Google Introduces MedLM AI Models for Healthcare to Streamline Clinical Processes

Positioned as a tool to assist clinicians and researchers in conducting complex studies and streamlining doctor-patient interactions, MedLM signals Google's commitment to carving a niche in the fiercely competitive healthtech landscape.

In a bid to tap into the booming healthcare AI market, Google has unveiled its latest-MedLM suite, a collection of healthcare-specific AI models.

Positioned as a tool to assist clinicians and researchers in conducting complex studies and streamlining doctor-patient interactions, MedLM signals Google's commitment to carving a niche in the fiercely competitive healthtech landscape.

Diversified AI Models for Tailored Solutions

Google's MedLM suite comprises both large and medium-sized AI models, both built on Med-PaLM 2, a language model trained on medical data. Aashima Gupta, Google Cloud’s global director of healthcare strategy and solutions, explained the rationale behind the suite, stating that different medically tuned AI models excel at specific tasks.

The larger MedLM model proves more adept at intricate, data-intensive tasks, such as comprehensive healthcare studies, while the medium-sized model is optimised for real-time functions including summarising doctor-patient interactions.

HCA Healthcare Harnesses MedLM for Efficiency Gains

Among the early adopters of Google's technology is HCA Healthcare, one of the largest health systems in the US.

Dr Michael Schlosser, HCA's Senior VP of Care Transformation and Innovation, speaking on the impact of MedLM, said its deployment in emergency rooms for automating physician-patient interaction documentation is highly crucial. He further stressed on its potential to streamline healthcare workflows and alleviate administrative burdens with plans for a nurse handoff tool in development.

Challenges in Implementation and Real-World Application

However, the road to integrating AI into healthcare is not without hurdles. Schlosser highlighted challenges faced by HCA, including the risk of incorrect information output by AI models and the need to establish best practices for minimising errors. He underlined the importance of a cautious approach to implementation to avoid any potential risks to patient care.

Looking ahead, Google aims to introduce healthcare specific versions of Gemini, its newest and most capable AI model, to MedLM in the future. While Gemini has generated excitement, Google acknowledges that it is still in the early stages and requires careful testing and evaluation in controlled healthcare settings before broader implementation.

Industry Responses to MedLM

Other companies, including BenchSci and Deloitte, have also tested and integrated MedLM into their operations. BenchSci's CEO, Liran Belenzon, highlighted the collaborative effort with Google to validate and improve the model, emphasising its role in accelerating specific scientific efforts.

Deloitte, on the other hand, is leveraging MedLM to assist health systems and health plans in providing accurate and efficient responses to members' inquiries about accessing care.