scholarly journals Eigenloss: Combined PCA-Based Loss Function for Polyp Segmentation

Mathematics ◽  
2020 ◽  
Vol 8 (8) ◽  
pp. 1316
Author(s):  
Luisa F. Sánchez-Peralta ◽  
Artzai Picón ◽  
Juan Antonio Antequera-Barroso ◽  
Juan Francisco Ortega-Morán ◽  
Francisco M. Sánchez-Margallo ◽  
...  
Keyword(s):  
Deep Learning ◽  
Loss Function ◽  
Survival Rates ◽  
Principal Component ◽  
Loss Functions ◽  
Cancer Death ◽  
Linear Combinations ◽  
The Individual ◽  
Deep Learning Model ◽  
Death Causes

Colorectal cancer is one of the leading cancer death causes worldwide, but its early diagnosis highly improves the survival rates. The success of deep learning has also benefited this clinical field. When training a deep learning model, it is optimized based on the selected loss function. In this work, we consider two networks (U-Net and LinkNet) and two backbones (VGG-16 and Densnet121). We analyzed the influence of seven loss functions and used a principal component analysis (PCA) to determine whether the PCA-based decomposition allows for the defining of the coefficients of a non-redundant primal loss function that can outperform the individual loss functions and different linear combinations. The eigenloss is defined as a linear combination of the individual losses using the elements of the eigenvector as coefficients. Empirical results show that the proposed eigenloss improves the general performance of individual loss functions and outperforms other linear combinations when Linknet is used, showing potential for its application in polyp segmentation problems.

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 Assessing the Impact of the Loss Function, Architecture and Image Type for Deep Learning-Based Wildfire Segmentation

2021 ◽  
Vol 11 (15) ◽  
pp. 7046
Author(s):  
Jorge Francisco Ciprián-Sánchez ◽  
Gilberto Ochoa-Ruiz ◽  
Lucile Rossi ◽  
Frédéric Morandini
Keyword(s):  
Deep Learning ◽  
Loss Function ◽  
State Of The Art ◽  
Fire Detection ◽  
Loss Functions ◽  
Wildfire Spread ◽  
Combine Information ◽  
The Impact ◽  
Image Type ◽  
Segmentation Models

Wildfires stand as one of the most relevant natural disasters worldwide, particularly more so due to the effect of climate change and its impact on various societal and environmental levels. In this regard, a significant amount of research has been done in order to address this issue, deploying a wide variety of technologies and following a multi-disciplinary approach. Notably, computer vision has played a fundamental role in this regard. It can be used to extract and combine information from several imaging modalities in regard to fire detection, characterization and wildfire spread forecasting. In recent years, there has been work pertaining to Deep Learning (DL)-based fire segmentation, showing very promising results. However, it is currently unclear whether the architecture of a model, its loss function, or the image type employed (visible, infrared, or fused) has the most impact on the fire segmentation results. In the present work, we evaluate different combinations of state-of-the-art (SOTA) DL architectures, loss functions, and types of images to identify the parameters most relevant to improve the segmentation results. We benchmark them to identify the top-performing ones and compare them to traditional fire segmentation techniques. Finally, we evaluate if the addition of attention modules on the best performing architecture can further improve the segmentation results. To the best of our knowledge, this is the first work that evaluates the impact of the architecture, loss function, and image type in the performance of DL-based wildfire segmentation models.

scholarly journals Deep Learning with Dynamically Weighted Loss Function for Sensor-Based Prognostics and Health Management

Sensors ◽  
2020 ◽  
Vol 20 (3) ◽  
pp. 723 ◽  
Author(s):  
Divish Rengasamy ◽  
Mina Jafari ◽  
Benjamin Rothwell ◽  
Xin Chen ◽  
Grazziela P. Figueredo
Keyword(s):  
Neural Network ◽  
Deep Learning ◽  
Loss Function ◽  
Short Term Memory ◽  
Health Management ◽  
Remaining Useful Life ◽  
Loss Functions ◽  
Pressure System ◽  
Failure Data ◽  
Prognostic And Health Management

Deep learning has been employed to prognostic and health management of automotive and aerospace with promising results. Literature in this area has revealed that most contributions regarding deep learning is largely focused on the model’s architecture. However, contributions regarding improvement of different aspects in deep learning, such as custom loss function for prognostic and health management are scarce. There is therefore an opportunity to improve upon the effectiveness of deep learning for the system’s prognostics and diagnostics without modifying the models’ architecture. To address this gap, the use of two different dynamically weighted loss functions, a newly proposed weighting mechanism and a focal loss function for prognostics and diagnostics task are investigated. A dynamically weighted loss function is expected to modify the learning process by augmenting the loss function with a weight value corresponding to the learning error of each data instance. The objective is to force deep learning models to focus on those instances where larger learning errors occur in order to improve their performance. The two loss functions used are evaluated using four popular deep learning architectures, namely, deep feedforward neural network, one-dimensional convolutional neural network, bidirectional gated recurrent unit and bidirectional long short-term memory on the commercial modular aero-propulsion system simulation data from NASA and air pressure system failure data for Scania trucks. Experimental results show that dynamically-weighted loss functions helps us achieve significant improvement for remaining useful life prediction and fault detection rate over non-weighted loss function predictions.

scholarly journals Deep Learning Model Based on Mobile-Net with Haar-like Algorithm for Masked Face Recognition at Nuclear Facilities

2021 ◽  
Vol 10 (7) ◽  
pp. 18-23
Author(s):  
Nadia.M. Nawwar* ◽  
Kasban . ◽  
Salama May
Keyword(s):  
Deep Learning ◽  
Face Recognition ◽  
Face Mask ◽  
Learning Model ◽  
Nuclear Facility ◽  
Nuclear Facilities ◽  
Novel Technique ◽  
The World ◽  
The Individual ◽  
Deep Learning Model

During the spread of the COVID-I9 pandemic in early 2020, the WHO organization advised all people in the world to wear face-mask to limit the spread of COVID-19. Many facilities required that their employees wear face-mask. For the safety of the facility, it was mandatory to recognize the identity of the individual wearing the mask. Hence, face recognition of the masked individuals was required. In this research, a novel technique is proposed based on a mobile-net and Haar-like algorithm for detecting and recognizing the masked face. Firstly, recognize the authorized person that enters the nuclear facility in case of wearing the masked-face using mobile-net. Secondly, applying Haar-like features to detect the retina of the person to extract the boundary box around the retina compares this with the dataset of the person without the mask for recognition. The results of the proposed modal, which was tested on a dataset from Kaggle, yielded 0.99 accuracies, a loss of 0.08, F1.score 0.98.

scholarly journals Kudo’s Classification for Colon Polyps Assessment Using a Deep Learning Approach

2020 ◽  
Vol 10 (2) ◽  
pp. 501 ◽  
Author(s):  
Sebastian Patino-Barrientos ◽  
Daniel Sierra-Sosa ◽  
Begonya Garcia-Zapirain ◽  
Cristian Castillo-Olea ◽  
Adel Elmaghraby
Keyword(s):  
Deep Learning ◽  
Fine Tuning ◽  
Learning Approach ◽  
Colon Polyps ◽  
Cancer Death ◽  
Timely Manner ◽  
The World ◽  
Feature Extractor ◽  
The University ◽  
Deep Learning Model

Colorectal cancer (CRC) is the second leading cause of cancer death in the world. This disease could begin as a non-cancerous polyp in the colon, when not treated in a timely manner, these polyps could induce cancer, and in turn, death. We propose a deep learning model for classifying colon polyps based on the Kudo’s classification schema, using basic colonoscopy equipment. We train a deep convolutional model with a private dataset from the University of Deusto with and without using a VGG model as a feature extractor, and compared the results. We obtained 83% of accuracy and 83% of F1-score after fine tuning our model with the VGG filter. These results show that deep learning algorithms are useful to develop computer-aided tools for early CRC detection, and suggest combining it with a polyp segmentation model for its use by specialists.

scholarly journals Application of Machine Learning Method of Data-Driven Deep Learning Model to Predict Well Production Rate in the Shale Gas Reservoirs

Energies ◽  
2021 ◽  
Vol 14 (12) ◽  
pp. 3629
Author(s):  
Dongkwon Han ◽  
Sunil Kwon
Keyword(s):  
Deep Learning ◽  
Hydraulic Fracturing ◽  
Physical Phenomenon ◽  
Principal Component ◽  
Learning Model ◽  
Gas Production ◽  
Data Driven ◽  
Percentage Error ◽  
Proxy Model ◽  
Deep Learning Model

Reservoir modeling to predict shale reservoir productivity is considerably uncertain and time consuming. Since we need to simulate the physical phenomenon of multi-stage hydraulic fracturing. To overcome these limitations, this paper presents an alternative proxy model based on data-driven deep learning model. Furthermore, this study not only proposes the development process of a proxy model, but also verifies using field data for 1239 horizontal wells from the Montney shale formation in Alberta, Canada. A deep neural network (DNN) based on multi-layer perceptron was applied to predict the cumulative gas production as the dependent variable. The independent variable is largely divided into four types: well information, completion and hydraulic fracturing and production data. It was found that the prediction performance was better when using a principal component with a cumulative contribution of 85% using principal component analysis that extracts important information from multivariate data, and when predicting with a DNN model using 6 variables calculated through variable importance analysis. Hence, to develop a reliable deep learning model, sensitivity analysis of hyperparameters was performed to determine one-hot encoding, dropout, activation function, learning rate, hidden layer number and neuron number. As a result, the best prediction of the mean absolute percentage error of the cumulative gas production improved to at least 0.2% and up to 9.1%. The novel approach of this study can also be applied to other shale formations. Furthermore, a useful guide for economic analysis and future development plans of nearby reservoirs.

scholarly journals An Empirical Study on Knowledge Aggregation in Academic Virtual Community Based on Deep Learning

2021 ◽  
Vol 5 (4) ◽  
pp. 372-388
Author(s):  
Liangfeng Qian ◽  
Shengli Deng
Keyword(s):  
Deep Learning ◽  
Online Community ◽  
Virtual Community ◽  
Principal Component ◽  
Community Resources ◽  
Community Based ◽  
Knowledge Resources ◽  
Network Community ◽  
Amount Of Knowledge ◽  
Deep Learning Model

Abstract Academic virtual community provides an environment for users to exchange knowledge, so it gathers a large amount of knowledge resources and presents a trend of rapid and disorderly growth. We learn how to organize the scattered and disordered knowledge of network community effectively and provide personalized service for users. We focus on analyzing the knowledge association among titles in an all-round way based on deep learning, so as to realize effective knowledge aggregation in academic virtual community. We take ResearchGate (RG) “online community” resources as an example and use Word2Vec model to realize deep knowledge aggregation. Then, principal component analysis (PCA) is used to verify its scientificity, and Wide & Deep learning model is used to verify its running effect. The empirical results show that the knowledge aggregation system of “online community” works well and has scientific rationality.

On the Nature of Zonal Jet EOFs

2006 ◽  
Vol 19 (24) ◽  
pp. 6409-6424 ◽  
Author(s):  
Adam H. Monahan ◽  
John C. Fyfe
Keyword(s):  
Time Series ◽  
Covariance Structure ◽  
Principal Component ◽  
Empirical Orthogonal Functions ◽  
Physical Processes ◽  
Linear Combinations ◽  
Orthogonal Functions ◽  
The Mean ◽  
Systematic Perturbation ◽  
The Individual

Abstract Analytic results are obtained for the mean and covariance structure of an idealized zonal jet that fluctuates in strength, position, and width. Through a systematic perturbation analysis, the leading empirical orthogonal functions (EOFs) and principal component (PC) time series are obtained. These EOFs are built of linear combinations of basic patterns corresponding to monopole, dipole, and tripole structures. The analytic results demonstrate that in general the individual EOF modes cannot be interpreted in terms of individual physical processes. In particular, while the dipole EOF (similar to the leading EOF of the midlatitude zonal mean zonal wind) describes fluctuations in jet position to leading order, its time series also contains contributions from fluctuations in strength and width. No simple interpretations of the other EOFs in terms of strength, position, or width fluctuations are possible. Implications of these results for the use of EOF analysis to diagnose physical processes of variability are discussed.

scholarly journals Driving Fatigue Detection from EEG Using a Modified PCANet Method

2019 ◽  
Vol 2019 ◽  
pp. 1-9 ◽  
Author(s):  
Yuliang Ma ◽  
Bin Chen ◽  
Rihui Li ◽  
Chushan Wang ◽  
Jun Wang ◽  
...  
Keyword(s):  
Principal Component Analysis ◽  
Feature Extraction ◽  
Deep Learning ◽  
Classification Accuracy ◽  
Traffic Accidents ◽  
Principal Component ◽  
Learning Model ◽  
Driving Fatigue ◽  
Fatigue Detection ◽  
Deep Learning Model

The rapid development of the automotive industry has brought great convenience to our life, which also leads to a dramatic increase in the amount of traffic accidents. A large proportion of traffic accidents were caused by driving fatigue. EEG is considered as a direct, effective, and promising modality to detect driving fatigue. In this study, we presented a novel feature extraction strategy based on a deep learning model to achieve high classification accuracy and efficiency in using EEG for driving fatigue detection. EEG signals were recorded from six healthy volunteers in a simulated driving experiment. The feature extraction strategy was developed by integrating the principal component analysis (PCA) and a deep learning model called PCA network (PCANet). In particular, the principal component analysis (PCA) was used to preprocess EEG data to reduce its dimension in order to overcome the limitation of dimension explosion caused by PCANet, making this approach feasible for EEG-based driving fatigue detection. Results demonstrated high and robust performance of the proposed modified PCANet method with classification accuracy up to 95%, which outperformed the conventional feature extraction strategies in the field. We also identified that the parietal and occipital lobes of the brain were strongly associated with driving fatigue. This is the first study, to the best of our knowledge, to practically apply the modified PCANet technique for EEG-based driving fatigue detection.

Sign in / Sign up

Export Citation Format

Share Document