Google Research and DeepMind Launch an AI-Based Healthcare Language Model

It is said that it can provide safer answers.

Google Research and DeepMind have launched an AI-based healthcare platform called MedPaLM, as reported by Interesting Engineering.

PORTUGAL-TECH-WEB SUMMIT
A Web Summit attendee passes by the Google logo during the Europe's largest tech conference, the Web Summit, in Lisbon on November 2, 2022. - The Web Summit will run until November 4, 2022. by PATRICIA DE MELO MOREIRA/AFP via Getty Images

MedPaLM aims to generate safe and helpful answers related to the medical field. It is a combination of HealthSearchQA and six existing open-question answering datasets that cover professional medical exams, research, and consumer queries.

What MedPaLM Does

Medical professionals, and even non-professionals, can use MedPaLM. The platform can address multiple-choice questions via the delivery of various datasets.

These datasets come from NedQA, MedMCQA, PubMedQA, LiveQA, MedicationQA, and MMLU. However, a new dataset was adder, the HealthSearchQA, which aims to improve MultiMedQA.

With the new dataset, it contains 3,375 frequently asked consumer questions gathered through seed medical diagnoses and their related symptoms. It was developed on PaLM.

Currently, MedPaLM claims to work well compared to Flan-PaLM. However, the AI still needs to outperform a medical expert's judgment. On the other hand, 92.6 percent of MedPaLM responses were on par with clinician-generated answers.

Also Read: AI in Healthcare: What it Can do and What it Can't

Safer Replies from MedPaLM

Surprisingly, only 61.9 percent of long-form Flan-PaLM answers are said to be accurate, while 5.8 of MedPaLM answers were said to have the potential to contribute to negative consequences compared to the 6.5 percent of answers generated by medical professionals and 29.7 percent from Flan-PaLM. Overall, it only goes to show that MedPaLM delivers safer answers.

It is a big deal to have an AI that can provide safer answers since even a little mistake could lead to big consequences, particularly in the field of healthcare. For instance, a wrong dose to a person undergoing chemotherapy may result in death.

Answering more complex medical questions may be beyond MedPaLM's capability. The platform is still a work in progress and researchers are constantly looking for improvements that can make it better.

AI in Healthcare

AI, particularly machine learning (ML) and artificial intelligence (AI), can provide better answers in the medical field. AI, or artificial intelligence, is a branch of computer science that deals with enabling machines to perform tasks that are normally done by humans.

The AI-backed platforms are still in its early stages, but researchers are making sure it will thrive.

AI has already been used to improve the quality of healthcare and most professionals, including medical professionals, are welcoming the change. AI is also expected to aid in accelerating the development of medical apps.

With AI, medical professionals can focus on actual medical care and not just clerical work. However, AI is not yet perfect as it may still produce incorrect answers. Nonetheless, it is undeniable that AI has been a great help in the medical field.

Related Article: AI in Healthcare: The Key To The Success In The Medical Industry.

April Fowell
April Fowell
ⓒ 2024 TECHTIMES.com All rights reserved. Do not reproduce without permission.
Join the Discussion
Real Time Analytics