Computers Being Trained To Predict The Outcome of Acute Diseases

Today, we have access to some groundbreaking technologies like Artificial Intelligence, Machine Learning and Computer Vision. While there’s still so much to learn and explore in this field, there is a quest to make computers as smart as the humans are. Machine Learning has a big a role to play in making computers smarter and learn to think and act like humans. The world is continuously making progress in the field of Machine Learning and Computer Vision and new aspects of these two technologies are coming into picture.

Researchers of the Purdue University, Indianapolis have recently developed a computer model based on Machine Learning which predicts the future outcome of a deadly disease called Acute Myelogenous Leukemia (AML). This model will help the computers decide which patient of AML will go into remission after the treatment and which one will relapse.

Also read How Machine Learning Will Aid Neural Science.
 

The Concept

The main idea is to train the computers using the advanced machine learning algorithms and make them adroit enough to understand the medical condition of a patient diagnosed with Acute Myelogenous Leukemia. Based on which, it will make accurate predictions regarding the future outcomes of AML post the treatment. For instance, if a patient is suffering from an AML that is very likely to relapse even after the treatment, it will be known to the patient or his/her family. After that, it is up to them to decide whether they would like to go for the treatment or not.

The Building Blocks

Training a computer to predict the future outcomes of a deadly disease like Leukemia was quite a challenge for the Researchers of the Indiana University. However, Murat Dundar, the Associate Professor of Computer Science at the University states that it wasn’t much difficult to teach a computer how to identify AML in Leukemia patients. According to him, the real challenge was to teach a computer to predict the direction of change that takes place in AML patients.

For this, the computer was fed with bone marrow data and medical history of the AML patients. Also, the blood data was collected from healthy individuals. After compiling this data, the computer was tested on certain AML patients for its accuracy and the results were startling. The predictions were made with 100 percent accuracy and relapse was predicted in 90% of the patients.  

Bridging The Gap Between Machines & Humans

The increasing use of computers in predictive analytics and medical sciences is no doubt a major leap for the entire human race. As astounding as it sounds, the computers are now being used in invasive surgeries and carrying out complex operations. Computers are helping the blind people see the world they never really saw their entire life which is no less than a blessing. All this has been done by the virtue of the cutting-edge technologies like Computer Vision, Machine Learning and Artificial Intelligence.