AI outperforms humans in identifying cervical cancer

A computer algorithm using artificial intelligence (AI) has accurately screened and identified cervical cancer better than trained experts, according to research from National Cancer Institute, an organisation that is part of the US Department of Health and Human Services.

The AI solution called automated visual evaluation has the potential to revolutionise cervical cancer screening, particularly in low-resource settings, NCI said in a press statement.

The researchers used comprehensive datasets to ‘train’ a machine learning algorithm to recognise patterns in complex visual inputs, such as medical images. The approach was created collaboratively by investigators at the NCI and Global Good, and the findings were confirmed independently by experts at the National Library of Medicine (NLM), the statement added.

"A deep learning algorithm can use images collected during routine cervical cancer screening to identify precancerous changes that, if left untreated, may develop into cancer. In fact, the computer analysis of the images was better at identifying precancer than a human expert reviewer of pap tests under the microscope (cytology)," said Mark Schiffman, senior author of the study and master of public health at NCI's Division of Cancer Epidemiology and Genetics.

Healthcare workers in low-resource settings currently use a screening method called visual inspection with acetic acid (VIA). In this approach, a health worker applies dilute acetic acid to the cervix and inspects the cervix with the naked eye, looking for ‘aceto whitening’, which indicates possible disease.

According to NCI, because of its convenience and low cost, VIA is widely used where more advanced screening methods are not available. However, this method is known to be inaccurate and needs improvement, the study said.

To create the algorithm, the research team used more than 60,000 cervical images from an NCI archive of photos collected during a cervical cancer screening study that was carried out in Costa Rica in the 1990s. More than 9,400 women participated in that population study, with follow up that lasted up to 18 years. The photos were digitised and then used to train a deep learning algorithm so that it could distinguish cervical conditions requiring treatment from those not requiring treatment.

Under the new method, health workers can use a cell phone or similar camera device for cervical screening and treatment during a single visit. In addition, this approach can be performed with minimal training, making it ideal for countries with limited healthcare resources, where cervical cancer is a leading cause of illness and death among women, the study noted.

"With advances in HPV vaccination, emerging HPV detection technologies, and improvements in treatment, it is conceivable that cervical cancer could be brought under control, even in low-resource settings," said Maurizio Vecchione, executive vice president of Global Good. 

AI applications are being used extensively in healthcare and cancer in particular. Recently, a computer program developed by New York University (NYU) School of Medicine was able to analyse images of lung tumours, specify cancer types, and even identify altered genes driving abnormal cell growth.

In another study, a team of researchers from the University of Central Florida had recently developed an AI system to detect cancerous tumours normally missed by radiologists.