Case Study

“AI provides extra information”: Saige Brain at Erasmus MC

With Meike Vernooij M.D., Neuroradiologist at the Alzheimercentrum Erasmus MC and Professor of Population Imaging at Erasmus MC.

DeepHealth

The challenges of neuroradiology and neurodegenerative diseases

Dr. Vernooij experiences an important challenge in her daily reporting work of brain scans of patients from the memory clinic or the Alzheimer center at Erasmus MC. The challenge is that it can be very difficult to distinguish neurodegenerative pathology from normal age-related changes to the brain, especially in young patients. Because of this, semi-quantitative visual rating scales are not very useful.

Introducing Saige Brain (formerly Quantib® ND) to gather quantitative insights

Saige Brain provides automatic quantifications of brain and brain structure volumes, and it provides a set of normative reference curves that enable the comparison of the values of your patient to those of a non-demented population. This helped Dr. Vernooij and the rest of the Erasmus team.

Dr. Vernooij states that they use the quantification and the relative information from the normative reference data, on top of their visual assessment, to add to their radiology reports. These features increase their understanding of whether the brain atrophy is normal or abnormal and enables a better diagnosis for each patient.

“Saige Brain can help us to indicate if there are any changes happening in the brain earlier in the disease process,” said Dr. Vernooij.

How Dr. Vernooij uses Saige Brain

In her work as a neuroradiologist, besides her visual assessment, Dr. Vernooij uses Saige Brain as an extra tool to help her make the best diagnosis for the individual patient. After the visual read of the brain MRI, she utilizes the software and integrates the insights gathered from both to create her radiology report.

Saige Brain adds value to her workflow as it indicates a finding she may have missed with only a visual assessment. That triggers a second visual assessment. The technology also provides information that would have been impossible to assess visually. Dr. Vernooij leverages Saige Brain to increase her diagnostic accuracy and her confidence in the radiology reports she delivers. 

A clinical case in which Saige Brain can be especially helpful

Dr. Vernooij expressed that a very specific illustration of how Saige Brain can have added value is in younger individuals (for example, in their 40s or 50s) who carry familial genetic mutations that will eventually lead to fronto-temporal dementia (FTD). In these people, the imaging scans and especially the quantification obtained from Saige Brain may show that the structural volumes are deteriorating, even when they don’t have any symptoms and the neuropsychological exams don’t manifest any disease development. Saige Brain helps radiologists detect changes earlier in the disease process.

“During our MDT meetings, we always pull up the report from Saige Brain. We use it as extra information to come to a joint decision on the best diagnosis and clinical management of the patient,” said Dr. Vernooij.

The added value of Saige Brain in the rest of the diagnostic process beyond radiology

As part of her job as neuroradiologist, Dr. Vernooij attends MTD meetings with referring clinicians, like neurologists and geriatricians. In those MTD meetings, she always shares the Saige Brain report. Dr. Vernooij also believes that the quantitative information indicates how the team arrived at diagnosis. Therefore it increases referring clinicians’ confidence in the radiology report.

Furthermore, referring clinicians use the Saige Brain report when communicating with patients to explain why a specific decision was made.

“The application of AI, like Saige Brain, in radiology, will provide us with extra information. Our task as radiologists will be to integrate this information and to communicate with our referring physicians in the diagnostic process.,” said Dr. Vernooij.

About Meike Vernooij, M.D.

  • Neuroradiologist at the Alzheimercentrum Erasmus MC
  • Professor of Population Imaging at Erasmus MC