The Endoscopy Computer Vision Challenge (EndoCV) is a crowd-sourcing initiative to address eminent problems in developing reliable computer aided detection and diagnosis endoscopy systems and suggest a pathway for clinical translation of technologies. Whilst endoscopy is a widely used diagnostic and treatment tool for hollow-organs, there are several core challenges often faced by endoscopists, mainly: 1) presence of multi-class artefacts that hinder their visual interpretation, and 2) difficulty in identifying subtle precancerous precursors and cancer abnormalities. Artefacts often affect the robustness of deep learning methods applied to the gastrointestinal tract organs as they can be confused with tissue of interest. EndoCV2020 challenges are designed to address research questions in these remits. In this paper, we present a summary of methods developed by the top 17 teams and provide an objective comparison of state-of-the-art methods and methods designed by the participants for two sub-challenges: i) artefact detection and segmentation (EAD2020), and ii) disease detection and segmentation (EDD2020). Multi-center, multi-organ, multi-class, and multi-modal clinical endoscopy datasets were compiled for both EAD2020 and EDD2020 sub-challenges. The out-of-sample generalization ability of detection algorithms was also evaluated. Whilst most teams focused on accuracy improvements, only a few methods hold credibility for clinical usability. The best performing teams provided solutions to tackle class imbalance, and variabilities in size, origin, modality and occurrences by exploring data augmentation, data fusion, and optimal class thresholding techniques.

Differentiation of small and atypical hemangiomas from other hepatic masses using imaging methods can be difficult, especially in patients with underlying malignant disease. Therefore, contrast-enhanced ultrasound was assessed in patients with histologically confirmed hemangiomas with respect to contrast-enhancing kinetics and tumor characteristics. In 58 patients with indeterminate hepatic lesions demonstrated with at least 2 imaging methods (ultrasound/computed tomography/magnetic resonance imaging), ultrasound-guided liver biopsy revealed hemangioma. In all patients a hepatic neoplasm had been suspected because of underlying malignant disease (n = 41), liver cirrhosis (n = 15), or growth of the lesion (n = 2). All patients underwent nonlinear, low mechanical index real-time contrast-enhanced ultrasound scanning with bolus injections of SonoVue. Peripheral nodular arterial enhancement was detected in 43 patients (74%), whereas the typical metastatic peripheral rim-like enhancement was not observed at all. Strong homogenous arterial enhancement was found in 9 of 58 (16%) patients. In 6 patients (10%), the arterial contrast enhancement pattern could not be determined because of the very small size of the lesions or fibrotic nodules. Forty-five (78%) of the hemangiomas showed homogenous centripetal filling within 180 seconds. Conclusion: Contrast-enhanced ultrasound demonstrates typical hemangioma imaging characteristics, that is, peripheral nodular contrast enhancement and iris-diaphragm sign in a high percentage of patients with undetermined lesions. This technique may therefore improve noninvasive functional characterization and differentiation of hemangiomas. (HEPATOLOGY 2007;45:1139–1145.)