Implementasi Fuzzy Inference System Tsukamoto dalam Mendiagnosis Penyakit Tuberkulosis Paru pada Tahap Awal

Abstract This study proposes the implementation of the Tsukamoto Fuzzy Inference System for early diagnosis of Pulmonary Tuberculosis. A web-based expert system was developed to analyze clinical symptoms, demonstrating the positive impact of artificial intelligence in the medical field. Symptom data were sourced from medical literature, medical records of Klinik Pratama Haji Medan Pancing, and interviews with medical practitioners at the clinic. Using 128 Tsukamoto fuzzy rules, the system predicts indications of Pulmonary Tuberculosis based on seven symptom variables. The research findings indicate that this technology implementation enhances early detection of Pulmonary Tuberculosis, offering a solution for Klinik Pratama Haji Medan Pancing and establishing an expert system accessible via a web interface, ensuring accessibility. The system's implementation also proves effective in analyzing clinical symptoms at early stages, providing a significant tool for medical practitioners. Validation results show a high accuracy rate in detecting potential diseases, supporting further potential developments in similar medical applications. The application of this system is expected to alleviate the workload of medical practitioners in expediting Pulmonary Tuberculosis management and enhancing healthcare service quality.