Artificial Intelligence Enhances Cervical Cancer Screenings

July 07, 2020 -A computer algorithm that leverages artificial intelligence (AI) to automate dual-stain (DS) evaluation improved the accuracy and efficiency of cervical cancer screening compared with Pap cytology, according to a recent study.

Researchers found that all DS strategies achieved equal or better sensitivity for detection of a clinical end point compared to Pap cytology while reducing unnecessary colposcopic referrals.

Specifically, automated DS evaluations reduced overall referral to colposcopy by one-third for the primary automated cutoff of two cells, a 41.9 percent reduction for automated DS evaluations versus a 60.1 percent reduction for cytology.

Automated DS evaluations at a cutoff of two or more cells also had the most favorable ratio of colposcopies per clinical end point detected compared to the least favorable ratio for the current standard, Pap cytology, at 6.8 versus 9.9, respectively.

The algorithm was developed and the study was conducted by investigators at the National Cancer Institute (NCI), part of NIH, in collaboration with Niels Grabe, PhD, and Bernd Lahrmann, PhD, of the Steinbeis Transfer Center for Medical Systems Biology.

The study aimed to uncover if a fully automated dual-stain test could match or exceed the performance of the manual approach. Researchers developed an imaging platform that could determine if any cervical cells were stained for both p16 and Ki-67 after being trained with deep learning, NIH said.

The Biopsy Study included 4,253 women aged 18 years or older who were referred to colposcopy at the University of Oklahoma Health Sciences Center between 2009 and 2011.

Automated evaluation of DS slides dramatically increases the efficiency of cervical cancer screening by substantially reducing unnecessary colposcopies compared with current standards and similarly achieves excellent performance in a simulated fully vaccinated population. Thus, CYTOREADER exceeds human diagnostic accuracy and serves as an example of AI achieving improvements beyond the automation of a human standard, researchers said in the study.

Our results demonstrate how automation and machine learning can transform cervical cancer screening that is currently undergoing major changes. HPV testing for cervical cancer screening is an objective and reliable approach directly linked to the carcinogenic process.

An NIH press release touched on the importance of the study findings.

The biomedical research agency said that in recent years, clinicians have hoped to take advantage of advances in digital imaging and machine learning to improve cervical cancer screening.

But the challenge providers have encountered is identifying which women with positive HPV test results are more likely to have precancerous changes in their cervical cells and, therefore, should have a colonoscopy to examine the cervix and take samples for biopsy or who need immediate treatment.

The current approaches to this care are not ideal, NIH said, because Pap cytology tests are time consuming, not very sensitive, and prone to false-positive findings.

The study showed that the automated tests surpassed the performance of the current standard, Pap cytology, which reduces the number of false positive results and reduces referral to unnecessary colposcopy procedures.

Were excited to show we have a fully automated approach to cervical cancer screening as a follow-up to a positive HPV test that outperformed the standard method in our study, Nicolas Wentzensen, MD, PhD, of NCIs Division of Cancer Epidemiology and Genetics, who led the study, said in the press release.

Based on our results, it could increase the efficiency of cervical cancer screening by finding more precancers and reducing false positives, which has the potential to eliminate a substantial number of unnecessary procedures among HPV-positive women.

See more here:
Artificial Intelligence Enhances Cervical Cancer Screenings