Lung Nodule Detection from Computed Tomography Images Using Stacked Deep Convolutional Neural Network

Nakrani, Mahender G.; Sable, Ganesh S.; Shinde, Ulhas B.

doi:10.1007/978-981-16-0730-1_16

Cited by 2 publications

(1 citation statement)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Author in this article [26], proposed a computer-aided detection system based on deep CNNs which are stacked for the detection of lung nodules. A CT image is given as input to the system and uses segmentation to extract region of interest and deliver extracted region of interest to the deep CNN stacked layers for lung nodule presence probability generations.…”

Section: Related Workmentioning

confidence: 99%

Bulletin of the Polish Academy of Sciences: Technical Sciences

Prasad,

Chakravarty,

Krishna

2021

View full text Add to dashboard Cite

Computer aided detection systems are used for the provision of second opinion during lung cancer diagnosis. For early-stage detection and treatment false positive reduction stage also plays a vital role. The main motive of this research is to propose a method for lung cancer segmentation. In recent years, lung cancer detection and segmentation of tumors is considered one of the most important steps in the surgical planning and medication preparations. It is very difficult for the researchers to detect the tumor area from the CT (computed tomography) images.The proposed system segments lungs and classify the images into normal and abnormal and consists of two phases, The first phase will be made up of various stages like pre-processing, feature extraction, feature selection, classification and finally, segmentation of the tumor. Input CT image is sent through the pre-processing phase where noise removal will be taken care of and then texture features are extracted from the pre-processed image, and in the next stage features will be selected by making use of crow search optimization algorithm, later artificial neural network is used for the classification of the normal lung images from abnormal images. Finally, abnormal images will be processed through the fuzzy K-means algorithm for segmenting the tumors separately. In the second phase, SVM classifier is used for the reduction of false positives. The proposed system delivers accuracy of 96%, 100% specificity and sensitivity of 99% and it reduces false positives. Experimental results shows that the system outperforms many other systems in the literature in terms of sensitivity, specificity, and accuracy. There is a great tradeoff between effectiveness and efficiency and the proposed system also saves computation time. The work shows that the proposed system which is formed by the integration of fuzzy K-means clustering and deep learning technique is simple yet powerful and was effective in reducing false positives and segments tumors and perform classification and delivers better performance when compared to other strategies in the literature, and this system is giving accurate decision when compared to human doctor's decision.

show abstract

Section: Related Workmentioning

confidence: 99%

Bulletin of the Polish Academy of Sciences: Technical Sciences

Prasad,

Chakravarty,

Krishna

2021

View full text Add to dashboard Cite

show abstract

Lung Nodules Localization and Report Analysis from Computerized Tomography (CT) Scan Using a Novel Machine Learning Approach

et al. 2022

View full text Add to dashboard Cite

A lung nodule is a tiny growth that develops in the lung. Non-cancerous nodules do not spread to other sections of the body. Malignant nodules can spread rapidly. One of the numerous dangerous kinds of cancer is lung cancer. It is responsible for taking the lives of millions of individuals each year. It is necessary to have a highly efficient technology capable of analyzing the nodule in the pre-cancerous phases of the disease. However, it is still difficult to detect nodules in CT scan data, which is an issue that has to be overcome if the following treatment is going to be effective. CT scans have been used for several years to diagnose nodules for future therapy. The radiologist can make a mistake while determining the nodule’s presence and size. There is room for error in this process. Radiologists will compare and analyze the images obtained from the CT scan to ascertain the nodule’s location and current status. It is necessary to have a dependable system that can locate the nodule in the CT scan images and provide radiologists with an automated report analysis that is easy to comprehend. In this study, we created and evaluated an algorithm that can identify a nodule by comparing multiple photos. This gives the radiologist additional data to work with in diagnosing cancer in its earliest stages in the nodule. In addition to accuracy, various characteristics were assessed during the performance assessment process. The final CNN algorithm has 84.8% accuracy, 90.47% precision, and 90.64% specificity. These numbers are all relatively close to one another. As a result, one may argue that CNN is capable of minimizing the number of false positives through in-depth training that is performed frequently.

show abstract

Lung Nodule Detection from Computed Tomography Images Using Stacked Deep Convolutional Neural Network

Cited by 2 publications

References 14 publications

Bulletin of the Polish Academy of Sciences: Technical Sciences

Bulletin of the Polish Academy of Sciences: Technical Sciences

Lung Nodules Localization and Report Analysis from Computerized Tomography (CT) Scan Using a Novel Machine Learning Approach

Contact Info

Product

Resources

About