Findings explored today at the ADLM 2026 meeting

ANAHEIM, CALIF., July 29, 2026 /PRNewswire/ — Breaking research presented today at ADLM 2026 in Anaheim, California suggests that machine learning (ML) could play an important role in enhancing the accuracy of a standard blood test used to diagnose rare tumors that form in or near the adrenal glands. The study also highlights the complexity of using ML models in laboratory medicine. As healthcare strives to harness the power of ML, this research could guide labs in rigorously validating their ML algorithms by illustrating the kinds of pitfalls they need to look out for.

Plasma-free metanephrines are the recommended first-line test for detecting masses known as pheochromocytomas and paragangliomas (PPGL), which cause the body to overproduce stress hormones. Left untreated, PPGL can cause heart problems, headaches, high blood pressure, and other issues.

While the metanephrine test effectively detects people who have PPGL, it does less well at ruling out everyone who doesn’t have the condition. That’s because mild elevations in metanephrines — which are metabolites derived from stress hormones — are often seen in patients who do not have these tumors, leading to false-positive results.

The researchers tested several ML algorithms to assess whether, and how much, they could bolster accuracy by reducing the likelihood of these false-positives. They analyzed data from 20,516 adults who underwent metanephrine testing at Samsung Medical Center between 2011 and 2024. Of the 19,797 patients tested who ultimately did not have PPGL, 25.2% demonstrated elevations in metanephrine that could trigger a false-positive result.

“Our initial machine-learning models suggested that combining plasma metanephrine results with structured clinical information from the electronic health record could improve real-world discrimination,” said Se-eun Koo, one of the study’s co-authors and a clinical chemistry fellow in the department of laboratory medicine and genetics at Samsung Medical Center in Seoul, South Korea.

Specifically, Koo and co-author Dr. Soo-Youn Lee found that, while metanephrine showed good reliability as a clinical marker, integrating ML to assess relevant clinical context — including kidney and urine biomarkers, patients’ medications, and the presence of other diseases — appeared to boost the test’s performance to make it excellent.

But the story doesn’t end there. After submitting their initial findings to the Association for Diagnostics & Laboratory Medicine (ADLM), Koo and Dr. Lee performed additional analyses to determine whether the ML algorithms were using any “shortcuts” to learning that could introduce errors or biases. (As an example: If ML consistently assesses images of computer hackers wearing hoodie sweatshirts, it might conclude that anyone wearing a hoodie is a hacker.)    

“After abstract submission, while testing the prototype app with the developed ML model, we realized that some of the apparent improvement might reflect patterns of clinical workup rather than independent biochemical information,” Koo said. “Additional robustness analyses showed that much of the improvement was explained by shortcut learning from informative missingness.”              

In other words, while the ML models seemed to improve diagnostic accuracy, they did so by learning which follow-up tests were ordered — something clinicians do when they already suspect PPGL.

“This study shows that machine learning in laboratory medicine should be evaluated not only by performance metrics, but also by whether the model is learning the intended clinical signal,” Koo said. “That is why external validation and careful audits for shortcut learning are so important.”

Koo plans to delve into this timely research project during her poster and oral presentations at ADLM 2026. The study itself is a fascinating case study of the unique issues researchers must confront when assessing artificial intelligence and ML.

“The findings show both the potential and the pitfalls of applying machine learning to laboratory data,” Koo said. “The key message is not that machine learning cannot help, but that routine-care models must be audited to confirm they are learning the intended clinical signal.”

Attend Koo’s abstract session at ADLM 2026 (details below) to learn more about how she and her co-investigators rebuilt their approach and developed a new ML algorithm based on their findings about shortcut learning.

Session information
ADLM 2026 registration is free for members of the media. Reporters can register online here: https://xpressreg.net/register/adlm0726/media/landing.asp

Abstract B-091: Machine learning-enhanced diagnostic performance of plasma metanephrines in PPGL: A large-scale real-world clinical cohort study will be presented during:

Scientific poster session
Wednesday, July 29
9:30 a.m. – 5 p.m. (presenting authors in attendance from 1:30 – 2:30 p.m.)

The session will take place in the Poster Hall on the Expo show floor of the Anaheim Convention Center in Anaheim, California.

About ADLM 2026
ADLM 2026 offers 5 days packed with opportunities to learn about exciting science from July 26-30 in Anaheim, California. Plenary sessions will explore landmark diabetes studies, the role of laboratory medicine in keeping astronauts healthy, the connection between Alzheimer’s and Down syndrome, advances in cervical cancer screening, and the search for new cancer biomarkers using mass spectrometry.

At the ADLM 2026 Clinical Lab Expo, more than 700 exhibitors will fill the show floor of the Anaheim Convention Center, with displays of the latest diagnostic technology, including but not limited to artificial intelligence, point-of-care, and automation.

About the Association for Diagnostics & Laboratory Medicine (ADLM)
Dedicated to achieving better health for all through laboratory medicine, ADLM unites more than 70,000 clinical laboratory professionals, physicians, research scientists, and business leaders from 110 countries around the world. Our community is at the forefront of laboratory medicine’s diverse subdisciplines, including clinical chemistry, molecular diagnostics, mass spectrometry, clinical microbiology, and data science, and is comprised of individuals holding the spectrum of lab-related professional degrees, certifications, and credentials. Since 1948, ADLM has championed the advancement of laboratory medicine by fostering scientific collaboration, knowledge sharing, and the development of innovative solutions that enhance health outcomes. For more information, visit www.myadlm.org.

Christine DeLong
ADLM
Director, Editorial and Media Relations
(p) 202.835.8722
cdelong@myadlm.org

Bill Malone
ADLM
Senior Director, Strategic Communications
(p) 202.835.8756
bmalone@myadlm.org

View original content to download multimedia:https://www.prnewswire.com/news-releases/large-scale-study-points-to-potential–and-challenges–of-using-machine-learning-as-diagnostic-aid-302834720.html

SOURCE Association for Diagnostics & Laboratory Medicine (ADLM)