Our team has contributed to a SCOPE magazine article with a focus on AI-driven auto-contouring for tumour segmentation. Here’s a look at the key conclusions.
Gamma Knife radiosurgery has been regarded as one of the most effective and advanced tools for treating a range of brain tumours and vascular malformations since it was first pioneered in the late 1960s. The technology has continued to evolve, and as we stand on the threshold of a new era of artificial intelligence, exciting possibilities are emerging.
These new developments were recently highlighted in an article published in the spring edition of SCOPE magazine, with contributions from our clinical team: Hannah, Radiosurgery Physicist at Thornbury Radiosurgery Centre, and Ian Paddick, Consultant Physicist at Queen Square.
The article has a particular focus on AI-driven auto-contouring for tumour segmentation, which allows for even greater precision and more consistent patient outcomes in stereotactic radiosurgery. Here’s a look at the key conclusions of the article.
A significant breakthrough in Gamma Knife technology that is highlighted in the article is AI-driven auto-contouring for tumour segmentation. This involves the automation of tumour delineations from MRI scans.
When this process is carried out manually, it is time consuming and subject to the variables of human interpretation, particularly in tumours or lesions without clearly defined boundaries. Clinicians are now able to use AI frameworks to segment tumour boundaries with the equivalent accuracy to human annotators. This has advantages including:
Reducing incidences of human error: AI automation maintains a consistent approach, which potentially lowers the risk of inaccurate results. Human error or variations in interpretation of the same data is impossible to fully mitigate against, and machine learning can significantly reduce variability and oversights.
Enhancing patient outcomes: The greater accuracy of the tumour segmentation afforded by AI tools can lead to the more precise delivery of radiation to the tumour cells. This reduces the risk of damage to the surrounding healthy brain tissue, and therefore also reduces the risk of side effects and complications.
Time efficiency: Manually segmenting tumour contours, particularly in the case of brain metastasis when multiple micro-tumours may be present, is a very time intensive task. Automated tools are highly sensitive and capable of detecting contours in much shorter timescales.
This will potentially allow for faster overall treatment times, reducing waiting lists. This could have significant benefits for patients and their families, reducing anxiety and uncertainty, and allowing patients to have potentially life saving or life extending treatment sooner.
However, the article notes that while an extremely valuable tool, AI models for auto-contouring still have some limitations and drawbacks to overcome. These include the limitations in training data for AI models, which may not generalise across diverse populations, and imaging noise leading to false positives.
For the foreseeable future, clinical oversight and judgement of the contouring process will remain necessary, particularly for complex cases such as brain metastasis.
Gamma Knife surgery is currently being used as part of ongoing research into treating conditions beyond brain tumours and vestibular schwannomas. For example, studies have shown that it is effective in treating essential tremor and Parkinson’s disease, with a low risk of side effects.
Stereotactic radiosurgery is also being explored as an option for treating various psychiatric disorders, in particular obsessive compulsive disorder in patients who have not responded to conventional treatments.
Research is ongoing into the use of AI for optimising biological effective dose (BED). This will allow for the delivery of the most accurate and effective radiation dose and exposure times for targeting the tumour cells, while minimising the risk of damage to the surrounding areas. This is crucial to the success of the overall treatment.
AI tools are trained by being exposed to vast amounts of data, which would be impossible for humans to process. This enables medical teams to put together more personalised treatment plans, which are more precisely tailored to the individual patient. Ultimately, this will potentially lead to better patient outcomes.
Tools such as augmented segmentation enable medical teams to make faster and more well-informed decisions, which can accelerate the progress of potentially life saving or extending treatment.
To sum up, Gamma Knife radiosurgery has always been a leading edge method of treating a range of brain tumours and other neurological conditions. Now, as a new chapter opens up with the involvement of AI technology, treatment planning and patient care is becoming more accurate, personalised, and effective than ever before.
Centres of Excellence for Stereotactic Radiosurgery treatment of complex Brain Tumours
Want to know more about Gamma Knife Treatment?
Our friendly staff are here to help you, get in touch with them today
