of-the-art detectors, both in one-stage and two-stage approaches, have struggled with detecting small objects. In this study, we evaluate current state-of-the-art models based on deep learning in both approaches such as Fast RCNN, Faster RCNN, RetinaNet, and YOLOv3. We provide a profound assessment of the advantages and limitations of models. Specifically, we run models with different backbones on different datasets with multiscale objects to find out what types of objects are suitable for each model along with backbones. Extensive empirical evaluation was conducted on 2 standard datasets, namely, a small object dataset and a filtered dataset from PASCAL VOC 2007. Finally, comparative results and analyses are then presented.

A literature review of localization methods and emergence of deep learning in single-molecule localization microscopy

2020 ◽  
Author(s):  
Anish Mukherjee
Keyword(s):  
Deep Learning ◽  
Single Molecule ◽  
Super Resolution ◽  
Point Sources ◽  
Learning Approaches ◽  
Localization Algorithm ◽  
Localization Microscopy ◽  
Trade Offs ◽  
Emitter Localization ◽  
Single Molecule Localization Microscopy

The quality of super-resolution images largely depends on the performance of the emitter localization algorithm used to localize point sources. In this article, an overview of the various techniques which are used to localize point sources in single-molecule localization microscopy are discussed and their performances are compared. This overview can help readers to select a localization technique for their application. Also, an overview is presented about the emergence of deep learning methods that are becoming popular in various stages of single-molecule localization microscopy. The state of the art deep learning approaches are compared to the traditional approaches and the trade-offs of selecting an algorithm for localization are discussed.

Improved YOLOv3 with duplex FPN for object detection based on deep learning

2021 ◽  
pp. 002072092098352
Author(s):  
Seokyong Shin ◽  
Hyunho Han ◽  
Sang Hun Lee
Keyword(s):  
Deep Learning ◽  
Object Detection ◽  
Autonomous Vehicles ◽  
Detection Accuracy ◽  
Small Object ◽  
Feature Maps ◽  
Low Level ◽  
Small Object Detection ◽  
High Level ◽  
Networks Structure

YOLOv3 is a deep learning-based real-time object detector and is mainly used in applications such as video surveillance and autonomous vehicles. In this paper, we proposed an improved YOLOv3 (You Only Look Once version 3) applied Duplex FPN, which enhanced large object detection by utilizing low-level feature information. The conventional YOLOv3 improved the small object detection performance by applying FPN (Feature Pyramid Networks) structure to YOLOv2. However, YOLOv3 with an FPN structure specialized in detecting small objects, so it is difficult to detect large objects. Therefore, this paper proposed an improved YOLOv3 applied Duplex FPN, which can utilize low-level location information in high-level feature maps instead of the existing FPN structure of YOLOv3. This improved the detection accuracy of large objects. Also, an extra detection layer was added to the top-level feature map to prevent failure of detection of parts of large objects. Further, dimension clusters of each detection layer were reassigned to learn quickly how to accurately detect objects. The proposed method was compared and analyzed in the PASCAL VOC dataset. The experimental results showed that the bounding box accuracy of large objects improved owing to the Duplex FPN and extra detection layer, and the proposed method succeeded in detecting large objects that the existing YOLOv3 did not.

Sign in / Sign up

Export Citation Format

Share Document