Advertisement

Scientists from Hong Kong, Beijing, South Korea create AI tool to predict brain cancer prognosis

  • Predictions can shed light on more precise treatments for patients to help improve survival rate, according to team led by Hong Kong University of Science and Technology
  • Scientists say AI-powered technology can accurately predict tumour recurrence and identify patients facing higher risks

Reading Time:2 minutes
Why you can trust SCMP
1
Researchers Chai Ruichao (left), Wang Jiguang and  Mu Quanhua. Photo: Elson LI

Researchers from Hong Kong, Beijing and South Korea have created an AI-powered tool and a web portal that can predict the prognosis for brain cancer patients, while providing publicly accessible information on the aggressiveness of individual cases.

The team led by the Hong Kong University of Science and Technology on Tuesday said the predictions could shed light on more precise treatments for patients to help improve their survival rate.

“The current treatment can usually only prolong patients’ overall survival time for around three months, and they almost inevitably suffer a recurrence,” said team leader Professor Wang Jiguan from the university’s division of life science.

Advertisement

“The recurrences of gliomas usually lead to a more aggressive and complex situation. The tumour cells in recurrence are also quite different from the ones in primary occurrences, which poses a great challenge to clinical treatment. ”

Glioma is a type of tumour that starts in the brain or spinal cord. Malignant diffuse gliomas are the most common primary brain tumours in adults.

Advertisement

The research – carried out by HKUST, Beijing Tiantan Hospital, Prince of Wales Hospital at Chinese University and the Samsung Medical Centre in South Korea – identified early predictors of glioma progression, such as the role of MYC gene amplification and CDKN2A gene deletion in accelerated tumour growth and increased risk of recurrence.

Advertisement
Select Voice
Choose your listening speed
Get through articles 2x faster
1.25x
250 WPM
Slow
Average
Fast
1.25x