with artificial intelligence to detect covid-19 in 2D lung ultrasound scans and quantify the severity of the disease

点击翻译按钮获取中文摘要

**Public overview**This project is to augment ultrasound as a screening tool with artificial intelligence to detect covid-19 symptoms in lung ultrasound scans and help doctors understand how risky patients are. It can also be used to monitor the progression of a patients' disease.CT scanning has been shown to outperform some of the other diagnostic tests to diagnoses corona virus but when a patient needs to be taken for an X-ray or CT scan, they are at risk for spreading the disease to other patients and the other doctors who look after them.Doctors in Italy found that corona virus can be detected using lung ultrasound.Ultrasound is a very portable way of checking patients' lungs and it can also be operated easily by nurses when a patient first comes into the hospital to decide how risky the patient is. Ultrasound can also be used next to the patient's bedside. This means decisions can be made quickly whilst the patient is not as exposed to other patients and / or doctors and so helps with both treatment for the individual patient and infection control.The current issue with lung ultrasound is that the image must still be analysed by a person, and people make mistakes, take time to do things, and get tired. If an algorithm can work alongside the doctors and nurses, it will improve the speed and quality of their work.This project has three objectives:1. To prove that corona virus can be automatically detected in lung ultrasound images2. To find a way to market this product3. To find a hospital to help supply data and demonstrate the algorithm worksIf this project is successful, this can potentially be a very useful tool to help doctors and nurses do a lot more with less resources. Making lung ultrasound more useful for doctors can be very helpful in the fight against this corona virus pandemic. As this project is software based, it will be very easy for other hospitals and care providers to take advantage of the benefits. This is also potentially very useful for more rural healthcare settings where the expertise to interpret these images may not be easy to access.