The researching team at Children’s Hospital of Philadelphia (CHOP) has officially announced the creation of a new AI technology called CelloType, which happens to be a spatial omics model designed toaccurately identify and classify cells in high-content tissue images.
To give you some context, spatial omics is a field of study that combines molecular profiling, such as genomics, transcriptomics or proteomics, with spatial information to map where different molecules are located within cells in complex tissues. In essence, the stated concept delivers at your disposal important, detailed insights in regards to how a disease develops and progresses at the cellular level, thus aiding the advancement of precise diagnostics and targeted treatments.Â
Now, in connection with the given development, we must also mention that, as the critical first step of spatial omics data analysis, researchers are required to facilitate cell segmentation (identifying cell boundaries) and classification (calling cell types).Another detail making this development important is a surge in spatial omics data, something which has spurred a need for more sophisticated computational tools for data analysis, prompting CHOP to go ahead and introduce its latest brainchild.
“We are just beginning to unlock the potential of this technology,” said Kai Tan, PhD, the study’s lead author and a professor in the Department of Pediatrics at CHOP. “This approach could redefine how we understand complex tissues at the cellular level, paving the way for transformative breakthroughs in healthcare.”
Talk about the new model on a slightly deeper level, it leverages a type of AI in the form of transformer-based deep learning. Here, deep learning automates the analysis of high-dimensional data, and therefore, makes it possible for the model to capture complex relationships and context. More on the same would reveal how technology boasts a marked amount of efficiency when it comes to handling large-scale tasks like natural language processing and image analysis, as well as subsequently learning patterns and making predictions or classifications. Not just that, it is also programmed to improve accuracy in cell detection, segmentation, and classification.
In the build up to the solution’s launch, Tan and his team analyzed how CelloType performed compared with a range of traditional methods using animal and human tissue datasets. During this analysis, unlike the typical two-stage approach which involves segmentation followed by classification, CelloType went for a multi-task learning strategy that was more efficient, considering it simultaneously integrated segmentation and classification.
As a result, it outperformed existing segmentation methods on various types of images, including natural images, bright light images, and fluorescence images.
Turning our attention towards cell type classification, it saw CelloType surpassing a model comprised of state-of-the-art methods for individual tasks and a high-performance instance segmentation model, which uses AI to precisely outline objects in an image.
 Furthermore, the researchers deployed a multiplexed tissue image, which would be an advanced biomedical image that displays multiple biomarkers within a single tissue sample to basically showcase how CelloType can be used for multi-scale segmentation and classification of both cellular and non-cellular elements in a tissue. Going by the available details, their technology was able to expedite the process of identifying and separating different size tissue elements within an image, allowing detailed analysis of both small and large cell structures.
Beyond the current setup, CHOP is also currently a collaborator in high profile projects, such as the Human Tumor Atlas Network, the Human BioMolecular Atlas Program (HuBMAP), and the BRAIN initiative, that use similar technologies to map spatial organizations of various types of healthy and diseased tissues.
“Our findings underscore the increasingly pivotal role technology plays in today’s biomedical research,” said Tan, who is also investigator in the Center for Childhood Cancer Research at CHOP. “CelloType advances spatial omics by providing a robust, scalable tool for analyzing complex tissue architectures, thereby expediting discoveries in cellular interactions, tissue function and disease mechanisms.”