Artificial Intelligence and the diagnosis of colorectal cancer

According to the National Cancer Institute, colorectal cancer is the second most common cancer in women and the third most common in mans. It is the second leading cancer that would be the cause of death in France, and accounts for almost 12% of all cancer deaths, especially among the aged 65 and over. Screening is a crucial step in reducing the mortality rate in patients. Artificial Intelligence (AI) allows you to explore the health data accumulated in recent years. This brings medicine into a new cycle of innovation that leads to improve diagnoses of many diseases including cancer before its appearance.


How? ‘Or’ What

Thanks to an artificial intelligence technique for the detection of Gi Genius colorectal cancer which have been associated directly with the usual equipment for a colonoscopy, it allows in real time to detect and report automatically, any abnormality suggestive of colorectal polyps. When the device detects an unusual outgrowth, it projects a green frame around the area and emits a beep. Two institutions of the Paris Public Assistance Hospital (AP-HP) in August and September 2019 tested this technique. Gi Genius acts as a second observer to increase  detection of anomalies. It has a sensitivity of 99.7% allowing it to miss almost no polyp.

Intelligence level of Gi Genius

According to Robert Benamouzig “On the theoretical level, this is the first pragmatic irruption of AI in endoscopy”. GI Genius would have three levels of intelligence possible:

– The aid for the detection of cancer, which is already done,

– The characterizations,

– And the proposal of a diagnosis to the health professional.

 

 

A propos de Massahalo DEGBE

Passionnée des nouvelles technologies et du e-commerce, je suis à la recherche de nouvelles opportunités.

Vous aimerez aussi...

1 réponse

  1. Boris dit :

    Vraiment intéressant.
    Comment peut-on entrer en contact avec les chercheurs qui travaille sur ce projet svp?

Laisser un commentaire

Votre adresse de messagerie ne sera pas publiée. Les champs obligatoires sont indiqués avec *

Ce site utilise Akismet pour réduire les indésirables. En savoir plus sur comment les données de vos commentaires sont utilisées.