Automatic Assessment of Colonoscopy Performance

Colonoscopy is an invasive procedure with significant risk of adverse events. Since 2012, CAMES has focused on developing tools enabling trainees and experts safely to develop and increase their competencies ensuring patient safety. 

​All pictures are from Vilmann AS, Lachenmeier C, Svendsen MBS, Søndergaard B, Park YS, Svendsen LB, Konge L. "Using computerized assessment in simulated colonoscopy: a validation study". Endosc Int Open. 2020 Jun;8(6):E783-E791.

Colonoscopy is the criterion standard for the diagnosis of several diseases, from colorectal cancer to inflammatory bowel disease. Colonoscopy is an invasive procedure, with significant risk of adverse events. Training in colonoscopy was previously based on supervised, hands-on training on patients. This apprenticeship model contributes to several problems: great need for resources, threat to patient safety, lack of standardization; and observer bias during assessment. Furthermore, the introduction of national colonoscopy screening programs for colorectal cancer has increased the number of colonoscopies; hence, increasing the need for training and insuring quality among practicing colonoscopists.

Vilmann AS, Lachenmeier C, Svendsen MBS, Søndergaard B, Park YS, Svendsen LB, Konge L. Using computerized assessment in simulated colonoscopy: a validation study. Endosc Int Open. 2020 Jun;8(6):E783-E791. doi: 10.1055/a-1132-5259. Epub 2020 May 25. PMID: 32490164; PMCID: PMC7247901.
Since 2012 CAMES has been focusing on the development of tools able assess colonoscopic performance in an automatic, unbiased and objective way. Furthermore, provide the operator with feedback during and after the procedure. This has led to the development of the Colonoscopy Progression Score (CoPS), a system using magnetic endoscopic imaging (MEI) to track the colonoscopeCoPS has proven to increase performance among trainees when the score was given as feedback which signifies the need for performance related feedback during trainingFurthermore, CoPs has found to be correlated with patient experienced pain and could provide the operator with a pain-sensitivity map indicating sites associated with painOlympus colonoscopes are provided with electromagnetic coils along the length of the scope. The system utilizes a Magnetic Endoscopy Imaging system called ScopeGuide (UPD-3, Olympus, Tokyo, Japan). For normal use, the UPD-3 system renders 2D images of the shape of the colonoscope to assist the operator in understanding and handling certain issues during a procedure. We collected XYZ-coordinates directly from the UPD-3 unit allowing the development of a system that assesses more than just the route of the tip. 

The system was tested in a simulation-based setup, and five different measures were created: 

  1. Travel length
  2. Tip progression
  3. Chase efficiency 
  4. Shaft movement without tip progression
  5. Looping  

A combined score based on the measures were created, the 3D-CoPS, and the results on 3D-CoPS and each of the five measurements revealed a statistically significant difference between groups with various experience, in performance.

Finally, based on XYZ coordinates we developed a Colonoscopy Retraction Score (CoRS) and explored the correlation between endoscopists’ historical performance measured by ADR and their immediate performance measured by CoRS, revealing a strong correlation with ADR. CoRS could potentially be used as a more precise substitute or addon to existing quality measures increasing ADR and therefore reducing post-colonoscopy cancer due to missed adenomas.

Current projects

With financial support from the Capital Region of Denmark and Danish Cancer Society the systems, 3D-CoPS and CoRS, are being implemented in a large-scale randomized setup providing feedback to the practicing colonoscopists, investigating the effect on diagnostic accuracy and scope handling.

Responsible for the Colonoscopy-projects

Responsible editor