scholarly journals Using Deep Learning for Classification of Early Lung Nodules on Computed Tomography Images

2018 ◽  
Author(s):  
Lucas Lima ◽  
Marcelo Oliveira

Lung cancer is the leading cause of cancer mortality, accounting for approximately 20% of all cancer-related deaths. Nevertheless, despite the development of new therapeutic agents and technologies, only 16% of lung cancer patients are diagnosed at early stages. Therefore, to diagnose in early stages, when the nodules are very small, is a complex task for specialists and presents some challenges. To assist the specialists, the main purpose of this work is to propose the use of Deep Learning to classify 25,200 small pulmonary nodules balanced with diameter 5-10mm. The result was of 0.992 (+/- 0.001) of area under ROC curve using 10-fold cross validation. The proposed method showed to be promising to assist the specialists in classification of small lung nodules.

2017 ◽  
Vol 2017 ◽  
pp. 1-7 ◽  
Author(s):  
QingZeng Song ◽  
Lei Zhao ◽  
XingKe Luo ◽  
XueChen Dou

Lung cancer is the most common cancer that cannot be ignored and cause death with late health care. Currently, CT can be used to help doctors detect the lung cancer in the early stages. In many cases, the diagnosis of identifying the lung cancer depends on the experience of doctors, which may ignore some patients and cause some problems. Deep learning has been proved as a popular and powerful method in many medical imaging diagnosis areas. In this paper, three types of deep neural networks (e.g., CNN, DNN, and SAE) are designed for lung cancer calcification. Those networks are applied to the CT image classification task with some modification for the benign and malignant lung nodules. Those networks were evaluated on the LIDC-IDRI database. The experimental results show that the CNN network archived the best performance with an accuracy of 84.15%, sensitivity of 83.96%, and specificity of 84.32%, which has the best result among the three networks.


Author(s):  
Nataliya Gusarova ◽  
Artem Lobantsev ◽  
Aleksandra Vatian ◽  
Anton Klochrov ◽  
Maxim Kabyshev ◽  
...  

Introduction: Lung cancer is one of the most formidable cancers. The use of neural networks technologies in its diagnostics is promising, but the datasets collected from real clinical practice cannot cover a variety of lung cancer manifestations.  Purpose: Assessment of the possibility of improving the classification of pulmonary nodules by means of generative augmentation of available datasets under resource constraints. Methods: We used part of LIDC-IDRI dataset,  the StyleGAN architecture for generating artificial lung nodules and the VGG11 model as a classifier. We generated pulmonary nodules using the proposed pipeline and invited four  experts to visually evaluate them. We formed four experimental datasets with different types of augmentation, including use of synthesized data, and we compared the effectiveness of the classification performed by the VGG11 network when training for each dataset. Results: 10 generated nodules in each group of characteristics were presented for assessment. In all cases, positive expert assessments were obtained with a Fleiss's kappa coefficient k = 0.6–0.9. We got the best values of ROCAUC=0.9604 and PRAUC=0.9625 with the proposed approach of a generative augmentation. Discussion: The obtained efficience metrics are superior to the baseline  results obtained using comparably small training datasets, and slightly less than the best results achieved using much more powerful computational resources. So, we have shown that one can effectively use for augmenting an unbalanced dataset a combination of StyleGAN and VGG11, which does not require large computing resources as well as a large initial dataset for training.


Author(s):  
Giovanni Da Silva ◽  
Aristófanes Silva ◽  
Anselmo De Paiva ◽  
Marcelo Gattass

Lung cancer presents the highest mortality rate, besides being one of the smallest survival rates after diagnosis. Thereby, early detection is extremely important for the diagnosis and treatment. This paper proposes three different architectures of Convolutional Neural Network (CNN), which is a deep learning technique, for classification of malignancy of lung nodules without computing the morphology and texture features. The methodology was tested onto the Lung Image Database Consortium and Image Database Resource Initiative (LIDC-IDRI), with the best accuracy of 82.3%, sensitivity of 79.4% and specificity 83.8%.


2021 ◽  
Vol ahead-of-print (ahead-of-print) ◽  
Author(s):  
Priyanka Yadlapalli ◽  
D. Bhavana ◽  
Suryanarayana Gunnam

PurposeComputed tomography (CT) scan can provide valuable information in the diagnosis of lung diseases. To detect the location of the cancerous lung nodules, this work uses novel deep learning methods. The majority of the early investigations used CT, magnetic resonance and mammography imaging. Using appropriate procedures, the professional doctor in this sector analyses these images to discover and diagnose the various degrees of lung cancer. All of the methods used to discover and detect cancer illnesses are time-consuming, expensive and stressful for the patients. To address all of these issues, appropriate deep learning approaches for analyzing these medical images, which included CT scan images, were utilized.Design/methodology/approachRadiologists currently employ chest CT scans to detect lung cancer at an early stage. In certain situations, radiologists' perception plays a critical role in identifying lung melanoma which is incorrectly detected. Deep learning is a new, capable and influential approach for predicting medical images. In this paper, the authors employed deep transfer learning algorithms for intelligent classification of lung nodules. Convolutional neural networks (VGG16, VGG19, MobileNet and DenseNet169) are used to constrain the input and output layers of a chest CT scan image dataset.FindingsThe collection includes normal chest CT scan pictures as well as images from two kinds of lung cancer, squamous and adenocarcinoma impacted chest CT scan images. According to the confusion matrix results, the VGG16 transfer learning technique has the highest accuracy in lung cancer classification with 91.28% accuracy, followed by VGG19 with 89.39%, MobileNet with 85.60% and DenseNet169 with 83.71% accuracy, which is analyzed using Google Collaborator.Originality/valueThe proposed approach using VGG16 maximizes the classification accuracy when compared to VGG19, MobileNet and DenseNet169. The results are validated by computing the confusion matrix for each network type.


2021 ◽  
pp. 1-13
Author(s):  
Malathi Murugesan ◽  
Kalaiselvi Kaliannan ◽  
Shankarlal Balraj ◽  
Kokila Singaram ◽  
Thenmalar Kaliannan ◽  
...  

Deep learning algorithms will be used to detect lung nodule anomalies at an earlier stage. The primary goal of this effort is to properly identify lung cancer, which is critical in preserving a person’s life. Lung cancer has been a source of concern for people all around the world for decades. Several researchers presented numerous issues and solutions for various stages of a computer-aided system for diagnosing lung cancer in its early stages, as well as information about lung cancer. Computer vision is one of the field of artificial intelligence this is a better way to detect and prevent the lung cancer. This study focuses on the stages involved in detecting lung tumor regions, namely pre-processing, segmentation, and classification models. An adaptive median filter is used in pre-processing to identify the noise. The work’s originality seeks to create a simple yet effective model for the rapid identification and U-net architecture based segmentation of lung nodules. This approach focuses on the identification and segmentation of lung cancer by detecting picture normalcy and abnormalities.


2015 ◽  
pp. 2015 ◽  
Author(s):  
Yu-Jen Yu-Jen Chen ◽  
Kai-Lung Hua ◽  
Che-Hao Hsu ◽  
Wen-Huang Cheng ◽  
Shintami Chusnul Hidayati

2019 ◽  
Vol 8 (4) ◽  
pp. 10893-10901

Mortality rate of lung cancer is increasing very day all over the world. Early stage lung nodules detection and proper treatment is solution to reduce the deaths due to lung cancer. In this research work proposed integrated CADe/CADx system segments and classifies lung nodules into benign or malignant. CADe phase segments Well Circumscribed Nodules (WCN), Juxta Vascular Nodules (JVN) and Juxta Pleural Nodules (JPN) of different size in diameter. This part uses algorithms proposed in our previous WCN, JVN and JPN lung nodules segmentation work. CADx performance classification of segmented WCNs, JVNs and JPNs nodules into benign or malignant. In first part of CADx system hybrid features of segmented lung nodules are extracted and features dimension vector is reduced with Linear Discrimination Analysis. Finally, Probabilistic Neural Network uses reduced hybrid features of segmented nodules to classify segmented nodules as benign or malignant. Proposed integrated system achieved high classification accuracy of 94.85 for WCNs, 97.65 for JVNs and 97.96 for JPNs of different size in diameter (nodules diameter< 10mm, nodules diameter >10mm and < 30mm, nodules diameter >30mm and <70mm). For small nodules achieved classification performance values are, accuracy of 94.85, sensitivity of 90 and specificity of 95.85. And nodules of size 10mm to 30mm obtained accuracy, sensitivity and specificity are 97.85, 97.65 and 94.15 respectively.


2022 ◽  
Vol 2022 ◽  
pp. 1-12
Author(s):  
Wenfa Jiang ◽  
Ganhua Zeng ◽  
Shuo Wang ◽  
Xiaofeng Wu ◽  
Chenyang Xu

Lung cancer is one of the malignant tumors with the highest fatality rate and nearest to our lives. It poses a great threat to human health and it mainly occurs in smokers. In our country, with the acceleration of industrialization, environmental pollution, and population aging, the cancer burden of lung cancer is increasing day by day. In the diagnosis of lung cancer, Computed Tomography (CT) images are a fairly common visualization tool. CT images visualize all tissues based on the absorption of X-rays. The diseased parts of the lung are collectively referred to as pulmonary nodules, the shape of nodules is different, and the risk of cancer will vary with the shape of nodules. Computer-aided diagnosis (CAD) is a very suitable method to solve this problem because the computer vision model can quickly scan every part of the CT image of the same quality for analysis and will not be affected by fatigue and emotion. The latest advances in deep learning enable computer vision models to help doctors diagnose various diseases, and in some cases, models have shown greater competitiveness than doctors. Based on the opportunity of technological development, the application of computer vision in medical imaging diagnosis of diseases has important research significance and value. In this paper, we have used a deep learning-based model on CT images of lung cancer and verified its effectiveness in the timely and accurate prediction of lungs disease. The proposed model has three parts: (i) detection of lung nodules, (ii) False Positive Reduction of the detected nodules to filter out “false nodules,” and (iii) classification of benign and malignant lung nodules. Furthermore, different network structures and loss functions were designed and realized at different stages. Additionally, to fine-tune the proposed deep learning-based mode and improve its accuracy in the detection Lung Nodule Detection, Noudule-Net, which is a detection network structure that combines U-Net and RPN, is proposed. Experimental observations have verified that the proposed scheme has exceptionally improved the expected accuracy and precision ratio of the underlined disease.


Sign in / Sign up

Export Citation Format

Share Document