AI Doubling Success Rate in Grading Rare Cancer Severity


In recent strides forward in the medical field, studies have indicated that artificial intelligence has nearly doubled the success rate in grading the severity of a certain type of rare cancer from scans. This intelligent system significantly outperforms the traditional method, displaying an accuracy rate of 82% versus the previous 44% for lab analysis.

This intelligent technology has the capacity to identify minute details that easily escape the human eye, hence, its superior results. Research experts from the Royal Marsden Hospital and Institute of Cancer Research are enthusiastic about the implications this could have for cancer treatment. An enhanced ability to accurately grade the aggressiveness of such cancers could drastically improve the effectiveness of care and treatment deputed to thousands of individuals annually.

Moreover, this innovative advancement is not limited to use in a single type of cancer. Scientists are excited about the potential it holds in early detection of other types of cancer as well. The promise AI technology displays in diagnosing breast cancers and shrinking treatment times is just the tip of the iceberg.

With artificial intelligence, large volumes of information can be imparted to computers, which they then utilize to identify patterns which aid in problem-solving, making predictions, and even learning from previous mistakes.

Renowned Professor Christina Messiou, a consultant radiologist at The Royal Marsden NHS Foundation Trust and professor in imaging for personalised oncology at The Institute of Cancer Research, London, shared her excitement. “We’re incredibly excited by the potential of this state-of-the-art technology. It could lead to patients having better outcomes, through faster diagnosis and more effectively personalised treatment.”

In a report published in Lancet Oncology, the researchers employed a technique called radiomics. This methodology enabled them to identify signs of retroperitoneal sarcoma, which develops in the connective tissue of the back of the abdomen, through scans of 170 patients. The signs were previously invisible to the human eye.

Empowered with this critical data, the AI algorithm successfully graded the severity of tumours in 89 patients from various European and US hospitals. This AI grading system proved to be remarkably more accurate than biopsy results, wherein a minuscule portion of the cancerous tissue is analyzed under a microscope.

In the case of Tina McLaughlan, a dental nurse who was diagnosed with a sarcoma at the back of her abdomen in June of last year, a CT scan was crucial in identifying the problem. Doctors determined that a needle biopsy was too risky. The 65-year-old from Bedfordshire had her tumour removed and is now under continual monitoring at the Royal Marsden with scans every three months.

Although McLaughlan herself was not part of the AI trial, she expressed positive sentiments about the implications it would have on other patients. Tina stated, “You go in for the first scan and they can’t tell you what it is – they didn’t tell me through all my treatment, until the histology, post-op, so it would be really useful to know that straight away. Hopefully, it would lead to a quicker diagnosis.”

Around 4,300 individuals in England are diagnosed with this type of cancer each year. Professor Messiou has high hopes that this technology would be used globally in the future, leading to specific treatments for high-risk patients, and sparing those at low risk unnecessary treatments and follow-up scans.

Dr Paul Huang, from the Institute of Cancer Research, London, shared, “This kind of technology has the potential to transform the lives of people with sarcoma – enabling personalised treatment plans tailored to the specific biology of their cancer. It’s great to see such promising findings.”


Please enter your comment!
Please enter your name here