AI Could Help Doctors Diagnose Lung Cancer Earlier
Researchers from The Royal Marsden NHS Foundation Trust, The Institute of Cancer Research, and Imperial College London have conducted a study that shows how Artificial Intelligence (AI) could help doctors diagnose lung cancer earlier. The study, named LIBRA and supported by several charities, used data from the CT scans of nearly 500 patients with large lung nodules to develop an AI algorithm, which was then tested to see if it could accurately identify cancerous nodules. According to the results published in the Lancet’s eBioMedicine, the AI model was able to identify each nodule’s risk of cancer accurately.
How Does the AI Model Help?
Lung nodules are abnormal growths that are common but mostly benign, but some can be cancerous, and large ones of 15-30mm in size are associated with the highest risk. The AI model which uses only two variables, compares favorably with other tests currently in use in clinics by using radiomics to extract information about the patient’s disease from medical images that the human eye could not easily see. As a result, the model could streamline and speed up nodule risk calculation, which would make it faster for clinicians to identify high-risk patients and recommend early intervention.
- Lung cancer accounts for just over a fifth (21%) of all cancer deaths in the UK and is the leading cause of cancer mortality worldwide.
- Recent data shows that over 60% of lung cancers in England are diagnosed at stage three or four, highlighting the need for initiatives to speed up detection.
- Using Herder, HScore is currently implemented as a test to categorize low-risk patients who score less than 10% and high-risk patients who score over 70%, while patients in the intermediate risk group of 10% to 70% can be considered for a broad range of tests or treatment options.
The study shows that AI algorithms, when combined with technological tools like radiomics, could help speed up lung cancer diagnoses by accurately identifying high-risk patients and recommending early intervention.
Early diagnosis of lung cancer is crucial in increasing the chances of survival, and this study has shown that AI could help in attaining this goal. The next step will be to test the technology on patients with a more significant number of large nodules and to develop it further for seamless integration into clinical practice.