Media Room

Cervical cancer is one of the leading causes of cancer mortality and morbidity across the globe According to recent statistics, 1 woman in India dies every 7 minutes because of cervical cancer. In some low resource settings, cervical cancer is the leading female malignancy, with lifetime cumulative incidence exceeding 5%. This is expected to increase in the decades ahead. Human papillomavirus vaccination (HPV) and cervical screening are mainstays of prevention programs used widely as a preventive measure which are successful in high-end resource settings. Automated visual evaluation algorithm that recognizes cervical cancer through images is a practical solution. Let’s find out more about its accuracy

Automated Visual Evaluation

Automated Visual Evaluation is a well-integrated program that combines a digital camera and visual algorithms to detect cervical lesions in the affected area. These lesions are indicative of cancer which is confirmed by the positive result of the test, while a negative result shows that cervix is not at increased risk of cancer. However, the accuracy seems to be a point of consideration. According to a study of National Cancer Institute, AVE was found to have a sensitivity of 97.7% and a specificity of 85% in women of reproductive age. It is evident from the fact that the visual algorithms are accurate, which can outperform both Pap cytology and examination by expert colposcopists.

Development of AVE Algorithm

AVE is developed by the National Institute of Health and Global Good which can analyze digital images of a woman’s cervix and accurately identify precancerous changes. To develop the method, researchers used comprehensive datasets to “train” a deep or machine learning algorithm to recognize patterns in complex visual inputs. The approach was developed collectively by investigators at the National Cancer Institute (NCI) and Global Good a fund at Intellectual Ventures. Experts at the National Library of Medicine (NLM) confirmed the findings. The results appeared in the Journal of the National Cancer Institute on January 10, 2019. NCI and NLM both are the parts of NIH.

“Our findings show that a deep learning algorithm can use images collected during routine cervical cancer screening, to identify precancerous changes if left untreated, may develop into cancer,” said Mark Schiffman, M.D., M.P.H., of NCI’s Division of Cancer Epidemiology and Genetics, and senior author of the study. “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).”

The research team used more than 60,000 cervical images from an NCI archive of photos collected during a cervical cancer screening study in Costa Rica in the 1990s. More than 9,400 women participated, and the follow up lasted up to 18 years. Because of the prospective nature of the study, the researchers gained insights as to which cervical changes became precancers and which did not. The photos were digitized and then used to train a deep learning algorithm that helped to understand cervical conditions requiring treatment.

Overall, the algorithm performed better than all standard screening tests during the Costa Rica study. The automated visual evaluation identified precancer with greater accuracy (AUC=0.91) than a human expert review (AUC=0.69) or conventional cytology (AUC=0.71). An AUC of 0.5 indicates the sensitivity and specificity, whereas an AUC of 1.0 represents a test with perfect accuracy in identifying disease.

“When this algorithm is combined 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.

AVE Technology in Today’s Clinical Setting

Health care workers currently use a screening method called visual inspection with acetic acid (VIA). Here a dilute acetic acid is applied to the cervix which is inspected with the naked eye. The “Aceto whitening” indicates possible disease. Because of its convenience and low cost, VIA is widely used in low resource settings.

AVE can replace the frequently inaccurate Pap tests as a first-line screening. Patients would benefit from immediate results at the point of care in a single visit. This could provide rapid risk-classification and immediately refer patients to secondary screening, to treatment, or routine screening.

Currently, all HPV 16 or 18 positive patients are sent to secondary screening (colposcopy), it will help in reducing the over-referral and proper resource allocation. Moreover, the user-friendly interface and artificial intelligence on the EVA device will allow even non-expert clinicians to provide the AVE test. Health workers can use an EVA system for automated visual evaluation in a single visit. It can be performed with minimal training, making it ideal for nations with limited health care resources.

What’s Next?

The researchers plan to further train the algorithm on a sample of representative images of cervical precancers and normal cervical tissue from women. This approach would work in communities around the world while using a variety of EVA systems and other imaging options. There could be subtle variations in the appearance of the cervix among women in different geographic regions. With the objective to create the best possible algorithm for common, open use it is best to use EVA systems.

Want to know more? Contact us at GenWorks.
For further updates stay tuned.