Monash Health librarians putting AI to the test for clinical research

31 December 2025
Library
5 min read

Monash Health librarians are leading the way in incorporating artificial intelligence (AI) into our clinician-researchers’ workflows.

As the artificial intelligence (AI) era booms, Monash Health librarians have taken on the role of expert local guides in a vast and complex information landscape.

They are road-testing the tools, determining which applications are safe, and teaching clinicians and researchers how to navigate AI quickly and confidently without getting lost.

At first glance, the library at Monash Medical Centre looks familiar: shelves of clinical handbooks, a row of 3D anatomy models lying on one of the back shelves, and quiet spaces for staff to read, relax, and reflect.

But behind the scenes, librarians here and across all five libraries have spent the past two years doing something that’s changing how doctors and researchers across Monash Health work: teaching them how to safely and responsibly use AI in their searches for information and evidence.

Entrance to Monash Health Library with bookshelves and seating visible in the background — Entrance to the library at Monash Medical Centre, Clayton.

‘The library has come a long way from recording bookings for the “internet computer” in an exercise book,’ said Head of Library Services Alice Anderson.

‘Today, we’re helping clinicians analyse vast amounts of information in minutes and then asking the hard questions about the source of that information and whether they should trust what they’re seeing.’

Taking the lead on AI

Once regarded mainly as the custodians of journals and textbooks, the library’s 8-person team now finds itself at the forefront of Monash Health’s AI adoption. They pilot new AI-powered search platforms, test tools that summarise research papers for time-poor clinicians and run workshops on everything from building search prompts to spotting hallucinated references.

Wall plaque with hands holding a book below a sign quoting patrons about the library as a peaceful, comfortable place to learn. — AI Search Tool Matrix Award by VALA

Manager of Access and Discovery Cassandra Gorton said the team began testing AI tools back in 2022, before ChatGPT became a household name.

‘Each of us was rostered two AI tools in a month to use them, demonstrate how they can be integrated in user workflows, and decide whether we would recommend them. We had a lot of internal learning, but we had to figure out how to make it useful to others,’ she said.

As a result, the library developed a list of recommended AI-powered search tools and citation mapping tools.

The team also offers tailored research consultation sessions to answer custom research questions, including guidelines on how to best use AI tools to complement their research.

Modern office space with desks, computers, red chairs, anatomical models, and large windows overlooking trees. — Desktops and tables by a large window is a popular spot among library patrons.

From manual searches to AI-summariser

For years, doctors have leaned on clinical decision support platforms and subscription databases to find answers to clinical questions. But the explosion of generative AI tools capable of summarising information has created both new opportunities and new risks.

Anatomical torso model displaying internal organs on a stand near informational posters in a meeting room. — A human anatomy model for educational purposes.

‘Clinicians don’t have the luxury of spending three hours on a literature review before a ward round,’ said Gorton.

‘If a tool can scan hundreds of articles and give them a starting point in 30 seconds, that’s incredibly attractive. But only if the output actually reflects the evidence. Importantly, the tool can only summarise evidence for a clinician – it doesn’t replace clinical judgement or actual clinical decision making.’

Anderson said that most AI search tools rely on freely available citation repositories, which can be incomplete, inconsistent, and contaminated by non-peer-reviewed content.

Another limitation is access and relevance.

Library shelf displaying anatomy books and anatomical models, including a model of the human larynx and brain. — Anatomy models available for loan.

‘Many AI tools cannot search subscription-based databases such as Medline or Embase, which are essential for comprehensive literature reviews. For example, using Copilot to build a search strategy with subject headings for databases it cannot access is inefficient and environmentally wasteful – and it will never match the quality of a librarian-led search,’ said Anderson.

Many of them are also not aligned with best practice guidelines for health research set by authorities, she added.

As a solution, her team is working with publisher partners to trial AI features embedded in trusted platforms that generate summaries for clinicians from known sources of evidence.

Skeleton model standing next to a digital library information kiosk in front of bookshelves. — A touch screen directory for library patrons.

In parallel, her team is also looking to develop custom prompts for Microsoft Copilot, an AI-powered assistant used across the organisation, among other initiatives.

The team’s innovative work has been recognised through the 2025 VALA Award for its AI search tools matrix and education program.

VALA – Libraries, Technology and the Future Inc. is an independent Australian-based not-for-profit organisation.

Beyond research: looking after people

The library’s innovation isn’t limited to researchers.

Display of colorful completed coloring pages above a table with coloring supplies and unfinished pages labeled Mindful Colouring. — Mindful colouring corner for patrons.

In partnership with council libraries in Clayton, Dandenong, and Berwick, the Monash Health Library now offers patients and families digital memberships to access eBooks, audiobooks, as well as films, TV shows, and digital magazines.

Anderson said it’s a simple way to support mental well-being and ease boredom during hospital stays.

The team also works on a bigger stage. Anderson and her team are working with the World Health Organization library team in the Philippines to set up a regional librarians’ network to improve access to reliable health information across the Western Pacific.

In October 2025, the Monash Health Research Repository, created by Anderson’s team, was used as a benchmark by a delegation from the Philippines as it works to develop its own country-wide research repository.

The Monash Health Research Repository collects publications by Monash Health researchers and provides access to a global audience.

Library study area with individual wooden desks and red chairs, a few people working at computers. — Library study area with individual desks and chairs.

Director of Research and Innovation Associate Professor Tam Nguyen, who oversees the library service, noted that the library is an integral part of the Research and Innovation Enabling Plan in the Monash Health Strategy 2026-2031.

‘By connecting people to evidence, expertise and emerging tools, the library helps turn research ambition into real-world impact for patients and staff.

‘It ensures clinicians and researchers can find, assess and apply the best available knowledge quickly, safely, and with confidence,’ said A/Prof Nguyen.

A quiet engine of innovation

From handwritten notebooks and CD-ROM towers to AI-powered tools and global networks, the team at Monash Health Library has always asked the same question:

‘What’s next, and how do we make it useful for our people and patients?’

The technology has changed. The pace has increased. But the mission remains the same: to ensure that the right information, and now, the right tools, are always within reach.