For humans, it is difficult and time-consuming to use Computed Tomography (CT) pictures to identify lung illnesses. This research work develops a novel algorithm for segmenting the cancer part in the lung and an algorithm is enhanced to classify the lung cancer images. MediNet is a segmentation-based algorithm, used to segregate the cancer-affected region from an unaffected region in the CT image whereas LuNet is the algorithm developed by enhancing an algorithm to achieve the classification of the type of cancer. This MediNet algorithm involves layers such as thirty convolutional layers, three maxpooling layers, and three bi-directional convolutional layers. By obtaining 97% accuracy, MediNet outperformed other models. Besides this, a web application is also developed to assist the doctor with a UI interface. This is developed in such a way that the doctor can upload a cancerous or non-cancerous scan in the application which will be linked to the developed classification and segmentation algorithm Medinet and LuNet. Thebackend will be running in a server that will take the input from the web app and process the output and return back the results to the web application in terms of the presence or absence of cancer as well as segmented cancerous areas in case of presence of cancer.


    Zugriff

    Zugriff prüfen

    Verfügbarkeit in meiner Bibliothek prüfen

    Bestellung bei Subito €


    Exportieren, teilen und zitieren



    Titel :

    Novel Lung Cancer Classification and Segmentation Algorithm with Web UI Interface


    Beteiligte:
    Divya, M. (Autor:in) / Sathya, S. (Autor:in)


    Erscheinungsdatum :

    2022-12-01


    Format / Umfang :

    3185357 byte




    Medientyp :

    Aufsatz (Konferenz)


    Format :

    Elektronische Ressource


    Sprache :

    Englisch



    Performance of Image Processing Algorithm for Lung Cancer Classification

    Subhash, K. B. V. / Revathi, R. / Ahammad, Sk Hasane et al. | IEEE | 2021


    Histological Classification of Lung Cancer

    Maggiore, C. / Mule, A. / Fadda, G. et al. | British Library Conference Proceedings | 2004


    New automatic defect classification algorithm based on a classification-after-segmentation framework

    Lee, S.-H. / Koo, H.-I. / Cho, N.-I. | British Library Online Contents | 2010


    A novel improved SLIC superpixel segmentation algorithm

    Chen, Zhijie / Zhong, Zhaoqi / Pan, Xiangyu et al. | IEEE | 2022