FOR IMMEDIATE RELEASE:
[San Francisco, November 17, 2016]– MobileODT, a connected health company enabling health providers around the world to diagnose and address disease using its Enhanced Visual Assessment (EVA) System, is collaborating with Intel and Kaggle to harness the power of artificial intelligence to assist health workers to better manage the treatment of cervical cancer. The competition, hosted on Kaggle, will enable data scientists around the world to access thousands of cervical images to build algorithms that may assist in determining the eligibility criteria for treating pre-cancerous lesions with cryoablation. This project, beginning December 2016, builds on the work MobileODT has already begun in partnership with Global Good to harness the power of machine learning to improve cervical cancer screening outcomes.
Cervical cancer is the number one cause of cancer death for women in low resource settings, worldwide – a tragedy because cervical cancer takes 20 years to develop, and can be treated for less than $28 and in a single visit if caught in the first five years.
MobileODT currently holds a bank of tens of thousands of cervical images taken with their EVA System by health providers in over 21 countries, which will serve as a data set to be used by competition participants in their coding. This data set will allow for the building of algorithms that use image recognition to screen for precancerous lesions and potential risk for cervical cancer.
“Kaggle is excited to work with Intel and MobileODT on this challenge. It’s a fabulous opportunity for our community to both push forward the cutting edge of image recognition while also solving an important public health problem.” Anthony Goldbloom, CEO of Kaggle said.
The competition will consist of two stages: In stage one, reviewers will look at 4,000 patient cases classified under the guidance of Professor Albert Singer, a world leader in Colposcopy education and research, by cervical state eligibility for treatment. In stage two, reviewers will classify images of the cervix as low, medium or high risk for cervical pre-cancer. Kaggle’s competition platform will provide the framework for this project and consultation for the participants.
“The opportunity to collaborate with Intel and Kaggle and to open part of our data to developers looking to change the world through machine learning is extraordinary,” Ariel Beery, CEO of MobileODT, said. “We’re looking forward to seeing what comes from this competition in order to save as many lives as possible.”
Developers and data scientists interested in joining the competition can sign up to be notified about the launch here: https://software.intel.com/en-us/ai/competition-signup
For Intel’s press release on this exciting project:
MobileODT- MobileODT is driven by a mission: to enable front line health providers in resource limited areas to deliver affordable and accessible visual based assessment and triage of diseases with connected medical devices. Their first product, The EVA System, functions on its own as an enhanced visual assessment screening or alongside any cervical cancer treatment. As a practical solution to the challenges of resource limited healthcare — and to enable the extension of care into medium-resource settings — the EVA System was built to be affordable and practical for the nurse or general practitioner at the point-of-care and point-of-treatment, everywhere.
Kaggle- Kaggle is the world’s largest online data science community. With more than 700,000 active members across 194 countries, the Kaggle community competes to build the best solutions for solving complex data science problems. Kaggle users come from diverse educational backgrounds and are often experts in their fields. Working as individuals or in teams, the winning competitors are awarded prizes and industry recognition for their accomplishments. Kaggle competitions have tackled a wide range of problems hosted by commercial, government & research organizations, including diagnosing heart failure, identifying right whales, forecasting sales, and improving search relevance.
For press inquires contact: