All you need to know about Google's new AI chatbot AMIE

Google has recently announced an extensive development in the artificial intelligence (AI) stream. The tech giant has introduced its new AI chatbot Articulate Medical Intelligence Explorer (AMIE), which will bring a revolutionary change in the world of medical science.

The AMIE chatbot is based on real-world datasets comprising medical reasoning, medical summarisation, and real-world clinical conversations. It is designed to interact with the patients, which would reflect quality in real-world clinical consultations with doctors.

The tech giant's latest launch in the AI world is a 'novel' based on a simulated diagnostic dialogue environment with automated feedback mechanisms to enrich and accelerate its learning process. The researchers at Google have also introduced an inference time chain-of-reasoning strategy to improve AMIE’s diagnostic accuracy and conversation quality.

The AMIE chatbot was tested by imitating consultations with both board-certified primary care physicians (PCPs) and trained actors (trained carefully to mimic a patient with a particular medical condition). The research included the study of 149 case scenarios from Objective Structured Clinical Examination (OSCE) providers in Canada, the UK and India in a diverse range of specialties and diseases, giving an optimistic approach towards the development.

The research found that AMIE had greater diagnostic accuracy and superior performance from the perspective of both specialist physicians, and patient actors. The comparison was drawn on the aspects like history-taking, diagnostic accuracy, clinical management, clinical communication skills, relationship fostering and empathy.

The researchers also call for further research before AMIE can be implemented in real-world settings. "Experimental performance under real-world constraints and dedicated exploration of health equity and fairness, privacy, robustness, and many more, to ensure the safety and reliability of the technology," the blogpost by Google Research added.