TAIPEI, Taiwan, August 4, 2023 —Taiwan Web Service (TWSC), a subsidiary of ASUS, today announced significant breakthroughs in the biomedical industry with its integration of the computational power of the Taiwania 2 supercomputer with NVIDIA®’s optimized GPU computing framework, resulting in seamless secondary gene analysis, drug development, and medical imaging. Plus, TWSC is leveraging federated learning technologies and flexible, diversified cloud data storage solutions to drive research and intelligent transformation of medical technologies.
"Since 2022, ASUS has partnered with NVIDIA to establish the AI Biomedical Innovation Center at the National Health Research Institute (NHRI), jointly driving significant developments in AI computing in biotechnology,” said Peter Wu, CEO of ASUS Cloud and TWSC, at the BIO Asia-Taiwan Exhibition 2023. “This time, in partnership with the Industrial Technology Research Institute (ITRI), we have incorporated AI applications into the entire biomedical engineering process to meet the needs of data processing, computing environment setup, AI biomedical model training, and technology tool creation.”
TWSC has integrated the needs of the medical industry with AI technology tools and introduced OneAI, a no-code development platform. OneAI allows users to perform secondary gene analysis in just three steps and without the need for complex programming. TWSC’s AI supercomputer uses the RAPIDS library for efficient GPU processing of large and highly complex data, reducing training time from hours to minutes. This is further boosted by NVIDIA Clara™ for Drug Discovery deep learning algorithms that can be used to perform computations for chemical informatics, molecular dynamics simulation, protein structure prediction, and molecular structure exploration to accelerate the development of new drugs. Secondary gene analysis is enabled by high-volume, high-precision NVIDIA Parabricks® GPU-accelerated analysis, which is 80 times faster than traditional CPU solutions and reduces computational costs by 50%. The platform also provides over 20 pre-trained medical image recognition models through the Medical Open Network for AI (MONAI) open-source deep learning framework, resulting in a six-fold improvement in processing efficiency.
TWSC also collaborated with ITRI, NHRI, NVIDIA, and the Taiwan Genomic Industry Alliance Inc. (TGIA) to host the AI Supercomputing and Biomedical Industry Integrated Application Forum during the BIO Asia-Taiwan Exhibition 2023. Together, they shared the latest applications of integrated AI supercomputing and biomedical technology. TWSC and ITRI also signed a Memorandum of Understanding (MOU) to combine high-efficiency AI computing with new technologies in biomedical research to optimize applications by bioinformatics software such as molecular docking and virtual screening. This collaboration expects to reduce the barrier of conduct complicated data analysis and AI model training in terms of the biomedical industry, showcasing Taiwan's R&D capabilities and the advantages of cross-domain integration in digital health.