Characterization of cervigram image sharpness using multiple self-referenced measurements and random forest classifiers

Published Online: February 13, 2018


Cervical cancer is the fourth most common cancer among women worldwide and is especially prevalent in low resource settings due to lack of screening and treatment options. Visual inspection with acetic acid (VIA) is a widespread and cost-effective screening method for cervical pre-cancer lesions, but accuracy depends on the experience level of the health worker. Digital cervicography, capturing images of the cervix, enables review by an off-site expert or potentially a machine learning algorithm. These reviews require images of sufficient quality. However, image quality varies greatly across users. A novel algorithm was developed to evaluate the sharpness of images captured with the MobileODT’s digital cervicography device (EVA System), in order to, eventually provide feedback to the health worker. The key challenges are that the algorithm evaluates only a single image of each cervix, it needs to be robust to the variability in cervix images and fast enough to run in real time on a mobile device, and the machine learning model needs to be small enough to fit on a mobile device’s memory, train on a small imbalanced dataset and run in real-time. In this paper, the focus scores of a preprocessed image and a Gaussian-blurred version of the image are calculated using established methods and used as features. A feature selection metric is proposed to select the top features which were then used in a random forest classifier to produce the final focus score. The resulting model, based on nine calculated focus scores, achieved significantly better accuracy than any single focus measure when tested on a holdout set of images. The area under the receiver operating characteristics curve was 0.9459.

Read Publication


Mayoore Jaiswal, Matt Horning, Liming Hu, Yau Ben-Or, Cary Champlin, Benjamin Wilson, David Levitz

More Buzz From IV

Intellectual Ventures Develops the First Offline Virtual Malaria Microscopy Training Course

The World Health Organization-approved eLearning tool can be used regardless of Internet access.

Read More
Why We’re Reinventing the Door and Window

A new study by Global Good and its partners explores how reinventing two common household items—the window and door—could save lives by keeping pesky mosquitoes out of African homes.

Read More
How New Cervical Cancer Screening Technology Could Save the Lives of Hundreds of Thousands of Women
On World Cancer Day, Global Good and QuantuMDx have announced a new partnership to develop a rapid, low-cost and mobile diagnostic test that could make a tremendous difference in the global fight against cervical cancer. Read More
We use cookies on this website to enhance your browser experience and to analyze your traffic. To learn more about cookies and how we use them view our cookie policy. By continuing to use our website you consent to the use of cookies.