NEC uses AI to Help Diagnose Pre-cancerous Cells
NEC's AI supports doctors to detect neoplasia in
Barrett's esophagus during endoscopic procedures
World's first CE mark-compliant
software of its kind; soon available in Europe
NEC announced the development of an
AI technology for supporting doctors to detect neoplasia in Barrett's esophagus
during endoscopic procedures. This is the world's first technology of its kind
to comply with the requirements for CE mark labeling, a European safety, health
and environmental protection standard (*1), and is being released as WISE
VISION® Endoscopy, where it will soon be available in Europe to help doctors in
the detection of Barrett's neoplasia (*2).
This software is connected to endoscopy processors and automatically notifies
users of potential Barrett's neoplasia from images captured during endoscopic
procedures. Professor Pradeep Bhandari: "Once the neoplasia is found, the
system takes a still image and transfers it to the top right corner of a screen
as a reference image for endoscopists. It also has a heat map which shows the
area of the AI-predicted neoplasia."
Barrett's esophagus is a
pre-cancerous condition in which the inner lining of the esophagus is replaced
by a lining that resembles that of the stomach. Worldwide prevalence of
Barrett's esophagus is 1% of the entire population but much higher (7%) in
patients with acid-reflux symptoms (*3). Patients with Barrett's esophagus have
30-40 fold higher risk of developing cancer of the esophagus as compared to the
normal population (*4).
However, if this cancer can be detected at an early stage, then it can be
removed through the endoscope and a patient can be cured. Patients with
Barrett's esophagus have to undergo regular endoscopy (every 2-3 years) to
detect early cancer, but this is still challenging as the early changes of
cancer are very subtle and difficult to identify. As a result, current
recommendations are to take multiple random biopsies throughout the length of
Barrett's esophagus to identify any focus of early cancer (*5, 6). However,
random biopsies are expensive and time consuming and are only performed in
accordance with official clinical guidelines by an estimated 30-51% of endoscopists
(*7, 8, 9, 10). Consequently, it has been pointed out that up to 40% of
neoplasia may be missed (*7) resulting in delayed cancer diagnosis and poor
outcomes for the patients.
To help resolve this issue, NEC collaborated with Professor Pradeep Bhandari
(Portsmouth, UK), Chair of the European Society of Gastrointestinal Endoscopy
(ESGE) Research Committee, in development of this new technology to support
doctors to detect Barrett's neoplasia during examination. WISE VISION®
Endoscopy has been trained with more than 1 million Barrett's esophagus
endoscopy images, together with the knowledge of expert endoscopists, enabling
it to meet the requirements for CE mark labeling. In developing this solution,
NEC applied its face recognition technology that has been highly evaluated
(*11) by the National Institute of Standards and Technology (NIST) in the
United States, and belongs to NEC's portfolio of cutting-edge AI technologies,
"NEC the WISE."
On clinical evaluation, WISE VISION® Endoscopy could find more than 90% of
Barrett's neoplasia (*12). WISE VISION® Endoscopy is expected to significantly
reduce the neoplasia miss rate during endoscopy and improve the outcome for
patients with Barrett's esophagus all around the world.
The abstract applying this technology has received the Gut Award from the
British Society of Gastroenterology (BSG), which was presented in the plenary
session of the BSG Campus. It was also selected one of the best endoscopy
abstracts in UEG Week Virtual 2020 held by United European Gastroenterology.
"I am delighted that NEC, as one of the world's leaders in AI technology,
has entered the field of endoscopy and developed WISE VISION® Endoscopy to
detect and help manage Gastrointestinal neoplasia. Barrett's neoplasia can be
very flat and easily missed during endoscopy, despite taking multiple biopsies.
I worked with NEC to develop the Barrett's neoplasia detection arm of WISE
VISION® Endoscopy and am super impressed by the speed with which NEC has
produced this as the World's first AI engine to obtain the CE mark for
notifying Barrett's neoplasia.
"WISE VISION® Endoscopy can notify very flat and subtle lesions and much
faster than most endoscopists. Once the neoplasia is found, this system takes a
still image and transfers it to the top right corner of a screen as a reference
image for endoscopists. It doesn't stop here, as it also has a heat map which
shows the area of the AI-predicted neoplasia. WISE VISION® Endoscopy is set to
revolutionize the endoscopic detection and management of Barrett's neoplasia in
the western world. I'm sure that clinicians involved in the management of
Barrett's esophagus are looking forward to the dawn of this new (Wise) era in
Endoscopy," said Professor Pradeep Bhandari, Chair of the ESGE Research
Committee.
About WISE VISION® Endoscopy
This software can be connected to endoscopes from three leading endoscope
manufacturers (*13). By simply connecting an existing endoscope to a monitor
and terminal equipped with the software, users can start using it immediately.
The system issues sound notifications when a candidate lesion has been located,
and the type and volume of the notification can be customized at any time
according to user preferences. Moreover, a highly visible user interface allows
intuitive operation, enabling specialists to smoothly proceed with
examinations.
Going forward, NEC aims to accelerate digital transformation in the medical
field by providing safe and secure products and services, including AI, to
promote a healthy, and sustainable society, where individuals can thrive from
the utilization of advanced IT technologies. Source: NEC media announcement Notes:
- (*1) Source: NEC Corporation
- (*2) "WISE VISION" is a registered trademark of
NEC Corporation in the United States of America, the European Union and
the United Kingdom.
- (*3) Marques de Sá I, Marcos P, Sharma P, Dinis-Ribeiro M.
The global prevalence of Barrett's esophagus: A systematic review of the
published literature. United European Gastroenterol J.
2020;8(9):1086-1105. doi:10.1177/2050640620939376
- (*4) Sharma P. Clinical practice. Barrett's esophagus. N
Engl J Med. 2009 Dec 24;361(26):2548-56. doi: 10.1056/NEJMcp0902173.
Erratum in: N Engl J Med. 2010 Apr 15;362(15):1450. PMID: 20032324.
- (*5) Weusten B, Bisschops R, Coron E, Dinis-Ribeiro M,
Dumonceau JM, Esteban JM, Hassan C, Pech O, Repici A, Bergman J, di Pietro
M. Endoscopic management of Barrett's esophagus: European Society of
Gastrointestinal Endoscopy (ESGE) Position Statement. Endoscopy. 2017
Feb;49(2):191-198. doi: 10.1055/s-0042-122140. Epub 2017 Jan 25. PMID:
28122386.
- (*6) ASGE STANDARDS OF PRACTICE COMMITTEE, Qumseya B, Sultan
S, Bain P, Jamil L, Jacobson B, Anandasabapathy S, Agrawal D, Buxbaum JL,
Fishman DS, Gurudu SR, Jue TL, Kripalani S, Lee JK, Khashab MA, Naveed M,
Thosani NC, Yang J, DeWitt J, Wani S; ASGE Standards of Practice Committee
Chair. ASGE guideline on screening and surveillance of Barrett's
esophagus. Gastrointest Endosc. 2019 Sep;90(3):335-359.e2. doi:
10.1016/j.gie.2019.05.012. PMID: 31439127.
- (*7) Kariv R, Plesec TP, Goldblum JR, et al. The Seattle
protocol does not more reliably predict the detection of cancer at the
time of esophagectomy than a less intensive surveillance protocol.
Clinical gastroenterology and hepatology : the official clinical practice
journal of the American Gastroenterological Association. 2009
Jun;7(6):653-8; quiz 06. PubMed PMID: 19264576. Epub 2009/03/07. eng.
- (*8) Shaheen NJ, Falk GW, Iyer PG, Gerson LB, American
College of G. ACG Clinical Guideline: Diagnosis and Management of
Barrett's Esophagus. Am J Gastroenterol. 2016;
- (*9) Spechler SJ, Sharma P, Souza RF, Inadomi JM, Shaheen
NJ. American Gastroenterological Association technical review on the
management of Barrett's esophagus. Gastroenterology. 2011
Mar;140(3):e18-52; quiz e13.
- (*10) Abrams JA, Kapel RC, Lindberg GM, Saboorian MH, Genta
RM, Neugut AI, et al. Adherence to Biopsy Guidelines for Barrett's
Esophagus Surveillance in the Community Setting in the United States. Clin
Gastroenterol Hepatol [Internet]. 2009 Jul 1 [cited 2018 Nov
16];7(7):736–42.
- (*11) NEC Face Recognition Technology Ranks First in NIST
Accuracy Testing
https://www.nec.com/en/press/201910/global_20191003_01.html - (*12) Presented at ESGE Days 2021 held by European Society of
Gastrointestinal Endoscopy
- (*13) Connectivity and operations have been confirmed with
the following endoscopes.
- Olympus
EVIS LUCERA ELITE Video System Center CV-290
- Olympus
EVIS EXERA III Video System Center CV-190
- FUJIFILM
ELUXEO video processor VP-7000
- PENTAX
Medical OPTIVISTA EPK-i7010 video processor
- PENTAX
Medical OPTIVISTA EPK-i7000 video processor
|