AI-based image analysis for enhancing Head and Neck Cancer detection during upper airway endoscopy (ENHANCE)
AI-based image analysis for enhancing Head and Neck Cancer detection during upper airway endoscopy (ENHANCE)

AI-based image analysis for enhancing Head and Neck Cancer detection during upper airway endoscopy (ENHANCE)

Using Artificial Intelligence and novel imaging techniques for improved diagnostic accuracy of Head & Neck Cancer

Periode
-
Looptijd
24 months
Deel van call / Programma
/
Projectpartners
UMCG
WSK Pink

Head and Neck Cancer (HNC) is an aggressive cancer type with poor survival rates in advanced stages. To prevent such late-stage detection, WSK Medical has developed an Artificial Intelligence (AI) tool for automatic detection of early-stage HNC during endoscopy. Within this project, WSK Medical and the Head and Neck Oncology Center (Dept of Otolaryngology) of the UMC Groningen (UMCG) join forces to further develop and clinically validate this tool for two highly innovative imaging techniques. This will significantly improve accurate early-stage HNC detection. 

The incidence of HNC rises every year with a current incidence of 3.000 new patients per year. Early-stage HNC frequently goes undetected in manual analysis efforts, resulting in invasive treatments (causing severe side effects such as speech problems and facial amputations) and high mortality rates. In addition, HNC is associated with high societal costs. This leaves high potential for automation initiatives. AI can significantly enhance diagnostic accuracy and objectivity, reduce the time-to-diagnosis and reduce clinicians’ current high workload. Early-stage HNC detection and treatment can lead to a significant reduction of HNC mortality and healthcare costs. Implementation of the ENHANCE tool from this project is therefore expected to result in €27.4M yearly averted costs. 

To realise this, WSK and UMCG will perform an observational cross-sectional study comparing innovative imaging techniques with the current industry standard (white light), with and without Artificial Intelligence. 

Deliverables of the project include: 

  1. Prototype tool for automatic HNC detection in the glottic larynx using traditional white light. 
  2. Prototype tool for automatic HNC detection in the oropharynx (i.e. base of tongue, pharyngeal wall, tonsils), using traditional white light. 
  3. Prototype tool for automatic HNC detection in the larynx and pharynx using Narrow Band Imaging (NBI). 
  4. Proof-of-concept tool for automatic HNC detection in the larynx and pharynx using Fluorescence-Guided Imaging (FGI). 
Thank you! Your submission has been received!
Oops! Something went wrong while submitting the form.