Khaberni - An international team led by researchers from St. Jude Children's Research Hospital in the United States has developed an innovative artificial intelligence algorithm called M-PACT (M-PACT), capable of diagnosing and classifying children's brain tumors with an accuracy of up to 92%, by analyzing only drops of cerebrospinal fluid, without the need for surgical intervention.
Published results of the study in the British journal Nature Cancer, where researchers explained that the technology relies on analyzing free tumor DNA in cerebrospinal fluid and detecting molecular "methylation" patterns using a deep neural network supported by artificial intelligence.
A less invasive alternative to surgery
Traditional diagnosis of brain tumors relies on taking surgical biopsies from tumor tissues, a procedure that may involve significant risks, especially in children. However, the new algorithm allows for the precise molecular identification of the tumor, monitoring its response to treatment, and even detecting any potential recurrence, through what is known as "liquid biopsy".
Researchers noted that the model was trained and tested on hundreds of samples from pediatric patients and proved highly capable of distinguishing between different types of central nervous system tumors, as well as differentiating between tumor recurrence and the emergence of a new tumor.
The lead author of the study, Dr. Paul Northcott, director of the Center of Excellence in Neuroscience Oncology (CENOS) and a member of the Department of Developmental Neurobiology says, "This is a new generation of analyses and the computational framework that we have refined and applied to a wide range of pediatric brain tumor patients. M-PACT aims to elevate liquid biopsy to a new level in the field of pediatric neuro-oncology, and to apply this technology in many different clinical cases."
International research collaboration
Research into the algorithm involved scientists from several international institutions, including the German Cancer Research Center and the Hopp Children’s Cancer Center Heidelberg, as part of a scientific collaboration aimed at developing more precise and less harmful diagnostic tools for children.
Promising therapeutic prospects
The study's contributors believe that this technology could represent a qualitative shift in the field of pediatric tumors, as it opens the door to precision medicine based on the molecular characteristics of the tumor, with the possibility of continuously monitoring the disease during and after treatment, without subjecting patients to painful or complicated procedures.
Researchers hope that the results will pave the way for the clinical adoption of this tool in the near future, following the completion of the necessary clinical trials.
