Penerapan Logika Fuzzy Dan Metode Profile Matching Pada Sistem Pakar Medis Untuk Diagnosis Covid-19 Dan Penyakit Lain

Abstract

COVID-19 masih menjadi masalah di seluruh dunia. Gejala infeksi COVID-19 berupa batuk, demam, sesak napas, pilek, dan diare mirip dengan penyakit lain, sehingga menyulitkan dokter untuk membedakan infeksi COVID-19 atau penyakit lain, sehingga diperlukan diagnosis. Kesalahan diagnosis menyebabkan kesalahan dalam upaya pengobatan. Berbagai macam upaya diagnosis telah dilakukan, seperti Rapid Diagnostic Test (RDT), RT-PCR, CT-Scan Image, dan machine learning, namun masih memiliki beberapa kelemahan. RTD dan RT-PCR membutuhkan waktu yang relatif lama dan biaya yang mahal. Pasien merasakan nyeri pada hidung dan tenggorokan serta mengalami batuk dan bersin. Diagnosis menggunakan CT-Scan menghasilkan radiasi yang dapat meningkatkan risiko kanker karena pasien menerima dosis zat yang mengandung yodium yang dapat menyebabkan reaksi alergi seperti rasa logam di mulut, gatal, dan sesak napas. Tujuan dari penelitian ini adalah untuk membangun sistem pakar medis dengan menggunakan metode yang berbeda yaitu dengan menggunakan logika Fuzzy-Profile Matching untuk diagnosis COVID-19 dan penyakit lainnya berdasarkan gejala klinis pasien seperti demam, batuk kering, batuk berdahak, lesu, sesak nafas, nyeri sendi, sakit kepala, bersin, pilek, hidung tersumbat, mata berair, sakit tenggorokan, dan diare. Hasil evaluasi menggunakan 30 data dari puskesmas di Bandar Lampung, metode usulan memiliki akurasi sebesar 90%. Metode usulan yang dikembangkan mampu memberikan hasil yang cepat, murah dan tanpa efek samping. AbstractCOVID-19 is still a problem all over the world. Symptoms of COVID-19 infection in the form of cough, fever, shortness of breath, runny nose, and diarrhea are similar to other diseases, making it difficult for doctors to distinguish between COVID-19 infection or other diseases, so a diagnosis is needed. Misdiagnosis leads to errors in treatment attempts. Various kinds of diagnostic efforts have been made, such as Rapid Diagnostic Test (RDT), RT-PCR, CT-Scan Image, and machine learning, but still have some weaknesses. RTD and RT-PCR take a relatively long time and are expensive. The patient feels pain in the nose and throat and experiences coughing and sneezing. Diagnosis using CT-Scan produces radiation that can increase the risk of cancer because the patient receives doses of substances containing iodine which can cause allergic reactions such as metallic taste in the mouth, itching, and shortness of breath. The purpose of this research is to build a medical expert sistem using different methods, namely by using Fuzzy-Profile Matching logic for the diagnosis of COVID-19 and other diseases based on the patient's clinical symptoms such as fever, dry cough, cough with phlegm, lethargy, shortness of breath, pain. joints, headache, sneezing, runny nose, nasal congestion, watery eyes, sore throat, and diarrhea. The results of the evaluation used 10 data from health centers in Bandar Lampung, the proposed method had an accuracy of 90%. The proposed method developed is able to provide fast, inexpensive and no side effects.