Free Hospital EMR and EHR Newsletter Want to receive the latest news on EMR, Meaningful Use, ARRA and Healthcare IT sent straight to your email? Join thousands of healthcare pros who subscribe to Hospital EMR and EHR for FREE!

AI May Be Less Skilled At Analyzing Images From Outside Organizations

Posted on November 26, 2018 I Written By

Anne Zieger is veteran healthcare branding and communications expert with more than 25 years of industry experience. and her commentaries have appeared in dozens of international business publications, including Forbes, Business Week and Information Week. She has also worked extensively healthcare and health IT organizations, including several Fortune 500 companies. She can be reached at @ziegerhealth or www.ziegerhealthcare.com.

Using AI technologies to analyze medical images is looking more and more promising by the day. However, new research suggests that when AI tools have to cope with images from multiple health systems, they have a harder time than when they stick to just one.

According to a new study published in PLOS Medicine, interest is growing in analyzing medical images using convolutional neural networks, a class of deep neural networks often dedicated to this purpose. To date, CNNs have made progress in analyzing X-rays to diagnose disease, but it’s not clear whether CNNs trained on X-rays from one hospital or system will work just as well in other hospitals and health systems.

To look into this issue, the authors trained pneumonia screening CNNs on 158,323 chest X-rays, including 112,120 X-rays from the NIH Clinical Center, 42,396 X-rays from Mount Sinai Hospital and 3,807 images from the Indiana University Network for Patient Care.

In their analysis, the researchers examined the effect of pooling data from sites with a different prevalence of pneumonia. One of their key findings was that when two training data sites had the same pneumonia prevalence, the CNNs performed consistently, but when a 10-fold different in pneumonia rates were introduced between sites, their performance diverged. In that instance, the CNN performed better on internal data than that supplied by an external organization.

The research team found that in 3 out of 5 natural comparisons, the CNNs’ performance on chest X-rays from outside hospitals was significantly lower than on held-out X-rays from the original hospital system. This may point to future problems when health systems try to use AI for imaging on partners’ data. This is not great to learn given the benefits AI-supported diagnosis might offer across, say, an ACO.

On the other hand, it’s worth noting that the CNNs were able to determine which organization originally created the images at an extremely high rate of accuracy and calibrate its diagnostic predictions accurately. In other words, it sounds as though over time, CNNs might be able to adjust to different sets of data on the fly. (The researchers didn’t dig into how this might affect their computing performance.)

Of course, it’s possible that we’ll develop a method for normalizing imaging data that works in the age of AI, in which case the need to adjust for different data attributes may not be needed.  However, we’re at the very early stages of training AIs for image sharing, so it’s anyone’s guess as to what form that normalization will take.

Effort Focuses On Better Ways For Hospitals To Detect Drug Diversion

Posted on May 17, 2018 I Written By

Anne Zieger is veteran healthcare branding and communications expert with more than 25 years of industry experience. and her commentaries have appeared in dozens of international business publications, including Forbes, Business Week and Information Week. She has also worked extensively healthcare and health IT organizations, including several Fortune 500 companies. She can be reached at @ziegerhealth or www.ziegerhealthcare.com.

Using a combination of machine learning technology and advanced analytics, a healthcare vendor has been working to find better ways to spot drug diversion in U.S. hospitals. The work done by the firm, Invistics, is funded by an NIH research grant.

The project has taken aim at a ripe target. According to a 2017 study by Porter Research, 96% of healthcare professionals who responded said that drug diversion happened often in their business. Also, sixty-five percent of respondents said that most diversion never gets detected. Clearly, there’s a hole you could drive a truck through in the drug dispensing process.

During the first stage of the research, Invistics worked with a pilot hospital to find opioid and drug theft across the entire facility. To get the job done, the vendor aggregated data from across the pilot hospital’s systems, including medical records, employee time clocks, wholesale purchasing, inventory and dispensing cabinets.

By leveraging data across several departments, Invistics got a much clearer view of potential problems than other efforts have in the past. The initiative was completely successful, with the technology picking out 100% of drug diversion happening within the project’s parameters, the company said. Since the completion of Phase I of the grant, Invistics has rolled out the solution at several other hospitals.

When it comes to avoiding opioid abuse, far morer attention has been focused on patterns of opioid prescribing, with the assumption that the opioid addiction epidemic can be stemmed at the source. For example, we recently covered a study looking at post hospital-discharge opioid use which centered on predicting which patients would be on chronic opioid therapy after discharge and planning for that discharge appropriately.

There’s no question that such research has a place in the battle against opioid misuse and abuse. After all, it seems likely that at least some needless addictive patterns stem from physician prescribing habits. It also makes sense that states are revising their guidelines for opioid prescribing, though to my knowledge these changes are being based more on ideology than rigorous research.

On the other hand, drug diversion creates a pipeline between drug supplies and drug abusers which must be addressed directly if the opioid abuse war is to be won. I for one was interested to learn about a solution that addresses this piece of the puzzle.

NYC Health Systems Get $7M To Share Data

Posted on January 29, 2014 I Written By

Anne Zieger is veteran healthcare branding and communications expert with more than 25 years of industry experience. and her commentaries have appeared in dozens of international business publications, including Forbes, Business Week and Information Week. She has also worked extensively healthcare and health IT organizations, including several Fortune 500 companies. She can be reached at @ziegerhealth or www.ziegerhealthcare.com.

Seven New York City health systems have gotten a delayed Christmas present — a $7 million grant designed to encourage data sharing initiatives and improve patient recruitment for clinical trials. The primary goal of the project is to use evidence-based research to help patients make good decisions about their healthcare.

The funding comes from a group known as the Patient-Centered Outcomes Research Institute, or PCORI. PCORI, which will create a clinical data research network in NYC, has already created 29 such networks across the nation, according to Healthcare IT News.

These networks, collectively, will form PCORnet, a $93.5 million patient-centered research initiative. The New York City Clinical Data Research Network (NYC-CDRN), a  consortium of 22 regional organizations, will work together to develop systems supporting data networking efforts and advance patient-centered research, Healthcare IT News reports.

NYC-CDRN will kick off their efforts by identifying patients with diabetes, obesity and cystic fibrosis. It will then partner with patients and clinicians by creating disease-specific community groups.

The NYC-CDRN network will connect medical records for 6 million New York City residents, then anonymize the records, and over the next 18 months, will work to standardize the data. Ultimately, the goal is to allow patients and providers to have access to evidence-based information they can use to make smart healthcare choices.

This should be an interesting project to watch over the next year and a half. PCORI is doing a lot of forward-thinking work with its money, including $5 million to the NIH for a tool called PROMIS designed to help with comparative effectiveness research. PROMIS has existed since 2004, but PCORI is now helping it move forward, making the $5 million in funds available  in research grants up to $500,000 for projects up to two years in length.