By accelerating the first step of spatial omics data analysis, a new artificial intelligence model was developed Children’s Hospital of Philadelphia provides detailed insight into how disease develops and progresses at the cellular level and can advance accurate diagnostics and targeted treatments.
CHOP’s open-source AI tool is now available in a public repository for non-commercial use, the hospital says.
WHY IT’S IMPORTANT
The pediatric researchers developed a deep learning biomedical imaging model, called CelloType, to accelerate the identification and classification of cells in tissue images. They then tested the biomedical imaging AI for a wide range of complex diseases, including cancer and chronic kidney disease.
CelloType is programmed to improve accuracy in cell detection, segmentation and classification, CHOP said, and is efficient at performing large-scale tasks such as natural language processing and image analysis.
Although the CHOP model requires training for segmentation and classification tasks, it can learn patterns and make predictions or classifications faster than previous approaches.
The researchers compared CelloType’s performance against models that segment multiplexed tissue images, including Mesmer and Cellpose2, and detailed their results from the National Institutes of Cancer-funded study in Nature methods.
“Unlike the traditional two-phase approach of segmentation followed by classification, CelloType adopts a multitask learning strategy that integrates these tasks while improving the performance of both,” they say in their report. report.
Certain cell types are large or irregularly shaped, posing challenges to conventional segmentation methods. CelloType, which uses transformer-based deep learning and automates the analysis of high-dimensional data, better captures complex relationships and context in tissue samples, they said.
CelloType uses AI to accurately outline objects in an image.
Kai Tan, the study’s lead author and a professor in CHOP’s Department of Pediatrics, said in a statement that the “approach could redefine the way we understand complex tissues at the cellular level, paving the way for transformative breakthroughs in health care.”
THE BIG TREND
There is an urgent need in spatial omics – a field that combines genomics, transcriptomics or proteomics with spatial information to map where different molecules are located in cells in complex tissues – for more advanced computational tools for data analysis, CHOP said.
Recent advances have led to the analysis of intact tissues at the cellular level, allowing unprecedented insights into the relationship between cellular architecture and functionality of various tissues and organs.
Using AI to improve the understanding of biomedical images can not only help doctors treat patients, but can also improve patients’ access to advanced imaging and even predict diseases such as cancer. That’s why healthcare systems are embracing AI imaging tools.
As researchers in Norway and Denmark use mammography images in national breast cancer screening programs to help predict diagnoses, Stamford Health’s Heart & Vascular Institute announced in October that its patients will be automatically screened for coronary artery disease during every CT scan without contrast on the chest. and when their future risk indicators are elevated.
“This tool increases our ability to detect early signs of cardiovascular disease and ensures patients receive the follow-up care they need to prevent serious health problems,” said Dr. David Hsi, head of cardiology and co-director of the institute, in a statement.
A senior physician and professor of pediatrics said he believes healthcare providers armed with AI and machine learning can turn the tide for patients battling complex diseases.
“Personalized genetic and epigenetic information can help tailor many medications to specific patients and specific diseases. All these omics involve vast amounts of data that information technology can now analyze in great detail and functionally assess through artificial intelligence and machine learning-derived data.” algorithms,” Dr. William Hay Jr., chief physician at Astarte Medical, a precision medicine company Healthcare IT news last year.
ON THE RECORD
“We are just beginning to unlock the potential of this technology,” Tan said in a statement.
Andrea Fox is editor-in-chief of Healthcare IT News.
Email: afox@himss.org
Healthcare IT News is a HIMSS Media publication.