Muhammad Aleem, a final year student at GIKI, has come up with an AI-based solution for detecting Coronavirus using chest X-rays. He claims that his AI Corona detector provides results with 96% accuracy. This news comes as a ray of hope for people who cannot afford expensive tests for the deadly virus.
Coronavirus has caused death of 6500 people worldwide and 170,000 cases overall. The pandemic has started spreading in Pakistan some days ago and has already affected nearly 200 people in the country. While it was found by American College of Radiology (ACR) that chest CT scans can be used to detect COVID-19, they are quite expensive for common people. However, chest X-rays can be done at any hospital or nearby dispensary.
Talking about his detector Aleem said, “With certificates in Artificial Intelligence from Stanford, IBM, and DeepLearning.ai. I love to apply engineering to the medical field which motivated me to use AI (Artificial Intelligence). I used this to diagnose COVID-19 since Pakistan doesn’t have proper resources for that. Other than this I’m also developing robotic arms for medical surgery.”
Neural Networks, a sub-branch of Artificial Intelligence, was used to develop the detector. Aleem used a dataset of X-ray images of COVID-19 positive from Dr. Joseph Paul Cohen’s GitHub repository and normal cases from Kaggle to train the system. Neural networks learn to make decisions by assigning weights to different factors and correcting them on the basis of wrong decisions made in the process.
Using the mentioned dataset, Aleem has been able to achieve 85-90% accuracy. More data will be required to make the predictions more accurate. Government agencies and medical authorities have been contacted for the provision of more data.
Aleem has requested people to send him X-ray pictures of Coronavirus patients so his model can be perfected. The more X-rays the systems has, the better it will be able to predict positive or negative cases of COVID-19. You can contact him here: email@example.com