Lung Cancer Detection Using CT Image Based on 3D Convolutional Neural Network

Ahmed, Tasnim; Parvin, Mst. Shahnaj; Haque, Mohammad Reduanul; Uddin, Mohammad Shorif

doi:10.4236/jcc.2020.83004

Cited by 31 publications

(8 citation statements)

References 6 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…As a modi cation of the existing long short-term memory (LSTM), ConvLSTM is characterized by the addition of a convolution layer to the input sequence to effectively learn the spatial information of the frame image [39]. Moreover, as time information can be considered through repetitive LSTM operations, the LSTM is suitable for video processing and weather prediction [37]. Figure 3 shows the inner structure of ConvLSTM.…”

Section: Model Descriptionmentioning

confidence: 99%

Protection from Particulate Matter and Infection: Thermal Imaging and Deep Learning-based Fit-checking of Quasi-drug Masks

Kim

Choi

et al. 2023

Preprint

View full text Add to dashboard Cite

Background Particulate matter and infectious diseases confer serious health risks, particularly in healthcare workers who experience occupational exposure risk. Masks can provide effective protection against such risks, although their efficacy is only as good as their fit. Therefore, a fit test is performed to ensure correct fit of the mask. In this study, we aimed to develop an artificial intelligence system to quickly and easily determine correct mask-wearing in real time using thermal videos that ascertained temperature changes caused by air trapped inside the mask. Methods We investigated the effectiveness of deep learning-based identification of the correct way to wear a mask based on thermal videos with five types of masks, which were approved as quasi-drugs by the Korean Ministry of Food and Drug Safety, and four ways of wearing these masks including one proper way and three improper ways. The same conditions were repeated five times, with a total of 100 videos per participant, and 5000 videos were obtained in this study. We used a 3D Convolutional Neural Network (3DCNN) and Convolutional Long Short-Term Memory (ConvLSTM) for data analysis. Both models performed binary and multi-classification to categorize mask-wearing. Results 3DCNN performed better than ConvLSTM by achieving higher scores in both binary and multi-classification tasks. The AUROC value for multi-classification using 3DCNN was the highest at 0.986 whereas the remaining parameters of accuracy, precision, recall, specificity, and F1-score were all better with the binary classification. All mask types showed AUROC values > 0.9, with KF-AD being the best classified. Conclusion This novel approach uses thermal imaging and deep learning techniques to effectively monitor correct mask-wearing and could be useful in high-risk environments, including in healthcare settings. This method can be applied to various mask types, which enables easy generalizability and advantages in public and occupational health and healthcare. Furthermore, integrating this novel technology into other screening methods can improve the safety and well-being of people, including healthcare workers, in various situations.

show abstract

Section: Model Descriptionmentioning

confidence: 99%

Protection from Particulate Matter and Infection: Thermal Imaging and Deep Learning-based Fit-checking of Quasi-drug Masks

Kim

Choi

et al. 2023

Preprint

View full text Add to dashboard Cite

show abstract

“…This section provides the summary of techniques commonly used by researchers for identification of lung cancer and their results. In a recent study, Ahmed et al [ 15 ] tested 3D convolutional neural network on LUNA16 (lungs nodule analysis) dataset of 100 patients to identify the effected nodules. First, preprocessing was performed using thresholding technique, which itself contained two stages, i.e., resizing the image and averaging it.…”

Section: Related Workmentioning

confidence: 99%

Identification and Classification of Lungs Focal Opacity Using CNN Segmentation and Optimal Feature Selection

JAVED

Liaqat

Meraj

et al. 2023

Computational Intelligence and Neuroscience

View full text Add to dashboard Cite

Lung cancer is one of the deadliest cancers around the world, with high mortality rate in comparison to other cancers. A lung cancer patient’s survival probability in late stages is very low. However, if it can be detected early, the patient survival rate can be improved. Diagnosing lung cancer early is a complicated task due to having the visual similarity of lungs nodules with trachea, vessels, and other surrounding tissues that leads toward misclassification of lung nodules. Therefore, correct identification and classification of nodules is required. Previous studies have used noisy features, which makes results comprising. A predictive model has been proposed to accurately detect and classify the lung nodules to address this problem. In the proposed framework, at first, the semantic segmentation was performed to identify the nodules in images in the Lungs image database consortium (LIDC) dataset. Optimal features for classification include histogram oriented gradients (HOGs), local binary patterns (LBPs), and geometric features are extracted after segmentation of nodules. The results shown that support vector machines performed better in identifying the nodules than other classifiers, achieving the highest accuracy of 97.8% with sensitivity of 100%, specificity of 93%, and false positive rate of 6.7%.

show abstract

“…Various researchers have proposed various models and methods globally in the past two decades [11] [12]. In [13], LUNA16 dataset, CT scan images with label nodules are used by the authors to detect cancer using 3D-CNN. Initially, the raw images are preprocessed using a threshold approach, and later vanilla 3D NN architecture is used to classify the images into cancerous and non-cancerous.…”

Section: Literature Studymentioning

confidence: 99%

Anomaly Detection on Medical Images using Autoencoder and Convolutional Neural Network

Siddalingappa¹,

Sekar²

2021

IJACSA

View full text Add to dashboard Cite

Detection of anomalies from the medical image dataset improves prognosis by discovering new facts hidden in the data. The present study aims to discuss anomaly detection using autoencoders and convolutional neural networks. The autoencoder identifies the imbalance between normal and abnormal samples. They create learning models flexible and accurate on training data. The problem is addressed in four stages: 1) training: an autoencoder is initialized with the hyperparameters and trained on the lung cancer CT scan images, 2) test: the autoencoder reconstructs the input from the latent space representation with a slight variation from the original data, indicated by a reconstruction error as Mean Squared Error (MSE), 3) evaluate: the MSE value of the training and test dataset are compared. The MSE values of anomalous data are higher than a base threshold, detecting those as anomalies, 4) validate: the efficiency metrics such as accuracy and MSE scores are used at both training and validation phases. The dataset was further classified as benign and malignant. The accuracy reported for outlier detection and the classification task is 98% and 97.2%. Thus, the proposed autoencoder-based anomaly detection could positively isolate anomalies from the CT scan images of lung cancer.

show abstract

Lung Cancer Detection Using CT Image Based on 3D Convolutional Neural Network

Cited by 31 publications

References 6 publications

Protection from Particulate Matter and Infection: Thermal Imaging and Deep Learning-based Fit-checking of Quasi-drug Masks

Protection from Particulate Matter and Infection: Thermal Imaging and Deep Learning-based Fit-checking of Quasi-drug Masks

Identification and Classification of Lungs Focal Opacity Using CNN Segmentation and Optimal Feature Selection

Anomaly Detection on Medical Images using Autoencoder and Convolutional Neural Network

Contact Info

Product

Resources

About