AI software could advance voicebox cancer treatment

Artificial intelligence improves outcomes for voicebox cancer patients, a new study has found, moving us one step closer to personalized treatment.

Experts at the University of Newcastle have discovered that artificial intelligence software could play a revolutionary role in guiding treatment of voice box cancer.

In a first-of-its-kind study into advanced laryngeal cancer (voicebox cancer), Amar Rajgol and his team at the University of Newcastle are harnessing the power of artificial intelligence (AI) to extract medical information from CT scans. The images were analyzed. The purpose is to predict survival.

Laryngeal cancer is a tough battle for many people in the UK, with around 2,400 new cases diagnosed each year, half of whom are at an advanced stage and do not survive for more than three years. But the research led by Rajgaard offers hope that patient care and treatment will improve.

“This allows us to maximize its potential.”

The AI ​​software works pixel by pixel, revealing hidden patterns within the tumor that are invisible to the human eye. This acts as a “virtual biopsy” and provides clues about the behavior of the cancer and the extent of the abnormality. This new way of looking at images in such detail is called radiomics.

Rajgaard, an ear, nose and throat specialist, spoke about his findings. “Radiomics can be thought of as a super-powerful magnifying glass for medical images such as CT scans. Radiomics carefully examines every detail that humans can't see or miss. You can spot patterns and irregularities that could be confusing.”

This study Journal of Laryngology and Otology, two important imaging markers were revealed: shape compactness and gray level nonuniformity (GLNU). High levels of these markers indicate that the patient is more likely to die early. An increase in shape compactness amplified the risk of death almost threefold, whereas an increase in GLNU doubled the risk.

The results also revealed that these imaging markers work better at predicting survival compared to traditional factors such as age and cancer stage that doctors currently use.

Although many studies have been conducted analyzing all head and neck cancers, these studies are not robust and clinically relevant because all head and neck cancers behave very differently. Also, from the clinician's point of view, he should not be combined into one group for analysis. .

Rajgaard added: “These developments are very exciting because this research could play a major role in guiding treatments and delivering precision medicine in the future. “There is also the possibility of ensuring that patients receive appropriate treatment based on this,” he added. We hope this will help patients make more informed decisions about their treatment plans.

“Another benefit is that this method enhances, rather than alters, patient pathways by analyzing medical images in a way that humans cannot.Currently, much of the information from scans is insufficient is underutilized, but this will enable us to realize its full potential.

“In the future, we could also use this technology to track how well patients are responding to treatment by looking at scans over time.”

“Voicebox has the potential to revolutionize cancer management”

Over the past 30 years, little progress has been made in improving survival rates for voicebox cancer patients. But a partnership between Newcastle University and Newcastle upon Tyne Hospitals NHS Foundation Trust brings hope to those battling laryngeal cancer. For this collaboration, Amar Rajgor was awarded a prestigious National Institute for Health Research (NIHR) Doctoral Fellowship.

Rajgaard concluded: “Further research will be carried out in this area. Following my award from the NIHR, we are now expanding this study to over 250 patients.” .

“We aim to create a risk prediction model that takes into account both survival and response to specific cancer treatments. This prototype model will include clinical factors, pathology details, and imaging markers. This Voicebox has the potential to truly revolutionize cancer management.”

Amar Rajgor's systematic review highlights the pitfalls of the current literature, and these are discussed in this study.