PICCOLO White-Light and Narrow-Band Imaging Colonoscopic Dataset: A Performance Comparative of Models and Datasets

Abstract

Colorectal cancer is one of the world leading death causes. Fortunately, an early diagnosis allows for effective treatment, increasing the survival rate. Deep learning techniques have shown their utility for increasing the adenoma detection rate at colonoscopy, but a dataset is usually required so the model can automatically learn features that characterize the polyps. In this work, we present the PICCOLO dataset, that comprises 3433 manually annotated images (2131 white-light images 1302 narrow-band images), originated from 76 lesions from 40 patients, which are distributed into training (2203), validation (897) and test (333) sets assuring patient independence between sets. Furthermore, clinical metadata are also provided for each lesion. Four different models, obtained by combining two backbones and two encoder–decoder architectures, are trained with the PICCOLO dataset and other two publicly available datasets for comparison. Results are provided for the test set of each dataset. Models trained with the PICCOLO dataset have a better generalization capacity, as they perform more uniformly along test sets of all datasets, rather than obtaining the best results for its own test set. This dataset is available at the website of the Basque Biobank, so it is expected that it will contribute to the further development of deep learning methods for polyp detection, localisation and classification, which would eventually result in a better and earlier diagnosis of colorectal cancer, hence improving patient outcomes.

Artificial Intelligence in Colorectal Cancer Diagnosis Using Clinical Data: Non-Invasive Approach

“…Several studies have been conducted on using deep learning for colorectal cancer detection in colonoscopy images. In [ 30 ], a novel dataset is presented that contains 3433 colonoscopy frames, divided into two categories: white-light and narrow-band images. Based on different deep learning approaches there are four different models constructed, trained and tested in the research and their performance on the PICCOLO and other two public datasets compared.…”

“…The models are either based on backbones or encoder-decoder architectures. The study concludes that the four deep learning models have the best performance in colorectal tumor recognition on the novel PICCOLO dataset [ 30 ].…”

Real-time polyp detection model using convolutional neural networks

“…Finally, the work presented here has some limitations that should be acknowledged and that will guide the future work. First, as explained before, flat polyps are underrepresented, similarly to what happens in other datasets [68]. Since these types of polyps are less frequently found, special efforts (e.g.…”

“…This way, the dataset described and used here will be publicly accessible for other researchers, allowing not only the development and evaluation of polyp detection systems, but also systems for automatic polyp classification. The publication of this dataset will increase the public datasets available, which has been recently also increased with the addition of the PICCOLO dataset [68].…”

Characterization of Optical Coherence Tomography Images for Colon Lesion Differentiation under Deep Learning

“…Convolutional neural networks (CNN) [28,29] have surpassed classical machine learning methods [30,31], and even medical expert capabilities [32][33][34]. They have been also successfully applied in colon cancer histopathological classification [35,36], MPM classification [37], polyp detection on colonoscopy [38][39][40], or histological colon tissue staining [41].…”