Guminta, Dinda Galuh (2019) Analisis Gejala Klinis Kanker Payudara Menggunakan Regresi Logistik Biner-Genetic Algorithm Di RS Nur Hidayah Bantul. Other thesis, Institut Teknologi Sepuluh Nopember.
Preview |
Text
06211540000093-Undergraduate_Theses.pdf Download (2MB) | Preview |
Abstract
Kanker payudara adalah penyakit seluler yang ditandai oleh pertumbuhan sel abnormal pada jaringan payudara. Penyakit ini menempati peringkat pertama jenis kanker penyebab kematian di Indonesia. Pada umumnya, wanita melakukan pemeriksaan kanker payudara pada stadium lanjut yang menyebabkan rendahnya angka harapan hidup, padahal kanker payudara dapat dicegah dengan mengenali gejala-gejala klinis seperti benjolan di payudara atau ketiak, bentuk dan ukuran payudara yang tidak sama, penarikan puting susu, dan lain-lain. Disamping itu, gangguan tidur perlu dipertimbangkan sebagai risiko kanker payudara. Pada penelitian ini gangguan tidur diukur dengan jam tidur responden setiap malam. Oleh karena itu, dilakukan analisis gejala klinis dan gangguan tidur dengan regresi logistik biner. Pemilihan variabel dilakukan menggunakan seleksi backward, forward, stepwise, dan genetic algorithm dengan tujuan mendapatkan gejala-gejala yang berpengaruh. Hasil penelitian menggunakan semua metode seleksi menunjukkan bahwa nyeri di bagian tertentu payudara berpengaruh terhadap status pemeriksaan kanker payudara pada taraf signifikan 5% dengan odds ratio 5,98. Gejala penarikan puting susu berpengaruh terhadap kanker payudara pada taraf signifikan 10% dengan odds ratio 6,13 menggunakan seleksi backward, forward, dan stepwise. Sementara itu, genetic algorithm menghasilkan gejala yang berpengaruh terhadap kanker payudara yaitu perubahan bentuk dan ukuran payudara pada tingkat signifikan 12% dengan odds ratio 4,19.
================================================================================================================================
Breast cancer is a disease caused by the uncontrolled growth of abnormal cells in breast tissue. This cancer is the most frequent cancer and the first leading cause of cancer death among women in Indonesia. A majority of women with breast cancer are diagnosed at an advanced stage which causes their survival rates to be low, whereas breast cancer can be detected through early diagnosis. Early diagnosis can be done by recognizing the clinical symptoms of breast cancer such as breast lump or a lump under the armpit, change the shape and size of the breast, nipple retraction, and so on. Besides that, sleep disorders also need to be considered as a risk of breast cancer. Sleep disorders were measured by the sleep duration of the respondent. Therefore, this research was conducted to analyze breast cancer symptoms and sleep disorders using binary logistic regression with variable selection using backward, forward, stepwise, and genetic algorithm. This study showed that all selection methods produce breast pain as a significant variable of breast cancer examination (α=5%) with odds ratio of 5,98. Backward, forward, and stepwise selection showed that nipple retraction whose correlation with the outcome is statistically significant (α=10%) and provide odds ratio of 6,13. Meanwhile, genetic algorithm showed that change shape and size of the breast have a statistically significant correlation with breast cancer examination (α=12%) and odds ratio of 4,19.
| Item Type: | Thesis (Other) |
|---|---|
| Additional Information: | RSSt 519.536 Gum a-1 2019 |
| Uncontrolled Keywords: | Durasi Tidur, Gejala Klinis, Genetic Algorithm, Kanker Payudara, Regresi Logistik Biner |
| Subjects: | Q Science > QA Mathematics > QA278.2 Regression Analysis. Logistic regression Q Science > QA Mathematics > QA402.5 Genetic algorithms. Interior-point methods. |
| Divisions: | Faculty of Mathematics, Computation, and Data Science > Statistics > 49201-(S1) Undergraduate Thesis |
| Depositing User: | GUMINTA DINDA GALUH |
| Date Deposited: | 31 May 2023 11:34 |
| Last Modified: | 31 May 2023 11:34 |
| URI: | http://repository.its.ac.id/id/eprint/63952 |
Actions (login required)
 |
View Item |