AI healthcare startups are pioneering thousands of applications and devices to build next-generation smart hospitals. The sheer number of applications and the pace of innovation bring new challenges to how these applications get deployed and where they live.
The cloud revolution built and delivered the infrastructure for AI and applications to be instantaneously deployed and accessed by millions of users. As the healthcare industry expands to use thousands of AI applications, it’s looking for a cloud-like experience inside the hospital.
To address critical industry challenges like securing patient data and delivering real-time applications, AI developers use NVIDIA NGC to bring solutions to market that are both on premises and cloud-ready, built atop an edge computing platform to scale to more customers and get to market faster.
Ease AI Deployment at the Edge Using Cloud-Native Infrastructure
NGC, a hub for GPU-accelerated software, offers containerized applications and AI frameworks designed to run on cloud-native Kubernetes clusters optimized for NVIDIA GPU acceleration.
Containerization is an industry-standard design pattern for application deployments, and with Kubernetes, it provides a consistent platform for deployment across edge, data center, cloud and hybrid environments.
The containers and Helm charts from NGC enable IT managers to quickly, easily and consistently deploy applications on premises or on remote systems centrally, giving users faster access to run their applications and workloads.
NGC containerized applications are tested and optimized to run on NVIDIA EGX, a software stack designed to run on a broad range of EGX systems, from Jetson-powered devices to NGC-Ready for Edge validated servers. System administrators can easily set up a fleet of edge servers remotely and securely, meeting hospitals’ needs for data security and real-time intelligence.
Many application providers are already on the NGC registry or launching soon:
- 16 Bit, an AI-powered medical image analysis systems, created the top-performing RSNA bone age prediction model back in 2017 and is working on algorithms to revolutionize breast cancer and osteoporosis screening paradigms.
- American College of Radiology AI-LAB provides a platform that lets ACR members create and share models using locally annotated data.
- CuraCloud offers AI professional services to medical technology vendors, healthcare organizations and pharmaceutical companies that need machine learning and bioinformatics research and development capacity.
- DDH provides a secure, intelligent healthcare solution that infuses the deep knowledge and experience of highly regarded clinicians into machines for performing fully automated medical image analysis.
- DeepTek.ai uses in-house innovated “assisted and augmented” imaging-focused algorithmic tools for medical imaging like radiographs, CT scans and MRI.
- Infervision provides AI-assisted diagnosis products in medical imaging that enable the healthcare industry to avail quick and accurate results of disease diagnosis and that help doctors with optimized decision making.
- Inform AI has a healthcare focus on AI solutions that speed up medical diagnosis at the point of care and improve radiologist productivity.
- Kaliber Labs assists surgeons by developing AI models to interpret video feeds of a patient’s anatomy and pathology, plus surgeons’ activities inside the operating room, in real time.
- LPixel develops advanced medical image diagnosis application and life science image analytics solutions, including a magnetic resonance angiography unruptured cerebral aneurysm detection application that has already cleared Japanese regulatory certification.
- Lunit develops AI solutions for precision diagnostics and therapeutics, focused especially on conquering cancer, with a goal to make data-driven medicine the new standard of care.
- QUIBIM has a proprietary software platform and develops AI algorithms, across imaging modalities, with a focus on neurology, chest, body and musculoskeletal analysis algorithms.
- Rad AI works with radiology groups to streamline radiology reporting, helping automate repetitive and time-consuming tasks for radiologists — boosting productivity while reducing the risk of radiologist burnout.
- Radiobotics is building AI algorithms that empower musculoskeletal radiologists in their daily work.
- Shukun develops AI technology to assist doctors in increasing diagnosis process efficiency and accuracy for cardiovascular and cerebrovascular diseases and tumors.
- Smart Reporting offers AI software for structured reporting and clinical guidance in medical diagnostics to enhance productivity, improve quality and fully digitize processes.
- TrainingData.io enables high-precision training data labeling for visual AI, offering engineers a speedup of several orders of magnitude and empowering data scientists with control of the quality of training data.
- United Imaging Intelligence develops and produces a full portfolio of advanced medical products covering the entire process of imaging diagnosis and treatment.
- Vuno develops a medical data analysis platform that uses AI algorithms to analyze patient imaging data and gain insights that help researchers and physicians quickly and accurately identify diseases and illnesses.
- Vysioneer empowers radiation oncologists to provide precision radiosurgery, with a deep learning-based system for brain tumor auto-contouring that can spot the three most common types of brain tumors, namely brain metastasis, meningioma and acoustic neuroma.
- XNAT provides an open-source imaging research platform that manages datasets, training and inference pipelines, integrating NVIDIA Clara SDKs.
Meet Us at RSNA 2019
More than 50 NVIDIA Inception partners will be exhibiting onsite at RSNA 2019 this week in Chicago, sharing their medical imaging applications for every phase of the radiology workflow at the RSNA AI Theater and NVIDIA booth 10939.
In our booth, we’ll be exhibiting the latest AI tools for medical imaging, from development to deployment. Come learn more about why and how to containerize your applications on NGC for EGX at our developer meetup on Tuesday, Dec. 3, from 11:30 a.m.-1:30 p.m. Central time.