Segmentation and classification of lung tumor from 3D CT image using K-means clustering algorithm

Sarker, Prionjit; Shuvo, Md. Maruf Hossain; Hossain, Zakir; Hasan, S. M. Kamrul

doi:10.1109/icaee.2017.8255451

Cited by 25 publications

(14 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Sarker et al in 2017, proposed a method that employed the strengths of k-means clustering along with the morphological image processing techniques for three-dimensional lung cancer segmentation with the ability to calculate size of tumor with diameter more than 7mm and also identified the stage that the cancer had reached. The proposed method demonstrated an experimental accuracy and specificity of 95.68% and 98%, respectively [13]. Moriya et al in 2018 employed the application of k-means clustering algorithm.…”

Section: Introductionmentioning

confidence: 94%

Improved Lung Cancer Segmentation Using K-Means and Cuckoo Search

Singh*,

Nanglia,

Mahajan

2019

IJITEE

View full text Add to dashboard Cite

Lung Cancer is among the deadliest disease that tolls mass every year. Information technology is playing an indispensable part in availing the most successful diagnosis and top treatment strategies to fight the condition when detected at an earliest stage. The work encompasses the improvement in the quality of image segmentation of Computer Tomography (CT) medical images of lung cancer. The paper evaluates the performance of two algorithms as a post segmentation process, namely, Artificial Bee Colony (ABC) and Cuckoo Search (CS). Support Vector Machine(SVM) is also used as a cross validator over the post segmentation algorithms ABC and CS. The experimental evaluation includes Accuracy, Precision, Error, Segmentation Time, Recall and F-measure to determine the success of the proposed hybrid model. The proposed results exhibit an improved Accuracy, Precision, Recall and F-measure by 5%, 6%, 3%, 4% and 10%, 11%., 12%, 11% for k-ABC and kCS respectively.

show abstract

Section: Introductionmentioning

confidence: 94%

Improved Lung Cancer Segmentation Using K-Means and Cuckoo Search

Singh*,

Nanglia,

Mahajan

2019

IJITEE

View full text Add to dashboard Cite

show abstract

“…The thresholding method can be applied to the lung-specific level, followed by morphological operation methods, including 3D connected component analysis, region erosion, region dilation, and lung volume thresholding for lung segmentation. Then, an improved image of the lung will be produced [27,28]. After these preprocessing steps, a patient's lung volume and its associated label are in pair, ready for training (https://github.com/booz-allen-hamilton/DSB3Tutorial).…”

Section: Preprocessing and Lung Segmentationmentioning

confidence: 99%

A Two-Stage Framework for Automated Malignant Pulmonary Nodule Detection in CT Scans

2020

View full text Add to dashboard Cite

This research is concerned with malignant pulmonary nodule detection (PND) in low-dose CT scans. Due to its crucial role in the early diagnosis of lung cancer, PND has considerable potential in improving the survival rate of patients. We propose a two-stage framework that exploits the ever-growing advances in deep neural network models, and that is comprised of a semantic segmentation stage followed by localization and classification. We employ the recently published DeepLab model for semantic segmentation, and we show that it significantly improves the accuracy of nodule detection compared to the classical U-Net model and its most recent variants. Using the widely adopted Lung Nodule Analysis dataset (LUNA16), we evaluate the performance of the semantic segmentation stage by adopting two network backbones, namely, MobileNet-V2 and Xception. We present the impact of various model training parameters and the computational time on the detection accuracy, featuring a 79.1% mean intersection-over-union (mIoU) and an 88.34% dice coefficient. This represents a mIoU increase of 60% and a dice coefficient increase of 30% compared to U-Net. The second stage involves feeding the output of the DeepLab-based semantic segmentation to a localization-then-classification stage. The second stage is realized using Faster RCNN and SSD, with an Inception-V2 as a backbone. On LUNA16, the two-stage framework attained a sensitivity of 96.4%, outperforming other recent models in the literature, including deep models. Finally, we show that adopting a transfer learning approach, particularly, the DeepLab model weights of the first stage of the framework, to infer binary (malignant-benign) labels on the Kaggle dataset for pulmonary nodules achieves a classification accuracy of 95.66%, which represents approximately 4% improvement over the recent literature.

show abstract

“…In some cases, however, the tumor location may not be detected by imaging techniques. This leads to delays for biopsy and the later start of treatment [7][8][9][10][11][12][13].…”

Section: Tc Akinci Is With Department Of Electrical Engineering Imentioning

confidence: 99%

A Convolutional Neural Network Application for Predicting the Locating of Squamous Cell Carcinoma in the Lung

Noğay¹,

Akıncı²

2018

Balkan Journal of Electrical and Computer Engineering

View full text Add to dashboard Cite

Squamous cell carcinoma, one of the most common types of lung cancer types, usually occurs in the middle, right or left bronchi. Squamous cell carcinoma can be easily detected by imaging methods to determine the location within the lung. However, rarely the location of some tumor types cannot be determined. In this case, it may be delayed to obtain the results of the assay such as biopsy. This possible delay also means delayed diagnosis and delayed start of treatment. In order to solve this problem, it is possible to perform applications with machine learning methods. In this study, convolutional neural networks method was used to determine the location of cancerous tumor in squamous cell carcinoma of lung. With the designed convolutional neural network model, squamous cell carcinoma tumor location in lung cancer was estimated with an accuracy rate close to 100%.

show abstract

Segmentation and classification of lung tumor from 3D CT image using K-means clustering algorithm

Cited by 25 publications

References 9 publications

Improved Lung Cancer Segmentation Using K-Means and Cuckoo Search

Improved Lung Cancer Segmentation Using K-Means and Cuckoo Search

A Two-Stage Framework for Automated Malignant Pulmonary Nodule Detection in CT Scans

A Convolutional Neural Network Application for Predicting the Locating of Squamous Cell Carcinoma in the Lung

Contact Info

Product

Resources

About