Skip to content Skip to footer
News

Human-Artificial intelligence collaborations best for skin cancer diagnosis

24 June 2020
Decorative

Artificial intelligence (AI) improved skin cancer diagnostic accuracy when used in collaboration with human clinical checks, an international study including University of Queensland researchers has found.

The global team tested for the first time whether a ‘real world’, collaborative approach involving clinicians assisted by AI improved the accuracy of skin cancer clinical decision making.

UQ’s Professor Monika Janda  said the highest diagnostic accuracy was achieved when crowd wisdom and AI predictions were combined, suggesting human-AI and crowd-AI collaborations were preferable to individual experts or AI alone

“This is important because AI decision support has slowly started to infiltrate healthcare settings, and yet few studies have tested its performance in real world settings or how clinicians interact with it,” Professor Janda said.

“Inexperienced evaluators gained the highest benefit from AI decision support and expert evaluators confident in skin cancer diagnosis achieved modest or no benefit.

“These findings indicated a combined AI-human approach to skin cancer diagnosis may be the most relevant for clinicians in the future.”

Although AI diagnostic software has demonstrated expert level accuracy in several image-based medical studies, researchers have remained unclear on whether its use improved clinical practice.

“Our study found that good quality AI support was useful to clinicians but needed to be simple, concrete, and in accordance with a given task,” Professor Janda said.

“For clinicians of the future this means that AI-based screening and diagnosis might soon be available to support them on a daily basis.

“Implementation of any AI software needs extensive testing to understand the impact it has on clinical decision making.”

Researchers trained and tested an artificial convolutional neural network to analyse pigmented skin lesions, and compared the findings with human evaluations on three types of AI-based decision support.

UQ’s Professor H. Peter Soyer and Dr Cliff Rosendahl were also part of the study.

The paper is published in Nature Medicine. (DOI: 10.1038/s41591-020-0942-0)

Media: Professor Monika Janda, m.janda@uq.edu.au; Faculty of Medicine Communications, med.media@uq.edu.au, +61 7 3365 5133, +61 436 368 746.

Related articles

An older person wearing a light blue cardigan sits at a dining table during a shared meal, facing another seated person in the foreground. A plate of food, a cup, glasses of water and a jug of orange juice are on the table, while a third person stands in a bright kitchen area in the background. The image conveys a social mealtime setting in a home-like environment.
Analysis

From cooked breakfast to their own TV, aged care residents need to watch for extra fees

Two residents in the same aged care facility can wake up to a very different day, depending on their budget and whether they and their families can afford to pay the new “Higher Everyday Living Fees” (HELF).
19 November 2026
Three people pose inside an elevator framed by red doors. One person in a dark shirt, light trousers, cap and glasses leans against the wall with one leg stretched across the lift, while two others stand behind. A purple sign on the back wall reads “Temporary Lift Replacement Works.” The image is brightly lit and composed as a casual group photo.

UQ researchers send an unblockable message using gravity

University of Queensland researchers have used gravity to transmit a message through a brick wall using an elevator and a sensitive gravimeter.
19 November 2026

Media contact

Subscribe to UQ News

Get the latest from our newsroom.