Social Interactions with CPS / Modelling and visualising precancerous lesions

Social Interactions with CPS

 

Autonomous, mobile cyber-physical systems are popular in the transportation, manufacturing and industry sectors, thanks to their ability to provide smart and twenty-four-seven services to society, production industries and companies. Fully autonomous systems, like CPS, require social interactions. These human beings’ interactions can be verbal conversations and non-verbal like hand signs and facial expressions with the system. Hand signs are required to communicate with the systems to provide a trustworthy and secure environment together with several cyber-physical systems.
The Handie system handles hand sign interactions with autonomous mobile cyber-physical systems, aka mobile robots. Both hand signs and facial expressions are handled by a deep learning recognition component in a humanoid, programmable robot Softbank Robotics’ NAO robot v6. The cyber-physical system can recognize hand signs and facial expressions to an acceptable extent to be useful.

 

handie-system-for-presentation.png

 

Modelling and visualising precancerous lesions.

 

The modelling and visualising precancerous lesions project applies real-time object detection algorithms and a revolutionary method with separate colour channels to improve the detection of precancerous lesions.

 

The project uses artificial intelligence to increase the detection of precancerous changes in the colon mucosa in patients with inflammatory bowel disease (IBD). The goal is to reduce the incidence of full-blown colon cancer (CRC) and reduce the need for major bowel surgery.

 

IBD often begins in the twenties and increases the risk of CRC 10 years after disease onset. To reduce the risk of cancer, patients are followed with regular colonoscopy checks. Despite this, serious changes are missed due to the difficulty of distinguishing these in normal mucosa or mucosa with scar changes. This delays diagnosis and treatment and results in young IBD patients receiving more complicated and burdensome surgical treatment and a worse prognosis. Early detection of precancerous lesions makes it possible to replace major surgical operations with minimally invasive treatment, improve the quality of life of young patients and reduce their morbidity and mortality.

 

Thanks to large polyp video databases, image analysis with AI algorithms shows good results in detecting polyps in the colon, but it is not applicable to precancerous lesions. But with real-time object detection algorithms and a revolutionary method separating colour channels improve the detection of precancerous lesions. The detection requirement is ≥90%. In addition, other medical data about the patients are linked to video recordings of precancerous lesions by creating appropriate lesion classifications.