by Amber Raub
FAU researchers tell CBS 12 News that the artificial intelligence work they’re doing could help detect this type of fraud sooner before so much damage is done. (WPEC)
WEST PALM BEACH, Fla. (CBS12) — A Florida dentist and her husband are charged with Medicaid fraud of over a million dollars appeared in court on Wednesday, and researchers with Florida Atlantic University (FAU) say artificial intelligence could help detect cases like this much faster.
Marianna Zadov and Shimon Leizgold, who goes by Simon appeared for a status hearing on Wednesday where they did not waive their right to a speedy trial.
FAU researchers tell CBS 12 News that the artificial intelligence work they’re doing could help detect this type of fraud sooner before so much damage is done.
“Those people who are committing this crime need to know, we have the tools to catch them. So that's what it is and fast to catch them,” said Taghi Khoshgoftaar, Ph.D., senior author and Motorola Professor, FAU Department of Electrical Engineering and Computer Science.
Researchers with Florida Atlantic University’s College of Engineering and Computer Science say Medicare and Medicaid are often compromised by fraudulent insurance claims.
According to the National Health Care Anti-Fraud Association, last year the estimated annual fraud topped $100 billion but they think that number is actually much higher.
“If that money could actually go for treatment for people who need it, who can't afford insurance and so on and so forth. Then that makes our country a better place,” said John Hancock, first author and a Ph.D. student in FAU’s Department of Electrical Engineering and Computer Science.
The team at FAU is analyzing a massive amount of government Medicare data like patient records and provider payments and running it through a trained AI program.
“There's a database of everybody's activity. So claims are processed, and when a pattern or fraudulent activity gets spotted by the algorithm, a person gets alerted to investigate. And if there is fraud that will be prosecuted,” said Hancock.
They’re trying to stop people like Dr. Zadov and Mr. Leizgold faster.
According to the state attorney general’s office, the two committed a complex scheme where they over-charged patients for services covered by Medicaid like extractions and fillings, but charged more than the authorized copayment amounts.
They’re also accused of collecting money from Medicaid for never-performed services.
And AI is built to process huge amounts of information like this, and quickly flag discrepancies.
“There's about two and a half million seconds in a month. Okay. And you think about there's millions of insurance claims that come in. But think about everybody over 65 who's in a doctor's office right now in the United States. Okay, so these are all claims that are going to go to Medicare. So you have millions of claims coming in, in a month,” said Hancock. ”So if you had a person looking at a claim one claim per second, and that's not much time to make a good judgment, it's going to take a month to do that so it's just it's not feasible. So this is where the artificial intelligence comes in. We can have machine learning algorithms that learn patterns in the data and detect the fraud for us and they can do it much faster. So that 2.5 million claims you know, we can process that in seconds,” said Hancock.
“Our technology is getting better. And so, we can detect faster and then hopefully reduce the amount from 100 billion hopefully in a year or so to 50 billion per year or maybe less,” said Khoshgoftaar.
On Thursday, the dental fraud case will start bright and early at the Palm Beach County Courthouse, while prosecutors make their argument to get their dozens of witnesses on the stand and determine how long they’ll need to for trial.