Muna, Nailatul (2020) Regresi Cox Dengan Fungsi B-Spline Pada Kasus Kanker Serviks. Undergraduate thesis, Institut Teknologi Sepuluh Nopember.
Preview |
Text
06211640000092-NAILATUL.MUNA.pdf Download (1MB) | Preview |
Abstract
Kematian yang disebabkan oleh kanker serviks di dunia terus meningkat. International Agency for Research on Cancer (IARC) merilis data kematian akibat kanker serviks tahun 2008 yaitu sebanyak 275.000 jiwa dan meningkat pada tahun 2018 mencapai 311.365 jiwa. Kematian yang terjadi akibat kanker serviks tidak hanya dikarenakan tumor primer melainkan juga akibat terjadinya invasi dan metastasis. Penelitian ini bertujuan untuk mengetahui faktor-faktor yang mempengaruhi pasien kanker serviks untuk tidak mengalami metastasis. Metode yang digunakan yaitu regresi Cox proportional hazard (PH) dengan asumsi log hazard ratio (HR) linear terhadap kovariat. Asumsi linearitas tersebut dapat dievaluasi menggunakan grafik martingale residual. Namun, seringkali diperoleh garis martingale residual yang tidak ideal, sehingga digunakan fungsi spline dalam regresi Cox PH dengan basis fungsi yaitu B-spline sebagai fungsi smoothing. Data penelitian yang digunakan yaitu data rekam medis pasien kanker serviks yang pernah menjalani perawatan di RS Onkologi Surabaya pada rentang waktu 1 januari 2008 sampai 31 Desember 2016. Hasil regresi Cox PH dengan fungsi B-spline pada model terbaik menunjukkan bahwa faktor yang mempengaruhi pasien kanker serviks untuk tidak mengalami metastasis yaitu faktor stadium yaitu stadium II dan stadium IV dan status pernikahan menikah dan janda.
==================================================================================================================
Deaths caused by cervical cancer in the world continue to increase. The International Agency for Research on Cancer (IARC) released data on deaths from cervical cancer in 2008, totaling 275,000, and increased in 2018 to reach 311,365 people. Deaths that occur due to cervical cancer are not only due to primary tumors but also due to invasion and metastasis. This study aims to determine the factors that influence cervical cancer patients not to experience metastasis. The method used is Cox proportional hazard (PH) regression where the assumption that log hazard ratio (HR) linear to covariates. The linearity assumption can be evaluated using a residual martingale graph. However, residual martingale lines are often obtained which is not ideal, so researchers apply the spline function in the Cox PH regression, B-spline as a smoothing function. The research data used were medical records of cervical cancer patients who had undergone treatment at the Surabaya Oncology Hospital in the period of 1 January 2008 to 31 December 2016. The results of the Cox PH regression with B-spline function of the best model suggest that the factors affecting cervical cancer patients did not experience metastasis are the patient' stage which stage II and stage IV and marital and widow marital status.
| Item Type: | Thesis (Undergraduate) |
|---|---|
| Additional Information: | RSSt 519.546 Mun r-1 • Muna, Nailatul |
| Uncontrolled Keywords: | B-spline, Kanker Serviks, Martingale Residual, Metastasis B-spline, Cervical Cancer, Martingale residual, Metastasis |
| Subjects: | Q Science > QA Mathematics > QA276 Mathematical statistics. Time-series analysis. Failure time data analysis. Survival analysis (Biometry) |
| Divisions: | Faculty of Mathematics, Computation, and Data Science > Statistics > 49201-(S1) Undergraduate Thesis |
| Depositing User: | Nailatul Muna |
| Date Deposited: | 26 Aug 2020 02:55 |
| Last Modified: | 12 May 2023 02:03 |
| URI: | http://repository.its.ac.id/id/eprint/81213 |
Actions (login required)
 |
View Item |