A real-time web-based telemedicine framework based on AI and IoMT for emergency triage and initial diagnostics: the...

Area of Science:

• Medical Informatics
• Digital Health
• Artificial Intelligence in Medicine

Background:

• Global challenge in rapid medical decision-making, particularly in under-resourced settings.
• Limitations of traditional triage models with narrow algorithms or limited sensor inputs.
• Hindered timely diagnosis and treatment due to existing triage system constraints.

Purpose of the Study:

• Introduce TeleMedQuick, a web-based telemedicine system for emergency triage and initial diagnosis.
• Utilize organized clinical rules based on medical guidelines, approved by physicians.
• Enhance prehospital care and support in low-access or high-demand contexts.

Main Methods:

• Integration of Internet of Medical Things (IoMT) devices with a rule-based expert inference engine.
• Development of 76,229 clinical rules through guideline reviews and physician consultations.
• Evaluation of vital signs, symptoms, demographics, and patient history for critical conditions.

Main Results:

• System evaluated on a dataset of 750 medically annotated patients.
• Achieved a high triage accuracy of 99.1%, validated by expert review.
• Demonstrated minimized symptom overlap and enabled rapid, understandable decisions with high F1-scores.

Conclusions:

• TeleMedQuick acts as an expert system bridging IoMT sensing and clinical reasoning in telemedicine.
• Enables scalable, real-time triage and initial diagnostic support with validated transparency.
• Suitable for prehospital care, addressing challenges in low-access or high-demand environments.