Tapping HPC and AI for Global Health and Wellness – HPCwire

Heres a look at how HPC, AI, and other technologies are being used throughout the world by organizations to enhance healthcare research, drug development, public health, and patient outcomes.

The ability to gather, process, and analyze data from genomics, bioinformatics, microscopy, medical imaging, and other areas in the life sciences has been supercharged with HPC systems and artificial intelligence (AI) algorithms. Researchers can sequence vast quantities of DNA data faster than ever before with supercomputer resources and use AI to identify patterns and make predictions. They can now use these available and affordable technologies to study genes and proteins, to predict health events, automate imaging analysis, and generate ideas for improving healthcare delivery. Heres a look at how HPC, AI, and other technologies are being used throughout the world by organizations to enhance healthcare research, drug development, public health, and patient outcomes.

The COVID-19 pandemic has provided a test case for the ability of HPC to accelerate genomic sequencing as scientists around the world seek to track, understand, and combat the SARS-CoV-2 virus. In England, researchers at the Wellcome Sanger Institute have tracked the spread and mutations of the virus by sequencing over 300,000 coronavirus genomes. The institutes HPC cluster has 38,000 cores of compute, 23.5 petabytes of file systems, and a 30-petabyte virtualized storage repository, all supported by a 60 Gbps network backbone. Its complemented with an OpenStack private cloud with more compute and storage resources.

With HPC architectures and the use of machine learning and AI constantly evolving, the institute has worked with companies like Dell Technologies to build their HPC environment. Genomic sequencing data is stored for computational analysis on Dell PowerScale scale-out storage. Researchers use the data to determine the relatedness of different viruses and help identify chains of transmission, super-spreader events, and fast-growing variants.

A similar collaboration for genomic research between the Texas Advanced Computing Center (TACC) at the University of Texas and Dell Technologies spawned the Lonestar6 supercomputer, which can perform almost three quadrillion mathematical operations per second. It is being used by faculty members from throughout the University of Texas system at other universities for COVID-19 drug discovery and genomic research.

In another pandemic-related role for HPC, the staff at the Ohio Supercomputer Center at Ohio State University designed the COVID-19 Analytics and Targeted Surveillance System (CATS) to help school administrators decide whether it was safe to bring students back to classrooms or if fully remote or hybrid learning should be used instead. Supported by a HPC system with Dell PowerEdge servers with Intel Xeon processors, CATS serves 21 school districts and 238,000 students and tracks data like school nurse visits, student and teacher absences, and other metrics to watch for outbreaks and inform decision making. Sixteen different dashboards are used daily by thousands of people to provide the rationale for decisions like closing or opening schools or specific buildings on campuses.

In yet another use of HPC for COVID-19 research, Swansea University in Wales has built an open platform for mathematical modeling of disease transmission. It provides comparisons of multiple models to help researchers determine demographic, socioeconomic, and clinical risk factors for COVID-19 infection, morbidity, and mortality, among other uses. Supercomputer resources at two hubs, built by Dell Technologies and Atos, contain more than 13,000 cores, tens of terabytes of memory, and hundreds of terabytes of high-performance storage, all interconnected by low-latency, high-bandwidth networking.

The Cineca Consortium is a national supercomputing facility in Italy that supports public and industry research institutions with HPC resources. Among Cinecas 4000 projects is the Human Brain Project, which, in conjunction with 90 European research institutes, aspires to be the worlds most detailed model of the brain. A dedicated supercomputer has been built for the project with HPC technology from Dell and Intel.

To-date, researchers participating in the Human Brain Project have used HPC to explore brain mechanisms behind cognition, learning, and plasticity. Their research has led to more than 1,400 journal articles, a new treatment for spinal cord injuries, a brain prosthesis for the blind, and better modeling and understanding of epilepsy and autism.

A collaboration among researchers at the Washington University School of Medicine, the Memorial Sloan-Kettering Cancer Center, and Temple University, the [emailprotected] distributed computing project uses HPC to simulate how proteins impact a variety of diseases. To visualize protein dynamics on a molecular level requires enormous computational power and the projects founders came up with an original HPC solution harnessing the unused processing power of PCs from volunteers around the world. Each volunteer downloads an application that runs small parts of much larger simulations for the project. On the backend servers, algorithms put the separate parts together to create composite simulations.

Today this distributed HPC network has the equivalent of 2.4 exaflops of computational power, making it the first exascale computer. Teams from Dell Technologies and VMware are part of the legions of volunteers and the [emailprotected] client software resides on a VMware vSphere appliance. In December 2020, [emailprotected] was awarded the HPCwire Readers Choice Award for Best Use of HPC in Response to Societal Plights for its simulations of SARS-CoV-2 proteins.

For more on Dell Technologies for healthcare, life sciences, and HPC, please visitDellTechnologies.com/healthcareandDellTechnologies.com/hpc.

Read more:

Tapping HPC and AI for Global Health and Wellness - HPCwire

Related Posts

Comments are closed.