Article Per Year (5 Year)
p-Index From 2021 - 2026
0.444
P-Index
This Author published in this journals
All Journal Innotech Innobiz: Jurnal Ilmu Manajemen, Bisnis dan Keuangan
Michael Sitorus
Universitas Siber Indonesia
Computer Science & IT

Published : 2 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 1 Documents
Search
Journal : innotech

 1 
Agnes Medika : Sistem Pakar Diagnosis Dan Rekomendasi Obat Berbasis Artificial Intelligence Melalui AI Agent Medis Michael Sitorus; Serenade Theodora Dwica
Innotech: Jurnal Ilmu Komputer, Sistem Informasi dan Teknologi Informasi Vol 3 No 2 (2026): Innotech Issue Juli 2026
Publisher : Universitas Siber Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar

Abstract

The low ratio of physicians to patients has contributed to delays in medical treatment and the delivery of health information. This study introduces Agnes Medika, a web-based AI medical agent developed using the Agnes.ai platform to facilitate independent disease symptom identification and provide recommendations for over-the-counter medications. By utilizing a structured medical knowledge base that adopts the model of leading telehealth platforms such as Halodoc, Agnes Medika employs Natural Language Processing (NLP) to analyze patient symptoms and conducts an interactive rule-based screening process. The system architecture was designed using flowcharts and use case diagrams to develop a clean and responsive conversational (chat) interface. Experimental results demonstrate that the AI agent is capable of providing accurate preliminary triage outcomes while minimizing the risk of data hallucination through a grounding mechanism based on a trusted medical knowledge base. Furthermore, the system generates an informative digital medical summary accompanied by an appropriate clinical disclaimer. Agnes Medika functions as a reliable health Decision Support System (DSS) for self-assessment prior to professional clinical examination. The evaluation results indicate that the system provides structured information, including patient identity and a visual representation of disease indications, with an accuracy rate of 92% in identifying Common Cold cases.
Co-Authors Ayi Wahid Dhiya Nada Adinda Serenade Theodora Dwica