AI is being used to pre-empt risk for colon cancer
Posted: 2 March 2017 | By Darcie Thompson-Fields
Artificial intelligence has made some great developments toward speeding up cancer diagnosis so far in 2017. Last month it was announced that AI from Sophia Genetics was helping to accelerate patient diagnosis across Latin America. Earlier this year researchers at Stanford University developed a deep learning algorithm that can analyse skin cancer as accurately as a human doctor.
Now, Israel-based company, Medial EarlySign has announced the ability of its AI tool to identify the top 1% at highest risk of undiagnosed colorectal cancer (CRC).
The machine learning developer announced the first-year results of its implementation with Maccabi Healthcare Services (MHS), for ColonFlag, a tool developed in collaboration with MHS to identify individuals with a high probability of having CRC.
ColonFlag uses machine learning algorithms crossed with big data to detect the likelihood of disease for subpopulations.
Medial EarlySign worked with integrated delivery network Maccabi to develop the tool. Maccabi fed its members’ electronic medical records into the ColonFlag tool to identify the top 1% at highest risk of undiagnosed CRC. All of the members analysed by the system were not up to date with their CRC screening. Once flagged, individuals are referred for further evaluation.
“Despite the high survival rates associated with the early detection of colorectal cancer, one-third of patients eligible for screening are not being addressed,” said Professor Varda Shalev, M.D., M.P.A., Director, Institute for Health Research and Innovation for Maccabi Healthcare.
“Medial EarlySign’s algorithms, based on machine learning, have enabled us to identify those patients at the highest risk of harbouring cancer earlier and prioritise resources accordingly.”
Over the first 12 months of implementation, the tool evaluated almost 80,000 individuals, of which 690 were flagged as those at highest risk for CRC. Of the 690 patients identified, 220 colonoscopies were performed and 42% had findings, including 20 cases (10%) of cancer, and further 20 individuals who had either intermediate or high-risk adenomas according to the company.
“These results with Maccabi illustrate the immense potential of implementing Medial EarlySign’s technology to accurately indicate subpopulations at risk of a range of serious health conditions including cancers, metabolic disorders and chronic diseases,” said Ori Geva, CEO, Medial EarlySign.
“While traditional systems react to visible symptoms, our prediction-based solution enables providers to identify patients at risk of harbouring or developing life-threatening illnesses, for earlier intervention and treatment. Colorectal cancer is just one of several potential clinical outcomes that our game-changing platform addresses.”