Artificial Intelligence proves superior at detecting spread of neck and throat cancer
Artificial Intelligence algorithm may help better guide oropharynx cancer treatment
CANCER DIGEST – April 21, 2023 – An artificial intelligence algorithm used to analyze MRI images of patients with cancer of the throat and neck proved better at detecting the spread of cancer to nearby lymph system and nearby tissues than expert radiologists.
The study showed the potential for using AI for improving the accuracy of cancer diagnoses. The researchers published their study online in the April 21, 2023 The Lancet Digital Health.
The spread of cancer from the throat and neck, or extranodal extension (ENE), is a common complication of oropharyngeal carcinoma leading the cancer to spread beyond the lymph nodes and into surrounding tissue. It can have a significant impact on a patient's prognosis and treatment plan, making accurate detection crucial for effective management of the disease.
In the study, the researchers trained an AI algorithm to analyze MRI images of patients with oropharyngeal carcinoma and identify cases of ENE. They then compared the AI's performance to that of expert radiologists and found that the algorithm was highly accurate in detecting ENE, with a sensitivity of 95.7% and a specificity of 91.3%, meaning the AI detected the ENE 95.7% of the time and the lesions identified by the AI did turn out to be cancer 91.3% of the time.
“What was important about this study is it tested the algorithm in the context of a very large randomized clinical trial, where patients who were enrolled, by definition, were supposed to be screened out for having ENE, and yet a significant portion still ended up having ENE,” said lead author Benjamin Kann, MD of Brigham and Women's Hospital, where the study was conducted.
“When we applied the algorithm to this population to see how it would have done in in predicting ENE, we found that it performed well with a high degree of accuracy — better than all four expert head and neck radiologists," Kann added.
The study is one of several recent examples of the growing use of AI in cancer research and treatment. AI has shown promise in a range of applications, from diagnosing cancer to developing personalized treatment plans and predicting patient outcomes.
The findings of the study hold implications for the diagnosis and management of oropharyngeal carcinoma, as well as for the broader field of cancer research and treatment. As AI continues to advance, it may play an increasingly important role in improving cancer care, treatment planning guidance, and ultimately better outcomes for patients.
This article was generated in part by ChatGPT and edited using source materials from The Lancet Digital Health, and Brigham and Women's press release.