scholarly journals Improving the Generalizability of Infantile Cataracts Detection via Deep Learning-Based Lens Partition Strategy and Multicenter Datasets

2021 ◽  
Vol 8 ◽  
Author(s):  
Jiewei Jiang ◽  
Shutao Lei ◽  
Mingmin Zhu ◽  
Ruiyang Li ◽  
Jiayun Yue ◽  
...  
Keyword(s):  
Deep Learning ◽  
Hough Transform ◽  
Training Sample ◽  
Training Dataset ◽  
Slit Lamp ◽  
Single Image ◽  
Diagnostic Systems ◽  
Training Sample Size ◽  
Specificity And Sensitivity ◽  
High Level

Infantile cataract is the main cause of infant blindness worldwide. Although previous studies developed artificial intelligence (AI) diagnostic systems for detecting infantile cataracts in a single center, its generalizability is not ideal because of the complicated noises and heterogeneity of multicenter slit-lamp images, which impedes the application of these AI systems in real-world clinics. In this study, we developed two lens partition strategies (LPSs) based on deep learning Faster R-CNN and Hough transform for improving the generalizability of infantile cataracts detection. A total of 1,643 multicenter slit-lamp images collected from five ophthalmic clinics were used to evaluate the performance of LPSs. The generalizability of Faster R-CNN for screening and grading was explored by sequentially adding multicenter images to the training dataset. For the normal and abnormal lenses partition, the Faster R-CNN achieved the average intersection over union of 0.9419 and 0.9107, respectively, and their average precisions are both > 95%. Compared with the Hough transform, the accuracy, specificity, and sensitivity of Faster R-CNN for opacity area grading were improved by 5.31, 8.09, and 3.29%, respectively. Similar improvements were presented on the other grading of opacity density and location. The minimal training sample size required by Faster R-CNN is determined on multicenter slit-lamp images. Furthermore, the Faster R-CNN achieved real-time lens partition with only 0.25 s for a single image, whereas the Hough transform needs 34.46 s. Finally, using Grad-Cam and t-SNE techniques, the most relevant lesion regions were highlighted in heatmaps, and the high-level features were discriminated. This study provides an effective LPS for improving the generalizability of infantile cataracts detection. This system has the potential to be applied to multicenter slit-lamp images.

The impact of training sample size on deep learning-based organ auto-segmentation for head-and-neck patients

2021 ◽  
Vol 66 (18) ◽  
pp. 185012
Author(s):  
Yingtao Fang ◽  
Jiazhou Wang ◽  
Xiaomin Ou ◽  
Hongmei Ying ◽  
Chaosu Hu ◽  
...  
Keyword(s):  
Deep Learning ◽  
Sample Size ◽  
Head And Neck ◽  
Training Sample ◽  
Training Sample Size ◽  
The Impact

Cross-domain gearbox diagnostics under variable working conditions with deep convolutional transfer learning

2020 ◽  
pp. 107754632093379
Author(s):  
Moslem Azamfar ◽  
Jaskaran Singh ◽  
Xiang Li ◽  
Jay Lee
Keyword(s):  
Fault Diagnosis ◽  
Transfer Learning ◽  
High Performance ◽  
Training Sample ◽  
Training Dataset ◽  
Target Domain ◽  
Training Sample Size ◽  
Vibration Data ◽  
Cross Domain ◽  
Testing Dataset

This study proposes a novel 1D deep convolutional transfer learning method that is able to learn the high-dimensional domain-invariant feature from the labeled training dataset and perform diagnosis tasks on the unlabeled testing dataset subjected to a domain shift. To obtain the domain-invariant features, the cross-entropy loss in the source domain classifier and the maximum mean discrepancies between the source and target domain data are minimized simultaneously. To evaluate the performance of the proposed method, an experimental study is conducted on a gearbox under significant speed variation. Because of inherent limitations of the vibration data, in this research, the effectiveness of torque measurement signals has been explored for gearbox fault diagnosis. Comprehensive studies on network parameters and the training sample size are performed to illustrate the robustness and effectiveness of the proposed method. A comparison study is performed on similar techniques to illustrate the superiority and high performance of the proposed diagnosis method. The achieved results illustrate the effectiveness of torque signal in multiclass cross-domain fault diagnosis of gearboxes.

scholarly journals NuSeT: A Deep Learning Tool for Reliably Separating and Analyzing Crowded Cells

2019 ◽  
Author(s):  
Linfeng Yang ◽  
Rajarshi. P. Ghosh ◽  
J. Matthew Franklin ◽  
Chenyu You ◽  
Jan T. Liphardt
Keyword(s):  
Deep Learning ◽  
Training Sample ◽  
Hybrid Network ◽  
Superior Performance ◽  
Biological Research ◽  
Imaging Data ◽  
Cell Nuclei ◽  
Low Contrast ◽  
Training Sample Size ◽  
Major Bottleneck

AbstractSegmenting cell nuclei within microscopy images is a ubiquitous task in biological research and clinical applications. Unfortunately, segmenting low-contrast overlapping objects that may be tightly packed is a major bottleneck in standard deep learning-based models. We report a Nuclear Segmentation Tool (NuSeT) based on deep learning that accurately segments nuclei across multiple types of fluorescence imaging data. Using a hybrid network consisting of U-Net and Region Proposal Networks (RPN), followed by a watershed step, we have achieved superior performance in detecting and delineating nuclear boundaries in 2D and 3D images of varying complexities. By using foreground normalization and additional training on synthetic images containing non-cellular artifacts, NuSeT improves nuclear detection and reduces false positives. NuSeT addresses common challenges in nuclear segmentation such as variability in nuclear signal and shape, limited training sample size, and sample preparation artifacts. Compared to other segmentation models, NuSeT consistently fares better in generating accurate segmentation masks and assigning boundaries for touching nuclei.

scholarly journals The Impact Of Training Sample Size On Deep Learning Based Organ Auto Segmentation For Head Neck

2020 ◽  
Vol 108 (3) ◽  
pp. e775-e776
Author(s):  
Y. Fang ◽  
J. Wang ◽  
S. Chen ◽  
Y. Guo ◽  
Z. Zhang ◽  
...  
Keyword(s):  
Deep Learning ◽  
Sample Size ◽  
Training Sample ◽  
Training Sample Size ◽  
The Impact ◽  
Head Neck

scholarly journals DEEP MULTI-TASK LEARNING FOR TREE GENERA CLASSIFICATION

2018 ◽  
Vol IV-2 ◽  
pp. 153-159 ◽  
Author(s):  
C. Ko ◽  
J. Kang ◽  
G. Sohn
Keyword(s):  
Deep Learning ◽  
Classification Accuracy ◽  
Network Performance ◽  
Training Sample ◽  
Training Data ◽  
Training Dataset ◽  
Main Task ◽  
Data Generation ◽  
Multiple View

The goal for our paper is to classify tree genera using airborne Light Detection and Ranging (LiDAR) data with Convolution Neural Network (CNN) &amp;ndash; Multi-task Network (MTN) implementation. Unlike Single-task Network (STN) where only one task is assigned to the learning outcome, MTN is a deep learning architect for learning a main task (classification of tree genera) with other tasks (in our study, classification of coniferous and deciduous) simultaneously, with shared classification features. The main contribution of this paper is to improve classification accuracy from CNN-STN to CNN-MTN. This is achieved by introducing a concurrence loss (<i>L</i><sub>cd</sub>) to the designed MTN. This term regulates the overall network performance by minimizing the inconsistencies between the two tasks. Results show that we can increase the classification accuracy from 88.7&amp;thinsp;% to 91.0&amp;thinsp;% (from STN to MTN). The second goal of this paper is to solve the problem of small training sample size by multiple-view data generation. The motivation of this goal is to address one of the most common problems in implementing deep learning architecture, the insufficient number of training data. We address this problem by simulating training dataset with multiple-view approach. The promising results from this paper are providing a basis for classifying a larger number of dataset and number of classes in the future.

scholarly journals Label3DMaize: toolkit for 3D point cloud data annotation of maize shoots

GigaScience ◽  
2021 ◽  
Vol 10 (5) ◽  
Author(s):  
Teng Miao ◽  
Weiliang Wen ◽  
Yinglun Li ◽  
Sheng Wu ◽  
Chao Zhu ◽  
...  
Keyword(s):  
Deep Learning ◽  
Point Cloud ◽  
Training Dataset ◽  
Growth Stages ◽  
3D Point Cloud ◽  
Plant Structure ◽  
Data Annotation ◽  
Cloud Data ◽  
Point Cloud Segmentation ◽  
Different Growth Stages

Abstract Background The 3D point cloud is the most direct and effective data form for studying plant structure and morphology. In point cloud studies, the point cloud segmentation of individual plants to organs directly determines the accuracy of organ-level phenotype estimation and the reliability of the 3D plant reconstruction. However, highly accurate, automatic, and robust point cloud segmentation approaches for plants are unavailable. Thus, the high-throughput segmentation of many shoots is challenging. Although deep learning can feasibly solve this issue, software tools for 3D point cloud annotation to construct the training dataset are lacking. Results We propose a top-to-down point cloud segmentation algorithm using optimal transportation distance for maize shoots. We apply our point cloud annotation toolkit for maize shoots, Label3DMaize, to achieve semi-automatic point cloud segmentation and annotation of maize shoots at different growth stages, through a series of operations, including stem segmentation, coarse segmentation, fine segmentation, and sample-based segmentation. The toolkit takes ∼4–10 minutes to segment a maize shoot and consumes 10–20% of the total time if only coarse segmentation is required. Fine segmentation is more detailed than coarse segmentation, especially at the organ connection regions. The accuracy of coarse segmentation can reach 97.2% that of fine segmentation. Conclusion Label3DMaize integrates point cloud segmentation algorithms and manual interactive operations, realizing semi-automatic point cloud segmentation of maize shoots at different growth stages. The toolkit provides a practical data annotation tool for further online segmentation research based on deep learning and is expected to promote automatic point cloud processing of various plants.

scholarly journals Study on Radar Echo-Filling in an Occlusion Area by a Deep Learning Algorithm

2021 ◽  
Vol 13 (9) ◽  
pp. 1779
Author(s):  
Xiaoyan Yin ◽  
Zhiqun Hu ◽  
Jiafeng Zheng ◽  
Boyong Li ◽  
Yuanyuan Zuo
Keyword(s):  
Deep Learning ◽  
Loss Function ◽  
Learning Algorithm ◽  
Weather Radar ◽  
Loss Functions ◽  
Training Dataset ◽  
Echo Intensity ◽  
Common Mean ◽  
Deep Learning Algorithm ◽  
Radar Beam

Radar beam blockage is an important error source that affects the quality of weather radar data. An echo-filling network (EFnet) is proposed based on a deep learning algorithm to correct the echo intensity under the occlusion area in the Nanjing S-band new-generation weather radar (CINRAD/SA). The training dataset is constructed by the labels, which are the echo intensity at the 0.5° elevation in the unblocked area, and by the input features, which are the intensity in the cube including multiple elevations and gates corresponding to the location of bottom labels. Two loss functions are applied to compile the network: one is the common mean square error (MSE), and the other is a self-defined loss function that increases the weight of strong echoes. Considering that the radar beam broadens with distance and height, the 0.5° elevation scan is divided into six range bands every 25 km to train different models. The models are evaluated by three indicators: explained variance (EVar), mean absolute error (MAE), and correlation coefficient (CC). Two cases are demonstrated to compare the effect of the echo-filling model by different loss functions. The results suggest that EFnet can effectively correct the echo reflectivity and improve the data quality in the occlusion area, and there are better results for strong echoes when the self-defined loss function is used.

scholarly journals Interpretation of Swedish Sign Language Using Convolutional Neural Networks and Transfer Learning

2021 ◽  
Vol 2 (3) ◽  
Author(s):  
Gustaf Halvardsson ◽  
Johanna Peterson ◽  
César Soto-Valero ◽  
Benoit Baudry
Keyword(s):  
Neural Networks ◽  
Sign Language ◽  
Transfer Learning ◽  
Convolutional Neural Networks ◽  
Web Application ◽  
Training Dataset ◽  
Motion Processing ◽  
Image Perception ◽  
Sign Languages ◽  
High Level

AbstractThe automatic interpretation of sign languages is a challenging task, as it requires the usage of high-level vision and high-level motion processing systems for providing accurate image perception. In this paper, we use Convolutional Neural Networks (CNNs) and transfer learning to make computers able to interpret signs of the Swedish Sign Language (SSL) hand alphabet. Our model consists of the implementation of a pre-trained InceptionV3 network, and the usage of the mini-batch gradient descent optimization algorithm. We rely on transfer learning during the pre-training of the model and its data. The final accuracy of the model, based on 8 study subjects and 9400 images, is 85%. Our results indicate that the usage of CNNs is a promising approach to interpret sign languages, and transfer learning can be used to achieve high testing accuracy despite using a small training dataset. Furthermore, we describe the implementation details of our model to interpret signs as a user-friendly web application.

scholarly journals Evaluation of the feasibility of explainable computer-aided detection of cardiomegaly on chest radiographs using deep learning

2021 ◽  
Vol 11 (1) ◽  
Author(s):  
Mu Sook Lee ◽  
Yong Soo Kim ◽  
Minki Kim ◽  
Muhammad Usman ◽  
Shi Sub Byon ◽  
...  
Keyword(s):  
Deep Learning ◽  
Diagnostic Performance ◽  
Absolute Error ◽  
Training Dataset ◽  
Computer Aided Detection ◽  
Test Dataset ◽  
Cardiothoracic Ratio ◽  
Computer Aided ◽  
Chest X Ray ◽  
Public Datasets

AbstractWe examined the feasibility of explainable computer-aided detection of cardiomegaly in routine clinical practice using segmentation-based methods. Overall, 793 retrospectively acquired posterior–anterior (PA) chest X-ray images (CXRs) of 793 patients were used to train deep learning (DL) models for lung and heart segmentation. The training dataset included PA CXRs from two public datasets and in-house PA CXRs. Two fully automated segmentation-based methods using state-of-the-art DL models for lung and heart segmentation were developed. The diagnostic performance was assessed and the reliability of the automatic cardiothoracic ratio (CTR) calculation was determined using the mean absolute error and paired t-test. The effects of thoracic pathological conditions on performance were assessed using subgroup analysis. One thousand PA CXRs of 1000 patients (480 men, 520 women; mean age 63 ± 23 years) were included. The CTR values derived from the DL models and diagnostic performance exhibited excellent agreement with reference standards for the whole test dataset. Performance of segmentation-based methods differed based on thoracic conditions. When tested using CXRs with lesions obscuring heart borders, the performance was lower than that for other thoracic pathological findings. Thus, segmentation-based methods using DL could detect cardiomegaly; however, the feasibility of computer-aided detection of cardiomegaly without human intervention was limited.

Sign in / Sign up

Export Citation Format

Share Document