Diagnosis, decision making and treatment outcome prediction are key areas where AI may have the most impact

Words: Dr Paul O’ Dwyer  BDS MSc (Healthcare Mgmt)

At the American Dental Association convention in Orlando, Florida, last October, there were two recurring themes: workforce planning and artificial intelligence (AI). It seems every facet of every industry and profession is seeing the effects of AI. It is timely to reflect on what this could mean for dentistry – and particularly dental practice.

Firstly, the term AI has been around for a long time. According to a recent paper by Ding et al (2023)1, one of the better definitions for AI includes “… the theory and development of computer systems able to perform tasks normally requiring human intelligence…”  Since the early 1950s, AI has been making headway.

The term AI is usually attributed to a workshop from 1955 led by John McCarthy, Marvin Minsky et al. Over the following decades, and particularly in the 1980s, two distinct pathways were established for AI development: machine learning (ML) and expert systems. Machine Learning allows computers to learn by experience, whereas Expert Systems need human experts to input all possible situations and solutions in advance.

Famous AI examples include the chess-playing expert system called Deep Blue which shocked the world when it defeated Gary Kasparov in 1997. Then some 20 years later, in 2017, Google’s own machine AlphaGo, a deep learning (DL) programme, defeated the No 1 ranked Go player Jie Ke in a Go match. If you don’t know the game of Go – it’s a terrific strategy game developed in China more than 2,000 years ago.

Then in 2022, OpenAI launched ChatGPT (Chat Generative Pre-trained Transformer) as a text-generation model that can generate human like responses based on text input.

But will ChatGPT eventually create ChatGDP? That is the question!

Among all the potential AI applications in dentistry, the key areas where most experts agree that it will have most impact are diagnosis, decision making and treatment outcome prediction. It’s helpful perhaps to look at some areas of dentistry where these impacts can be harnessed and ideally lead to better quality patient care.

Operative dentistry: Traditionally dental caries (as we know) is diagnosed through a combination of visual, tactile and radiographic examination. Early lesion detection Is often challenging with deep fissures, tight interproximal contacts and with secondary lesions present. It is suggested that research into accurate radiographic diagnosis with further follow up could add vital information to collective learning here and –after many hundreds of thousands of data collected – would help to accurately diagnose early caries.

Periodontology: Similarly, there is a suggestion that an algorithm can be developed based on extent data which could help to better diagnose early periodontal disease.

Orthodontics: AI appears to be the ideal tool here for treatment outcome/prediction with utilisation in planning and simulating changes. These smart algorithms combined with the latest CBCT imaging and 3D modelling hold out much hope for powerful tools in the orthodontic world.

Oral and maxillofacial surgery: The many applications being studied include one piece of software which can help distinguish between ameloblastoma and keratocystic odontogenic tumour. The two oral tumours can have similar radiographic features. A recent study using computer generated images (with biopsy results confirming diagnosis) showed an encouraging 83 per cent AI success rate in diagnosis – with a diagnostic time of under one minute.

Prosthodontics: CAD/CAM has already changed the way we now deliver crowns and bridges. The future for this and other technologies is in its infancy.

Some say that this is beginning of a new industrial age, with many time and labour-saving technologies such as AI changing the face of many industries, much like the assembly line for Ford or the so called “modern conveniences” that we have learned to accept as part of everyday life.

It is certain that whatever AI may bring to dentistry, the physical delivery of treatment will always remain – in some shape or form. The incorporation of an AI module in the training of tomorrow’s dentists or recently graduated dentists, might be one suggestion to better keep ahead of this fast changing field.

Reference

1. Ding H, Wu J, Zhao W, Matinlinna JP, Burrow MF and Tsoi JKH (2023) Artificial intelligence in dentistry—A review. Front. Dent. Med 4:1085251. doi: 10.3389/fdmed.2023.1085251

Tags: AI, American, Association, Dental, Dentistry