• AI
  • Molecular Imaging
  • CT
  • X-Ray
  • Ultrasound
  • MRI
  • Facility Management
  • Mammography

MRI Liver Lesion Detection Gets Boost from AI

Article

Algorithm can be used to detect changes in liver lesions after treatment initiation.

Radiologists can improve how they track changes in liver lesions over time by adding an artificial intelligence (AI) algorithm to MRI scans.

In a study published July 25 in the Journal of the American College of Radiology, researchers from Beth Israel Deaconess Medical Center and Harvard Medical School discussed how incorporating the AI algorithm can speed up the radiologist’s workflow. The tool makes it much easier to keep tabs on the behavior of these lesion after treatment has been initiated, freeing up the provider's time.

“With recent advances in neural networks for medical imaging, the possibility to automate segmentation tasks offers the promise of helping clinical radiologists by completing tedious tasks more quickly and comprehensively and, thus, allowing them more time to focus on higher-level interpretive and cognitive tasks,” wrote a team led by Alexander Goehler, M.D., Ph.D., a radiologist with Beth Israel Deaconess Medical Center.

Related Content: New Radiopharmaceutical Offers Better Tumor Imaging and Staging for Patients with Liver Cancer

Even though MRI is typically used to track and measure changes in liver lesion after therapy has begun, it has a downside – these scans take a long time. To make the process easier, Goehler’s team launched an investigation into whether AI tools that are used to detect liver lesions can also accurately detect changes in lesion size. If so, that information could be used to determine how a patient is responding to treatment, the team said. Data from 691 liver lesions – 160 malignant and 141 benign – were used to train the algorithm.

To evaluate the efficacy of the tool, the team analyzed MRI scans from 64 patients who had neuroendocrine tumors. These patients had undergone at least two consecutive liver scans with gadoxetic acid. Using the tool, the team identified and segmented the liver, and, then, used it to pinpoint any lesions present.

As a second step, they compared the algorithm’s performance to that of radiologists, and they found agreement in 91 percent of cases. In addition, they tool was able to distinguish between malignant and benign lesions with high sensitivity (85 percent), specificity (92 percent), and accuracy (88 percent).

Ultimately, the team said, the AI algorithm will be useful in providing a greater level of detail than radiologists are able to focus on at any given time.

“The true potential of the algorithm goes beyond its faster performance compared with a human interpreter and includes the potential to perform an entire volumetric tumor burden assessment,” Goehler said.

Related Videos
Can Diffusion Microstructural Imaging Provide Insights into Long Covid Beyond Conventional MRI?
Emerging MRI and PET Research Reveals Link Between Visceral Abdominal Fat and Early Signs of Alzheimer’s Disease
Nina Kottler, MD, MS
Practical Insights on CT and MRI Neuroimaging and Reporting for Stroke Patients
Related Content
© 2024 MJH Life Sciences

All rights reserved.