NOVITASARI, DIAH AYU (2014) ANALISIS SURVIVAL PADA DATA REKURENSI DENGAN MENGGUNAKAN COUNTING PROCESS APPROACH DAN MODEL PWP-GT STUDY KASUS: DATA KANKER SERVIK DI RUMAH SAKIT DR.SOETOMO SURABAYA. Masters thesis, Institut Teknologi Sepuluh Nopember.
Preview |
Text
1312201026-Abstract.pdf - Published Version Download (214kB) | Preview |
Preview |
Text
1312201026-Master Thesis.pdf - Published Version Download (957kB) | Preview |
Preview |
Text
1312201026-Conclusion.pdf - Published Version Download (786kB) | Preview |
Abstract
Kanker Serviks merupakan jenis kanker dengan jumlah pasien
wanita paling banyak nomer dua di dunia. Penyebab utama kanker
Serviks adalah infeksi Human Papilloma Virus (HPV). Pada stadium
akhir atau kasus yang parah maka terpaksa dilakukan histerektomi, yaitu
bedah pengangkatan rahim (uterus) secara total agar sel-sel kanker yang
sudah berkembang dalam kandungan tidak menyebar ke bagian lain
dalam tubuh. Namun pengobatan pada stadium awal tidak serta merta
menyembuhkan kanker Serviks. Analisis yang sering digunakan untuk
menganalisis kanker serviks adalah analisis survival overall. Namun,
analisis yang dilakukan dengan survival overall menggunakan data 1-5
tahun setelah pasien terdiagnosa kanker Serviks. Padahal kenyataannya
angka rekurensi kanker Serviks sangat tinggi meskipun pasien sudah
menjalani operasi, sehingga diperlukan juga analisis survival untuk data
rekurensi. Banyak model cox proporsional hazard yang telah
dikembangkan untuk menganalisis data yang berulang atau rekurensi
data. Namun, model yang direkomendasikan untuk menganalisis data
rekurensi adalah model Prentice William Peterson-Gap Time (PWP-GT)).
Tujuan dari penelitian ini adalah mengkaji estimator model Prentice
William Peterson-Gap Time (PWP-GT) dan hasil model akan
dibandingkan dengan hasil dari metode pendekatan cox proporsional
hazard yaitu Counting Process Approach serta mengaplikasikannya pada
data kanker Serviks di Rumah Sakit DR.Soetomo Surabaya. Pada model
Counting Process Approach, Stadium merupakan variabel prediktor yang
berpengaruh signifikan terhadap variabel respon dengan nilai estimasi
parameter sebesar -0.54517. Pada Model PWP-GT, variabel prediktor
yang berpengaruh signifikan terhadap variabel respon yaitu variabel jenis
kanker dengan nilai estimasi parameter sebesar -1.46399.
=====================================================================================================
Cervical cancer is a type of cancer with the number of female patients most
number two in the world. In Indonesia, Cervical cancer is the number one cause of
death in women. Primary cause of cervical cancer is infection Human Papilloma
Virus (HPV) or human papilloma virus. In the later stages or severe cases it is forced
to do a hysterectomy, the surgical removal of the uterus (womb) in total so that
cancer cells that have developed in the womb does not spread to other parts of the
body. However, treatment at an early stage and m erta not cure cervical cancer. Cases
of recurrence often occurs after initial treatment has been carried out. Analysis is
often used to analyze cervical cancer is overall survival analysis. However, the
analysis performed by using the data overall survival of 1-5 years after patient
diagnosed with cervical cancer. Although in the fact, recurrent rate of cervical cancer
is very high even though the patient had undergone surgery, so it is also necessary for
the survival analysis of data recurrent. Many cox proportional hazard models that
have been developed to analyze the data are repeated or recurrent data. However, the
recommended models for analyzing the data is modeled recurrence William Prentice
Peterson- Gap Time (PWP-GT)). The purpose of this study is to examine the model
estimator William Prentice Peterson- Gap Time (PWP-GT) and the model results will
be compared with the results of Cox proportional hazard approach is Counting
Process Approach and applying it to the data of cervical cancer in Dr.Soetomo
Hospital Surabaya. On Counting Process Approach, Stadium is a predictor variables
that significantly influence the response variable with the value of parameter
estimation is -0. 54517. the PWP-GT model, predictor variables that significantly
influence the response variable is the variable type of cancer with the estimated value
of the parameter is -1.46399.
| Item Type: | Thesis (Masters) |
|---|---|
| Additional Information: | RTSt 519.546 Nov a |
| Uncontrolled Keywords: | Analisis Survival, Rekurensi, Counting Process Approach, Kanker Servik, PWP-GT. |
| Subjects: | Q Science > QA Mathematics Q Science > QA Mathematics > QA278.2 Regression Analysis. Logistic regression |
| Divisions: | Faculty of Mathematics and Science > Statistics > 49101-(S2) Master Thesis |
| Depositing User: | Mr. Tondo Indra Nyata |
| Date Deposited: | 05 Jan 2017 07:49 |
| Last Modified: | 14 Nov 2018 05:26 |
| URI: | http://repository.its.ac.id/id/eprint/1338 |
Actions (login required)
 |
View Item |