Traditionally, pathologists diagnose cancer and rare diseases by looking for abnormalities in tumor tissue and cells under a microscope, however, that is a time-consuming process often prone to errors. With the advent of whole slide imaging, pathologists are slowly migrating to a fully digital workflow. Whole slide imaging is the process of digitization of tissue samples on glass slides using digital whole slide scanners, enabling clinicians in histopathology, immunohistochemistry, and cytology to view, manipulate, interpret and digitally store the digital images, optimizing their workflow. Digitization also allows pathologists to interpret images using computational approaches, with the potential to improve accuracy, reduce inter-observer variability and provide new insights from a patient’s biopsy. Ultimately digital and computational pathology workflows will help clinicians and researchers discover, diagnose and treat diseases like cancer faster.
Computational pathology requires meticulous preprocessing and algorithmic processing for interpretation, which represents a challenge for traditional hardware. Loading whole slide images from disk into memory and then processing the tiled image can be immensely time-consuming. Whole slide images are often very large in data file size with resolutions higher than 100,000 x 100,000 pixels.
The turn around time for most algorithmic diagnostic analysis solutions using the cloud can range from an hour to multiple days. When a cancer patient is in a hospital or nervously at home waiting for diagnosis results. Every hour and day is critical, and the cost to health systems and insurance companies can be upwards to tens of billions of dollars a year.
ALAFIA leverages a high-performance computing (HPC) architecture to accelerate widely used and highly parallel workloads and algorithms in digital, computational pathology and spatial biology. Using QuPath, an open source application for bioimage analysis. ALAFIA Supercomputers can execute a standard open source benchmark (QuMark) developed by Dr. Mark Zaidi, based on the QuPath open source application for bioimage analysis. QuMark leverages a publicly available WSI dataset by OpenSlide from an Aperio whole slide scanner (CMU-2) and performs annotations from a pixel classifier to detect, classify, and export cell measurements in record breaking time under 35 seconds.
Below is an example of the performance our ALAFIA supercomputer can deliver for operations reflective of what a pathologist or bioinformatician would typically perform in QuPath:
To learn more about how to accelerate and execute a digital pathology migration and leverage computational pathology tools, reach out to us to set up a private suite meeting overlooking the beautiful Seaport Harbor and see a demonstration of the real-time cancer diagnostic during the 114th Annual Meeting of the United States and Canada Academy of Pathology (USCAP) at the Boston Convention and Exhibition Center March 22-27.