Diagnostic company Droplet Biosciences announced on Tuesday that it has entered into a collaboration with NVIDIA to utilize the chipmaker's AI infrastructure for speeding up the delivery of post-operative cancer test results.
The company has been using NVIDIA Parabricks, a GPU-accelerated software suite, to significantly accelerate the analysis of genomic data from DNA sequencing.
Droplet stated that its testing method can detect minimal residual disease within 24 hours by analyzing lymph fluid collected after surgery, whereas traditional blood tests typically require 4 to 6 weeks to identify tumor remnants.
"With the acceleration capabilities of NVIDIA Parabricks, we have reduced some of the most computationally intensive steps from over a day to just a few hours," said Wendy Winkler, Chief Scientific Officer at Droplet.
The company noted that although GPU computing incurs a higher hourly cost, it still achieves overall operational benefits: "The significant reduction in runtime results in a lower total cost per sample."
Faster test turnaround times enable patients to receive results during their hospital stay, eliminating the need for multiple follow-up visits or extended waiting periods associated with traditional blood tests.
"We are using Parabricks to accelerate genomic analysis, cutting the turnaround time from 10 days to under 5 days," said Zhuosheng Gu, Senior Director of Research Informatics at Droplet.
The diagnostic startup's first clinical test, targeting HPV-negative head and neck cancer, has received Clinical Laboratory Improvement Amendments certification.
Droplet is a member of the NVIDIA Inception startup accelerator program and is also an NVIDIA AI Enterprise customer.
Comments