Using Machine Learning To Make Better Medical Decisions
Researchers are studying how machine learning systems make decisions and how they can choose the wrong one.
Source: Getty Images
By - While artificial intelligence has become essential in advancing medical research, scientists have yet to work out all the technology’s flaws. A University of Texas at Arlington computer scientist has received a three-year, $385,000 grant from the National Institute of Standards and Technology to study how machine learning systems make decisions and what happens when they make wrong ones.
Jeff Lei, a professor in the Computer Science and Engineering Department, will examine why machine learning systems make specific decisions and how to fix incorrect decisions.
Machine learning systems are increasingly assisting humans in the decision-making process, especially in the medical field. However, with the reliance on the systems, researchers must ensure they are properly trained to make the correct decisions.
The system makes decisions by evaluating a large set of data points. Data points that are closer to a decision point are more influential than those further away. Lei will engage in “neighborhood exploration” by studying data points in the vicinity of a decision point rather than looking at the entire training set. According to Lei, this method will significantly reduce the computational complexity.
Data can be biased, or individuals can make mistakes in data collection. However, what leads to an incorrect decision can be identified by examining the data points that influenced the decision most.
“Artificial intelligence is helpful in making decisions, but because of the complexity of the process, it isn’t quite transparent,” Lei said in a press release.
“This is a serious concern in domains where decisions have important consequences. We must provide good explanations for why decisions are made, pinpoint the root cause of any incorrect decisions, and suggest changes to correct them to maintain public trust and ensure that the systems are working as intended.”
According to the chair of the Computer Science and Engineering Department at the University of Texas at Arlington Hong Jiang, Lei’s work could potentially increase the use and usability of artificial intelligence technology for future applications.
“One of the main attractions of AI technology is its apparent power in automating the decision-making process by providing accurate predictions via training on massive amounts of data,” Jiang said.
“Ironically, however, one of the biggest challenges facing AI is the inability to explain predictions and their accuracy, because how AI algorithms reach their conclusions has long been considered a mysterious black box. Professor Lei’s work on explaining decisions made by AI is very timely and potentially highly impactful.”
