Unlock the Potential of AI in Spatial Omics

How the Model Works

The hospital's open-source AI tool, CelloType, is now accessible in a public repository for noncommercial use. Pediatric researchers developed this deep learning-enhanced biomedical imaging model to speed up the identification and classification of cells in tissue images. It has been tested across a wide range of complex diseases such as cancer and chronic kidney disease.

CelloType is programmed to enhance accuracy in cell detection, segmentation, and classification. It is highly efficient in handling large-scale tasks like natural language processing and image analysis. While it requires training for segmentation and classification tasks, it learns patterns and makes predictions or classifications faster than previous approaches.

Comparing with Other Models

The researchers compared CelloType's performance against models that segment multiplexed tissue images, including Mesmer and Cellpose2. In their report published in Nature Methods, they detailed the results of this National Institutes of Cancer-funded research. "Unlike the traditional two-stage approach of segmentation followed by classification, CelloType adopts a multitask learning strategy that integrates these tasks, simultaneously enhancing the performance of both," they stated.

Conventional segmentation methods face challenges with certain cell types that are either large or of irregular shape. CelloType, which utilizes transformer-based deep learning and automates the analysis of high-dimensional data, better captures the complex relationships and context in tissue samples.

The Larger Trend in Spatial Omics

There is a growing need in the field of spatial omics for more sophisticated computational tools for data analysis. Recent advancements have enabled the analysis of intact tissues at the cellular level, providing unparalleled insights into the link between cellular architecture and the functionality of various tissues and organs.

Using AI to improve the understanding of biomedical images is not only beneficial for clinicians in treating patients but also enhances patient access to advanced imaging. It even has the potential to predict diseases like cancer. As a result, health systems are increasingly embracing AI imaging tools.

For example, in Norway and Denmark, researchers are using mammography images in national breast cancer-screening programs to predict diagnoses. Stamford Health's Heart & Vascular Institute announced in October that its patients will automatically receive coronary artery disease screening during non-contrast chest CT scans when their future risk indicators are elevated. "This tool enhances our ability to detect early signs of cardiovascular disease and ensures that patients receive the follow-up care they need to prevent serious health outcomes," said Dr. David Hsi, chief of cardiology and the institute's co-director.

The Impact on Healthcare

One chief medical officer and pediatrics professor believes that with the help of AI and machine learning, healthcare providers can make a significant difference in treating patients with complex diseases. "Personalized genetic and epigenetic information can help tailor many medications to specific patients and diseases. All of these omics involve huge amounts of data that information technology can now analyze in exquisite detail and assess functionally through artificial intelligence and machine learning-derived algorithms," Dr. William Hay Jr., chief medical officer at Astarte Medical, told Healthcare IT News last year.

As Tan said in a statement, "We are just beginning to unlock the potential of this technology." The future holds great promise as this AI model continues to revolutionize spatial omics data analysis and its applications in healthcare.