AI Tools Accurately Predict Tumor Re-Growth in Cancer Patients | Cancer

Doctors and scientists have developed artificial intelligence tools that can accurately predict how likely a cancer patient’s tumor will grow after being treated.

Breakthroughs described by clinical oncology as “exciting” can revolutionize patient surveillance. Although recent advances in treatment have increased the chances of survival, there remains the risk that the disease may recur.

Monitoring patients after treatment is essential to ensure that cancer recurrence is urgently addressed. But now, doctors tend to have to resort to traditional methods, including methods that focus on the amount and spread of the original cancer, to predict how the patient will be carried in the future.

Currently the world’s first study by the British Royal Marsden NHS Foundation Trust cancer Research, London, and Imperial College London used machine learning (a type of AI) to predict the risk of cancer recurrence and identify better models than existing methods.

“This is because AI can be used to understand which patients are at greatest risk of cancer recurrence so that this recurrence can be detected earlier and retreatment more effective. This is an important step in the process, “said Dr. Richard Lee, a consultant doctor. Responsible for respiratory medicine and early diagnosis at the Royal Marsden NHS Foundation Trust.

Lee, Principal Investigator for OCTAPUS-AI Research, is essential to guardians not only to improve the outcome of cancer patients, but also to reduce their fear of recurrence of the “major cause of anxiety” for many. He said he could prove that. “We want to push the boundaries to improve the care of cancer patients, help them live longer, and reduce the impact of illness on their lives.”

AI tools can be used to detect recurrence earlier in patients considered to be at high risk and receive treatment more urgently, but unnecessary follow-up scans and outpatient visits to patients considered to be at low risk. May be less.

“Reducing the number of scans required in this setting can help, reduce radiation exposure and hospital visits, and make more efficient use of valuable NHS resources,” Lee said. I am saying.

In a retrospective study, doctors, scientists, and researchers developed a machine learning model to determine if they could accurately identify patients with non-small cell lung cancer (NSCLC) who were at risk of recurrence after radiation therapy. Machine learning is a type of AI that allows software to automatically predict outcomes.

Lung cancer is the leading global cause of cancer deaths, accounting for just over one-fifth (21%) of cancer deaths in the United Kingdom. NSCLC accounts for almost five-sixths (85%) of lung cancer cases, and if detected early, the disease is often cured. However, more than one-third (36%) of NSCLC patients experience recurrence in the United Kingdom.

The researchers used clinical data from 657 NSCLC patients treated in five UK hospitals to feed the model and added data on various prognostic factors to indicate the potential for patient recurrence. I predicted more accurately.

These included the patient’s age, gender, body mass index, smoking status, radiation therapy intensity, and tumor characteristics. The researchers then used an AI model to classify patients into low-risk and high-risk recurrence, time to recurrence, and overall survival of 2 years after treatment.

This tool turned out to be more accurate in predicting results than traditional methods.Results of research supported by Royal Marsden Cancer Charity and National Institutes health The study was published in The Lancet’s eBioMedicine Journal.

“Currently, there is no clear framework for monitoring patients with non-small cell lung cancer after radiation therapy treatment in the UK,” said Dr. Sumeet Hindocha, clinical oncology specialist registrar at Royal Marsden University in London and Imperial College London. .. “This means that patients receive varying types and frequencies of follow-up … Using AI in healthcare data may be the answer.

“Because of the easy access to this type of data, this methodology can be replicated to different healthcare systems.”

Hindocha added that the study is an “exciting first step” in deploying tools to guide post-treatment monitoring of cancer patients nationally and internationally.