New AI Tool Can Identify Cancer More Efficiently, Better Than Current Methods: Study

According to World Health Organization, cancer is a leading cause of death worldwide and results in about 10 million deaths annually.

Advertisement
Read Time: 3 mins
Researchers say it can speed up the diagnosis and also fast-track treatment (Representational pic)

An artificial intelligence (AI) model has been created by scientists, doctors and researchers that can identify cancer accurately, Guardian reported. Researchers say it can speed up the diagnosis of the disease and also fast-track treatment given to the patients.

The AI tool, which ''performs more efficiently and effectively than current methods'' has been designed by experts at the Royal Marsden NHS foundation trust, the Institute of Cancer Research, London, and Imperial College London.  The tool can identify if abnormal growths seen on a person's CT scans are cancerous.

The research's findings were published in the ebiomedicine journal of the Lancet. 

''In the future, we hope it will improve early detection and potentially make cancer treatment more successful by highlighting high-risk patients and fast-tracking them to earlier intervention,” said Dr Benjamin Hunter, a clinical research fellow at Imperial and a clinical oncology registrar who works at the Royal Marsden.

For the research, the team used CT scans of about 500 patients with large lung nodules to develop an AI algorithm using radiomics. The technique can extract important information from medical images not easily spotted by the human eye. 

The study then used a measure called area under the curve (AUC) to see how effective the model was at anticipating cancer.  If a model gets an AUC of 1, it indicates a perfect model, while 0.5 would be expected if the model was randomly guessing. 

The results showed the AI model could spot each nodule's risk of cancer with an AUC of 0.87. The performance improved on the Brock score, a test used in the clinic, which scored 0.67. 

''According to these initial results, our model appears to identify cancerous large lung nodules accurately. Next, we plan to test the technology on patients with large lung nodules in the clinic to see if it can accurately predict their risk of lung cancer,'' Mr Hunter added.

Libra study's chief investigator, Dr Richard Lee said, ''Through this work, we hope to push boundaries to speed up the detection of the disease using innovative technologies such as AI.''

Advertisement

According to World Health Organization, cancer is a leading cause of death worldwide and results in about 10 million deaths annually, or nearly one in six deaths.

Featured Video Of The Day
Amazon, Starbucks Workers Go On Massive Strike In US During Christmas Season
Topics mentioned in this article