Liver disease screening based on densely connected deep neural networks

Yao, Zhenjie; Li, Jiangong; Guan, Zhaoyu; Ye, Yancheng; Chen, Yixin

doi:10.1016/j.neunet.2019.11.005

Cited by 40 publications

(20 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…To highlight the advantages of the proposed model, we compared with some of the state-of-the-art baseline models used for CV task and medical images applications as VGG-16 [13], ResNet50v2 [14], DenseNet121 [15] and the LZNet2019 [37] that was the top result until the development of this research.VGG-16 and DenseNet121 use images of dimensions 224 × 224 as input. Moreover, the LZNet2019 uses 150 × 150 as input.…”

Section: Test Resultsmentioning

confidence: 99%

“…Since then, and with the publication of the Deep Learning article [19], baseline models of CNN designed for CV tasks and normally trained in the ILSVRC, have been released to reuse the models. Examples of baseline models are VGG-16 network [13], Inception-v3 network [20], Residual Networks versions 1 and 2 (ResNet) [14,21], depth-wise separable convolutions networks (Xception) [22], Densely Connected Networks (DenseNet) [15], among others. These CNN baseline models are often used to implement new systems for CV task and for CADe and CADx [23].…”

Section: Convolutional Neural Networkmentioning

confidence: 99%

“…Among DL models, there exist Convolutional Neural Networks (CNN) that are widely used in Computer Vision (CV) tasks due to their capabilities for feature extraction [9].Although several works for CNN have been developed in this area in CAD to classify medical images between lesions and non-lesions [6], for automated brain abnormality classification [7], multiple fetal organs segmentation [10], for brain tumor segmentation [11,12], there are possibilities to get better results to help the medical field using CNNs.Therefore, we propose the use of a pre-trained model of CNN that uses Transfer Learning (TL) technique to use it as initialization for the training of the CNN used for classification of CXRAY images of pneumonia and visualization of the possible localization of manifestations like fluid and pus.The main contribution of this research is a TL model of CNN for the classification of CXRAYs in conjunction with a preprocessing technique, and two approaches to solve the imbalance problem of the dataset. The proposed model is competitive compared to baseline architectures of CNN like the VGG-16 network [13], Residual Networks (ResNet-v2) [14], Densely Connected Networks (DenseNet) [15], and some of the state-of-the-art proposals.…”

mentioning

confidence: 99%

“…The main contribution of this research is a TL model of CNN for the classification of CXRAYs in conjunction with a preprocessing technique, and two approaches to solve the imbalance problem of the dataset. The proposed model is competitive compared to baseline architectures of CNN like the VGG-16 network [13], Residual Networks (ResNet-v2) [14], Densely Connected Networks (DenseNet) [15], and some of the state-of-the-art proposals.…”

mentioning

confidence: 99%

See 3 more Smart Citations

A Transfer Learning Method for Pneumonia Classification and Visualization

et al. 2020

View full text Add to dashboard Cite

Pneumonia is an infectious disease that affects the lungs and is one of the principal causes of death in children under five years old. The Chest X-ray images technique is one of the most used for diagnosing pneumonia. Several Machine Learning algorithms have been successfully used in order to provide computer-aided diagnosis by automatic classification of medical images. For its remarkable results, the Convolutional Neural Networks (models based on Deep Learning) that are widely used in Computer Vision tasks, such as classification of injuries and brain abnormalities, among others, stand out. In this paper, we present a transfer learning method that automatically classifies between 3883 chest X-ray images characterized as depicting pneumonia and 1349 labeled as normal. The proposed method uses the Xception Network pre-trained weights on ImageNet as an initialization. Our model is competitive with respect to state-of-the-art proposals. To make comparisons with other models, we have used four well-known performance measures, obtaining the following results: precision (0.84), recall (0.99), F1-score (0.91) and area under the ROC curve (0.97). These positive results allow us to consider our proposal as an alternative that can be useful in countries with a lack of equipment and specialized radiologists.

show abstract

Section: Test Resultsmentioning

confidence: 99%

Section: Convolutional Neural Networkmentioning

confidence: 99%

mentioning

confidence: 99%

mentioning

confidence: 99%

See 2 more Smart Citations

A Transfer Learning Method for Pneumonia Classification and Visualization

et al. 2020

View full text Add to dashboard Cite

show abstract

“…Liver disease is an important public health problem that mostly manifested as abnormal Liver Function Test (Yao, Li, Guan, Ye, & Chen, 2020). Patient with liver disease have been continuously increasing because of many causes such as excessive alcohol consumption (McDermott & Forsyth, 2016), inhale of harmful gases, intake of contaminated food, pickles, drugs (Venkata Ramana, Babu, & Venkateswarlu, 2011), inherited metabolic liver disease (Kelly, 2019), epidemic, and endemic outbreaks by geographical settings (Beeching & Dassanayake, 2019).…”

Section: Introductionmentioning

confidence: 99%

Support Vector Machine Parameter Optimization to Improve Liver Disease Estimation with Genetic Algorithm

Harafani

2020

SinkrOn

View full text Add to dashboard Cite

Liver disease is an important public health problem. Over the past few decades, machine learning has developed rapidly and it has been introduced for application in medical-related. In this study we propose Support Vector Machine optimization parameter with genetic algorithm to get a higher performance of Root Mean Square Error value of SVM in order to estimate the liver disorder. The experiment was carried out in three stages, the first step was to try the three SVM kernels with different combination of parameters manually, The second step was to try some combination of range parameters in the genetic algorithm to find the optimal value in the SVM kernel. The third step is comparing the results of the GA-SVM experiment with other regression methods. The results prove that GA has an influence on improving the performance of GA-SVM which has the lowest RMSE value compared to another regression models.

show abstract

Using a machine learning model for the optimal design of simulated moving bed processes and its application to separate rebaudioside A and stevioside

Yao

Zheng

Chen

et al. 2021

J of Chemical Tech & Biotech

View full text Add to dashboard Cite

BACKGROUND Solving a mechanistic model of simulated moving bed (SMB) is extremely time consuming, therefore, it is inefficient for online control using the optimization method based on such a mechanistic model. To this end, the machine learning model can be regarded as a potential substitution to accelerate the optimization process. RESULTS Several machine learning algorithms were applied to predict the product purities under various operation conditions using two SMB processes as case studies: i.e., a sugar separation of rebaudioside A and stevioside, as well as an enantioseparation of 1,1′‐bi‐2‐naphthol racemate. The results indicate that the random forest (RF) model and the deep neural network (DNN) model provide a satisfactory accuracy with a mean absolute error (MAE) lower than 0.19% (RF) and 0.08% (DNN), respectively. During the optimization process to maximize the feed flowrate under specific purity requirements, among the two selected models, DNN model showed a better generalization ability than RF model and gave a feed flowrate 10% higher than the highest value in the training dataset, which was consistent with the result obtained by using the mechanistic model. The optimized operation conditions for sugar separation were verified experimentally and the achieved purities of rebaudioside A and stevioside were 99.2% and 98.8%, respectively. CONCLUSION The DNN model was successfully used to substitute the mechanistic model to reach a rapid optimization of SMB processes with an improved efficiency of 103 ~ 104 times. © 2021 Society of Chemical Industry (SCI).

show abstract

Liver disease screening based on densely connected deep neural networks

Cited by 40 publications

References 12 publications

A Transfer Learning Method for Pneumonia Classification and Visualization

A Transfer Learning Method for Pneumonia Classification and Visualization

Support Vector Machine Parameter Optimization to Improve Liver Disease Estimation with Genetic Algorithm

Using a machine learning model for the optimal design of simulated moving bed processes and its application to separate rebaudioside A and stevioside

Contact Info

Product

Resources

About