HEALTHCARE

Medical AI is revolutionizing healthcare by enabling more accurate and efficient diagnoses, particularly in medical imaging. For example, physicians and technologists leverage AI in medical imaging to reduce doctors’ workloads and improve disease detection. From identifying subtle abnormalities to accelerating diagnoses, these AI-driven advancements hold immense potential. But deploying AI models outside the data center—closer to where data is collected—remains a challenge.

One major hurdle is the sheer scale of data in medical imaging. Techniques like Optical Coherence Tomography (OCT), used to capture detailed 3D images of the eye, generate massive files with resolutions as fine as 2 to 5 microns. These high-resolution images require significant storage and computational resources, creating bottlenecks for AI processing.

While data centers are equipped to handle such workloads, transferring large image data sets introduces complications, including network bandwidth limitations and concerns over patient privacy. To overcome these challenges, developers explore innovative ways to integrate deep learning directly at the point of data collection. This approach not only addresses storage and bandwidth constraints but also enables faster, more secure medical AI applications, empowering clinicians with real-time insights.

Here the challenge becomes one of compute requirements. To handle the workloads associated with imaging, developers have turned to a heterogeneous mix of CPUs, GPUs, FPGAs, and other processors. But these complex architectures can be difficult to configure and program.

As an alternative, developers can leverage off-the-shelf technologies. A new generation of AI hardware and software designed specifically for edge applications can greatly simplify development. Developers should consider using these APIs and frameworks to maximize the portability, flexibility, and scalability of their deep learning efforts.

Deploying AI in Medical Imaging

One compelling example comes from IEI, whose deep learning solution is built on ready-made technologies like the Intel^® FPGA Deep Learning Acceleration Suite and the Intel^® OpenVINO^™ toolkit. These tools are part of a quick-to-deploy AI medical diagnosis system tailored for edge applications in medical imaging.

The system includes a workstation, deep learning software development kit, and powerful network-attached storage (NAS) sourced from the company’s QNAP storage division. It is designed to enhance the efficiency and accuracy of medical imaging workflows, particularly for applications like OCT.

OCT is a critical diagnostic tool for identifying eye diseases such as glaucoma, age-related macular degeneration (AMD), and diabetic eye disease. Traditionally, analyzing OCT images requires highly trained specialists and significant time. By integrating AI and deep learning at the edge, IEI's solution accelerates both training and inferencing processes, streamlining diagnoses and enabling faster, more precise decision-making.

Medical AI Technology Enables Early Detection and Better Outcomes

Diagnosing AMD quickly delivers real benefits to patients. AMD, like many diseases, does not present obvious symptoms in the early stages, and often goes unnoticed until a patient’s sight is failing. By that time, the disease has reached the middle or late stages, making treatment less effective.

But if the disease is discovered early, treatment can begin. And the earlier treatment starts, the more likely a patient will retain their sight. The same is true for many other diseases, where early detection results in significantly better outcomes.

Traditional approaches to interpreting medical images, while highly accurate, take considerable time to complete. They require multiple viewings and discussion by more than one specialist. In areas without such specialists, medical images must be sent out for diagnosis. And wait times often take several weeks before a diagnosis can be made and for treatment to begin.

By using an AI medical diagnosis solution, doctors in a central location can make accurate diagnoses in less time. Therefore, patients can start their treatment earlier, often leading to better outcomes. 

Leveraging AI for Efficient Medical Imaging

Deep learning success in medical AI relies on three key factors: data, computing power, and algorithms. The solution’s network-attached storage (NAS) is optimized for medical data, providing the ideal foundation for efficient storage and access. To enhance computational capabilities, the NAS supports add-in AI cards, enabling more efficient processing of complex tasks. For example, the IEI Mustang series acceleration card, powered by the Intel FPGA Deep Learning Acceleration Suite, meets the extreme compute demands of AI in medical imaging.

The platform also supports a variety of AI algorithms and frameworks, including Caffe, MXNet, TensorFlow, and CNTK, making it easier to deploy and scale deep learning models. Existing containerized solutions can be seamlessly migrated to the platform, and new ones can be quickly started. The included container station supports Docker and other container technologies, offering flexibility in development. Additionally, the system is compatible with QuAI libraries, providing a robust foundation for AI model training and enabling a wide range of applications in medical imaging.

Medical AI, Deep Learning, and Improved Outcomes

To deliver value to physicians and patients, an AI-embedded deep learning edge solution must:

• Provide the enormous amount of compute power, storage, and network connectivity that medical professionals require to manage a massive amount of medical images, in addition to genomic and patient data.
• Combine a range of highly technical, cross-disciplinary skill sets—not held by most medical professionals and researchers—to build a deep learning platform and data management system for healthcare.
• Set up drivers, containers, data backup/transfer, and network configurations that are beyond the expertise of most data scientists.
• Offer scalability to work in hospitals and clinics of all sizes.
• Meet compliance for HIPAA and other regulations.

IEI’s innovative solution illustrates how developers can meet these goals with off-the-shelf solutions. That’s excellent news for doctors and patients alike. The ability for medical AI and deep learning to help diagnose disease in less time and increase accuracy makes a significant improvement in people’s lives. In the very near future, AI-embedded deep learning edge solutions will become common in medical facilities.

This article was edited by Georganne Benesch, Editorial Director for insight.tech.

This article was originally published on December 4, 2018.