AI Continues Revolutionising Cancer Diagnosis: iStar and Precision Pathology

In a significant advancement in medical imagine technology, researchers at the Perelman School of Medicine, University of Pennsylvania, have developed a state-of-the-art artificial intelligence tool named iStar (Inferring Super-Resolution Tissue Architecture). This AI imaging Architecture is set to bring new focus to the field of cancer diagnosis and treatment by offering unprecedented clarity in interpreting medical images.

iStar’s capabilities extend imagine methods, allowing clinicians to see both the minute details of both individual cells and the broader scope of gene operations. This dual perspective is crucial in diagnosing and treating various cancers, especially those that might go undetected with current technologies. iStar is its ability to identify anti-tumour immune formations and predict gene activities with near-single-cell resolution.

Prediction accuracy evaluation and gene-based segmentation comparison in the Xenium-derived pseudo-Visium mouse brain data

The above iStar images are made using hierarchical image feature extraction that integrates ST data and high-resolution histology images to predict spatial gene expression with super-resolution.

The tool’s high-speed processing is not just an obvious advantage for individual patient care but also promises to significantly impact large-scale biomedical studies. The team from the Perelman School of Medicine at the University envisage its potential application in 3D and biobank sample prediction, opening new avenues in medical research and precision medicine.

Funded by the National Institutes of Health, this innovative research has been published in Nature Biotechnology, hopefully suggesting a new era in precision pathology. You can read the full publication here.

Staff Writer

Our in-house science writing team has prepared this content specifically for Lab Horizons

Leave a Reply

Your email address will not be published. Required fields are marked *