From October 20th-24th the annual UEG Week took place in Vienna, Austria. Beside many interesting workshops and conferences, one of the most interesting topics has been the potential of artificial intelligence for quick and reliable diagnosis of many diseases.
AI is becoming more and more popular!
It is amazing to see first results. All around the world, teams started cooperations to automatically detect different diseases such as gastric cancer (OP250, Japan), Barrett’s esophagus (OP251 NL and Germany) and esophageal cancer (OP251, Japan). Also detection of sessile serrated adenomas (P0766) neoplastic colorectal lesions (COACH – Study, P1412, Germany) or colorectal polyps (P0798, Japan) were revealed with promising numbers. Every team aims to detect findings quicker and with a higher precision to guide good follow-up or even to reduce unnecessary surgery (P1058, Japan).
Congratulations to Professor Xavier Dray!
He and his team from France have been able to show for the first time an automated detection of angiodysplasia (P0231, P0232). The findings were based on a French national database (CAD-CAP) and the automated evaluation of the cleansing level of SB capsule videos (P0230). To improve quality, it is crucial to understand whether a video was clean enough – and judging it by yourself is quite time-intensive and cumbersome.
On the UEGW the team concluded: “We propose a novel, efficient, perfectly reproducible, automatic and rapid multi-criterion electronic score to determine the level of cleanliness of SBCE still frames. This tool will be able to serve in clinical practice (to determine if the quality of preparation of SB-CE is acceptable) and research (testing different modalities of bowel preparation).“
Find the full publication here!
It is good to see the cooperation between engineers and clinicians and I hope they will be able to support patients, payers and health care to improve the intelligent diagnostic and support decision making during the patient pathway.
