New AI Diagnosis Method Surpasses Traditional Brain Tumor Detection

In a groundbreaking development, researchers from the Vall d’Hebron Institute of Oncology (VHIO) and the Bellvitge University Hospital’s Neuroradiology Unit have unveiled a novel, non-invasive AI-based tool named DISCERN (Diagnosis in Susceptibility Contrast Enhancing Regions for Neuroncology). This advanced technology leverages deep learning models trained on standard magnetic resonance imaging (MRI) data to significantly enhance the accuracy of brain tumor diagnoses, outshining traditional methods.

Published in Cell Reports Medicine, the study showcases DISCERN’s ability to accurately distinguish between the three most common types of malignant brain tumors: glioblastoma multiforme, brain metastasis from solid tumors, and primary central nervous system lymphoma. Unlike current diagnostic practices that often necessitate invasive neurosurgical interventions, DISCERN offers a promising non-invasive alternative, relying on the analysis of MRI scans.

Raquel Perez-Lopez, the head of VHIO’s Radiomics Group and the corresponding author of the study, explained that definitive diagnosis of brain tumors has historically required procedures that could compromise patients’ quality of life. The development of DISCERN represents a significant step forward, offering a more patient-friendly diagnostic option.

DISCERN’s deep learning model is trained using approximately 50,000 voxels (the smallest volume measurement in MRI, akin to three-dimensional pixels) from 40 patients with previously diagnosed tumors. This training enables the AI to recognize specific imaging patterns associated with each tumor type. The study’s results, validated in over 500 additional cases, demonstrate DISCERN’s impressive 78% accuracy rate in tumor classification, surpassing conventional diagnostic methods.

Carles Majós, a clinical neuroradiologist and co-author of the study, highlighted DISCERN’s potential to streamline medical decision-making processes. By accurately classifying brain tumors, DISCERN can assist multidisciplinary teams in determining the most appropriate surgical or therapeutic interventions for patients.

Albert Pons-Escoda, another key contributor to the study, emphasized the importance of this research in the development of innovative biomarkers for non-invasive brain tumor diagnosis. The creation of DISCERN, which features a user-friendly interface for clinicians, is the culmination of over five years of dedicated research.

For those interested in exploring DISCERN’s capabilities, the tool is available as an open-access application, promoting study reproducibility and encouraging its adoption in clinical settings. This innovative diagnostic tool holds the promise of transforming the landscape of brain tumor diagnosis, offering a more accurate, non-invasive, and patient-friendly alternative to traditional methods.

Source: Alonso Garcia-Ruiz et al, An accessible deep learning tool for voxel-wise classification of brain malignancies from perfusion MRI, Cell Reports Medicine (2024). DOI: 10.1016/j.xcrm.2024.101464

---

Like this article?  Keep up to date with AI news, apps, tools and get tips and tricks on how to improve with AI.  Sign up to our Free AI Newsletter

Also, come check out our free AI training portal and community of business owners, entrepreneurs, executives and creators. Level up your business with AI ! New courses added weekly.