The US Food and Drug Administration (FDA) has cleared BrainSpec’s magnetic resonance spectroscopy (MRS) platform, meaning clinicians will now have access to the artificial intelligence (AI)-powered software that measures brain chemistry.
BrainSpec’s Core platform uses non-invasive imaging to gather data on chemical concentrations in the brain. The US company says its software gathers signals and feeds them into AI models for insights.
MRS, the biological application of nuclear magnetic resonance spectroscopy, is conducted on the same machines as conventional magnetic resonance imaging (MRI). It has the ability to provide a virtual insight into brain chemistry to assist in diagnosing neurodegenerative conditions such as Alzheimer’s and multiple sclerosis, and diseases like brain tumours or the presence of epilepsy.
Whilst MRI produces anatomical images that reflect the magnetic properties of tissue, the spectra produced by MRS reveal chemical composition and potential changes to metabolic pathways in brain tissue.
BrainSpec says its platform is compatible with MRI scanners made by GE HealthCare, Siemens, and Philips who, according to a market model by GlobalData, occupy a combined 77.4% share of the MRI device market.
BrainSpec’s platform previously received FDA breakthrough device designation for estimating isocitrate dehydrogenase (IDH) status in glioma patients. The recent clearance expands its use for measuring brain metabolites for other brain indications, such as gliomas. The software’s quantitative measurement of 2-hydroxyglutarate – the oncometabolite synthesised from IDH mutation – could help with a speedier and more accurate diagnosis of tumours. According to the company, the platform shows 85% sensitivity and 90% specificity in detecting 2-hydroxyglutarate.
BrainSpec is joining the wave of AI software that are being implemented in the medical imaging sector. Though the benefits of using AI in this field are clear, some in the industry have said that further digital infrastructure is needed before its long-term roll-out is secure.
A 2023 report by GlobalData predicts that global revenue for AI platforms across healthcare will reach $18.8bn by 2027.
BrainSpec CEO and co-founder Alex Zimmerman said: “This [clearance] signals a new era of software enhancements that will not only complement but also speed up the complicated and lengthy process of diagnosis. Our aim is not to replace the critical role of physicians but to augment their expertise with invaluable information previously unavailable.”
In October, the FDA cleared a new AI tool from South Korean company VUNO that characterises brain structures from MRI scans.
