Distinguishing spinal cord tumors from non-tumor conditions in advanced imaging

What is it about?

Spinal cord lesions can be challenging to detect, given the wide range of diseases. Each of the cord diseases has a unique imaging pattern on advanced imaging using magnetic resonance imaging (MRI). In this report, we present a collection of classic patterns in imaging to distinguish tumors from non-tumoral conditions of the spinal cord. By doing so, we can narrow down a patient's disease to a smaller number of possibilities to help doctors in disease treatment. The imaging patterns related to the cord diseases are not complicated and can be made easy to understand. Recent advances in artificial intelligence (AI) has been applied to advanced imaging. The AI work-flow, has made remarkable progress in image recognition tasks. By training AI to recognise a particular pattern of a spinal cord disease on MRI, we may in future, be able to design or programme it to distinguish tumoral from non-tumoral diseases of the spinal cord.

Featured Image

Why is it important?

AI is at its horizon and the possibilities are unimaginable. One of the possible applications is incorporating AI into advanced imaging including MRI. There are very few papers written on AI application in MRI of the spine. Recognizing the different spinal cord diseases by recognizing their classic patterns on imaging is vital for radiologists in spine practices. Incorporating AI in such practices may help radiologists in their daily work rather than replacing them.

Perspectives