A Deep Analysis of Google Net and AlexNet for Lung Cancer Detection
B.Almas1, K.Sathesh2, S.Rajasekaran3
1B.Almas, Electronics & Communication Engineering, Madanapalle Institute of Technology and Science, Madanapalle ,Andra Pradesh, India.
2Dr. K.Sathesh, Associate Professor, Department of Electronics &Communication Engineering, Madanapalle Institute of Technology and Science, Madanapalle ,Andra Pradesh, India.
3Dr.S.Rajasekaran, Associate Professor, Department of Electronics &Communication Engineering, Madanapalle Institute of Technology and Science, Madanapalle ,Andra Pradesh, India.
Manuscript received on November 25, 2019. | Revised Manuscript received on December 08, 2019. | Manuscript published on December 30, 2019. | PP: 395-399 | Volume-9 Issue-2, December, 2019. | Retrieval Number: B3226129219/2019©BEIESP | DOI: 10.35940/ijeat.B3226.129219
Open Access | Ethics and Policies | Cite | Mendeley
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: Lung cancer is the major cancer that cannot be disregarded intentionally and causes deceased with late healthcare. Now, Computed Tomography(CT) scan allows the doctors to recognize the lung cancer in the beginning of the stage. Majority of cases are tends to be failed in diagnosis of determining the lung cancer eventhough the doctors are experienced, they failed to detect the cancer. Deep learning is the important technique that can be applicable in medical imaging diagnosis. In this paper, the implementation of Convolutional Neural Networks such as GoogleNet (Inception) and AlexNet are analyzed for the lung cancer detection. The cancer images from LIDC-IDRI dataset is used for this research work. The Preprocessed cancer images are trained using GoogleNet and AlexNet to determine the cancer affected part of the lungs. The identification of lung cancer by using GoogLeNet and AlexNet are used for training the network, and image classification. These networks are provided with layered architecture for classification. We have found that AlexNet and GoogLeNet provides the comparable results by including parameters like time, initial learning rate and accuracy.
Keywords: Alex Net, Accuracy, Convolutional NeuralNetwork, Diagnosis, GoogleNet, Learning rate