Sponsored By

Scientists upgrade AI-aided computational tool to better assess gut microbiome healthScientists upgrade AI-aided computational tool to better assess gut microbiome health

Researchers have revealed details of an upgrade to a digital-based tool that now achieves an 80% accuracy in assessing gut microbiome health simply by analysing stool samples.

Will Chu, Editor & Writer

October 30, 2024

3 Min Read
Scientists upgrade AI-aided computational tool to better assess gut microbiome health
© iStock/Blue Planet Studio

Developed by a team from Mayo Clinic, the Gut Microbiome Wellness Index 2 (GMWI2) uses bioinformatics and machine learning methods to detect subtle gut changes, indicating whether a person may improve or recover from a health condition.

“GMWI2 is not intended for confirming specific disease diagnoses but rather serves as an early warning system, akin to a ‘canary in a coal mine’,” said the research team, led by senior author Dr Jaeyun Sung, computational biologist at Mayo Clinic Center for Individualized Medicine's Microbiomics Program.

“Such detection could inform dietary or lifestyle modifications to prevent mild issues from escalating into severe health conditions, or prompt further diagnostic tests.

“This broad applicability could be particularly useful in clinical scenarios such as selecting faecal microbiota transplant (FMT) donors, where gut health could be taken as a reflection of overall health in conditions like rheumatoid arthritis and other autoimmune inflammatory disorders.”

GMWI2 improves on classification bias, accuracy, and general applicability

The GMWI2 is the latest version of the researchers’ original tool that, despite its potential, contained limitations such as classification bias, accuracy, and general applicability that impeded its effectiveness.

It builds upon the original model by including a wider range of data and using refined computational methods to better assess gut health and monitor gut microbiome changes.

Writing in the journal Nature Communications, the team said that GMWI2 addresses these limitations and has improved classification accuracy that extends to better distinguishing between healthy and non-healthy phenotypes.

The research team first pooled 8,069 stool samples that represented various diseases, geographic regions, and demographic groups, gathered from 54 published studies spanning 26 countries and six continents.

This was followed up with a process of validation using a new cohort of 1,140 samples, along with tests spanning a number of clinical scenarios.

These included FMT cases and individuals who had made changes in their consumption of dietary fibre or followed a course of antibiotics, to test the tool’s ability to identify changes in gut health.

GMWI2 better defines a ‘healthy’ and ‘unhealthy’ microbiome

Findings revealed that GMWI2 achieved an accuracy of 80% in distinguishing healthy (no disease) from non-healthy (diseased) individuals and surpassed a 90% accuracy for samples with higher confidence.

According to the team, the tool’s performance addresses longstanding challenges in human microbiome research, including defining what constitutes a “healthy” microbiome and identifying early indicators of potential health issues.

Together with researchers from the University of Minnesota and South Korea, they believe the tool also fills a significant gap in existing measurement tools of health and wellness.

An early warning system to prevent escalation of health conditions

“Finally, we have a standardised index to quantitatively measure how 'healthy' a person's gut microbiome is,” said Sung. 

“Our tool is not intended to diagnose specific diseases but rather to serve as a proactive health indicator. By identifying adverse changes in gut health before serious symptoms arise, the tool could potentially inform dietary or lifestyle modifications to prevent mild issues from escalating into more severe health conditions, or prompt further diagnostic testing.”

Sung added that by being able to answer whether a person's gut is healthy or trending toward a diseased state, the researchers ultimately aim to empower individuals to take proactive steps in managing their own health.

They now plan to develop the GMWI2 by expanding its dataset to include a broader range of microbiome samples from diverse populations. Further approaches include adding more advanced artificial intelligence (AI) techniques to enhance the tool’s predictive accuracy and adaptability.

In a LinkedIn post, Sudarshan Shetty, senior scientist at Danone Research and Innovation in Utrecht, the Netherlands, wrote: “GMWI2 is a step towards incorporating microbiome-centric approaches to promoting gut health.”

About the Author

Will Chu

Editor & Writer

As a seasoned Editor and Writer, Will has extensive experience in content creation for the pharmaceutical, biotechnology, and nutrition sectors in Europe and the US.

Holding a BSc in Biological Sciences and an MA in Print Journalism, he combines scientific expertise with storytelling to produce impactful content.

He has collaborated with organisations such as Informa, Roche, and GE Healthcare, addressing topics such as food sustainability, personalised healthcare, and biotechnology trends.

Will’s expertise also extends to leading editorial teams, managing digital publications, and developing multimedia strategies.

Passionate about science communication, he is particularly interested in advancing microbiome research and its impact on health and wellbeing.

SUBSCRIBE TO OUR FREE NEWSLETTERS
Get your dose of nutrition science, health ingredient innovations, and nutraceutical R&D!