Triple-Crossing 2.5D Convolutional Neural Network for Detecting Neuronal Arbours in 3D Microscopic Images

2017 ◽  
pp. 185-193 ◽  
Author(s):  
Siqi Liu ◽  
Donghao Zhang ◽  
Yang Song ◽  
Hanchuan Peng ◽  
Weidong Cai
Keyword(s):  
Neural Network ◽  
Convolutional Neural Network ◽  
Microscopic Images

scholarly journals Automatic Zebrafish Egg Phenotype Recognition from Bright-Field Microscopic Images Using Deep Convolutional Neural Network

2019 ◽  
Vol 9 (16) ◽  
pp. 3362 ◽  
Author(s):  
Shang Shang ◽  
Ling Long ◽  
Sijie Lin ◽  
Fengyu Cong
Keyword(s):  
Neural Network ◽  
Image Analysis ◽  
Convolutional Neural Network ◽  
Classification Accuracy ◽  
Deep Convolutional Neural Network ◽  
Microscopic Image ◽  
Bright Field ◽  
Microscopic Images ◽  
Phenotype Classification ◽  
Microscopic Image Analysis

Zebrafish eggs are widely used in biological experiments to study the environmental and genetic influence on embryo development. Due to the high throughput of microscopic imaging, automated analysis of zebrafish egg microscopic images is highly demanded. However, machine learning algorithms for zebrafish egg image analysis suffer from the problems of small imbalanced training dataset and subtle inter-class differences. In this study, we developed an automated zebrafish egg microscopic image analysis algorithm based on deep convolutional neural network (CNN). To tackle the problem of insufficient training data, the strategies of transfer learning and data augmentation were used. We also adopted the global averaged pooling technique to overcome the subtle phenotype differences between the fertilized and unfertilized eggs. Experimental results of a five-fold cross-validation test showed that the proposed method yielded a mean classification accuracy of 95.0% and a maximum accuracy of 98.8%. The network also demonstrated higher classification accuracy and better convergence performance than conventional CNN methods. This study extends the deep learning technique to zebrafish egg phenotype classification and paves the way for automatic bright-field microscopic image analysis.

Convolutional neural network-based registration for mosaicing of microscopic images

2019 ◽  
Vol 28 (04) ◽  
pp. 1
Author(s):  
Junhua Zhang ◽  
Yihua Huang ◽  
Yingchao Song ◽  
Yi Jiang ◽  
Lun Zhang ◽  
...  
Keyword(s):  
Neural Network ◽  
Convolutional Neural Network ◽  
Microscopic Images

scholarly journals Automated detection of superficial fungal infections from microscopic images through a regional convolutional neural network

PLoS ONE ◽  
2021 ◽  
Vol 16 (8) ◽  
pp. e0256290
Author(s):  
Taehan Koo ◽  
Moon Hwan Kim ◽  
Mihn-Sook Jue
Keyword(s):  
Neural Network ◽  
Deep Learning ◽  
Object Detection ◽  
Convolutional Neural Network ◽  
Sensitivity And Specificity ◽  
Real World ◽  
Data Model ◽  
Fungal Infections ◽  
Curve Analysis ◽  
Microscopic Images

Direct microscopic examination with potassium hydroxide is generally used as a screening method for diagnosing superficial fungal infections. Although this type of examination is faster than other diagnostic methods, it can still be time-consuming to evaluate a complete sample; additionally, it possesses the disadvantage of inconsistent reliability as the accuracy of the reading may differ depending on the performer’s skill. This study aims at detecting hyphae more quickly, conveniently, and consistently through deep learning using images obtained from microscopy used in real-world practice. An object detection convolutional neural network, YOLO v4, was trained on microscopy images with magnifications of 100×, 40×, and (100+40)×. The study was conducted at the Department of Dermatology at Veterans Health Service Medical Center, Seoul, Korea between January 1, 2019 and December 31, 2019, using 3,707 images (1,255 images for training, 1,645 images for testing). The average precision was used to evaluate the accuracy of object detection. Precision recall curve analysis was performed for the hyphal location determination, and receiver operating characteristic curve analysis was performed on the image classification. The F1 score, sensitivity, and specificity values were used as measures of the overall performance. The sensitivity and specificity were, respectively, 95.2% and 100% in the 100× data model, and 99% and 86.6% in the 40× data model; the sensitivity and specificity in the combined (100+40)× data model were 93.2% and 89%, respectively. The performance of our model had high sensitivity and specificity, indicating that hyphae can be detected with reliable accuracy. Thus, our deep learning-based autodetection model can detect hyphae in microscopic images obtained from real-world practice. We aim to develop an automatic hyphae detection system that can be utilized in real-world practice through continuous research.

scholarly journals Intellectual Acute Lymphoblastic Leukemia (ALL) Detection Model for Diagnosis of Blood Cancer from microscopic images using Hybrid Convolutional Neural Network

2019 ◽  
Vol 8 (6) ◽  
pp. 2972-2981
Keyword(s):  
Neural Network ◽  
Image Processing ◽  
Feature Extraction ◽  
Blood Cell ◽  
Convolutional Neural Network ◽  
Blood Cancer ◽  
Microscopic Images ◽  
Feature Extraction And Selection ◽  
Evaluation Parameters ◽  
State Of Art

Blood cell malignantly growth has been accounted for to be one of the most transcendent types of disease maladies. ALL (Acute Lymphoblastic Leukaemias) is the malignant types of blood cancer and their detection and classification in earlier stage is biggest issue. Automatic detection and classification of ALL from microscopic images is a challenging and intellectual assignment in medical science. Existing techniques for ALL detection and classification are an understandable alternative for real-time dermoscopic data analysis. Existing microscopic image processing approaches are unable to analyze the ALL data with non-stationary nature. In this perspective, the focus of this research is to design hybrid Convolutional Neural Network (CNN) architecture by utilizing Firefly Optimization Algorithm (FOA/FFA) to detect the ALL from microscopic images of human blood cell into malignant or normal blood cell. Methods: For training and testing of proposed ALL Detection and Classification (ALL-DC) Model, Standard ALL-IDB (Acute-Lymphoblastic-Leukaemias Image Database for Image Processing) is used with hybrid CNN architecture based on the FOA. Here, Histogram of Oriented Gradients (HOG) descriptor with FOA is used as feature extraction and selection mechanism from the Region of Blood Cell (ROBC).Feature extraction approach plays an important responsibility to classify lots of blood diseases. On the way to achieve this goal, we proposed ALL-DC model that combines recent developments in deep learning with fuzzy based CNN structure and for ROBC segmentation, hybridization of K-means segmentation algorithm with FOA that are capable to segment the accurate blood cell region from microscopic images. Using k-means segmentation technique, the foreground and background component is separated into two regions and after that to improve the segmentation results; FOA is used with the novel concept of image enhancement approach. Results: The proposed ALL-DC system is evaluated using the largest publicly accessible standard ALL-IDB dataset, containing 600 training and 400 testing microscopic images. When the evaluation parameters of proposed work is compared with a number of other state-of-art schemes, the proposed scheme achieves the most excellent performance of 98.5% in terms of accuracy which also known as area under the curve (AUC) in differentiating ALL from benign cell using only the extracted and optimized HOG feature. Conclusion: When the proposed model is tested on different microscopic images, evaluation parameters is calculated and compared with a few other state-of-art methods and we obtained the proposed method achieves the best performance in terms of classification accuracy. ALL-DC model is implemented and constructed using the concept of Image

Sign in / Sign up

Export Citation Format

Share Document