16 September 2016
A newly developed computer program has shown potential to outperform human physicians in diagnosing brain cancer.
Researchers from Case Western Reserve University have developed an algorithm to identify radiomic features that discriminate between brain cancer and dead brain cells caused by radiation, called radiation necrosis, using routine follow-up MRI scans from 43 patients, including features that cannot be seen through an assessment with the human eye.
The program was shown to be nearly twice as accurate as two neuroradiologists in determining whether abnormal tissue seen on the MRI scans were dead brain cells caused by radiation necrosis, or if brain cancer had returned.
A total of 15 patients were assessed, with one neuroradiologist diagnosing seven patients correctly, and the second correctly diagnosing eight patients. By contrast, the computer program was correct in 12 of 15 cases.
As such, this could be a useful aid in helping neuroradiologists to improve their confidence in diagnosing a suspicious lesion.
Study leader Pallavi Tiwari, assistant professor of biomedical engineering at Case Western Reserve, said: "What the algorithms see that the radiologists don't are the subtle differences in quantitative measurements of tumour heterogeneity and breakdown in microarchitecture on MRI, which are higher for tumour recurrence."
Posted by Edward Bartel
Health News is provided by Axonn Media in collaboration with Spire Healthcare. Please note that all copy above is ©Axonn Media and does not reflect views or opinions of Spire Healthcare unless explicitly stated. Additional comments on the page from individual Spire consultants do not necessarily reflect the views or opinions of other consultants or Spire Healthcare.