Akurasi Tes Medis untuk Mendiagnosa Penyakit Tinea Unguium Menggunakan Metode Bayesian

Selfia Aprianti Alifa, Suwanda Suwanda, Suliadi Suliadi

Abstract


Abstract. Tinea unguium is fungal infection of upon the nail.Usually affected by is toenails, who become thick and scaly with debris subungual.An instrument for diagnose disease tinea unguium uses the standard is a dermatophyte strip test (DST) and instrument to diagnose disease tinea unguium used a gold standard is through a method of biakan.In this paper want to see the level of accuracy of of a tool standard but knowledge same like the gold standard. In general known test accuracy used that is true and false positive fraction, positive and negative predictive value, and positive and negative distribution of diagnostic likelihood ratios produced is the damage point, and often we want to see the estimation the interval, to get the estimation the interval we need to the distribution of the estimate. The Bayes method can be used to obtain distributions from estimates. From the examination, the results of 89.35% True Positive Fraction, 12.71% False Positive Fraction, 83.98% Positive Predictive Value, 91.66% Negative Predictive Value, Positive of Diagnostic Likelihood Ratio 7.899, Negative of Diagnostic Likelihood Ratio 0.1223.

Keywords: Bayesian Method, Gold Standard, Diagnostic Accuracy Test.

Abstrak. Tinea Unguium adalah infeksi jamur pada kuku. Biasanya yang terkena adalah kuku kaki, yang menjadi tebal dan bersisik dengan puing-puing subungual. Alat untuk mendiagnosa penyakit Tinea Unguium menggunakan metode standard adalah Dermatophyte Strip Test (DST) dan alat untuk mendiagnosa penyakit Tinea Unguium menggunakan alat gold standard adalah melalui metode Biakan. Pada skripsi ini ingin melihat tingkat akurasi dari alat standard tetapi perlakuannya sama seperti alat gold standard. Pada umumnya ukuran-ukuran uji akurasi yang digunakan yaitu True and False Positive Fraction, Positive and Negative Predictive Value, dan Positive and Negative Distribution of Diagnostic Likelihood Ratios yang dihasilkan adalah taksiran titik, padahal sering kali kita ingin melihat taksiran intervalnya, untuk mendapatkan taksiran intervalnya kita perlu distribusi dari taksiran. Metode Bayes dapat digunakan untuk mendapatkan distribusi dari taksiran. Dari pemeriksaan didapat hasil True Positive Fraction sebesar 89,35%, False Positive Fraction sebesar 12,71%, Positive Predictive Value sebesar 83,98%, Negative Predictive Value sebesar 91,66%, Positive of Diagnostic Likelihood Ratio 7,899, Negative of Diagnostic Likelihood Ratio 0,1223.

Kata Kunci: Metode Bayesian,Gold Standard, Ukuran Uji Akurasi Diagnosik.


Keywords


Metode Bayesian,Gold Standard, Ukuran Uji Akurasi Diagnosik

Full Text:

PDF

References


Ambrosius, T. Walter. (2007). Topics in Biostatistics. New Jersey:Humana Press.

Broemeling, Lyle D. (2011). Bayesian Methods for Medical Test Accuracy, USA.

Broemeling, Lyle D. (2012). Advanced Bayesian Methods for Medical Test Accuracy, London:Taylor and Francis Group.

Ferri, F. Fred. (2018). Clinical Advisor. United States:Elsevier.

Hajarisman, N. (2009). Analisis Data Kategorik. Bandung: Pusat Penerbitan Unisba.

Hogg, Robert V., Allent T Craig. (1995). Introduction to Mathematical Statistics. New jersey:Prentice Hall.

Mutaqin, A. K. (2009). Pengantar Ilmu Peluang, Bandung:Pustaka Ceria Yayasan Pena.

Ng, Kai Wang., Guo-Liang Tiang., Man Lai Tang. (2011). Dirichlet and Related Distributions, Hongkong.

Peterson, M. Andrew. (2013). Pharmacotherapeutics for Advanced Practice. China: Lippincott Williams & Wilkins.

Rizkya, Astrid., dkk. (2015). Nilai Diagnostik Dermatophyte Strip Test pada Pasien Tinea Unguium. Jurnal Kedokteran dan Kesehatan Volume 2, No. 1, Januari 2015: 20-24.

Robert, Christian P. (1996). The Bayesian Choice A Decision-Theoretic Motivation , Paris:Springer.

Siska, Ade Candra. (2011). Inferensi Statistik Distribusi Binomial dengan Metode Bayes Menggunakan Prior Konjugat, Semarang.

Wilks, Samuel S. (1962). Mathematical Statistics. New york:John Wiley.

Wong, Tzu-Tsung. (2008). Alternative Prior Assumptions for Improving the Performance of Naϊve Bayesian Classifiers. Data Min Knowl Disc (2009) 18:183–213.

Zhou, X.H., Nancy A. Obuchowski., Donna K. McClish. (2002). Statistical Methods in Diagnostic Medicine, USA.




DOI: http://dx.doi.org/10.29313/.v0i0.11961

Flag Counter