AI technology develops early detection of glioma’s progress

Scientists from the University of Virginia School of Medicine knock the power of artificial intelligence to increase and accelerate the treatment of glioma, the most deadly brain cancer.

UVA BIJOY KUNDUM researcher and colleagues are developing an AI imaging approach to distinguish between tumor progress and brain changes caused by tumor treatment. This distinction may take months, leaving doctors uncertain whether the tumor grows and stops important care decisions.

The AI Kunduj approach already exceeds the standard clinical option when pre -testing. Presented for the test in 26 patients with glioma immediately after treatment, artificial intelligence was able to correctly distinguish in 74% of cases.

The aim of the project is to train additional patient data and increase this accuracy by more than 80% for clinical use. “

Dr. Bijoy Kundu, UVA Cancer Center, Department of UVA Health Radiology and medical imaging and UVA biomedical engineering department

This can bring real benefits to patients. “Early distinction would allow prior modifications to the treatment of cancer recurrence in brain cancer patients,” said David Schiff, MD, some of the branches of neurology, neurology and UVA medicine. Schiff is also a co-director of the UVA Health Neuro-Ononology Center.

Better glioma care

The glioma is more than half of all primary brain tumors. It is a highly aggressive and rapidly developing RAK-Typical experience from the diagnosis is only 15 months. This makes extremely important that doctors work quickly. In this way, it can extend the survival time and improve the quality of life of patients.

Now, however, doctors must wait three to four months after treatment to assess the progress of tumor. They do it using magnetic resonance imaging (MRI) or in some cases of brain surgery.

Kunduju approach combines MRI with a different form of imaging, dynamic PET (positron emission tomography). This gives sophisticated, multi -dimensional views in the brain that artificial intelligence can analyze – everything without the need to cut in a skull.

Kundu and his colleagues received USD 90,000 with Ivy Biomedical Innovation Fund UVA to develop and improve their approach. They will use this money to improve the accuracy of deep learning algorithms, fundamentally teaching AI to better distinguish the symptoms of tumor progression from the effects of chemotherapy and radiation.

They hope that their work will ultimately help doctors get the information needed earlier, improving the care of patients with glioma.

“We hope that this work helps patients and families get answers faster. If our artificial intelligence can give doctors more confidence, this may mean faster treatment decisions and better results,” said Kundu. “Our goal is to provide doctors with better tools so that they can focus less on guessing and more care. We are still at an early stage, but even now our approach turns out to be a real promise. We are working on the future in which patients receive brightness faster and where this transparency helps save lives.”

The latest UVA cancer research

Finding new ways to improve patient care is the basic mission of both UVA Cancer Center and UVA Paul and Diane Manning Institute of Biotechnology. The UVA Cancer Center is one of only 57 cancer centers in the country designated by the National Cancer Institute for exceptional patient care and the most modern research on cancer.

Meanwhile, the Manning Institute was launched to accelerate the development of new treatment and medicine for the most difficult diseases. This will be supplemented by a national clinical research network, which extends access to potential new treatment methods during their development and testing.

The UVA Biomedical Engineering Department is a joint program of the School of Medicine and School of Engineering and Applied Science.