Dataset Open Access


Sara Moccia; Gabriele Omodeo Vanone; Elena De Momi; Leonardo S. Mattos

Dublin Core Export

<?xml version='1.0' encoding='utf-8'?>
<oai_dc:dc xmlns:dc="" xmlns:oai_dc="" xmlns:xsi="" xsi:schemaLocation="">
  <dc:creator>Sara Moccia</dc:creator>
  <dc:creator>Gabriele Omodeo Vanone</dc:creator>
  <dc:creator>Elena De Momi</dc:creator>
  <dc:creator>Leonardo S. Mattos</dc:creator>
  <dc:description>The NBI-InfFrames dataset aims to provide the surgical data science community with a labeled dataset for the identification of informative endoscopic video frames. 

It is composed of 720 video frames. The frames are manually extracted and labeled from 18 narrow-band laryngoscopic videos of 18 different patients affected by laryngeal spinocellular carcinoma (diagnosed after histopathological examination). 

The frames include 180 informative (I) video frames, 180 blurred (B) frames, 180 frames with saliva or specular reflections (S) and 180 underexposed (U) frames.

The dataset was created for testing the method proposed in S. Moccia, et al. "Learning-based classification of informative laryngoscopic frames." COMPUTER METHODS AND PROGRAM IN BIOMEDICINE, (accepted for publication).

The folder contains 3 subfolders (FOLD1, FOLD2, FOLD3), which are the 3 folds used for cross-validation purpose in the frame classification performance assessment. Data separation in the folds is performed both at image- and patient-level.

Each subfolder contains 4 folders relative to the four frame classes, i.e., I, B, S and U.</dc:description>
  <dc:subject>Frame selection, NBI endoscopy, machine learning, classification</dc:subject>
All versions This version
Views 1,7361,734
Downloads 1,0221,022
Data volume 175.2 GB175.2 GB
Unique views 1,5671,565
Unique downloads 615615


Cite as