GoogLeNet-Based Ensemble FCNet Classifier for Focal Liver Lesion Diagnosis

Balagourouchetty, Lakshmipriya; Pragatheeswaran, Jayanthi K.; Pottakkat, Biju; Ramkumar, G.

doi:10.1109/jbhi.2019.2942774

Cited by 69 publications

(29 citation statements)

References 22 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In the first method ( GLCM + SVM ) based on Gletsos et al, 11 the SVM classifier was trained on the 22‐dimensional GLCM texture features extracted from the original image patches. In the second method ( GoogleNet + Real ) based on Balagourouchetty et al, 18 the GoogleNet 29 was trained on the original image patches with the real training data augmentation. In the third method ( AlexNet + GAN ) based on Frid‐Adar et al, 17 the AlexNet 30 was trained on the original image patches with the mixed training data augmentation (

r

=50%).…”

Section: Resultsmentioning

confidence: 99%

“…Frid‐Adar et al trained a convolutional neural network (CNN) on PVP CT images of liver lesions to classify them into cysts, hemangiomas, and metastases and used a generative adversarial network (GAN) to generate synthetic lesion samples for data augmentation in the training process 16,17 . Balagourouchetty et al trained a GoogleNet‐based CNN with an image fusion of hepatic arterial phase and PVP CT images to classify data into cysts, hemangiomas, metastases, HCCs, abscesses, and normal types 18 . Various approaches ranging from classical texture analysis to recent deep learning techniques have been applied to liver lesion classification tasks.…”

Section: Introductionmentioning

confidence: 99%

“…16,17 HCCs, abscesses, and normal types. 18 Various approaches ranging from classical texture analysis to recent deep learning techniques have been applied to liver lesion classification tasks. However, most of those studies mainly focused on finding helpful imaging features for classifying lesions, whereas the challenges associated with the small lesion classification and the class imbalance problem were rarely discussed.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Classification of focal liver lesions in CT images using convolutional neural networks with lesion information augmented patches and synthetic data augmentation

Lee

Hong

et al. 2021

Medical Physics

View full text Add to dashboard Cite

Purpose We propose a deep learning method that classifies focal liver lesions (FLLs) into cysts, hemangiomas, and metastases from portal phase abdominal CT images. We propose a synthetic data augmentation process to alleviate the class imbalance and the Lesion INformation Augmented (LINA) patch to improve the learning efficiency. Methods A dataset of 502 portal phase CT scans of 1,290 FLLs was used. First, to alleviate the class imbalance and to diversify the training data patterns, we suggest synthetic training data augmentation using DCGAN‐based lesion mask synthesis and pix2pix‐based mask‐to‐image translation. Second, to improve the learning efficiency of convolutional neural networks (CNNs) for the small lesions, we propose a novel type of input patch termed the LINA patch to emphasize the lesion texture information while also maintaining the lesion boundary information in the patches. Third, we construct a multi‐scale CNN through a model ensemble of ResNet‐18 CNNs trained on LINA patches of various mini‐patch sizes. Results The experiments demonstrate that (a) synthetic data augmentation method shows characteristics different but complementary to those in conventional real data augmentation in augmenting data distributions, (b) the proposed LINA patches improve classification performance compared to those by existing types of CNN input patches due to the enhanced texture and boundary information in the small lesions, and (c) through an ensemble of LINA patch‐trained CNNs with different mini‐patch sizes, the multi‐scale CNN further improves overall classification performance. As a result, the proposed method achieved an accuracy of 87.30%, showing improvements of 10.81%p and 15.0%p compared to the conventional image patch‐trained CNN and texture feature‐trained SVM, respectively. Conclusions The proposed synthetic data augmentation method shows promising results in improving the data diversity and class imbalance, and the proposed LINA patches enhance the learning efficiency compared to the existing input image patches.

show abstract

r

=50%).…”

Section: Resultsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Classification of focal liver lesions in CT images using convolutional neural networks with lesion information augmented patches and synthetic data augmentation

Lee

Hong

et al. 2021

Medical Physics

View full text Add to dashboard Cite

show abstract

“…GoogLeNet proposed nine inception modules to overcome the drawback that high accuracies were usually achieved by deepening the layers, and the computational complexity exponentially increases as the layers become deeper. Multiple convolutions with multiple kernels and max pooling was permitted to take place simultaneously within a single layer by inception structures, to ensure that network selects more useful features and trains with optimal weights [54], [55].…”

Section: B Pretrained Cnnmentioning

confidence: 99%

Bayesian-Convolutional Neural Network Model Transfer Learning for Image Detection of Concrete Water-Binder Ratio

2020

View full text Add to dashboard Cite

Since concrete is still the most widely used building material at present, the phenomenon of engineering accidents caused by inaccurate mix proportion is extremely prominent. This makes the contradiction between timely and effective detection and test results cannot be given quickly by traditional technology become particularly outstanding. In this paper, a new method based on Bayesian-convolutional neural network model transfer learning is proposed to detect water-binder ratio, the most important parameter in mix proportion of concrete mixtures. Bayesian optimization was applied to pretrained convolutional neural networks to establish Bayesian-convolutional neural network models, avoiding tuning hyperparameters manually. The authors performed several experiments and obtained large numbers of images of freshlymixed concrete mixtures, which were used as datasets to carry out water-binder ratio detection. These models achieved high accuracies on training, validation and testing sets. Applying these models, we could implement real-time and high sampling rate water-binder ratio detection. The authors integrated the models and developed a detection system of concrete mix proportion. Equipped with definite hardware facilities, this system can effectively monitor the quality of concrete in production process and prevent engineering accidents. According to training curves of the models, a new parameter was introduced to discuss how mix proportion influence the sensitivity of concrete mixtures apparent state to the change of water-binder ratio, which is an important consideration to preliminary assess the service behaviors of concrete. Through this parameter, we also explored the essence of image features learned by models is the fluidity of concrete mixtures. INDEX TERMS Bayesian-convolutional neural network model, computer aided engineering, image detection, mix proportion, transfer learning, water-binder ratio.

show abstract

“…EMG signals are a way for estimating the electrical signal that relate to skeletal muscle; the electrical signal consists of some motor unit action potentials (MUAPs) [4]. The combination of muscle fibre action potentials from all the muscle fibres of a single motor unit is called the MUAP [5]. The pattern recognition (PR) system based on electromyography signals is divided into three main parts; data collection, feature extraction, and classification of the input data.…”

Section: Introductionmentioning

confidence: 99%

EMG Signals Classification of Wide Range Motion Signals for Prosthetic Hand Control

Mahmood¹

2021

IJIES

View full text Add to dashboard Cite

The recent revolution in the biomedical field carried out the researchers to work on the prosthetic technique because it reflects the amputee's need. Therefore, the electromyography (EMG) signals generated by muscle contractions are used to implement the prosthetic human body parts. This paper presents a pattern recognition system based on two EMG data; the first EMG data represents the general body movements collected from biceps and triceps muscles for six different motions, i.e., bowing, clapping, handshaking, hugging, jumping, and running. The Root Mean Square, Difference Absolute Standard Deviation Value, and Principal Component Analysis extract the raw signal data features and enhance classification accuracy. The machine learning method is applied, i.e., Support Vector Machine (SVM) and K-Nearest Neighbours (KNN) are used for data classification; the results show high accuracy reached 94.8% and 98.9%, respectively. Whereas, the second EMG data is selected to be more specific in hand movements, i.e., cylindrical, spherical, palmar, lateral, hook, and tip motions, because these significant motions are the first step implementing any prosthetic hand. Consequently, the mean, Standard Deviation Value, and Principal Component Analysis extract the raw signal feature. Meanwhile, the same algorithm used in the first data classification is also used to classify the second data because it shows high accuracy and good performance. SVM algorithm is used to classify the data and achieved high training accuracy, reaching 89%. The high training accuracy for different hand movements is considered an essential step toward implementing human prosthetic parts to help the people who suffer from an amputee.

show abstract

GoogLeNet-Based Ensemble FCNet Classifier for Focal Liver Lesion Diagnosis

Cited by 69 publications

References 22 publications

Classification of focal liver lesions in CT images using convolutional neural networks with lesion information augmented patches and synthetic data augmentation

Classification of focal liver lesions in CT images using convolutional neural networks with lesion information augmented patches and synthetic data augmentation

Bayesian-Convolutional Neural Network Model Transfer Learning for Image Detection of Concrete Water-Binder Ratio

EMG Signals Classification of Wide Range Motion Signals for Prosthetic Hand Control

Contact Info

Product

Resources

About