Desain Sistem Deteksi Kanker Serviks menggunakan Algoritma Recurrent Neural Network pada Metode Deep Learning

    Fairuz Fernanda Hermawan, - (2023) Desain Sistem Deteksi Kanker Serviks menggunakan Algoritma Recurrent Neural Network pada Metode Deep Learning. S1 thesis, Universitas Pendidikan Indonesia.

    Abstract

    Kanker merupakan salah satu penyakit berbahaya dan mematikan. Kanker dapat menyerang siapapun tanpa mengenai batas usia dan jenis kelamin seseorang. Kanker serviks merupakan kanker yang menginfeksi daerah yang menghubungkan rahim dengan vagina. Kanker serviks merupakan kanker terbanyak kedua dunia yaitu sebanyak 36.633 kasus mengacu pada Global Burden of Cancer Study (Globocan). Dalam mendeteksi kanker serviks dokter memerlukan waktu lama sebelum akhirnya dapat melakukan diagnosa. Selain itu, terdapat masalah lain yaitu mahalnya biaya untuk pemeriksaan secara visual serta memakan waktu. Untuk mengatasi masalah itu peneliti mengajukan metode Deep Learning terutama Recurrent Neural Network (RNN) untuk melakukan prediksi pada kanker serviks. Diharapkan dari pembuatan model RNN dapat memangkas waktu dalam melakukan diagnosa kanker serviks. Adapun metode yang digunakan dalam penelitian ini adalah Research and Development (R&D), data penelitian diperoleh dari rumah sakit CMI Hospital bandung sebanyak 600 sampel data dengan data yang diprediksi benar sebesar 476 dan data yang gagal prediksi sebanyak 92 sampe data dengan akurasi dari model RNN adalah 83%. adapun metrik evaluasi yang digunakan untuk menguji hasil pelatihan model RNN adalah Mean Squared Error (MSE) sebesar 0,04, Root mean Squared Error (RMSE) sebesar 0,0016, dan Mean Absoluter Error (MAE) nya adalah 0,03. grafik data aktual dan data prediksi prediksi pasien kanker serviks, serta tabel perbandingan nilai data aktual dan data prediksi yang diolah oleh model RNN.
    -----
    Cancer is one of the most dangerous and deadly diseases. Cancer can affect
    anyone regardless of age and gender. Cervical cancer is cancer that infects the
    area that connects the uterus with the vagina. Cervical cancer is the second most
    common cancer in the world, which is 36,633 cases referring to the Global Burden
    of Cancer Study (Globocan). In detecting cervical cancer, doctors need a long time
    before they can finally make a diagnosis. In addition, there are other problems,
    namely the high cost of visual inspection and time consuming. To overcome this
    problem, researchers proposed Deep Learning methods, especially Recurrent
    Neural Network (RNN) to predict cervical cancer. It is hoped that the creation of
    the RNN model can decrease time in diagnosing cervical cancer. as for the method
    used in this research is Research and Development (R&D), research data obtained
    from CMI Hospital Bandung as many as 600 data samples with correctly predicted
    data of 476 and data that failed the prediction of 92 data samples with the accuracy
    of the RNN model is 83%. The evaluation metrics used to test the training results
    of the RNN model are Mean Squared Error (MSE) of 0,04, Root mean Squared
    Error (RMSE) of 0,0016, and Mean Absoluter Error (MAE) is 0,03. graphs of
    actual data and predicted data prediction of cervical cancer patients, as well as a
    comparison table of actual data values and predicted data processed by the RNN
    model.

    [thumbnail of S_SISTEL_1903601_Title.pdf] Text
    S_SISTEL_1903601_Title.pdf

    Download (582kB)
    [thumbnail of S_SISTEL_1903601_Chapter1.pdf] Text
    S_SISTEL_1903601_Chapter1.pdf

    Download (129kB)
    [thumbnail of S_SISTEL_1903601_Chapter2.pdf] Text
    S_SISTEL_1903601_Chapter2.pdf
    Restricted to Staf Perpustakaan

    Download (630kB)
    [thumbnail of S_SISTEL_1903601_Chapter3.pdf] Text
    S_SISTEL_1903601_Chapter3.pdf

    Download (435kB)
    [thumbnail of S_SISTEL_1903601_Chapter4.pdf] Text
    S_SISTEL_1903601_Chapter4.pdf
    Restricted to Staf Perpustakaan

    Download (1MB)
    [thumbnail of S_SISTEL_1903601_Chapter5.pdf] Text
    S_SISTEL_1903601_Chapter5.pdf

    Download (58kB)
    [thumbnail of S_SISTEL_1903601_Apendix.pdf] Text
    S_SISTEL_1903601_Apendix.pdf
    Restricted to Staf Perpustakaan

    Download (134kB)
    Official URL: http://repository.upi.edu
    Item Type: Thesis (S1)
    Additional Information: Link Google Scholar: Ahmad Fauzi : https://scholar.google.com/citations?user=b6BGJbEAAAAJ&hl=en Ichwan Nul Ichsan : https://scholar.google.co.id/citations?user=uIGskD0AAAAJ&hl=en ID SINTA: Ahmad Fauzi : 6122861 Ichwan Nul Ichsan : 6721201
    Uncontrolled Keywords: kanker serviks , Deep Learning, Recurrent Neural Network
    Subjects: L Education > L Education (General)
    T Technology > T Technology (General)
    Divisions: UPI Kampus Purwakarta > S1 Sistem Telekomunikasi
    Depositing User: FAIRUZ FERNANDA HERMAWAN
    Date Deposited: 31 Aug 2023 09:13
    Last Modified: 31 Aug 2023 09:13
    URI: http://repository.upi.edu/id/eprint/100858

    Actions (login required)

    View Item View Item