A group of scientists together with doctors have presented an AI model that can accurately detect cancer. Such a breakthrough is expected to help the acceleration of cancer diagnosis and speed up treatment. Right now cancer is responsible for nearly 1/6 of deaths worldwide mostly because of the lack of technology to spot the disease in its early stages. Prompt detection and swift treatment of cancer at the very beginning of the illness, in many cases, results in full recovery.
The AI tool was created by experts at the Royal Marsden NHS Foundation Trust, the Institute of Cancer Research in London, and Imperial College London. It can identify cancerous abnormal growths found on CT scans. According to a study published in the Lancet’s eBioMedicine journal, the algorithm performs with greater efficiency than currently available methods.
Dr. Benjamin Hunter, a clinical oncology registrar at the Trust and a clinical research fellow at Imperial College, said, "In the future, we hope it will improve early detection and potentially make cancer treatment more successful by highlighting high-risk patients and fast-tracking them to earlier intervention."
During the study, the team used scans of about 500 patients with large lung nodules to develop an AI algorithm that would be able to accurately identify cancerous nodules.
The researchers used a measure called area under the curve (AUC) to see how effective artificial intelligence was at predicting cancer. The model was able to detect each nodule's risk of cancer with an AUC of 0.87. They also experimented with other currently used tests like the Brock score, which resulted in a 0.67 score, and the Herder score, which had an AUC of 0.83. Both tests indicated that the performance of the AI tool improved.
The AI is expected to also help doctors make swift decisions about patients with abnormal growths that are under the medium-risk at the moment. According to the study, combined with Herder, the AI model was able to identify high-risk patients in the research group and suggested early intervention for 18 out of 22 (82%) of the nodules that could be cancerous.
Despite its grand progress, the AI tool is still in the early stage of research, so naturally, it cannot be implemented into the healthcare system.
However, the potential of such technology and the overall AI use in clinics is clearly worthy of further study.
"Through this work, we hope to push boundaries to speed up the detection of the disease using innovative technologies such as AI," said the Libra study's chief investigator, Dr. Richard Lee.
Summing up, we can say that the opportunity to quicken the detection of cancer by helping to fast-track patients to treatment and streamlining the analysis of CT scans will utterly change the unfortunate statistics of cancer-cased deaths.
Undoubtedly, using AI for such an important matter is a great way to improve the reputation of this technology, especially after so many scientists and developers expressed their concerns about artificial intelligence development courses.
Even ordinary people start noticing the presence of AI chatbots that pollute the social media space with fake reviews and comments.
Moreover, to successfully develop AI models for the healthcare system and other industries we need to overcome the issue of enormous energy consumption. Overall, there are still a lot of things that scientists need to take care of and more experiments to conduct before we’ll be able to use the AI potential to its full extent.