How accurate are artificial intelligence systems at identifying tooth decay?

Artificial intelligence systems are able to identify tooth decay more accurately than human dentists, new research suggestsArtificial intelligence systems are able to identify tooth decay more accurately than human dentists, new research suggests.

The study, carried out by Pearl, compares the diagnostic abilities of three human dentists to that of an artificial intelligence (AI) diagnostic system.

It looked at the agreement between the human and AI diagnoses after they analysed more than 8,700 bitewing and periapical radiographs.

Additionally, the research revealed there was a lack of diagnostic agreement among the three human dentists.

Diagnostic discrepancies

They displayed a 79% unanimous agreement when it came to the absence of decay. But this dropped to 4.2% – around 370 X-rays – when it came to presence of decay.

And in nearly one in five instances, when two dentists identified decay in an X-ray, the third did not.

‘Our intention in producing this study was simply to demonstrate the efficacy of computer vision machine learning diagnostics in dental radiology,’ said Pearl CEO Ophir Tanz.

‘Diagnostic inconsistency may be a natural byproduct of human professionals making case-by-case judgements. But a scale inconsistency in the standard of care could result in suboptimal patient treatment. With broader implications for population health at large.

‘Fortunately, the study’s findings also point to a solution; the machine’s diagnostic performance shows that AI is capable of infusing consistency into the bedrock of dental care.’

Further research necessary

The study’s authors acknowledged the limitations of the research, adding that further work is needed. This could look into more precise conclusions, such as the level of decay. It could also use a larger number of human dentists.

Dr Sanjay Mallya is the chair and associate professor in oral and maxillofacial radiology in the division of diagnostic and surgical sciences at the UCLA School of Dentistry.

‘More studies like this are in order,’ he said.

‘It would be useful, for instance, to drill deeper to see if the discrepancy evident in this study persists for caries across varying degrees of severity.’

Follow on Instagram to keep up with all the latest dental news and trends.

Get the most out of your membership by subscribing to Dentistry CPD
  • Access 600+ hours of verified CPD courses
  • Includes all GDC recommended topics
  • Powerful CPD tracking tools included
Register for webinar
Add to calendar