Pcf q180al

The dataset was gathered from screening colonoscopy procedures performed in 3 Israeli hospitals. Each case consisted of a single video of an entire procedure (including the insertion phase) recorded at 30 frames per second. The following endoscope models were used: CF-H180AL, CF-HQ190L, and PCF-Q180AL (Olympus, Tokyo, Japan); EC-760R-V/L, EG-760R, and EC-530LP (Fujifilm, Tokyo, Japan); and EC-3890LK (Pentax, Tokyo, Japan). All videos and metadata were deidentified, according to the Health Insurance Portability and Accountability Act Safe Harbor. For both train and validation sets, procedures represented a sampling of the procedures performed at each institution over a period of time. The training dataset was obtained from 2 different university hospitals; the validation data were obtained from a third unrelated community hospital. Then we performed a small prospective clinical validation study in 100 patients that represented new videos never "seen" by DEEP 2 before.

Participants Anthropometric data is shown in electronic supplementary material (ESM) Table 1. Twelve obese participants with type 2 diabetes and 11 age-and BMI-matched normoglycaemic participants referred to elective RYGB participated in the study. All gave informed consent and the study was carried out in accordance with the Declaration of Helsinki. The participants with type 2 diabetes had been diagnosed with diabetes for 7.0±1.6 (mean±SEM) years and were all on oral glucose-lowering agents and/or insulin at the time of inclusion.
Experimental procedures On experimental day 1, all participants underwent RYGB using standard laparoscopic RYGB technique as previously described [3] . During the RYGB procedure, a 1 cm segment was excised from the jejunal division site (proximal end) and designated area A. Some mucosal biopsies were placed in RNAlater (Life Technologies, Carlsbad, CA, USA) while others were placed in 4% formaldehyde,.
On experimental day 2 (≥4 months after RYGB) after 8 h of fasting, an upper enteroscopy was performed in sedated participants using a colonoscope (PCF-Q180AL; Olympus, Tokyo, Japan). Nine mucosal biopsies were taken from each of three different anatomical locations (the alimentary limb, the common limb and the biliopancreatic limb), designated areas B, C and D, respectively. The biopsies were placed directly in RNAlater and 4% formaldehyde, respectively.