Danniell Domingo

In an operating room where tensions run high and every second counts, doctors are left with a highly important decision that can have a consequence of a lifetime for their patients.


Luckily, with the emergence of artificial intelligence (AI), researchers have innovated Sturgeon, an AI-powered tool designed to analyze and classify tumors found in the central nervous system (CNS), which can speed up doctors’ decision-making.

Sturgeon has the power to produce highly accurate diagnosis of tumors, helping doctors to make a decision in a short amount of time while the brain is exposed. 

While it is invasive to cut open the skull and get a tissue sample from the brain, Sturgeon’s accuracy in classifying subtypes of tumors further reduces the risk of misdiagnosis and neurological comorbidities.

The research was developed by 14 doctors, which started in 2022 when molecular pathologist Bastiaan Tops saw bioinformatics professor Jeroen de Ridder use a nanopore sequencer, an affordable device that reads strands of DNA for tumor identification.

“We sat together, and we started brainstorming what that might entail,” de Ridder, the project leader, said on integrating the nanopore sequencer with an algorithm-based machine.

Sturgeon, at its core, relies on machine learning which works by banking on a heap of datasets of tumor samples to produce an outcome–a diagnosis of what subtype the tumor is.

However, since data on tumor samples are limited and sparse, the AI has been trained to overcome this challenge by looking at the methylation patterns in the patient’s cells.

Assessing the methylation patterns can provide a better picture of the tumor cells, such as its origin, type, and prognosis, the likely outcome of a disease.

In just 40 minutes, Sturgeon accurately classified 90% of the 50 tumor samples on its trial, but when tested real-time, the AI tool made a correct diagnosis of 72% of the 25 brain surgeries, which took 90 minutes or less only.
 
It’s an ample time already to help doctors make quicker decisions during critical moments.

The AI tool helps doctors in deciding whether to perform a surgery to remove the tumor, or opt for non-invasive treatments like radiation or chemotherapy.

Currently, researchers at the Princess Maxima center are doing clinical trials to know more of Sturgeon’s capabilities and ensure it is ethically safe for medical regulation.

“What we have now uniquely enabled is to allow this very fine-grained, robust, detailed diagnosis to be performed already during the surgery,” de Ridder said.