Automatic Polyp Detection System

Project Information

MagentiqEye

Grant agreement ID: 854709

DOI

10.3030/854709

Project closed

EC signature date 21 January 2019

Start date 1 February 2019

End date 31 July 2019

Funded under

• INDUSTRIAL LEADERSHIP - Innovation In SMEs

Total cost

€ 71 429,00

EU contribution

€ 50 000,00

Coordinated by

MAGENTIQ EYE LTD
Israel

Periodic Reporting for period 1 - MagentiqEye (Automatic Polyp Detection System)

Reporting period: 2019-02-01 to 2019-07-31

Summary of the context and overall objectives of the project

The World Cancer Research Fund estimates the colorectal cancer to be the third-most occurring form of cancer around the world, with over 1.4 million cases. It is estimated that 40 million colonoscopy procedures are performed every year[1]. The ability of colonoscopy procedures to decrease the number of colon cancer cases has been proven without a doubt and their number expected to grow. However, despite this progress the problem of undetected or missed polyps during colonoscopy procedures remains a significant problem at 22-27% [2-4]. Magentiq Eye has developed an Automatic Polyp Detection System (APDS) which uses mainly Deep Learning and Computer Vision techniques in order to detect the polyps which appear on the physician screen during the test. APDS is an AI system which works as an additional eye, helping doctors to detect polyps and thus to reduce the miss-rate.

1) https://www.sciencedirect.com/science/article/pii/S2451865417300236
2) H. Gilbert Welch, M.D. M.P.H. and Douglas J. Robertson, M.D. M.P.H. N Engl. J. Med. 2016; 374:1605-1607, April 28, 2016. DOI: 10.1056/NEJMp1600448
3) Jun Pan, MD et. al; Colonoscopy Reduces Colorectal Cancer Incidence and Mortality in Patients with Non-Malignant Findings: A Meta-Analysis. Am. J. Gastroenterol. 2016 Mar; 111(3): 355-365
4) Pullens HJ, van Leeuwen MS, Laheij RJ, Vleggaar FP, Siersema PD. CT-colonography after incomplete colonoscopy: what is the diagnostic yield? Dis Colon Rectum. 2013;56:593–599

Work performed from the beginning of the project to the end of the period covered by the report and main results achieved so far

Following a successful application to Phase 1 of the EC Horizon 2020 SME Instrument, Magentiq Eye has carried out a feasibility study for its flagship APDS technology. During this SME Instrument PH1 project, the company has registered three clinical trials (IRB approved, two in Israel, one of them already finished, and one in the USA) and has planned large scale clinical trials which will be carried out during 2019-20, including objectives, methodology, number of participants, inclusion and exclusion criteria, diagnosis protocol and reference points needed to validate the APDS. In addition, during this PH1 project, the company has tightened the connection with Key Opinion Leaders (KOL’s), especially to better understand the users’ requirements for APDS and to better define the product development plan, and with potential strategic partners to review in more details the options for the future commercialization and distribution process. The achievements of these results helped us to start define the additional technology features needed to support the main detection capability of our system, to start define the protocol of the our real-time clinical trial which shall start soon in order to get needed regulatory approvals, and to pick and have additional interactions with the most appropriate strategic partner with whom we would like to cooperate.

Progress beyond the state of the art and expected potential impact (including the socio-economic impact and the wider societal implications of the project so far)

With our innovative and preceding (compared to others) patents and with the current status of our technology, together with the clinical data that we have collected so far, the agreements we already have for the following clinical trials and our current business activity, we expect to be the first (or one of the firsts) in the market with our AI (Artificial Intelligence) add-on to the colonoscopy test, and thus to have a significant impact on the improvement of the quality of the test and on the better liability that pateints in Europe and all over the world will get when undergoing the colonoscopy test, which is the most effective screening test in order to prevent the currently preventative colorectal cancer disease. On top of that, we are going to add additional AI features, on top of the automatic detection, which will help to inspect the different performances parameters of the colonoscopy test and thus make it of even higher quality. Most of these features and capabilities of the system would be relevant also to other type of endoscopy procedures, such gastroscopy and uroscopy.

A polyp surrounded by the detection bounding box and the resection contour

Left side is the physician screen (detected polyp), right side is the corresponding heat-map