Samsung Electronics Co Ltd (KRX: 005930) Medical Center in Seoul has pioneered a life-saving model of artificial intelligence.
Major Korean hospitals have created deep learning tools that can accurately predict the likelihood of recurrence of lung cancer in patients following surgery. You can divide them into different risk categories and determine the likelihood of illnesses returning from a year ago. High-risk groups have only 10% chances of returning status, 5% in the intermediate cohort, and 1% in the low-risk group.
Radar care research (initial stage non-small cell lung cancer in a comprehensive, real-time risk-adapted surveillance, comprehensive strategic AI model) was published in late July and was under the radar until this month. The study analyzed data from over 14,000 subjects who underwent the procedure between 2008 and 2022.
“In this proof-of-concept study, radar scores successfully predicted disease recurrence one year ago in patients with early stage NSCLC who underwent surgical resection with curative intent,” the authors concluded.
They also argued that previous attempts to achieve the same results were inferior because of the small datasets used by other researchers.
“Previous prediction models rely on static and single time point data and did not incorporate temporal or longitudinal information,” the investigator noted, adding that the distinction could potentially allow the tool to be applied more robust and clinically to risk prediction.
Deep learning is a subset of machine learning AI to utilize multi-layered neural networks to analyze large, diverse and complex data sets. It aims to mimic the complex decision-making capabilities of the human brain.
read more: Breath Diagnosis employs cutting-edge mass spectrometer devices for lung screening
limit
Medical experts say despite the strengths of research, they also have weaknesses.
“This model only contains CT imaging interpretation data from the radiologist, not the original image data,” he explained.
Furthermore, the authors state that deep learning models need to be assessed more extensively using multicenter and multinational data. In consideration, they determined that follow-up studies were guaranteed.
Samsung Medical Center employs more than 1,200 doctors and completes approximately 160 complicated surgeries every day. It is the largest comprehensive cancer centre in the country and is renowned for its highly advanced oncology expertise.
The health facility ranked 18th in Newsweek's list of world's best smart hospitals this year.
read more: Breath Diagnostic Technology Achieves a Breakthrough in Pneumonia Prediction in Peer Review Research
Like Mugglehead on Facebook
Follow Rowan Dunne on LinkedIn
rowan@mugglehead.com
