Diagnosis and treatment with artificial intelligence: A new approach in healthcare

Artificial intelligence technologies continue to offer groundbreaking innovations in the healthcare sector. In China, a system developed by the Institute for AI Industry Research at Tsinghua University, called the AI-Powered Virtual Hospital, hosts 42 AI doctors working across 21 different specialties.

This virtual hospital facilitates interaction between doctors, nurses, and patients through intelligent agents powered by large language models (LLMs). The system is built on an AI architecture that covers all processes from the onset of a disease to treatment.

Researchers state that this virtual hospital can diagnose 10 thousand virtual patients in just a few days. This is equivalent to the two-year workload of a doctor who sees an average of 100 patients per week. Furthermore, it is emphasized that the AI doctors achieved a 93.06 percent accuracy rate on the MedQA dataset.

Liu Yan, the lead researcher from Tsinghua University, stated that this technology will create a significant transformation in diagnosis and treatment processes and will provide great benefits to both healthcare professionals and the public. However, it is noted that for this technology to be implemented, it must comply with national health regulations, and the most effective ways for AI and human collaboration must be explored.

The research team continues its work to increase the variety of diseases that AI doctors can diagnose and to improve their competencies. Liu Yan indicated that the system is nearly ready for real-world application.

LLM, or 'Large Language Model', refers to AI systems trained on vast amounts of text that can understand and respond to human language. These models have the ability to process natural language, write, understand questions, and summarize texts. For example, ChatGPT is an LLM and can be used in various fields.