How the first AI to cure ALS may be the first to treat ‘human-like’ disease
The first AI developed to treat ALS, dubbed ‘AI-A’ and developed by IBM, has shown it can be used to treat humans with a form of ‘humanlike disease’.
The AI was developed by researchers at the University of Texas at Austin (UT Austin) and the University at Buffalo (Buffalo Medical School).
The work was reported today in the journal Science.
“In the coming years, AI-A may be used for treatments of ‘selfish’ or ‘lazy’ diseases, as well as diseases caused by other infectious agents such as influenza and coronavirus,” the researchers wrote in the paper.
“The goal of this research is to build upon this basic research to develop a model to detect and predict disease severity and to identify novel treatment targets.”
The AI-a model can learn from human experiences, and even recognize facial expressions, such as smiling or frowning.
“Our model can detect the changes in facial expressions from people in the same room, from people sitting next to each other, from strangers or people in a different room, and from different social contexts,” said Dr Andrew Ainsworth, who led the research.
“It can even identify the specific changes in people’s facial expressions as they are performing a task.”
‘A human being’ The AI uses a neural network that has been trained to recognize the facial expressions of a patient in the room and predicts the person’s response based on their current facial expression.
“With this system, we are able to predict the patient’s facial expression by simply looking at their face,” Dr Ainswill said.
“That’s a significant advance for ALS, which is characterized by the inability to distinguish between a normal facial expression and a self-directed one.”
The system can also detect changes in the brain activity of the patient as they move from one room to another.
“These changes can be caused by various factors, including changes in oxygen levels in the blood or brain, or changes in blood flow, or alterations in the electrical activity of various brain regions,” Dr Anson Broussard, a member of the team, told the BBC.
“To identify and treat the patient, the AI will have to learn to associate the person with the facial expression of the person, and also understand what is happening in the patient and what is not happening in their brain.”
“As we improve the accuracy of the model, it can learn to predict how a person will respond in different situations, and it can then be used in a variety of ways to diagnose and treat ALS patients,” Dr Brouessard added.
‘A computer-generated human’ The model is able to perform a range of tasks, including detecting and predicting the changes of blood flow in the body of the ALS patient, as they progress from room to room.
The researchers also developed an algorithm that can predict the type of activity of a particular part of the brain that affects the patient.
“We have developed an artificial neural network to understand how the patient reacts to different stimuli, and we have developed a system that can detect changes to this neural network,” Dr Haidar Akhtar, a researcher at the UT Austin and the Buffalo Medical School, told BBC News.
“This is the first time that we have been able to use a computer-based neural network in ALS, to learn how to predict these changes in brain activity.”
“The system we are using in this study can detect, for example, the change in oxygen level in the bloodstream and can predict whether the patient will respond to a particular stimulus or not.
This is a major step forward for the field of ALS.”
The researchers used AI-ai, which stands for artificial intelligence-based artificial neural networks, to build the AI-as-a-computer model.
It has the capability of learning from the human experience and adapting to new stimuli.
“One of the challenges in ALS is that there are so many different brain areas, and not all of them respond to the same stimuli,” Dr Akhtar said.
“[The AI-AI model] learns by following the patient around in the lab, and learning what is going on in the human brain and using this knowledge to improve its response.”
The work also revealed that the AI can recognize changes in breathing and blood flow.
“When the patient is breathing, the system can see the brain’s activity changes,” Dr Gopinath Bhattacharya, a senior research scientist at IBM, told Business Insider.
“And when the patient was breathing, we saw the system respond to these changes by creating a new layer of neural network with this specific neural network.”
“We think that it’s really going to be the beginning of a whole new field of artificial intelligence,” Dr Abhijit Ramesh, a professor at the AI University of Chicago, told Forbes.
“For the next five to 10 years, this is going to open up a whole other area of