search

cosmoimd / real-colon-dataset

Helper function for working with the REAL-Colon Dataset
https://www.nature.com/articles/s41597-024-03359-0
26 stars 5 forks source link

Inquiry About Polyp Size Measurements in the Dataset #2

Open jangbi1 opened 1 month ago

jangbi1 commented 1 month ago

Dear @carlobiffi,

Hello!

First of all, thank you for sharing your astonishing dataset with the research community.

I have opened an issue regarding the size measurements in your annotations, because I wanted to clarify some details regarding the annotations, specifically about the "size" of the polyps. In the paper, I noticed the following statement:

"Polyp information, including histology, size, and anatomical site, has been recorded, double-checked by annotation specialists and at least an experienced gastroenterologists, and reported with several other clinical variables."

As you may know, polyp size measurements typically involve either comparative methods (such as using forceps as a reference) or visual estimation. However, I noticed that the documentation doesn't specifically detail:

  1. The measurement methodology used for size determination
  2. The validation process or double-checking procedures implemented

Could you please provide more information about how these measurements were conducted and verified? This information would be extremely helpful for those of us working with your dataset.

Looking forward to your response.

Best regards,

Jangbi

carlobiffi commented 5 days ago

Dear Jangbi,

Thank you for your interest in our work. To address your question:

  1. Each polyp’s size was estimated live through visual assessment by the endoscopist who performed the polyp resection and documented in an electronic Case Report Form (eCRF).

  2. The histopathological information we released, including size, was cross-verified against the corresponding patient eCRF to ensure it matched the reported polyp and not another from the same video.