Deshang Yunxing Medical Technology Co., Ltd. (Deshang Yunxing) is applying Intel® Software Protection Extension (Intel® SGX) and Intel® oneAPI Math Kernel Library (oneMKL) to protect its medical artificial intelligence algorithms on edge medical devices And intellectual property rights. As a leader in the research and development of artificial intelligence ultrasound technology in China, Deshang Yunxing adopts its self-developed deep learning framework DE-Light. Under the open source framework, the framework has demonstrated excellent performance and improved the accuracy of thyroid nodule detection. 30-40%.
Liang Yali, vice president of Intel’s marketing group and general manager of the industry solutions department in China, said, “Based on the characteristics of the medical industry, many artificial intelligence vendors have to deploy their solutions on the client side. Intelligent model for protection has become an urgent need. Trusted execution environment technology based on Intel SGX technology has become the most cost-effective and feasible solution that can provide runtime protection. With Intel’s support, Deshang Yunxing has made SGX technology strategic It has been applied to a series of artificial intelligence products including its ultrasonic artificial intelligence products, greatly enhancing its market competitiveness and innovation capabilities.”
At this stage, ultrasound imaging diagnosis and analysis are widely used and are an important part of medical diagnosis and treatment. However, imaging examination and analysis depend to a large extent on the doctor’s experience, and the efficiency is low.
The greatest value of artificial intelligence in medicine lies in its ability to quickly analyze large amounts of data and obtain accurate results. By making image diagnosis intelligent, AI technology in diagnosis and analysis is helping medical institutions accelerate detection and treatment. Doctors can improve the efficiency of image reading, reduce the possibility of misdiagnosis and obtain diagnostic assistance, while patients can obtain more accurate diagnostic recommendations and personalized treatment recommendations.
Deshang Yunxing’s AI medical application products have been successfully deployed in more than 400 hospitals in China, and plans to promote and deploy in more than 1,000 hospitals in the coming year.
Deshang Yunxing has launched the AI-SONIC® computer-aided diagnosis system for ultrasound imaging, and has also developed more artificial intelligence products that use Intel SGX to protect algorithms. In the ultrasonic AI model solution, in order to ensure that the algorithms in Intel SGX can make full use of the computing power of Intel? processors, the project team also artificially optimized the algorithm built on oneMKL based on the Intel? C++ compiler. The core calculation of point matrix multiplication.
oneMKL is a parallel mathematical function library highly optimized for Intel and compatible processors. In different operating environments, oneMKL can perform automatic runtime processor detection, thereby running different optimized versions of programs on different processors, so as to ensure that it can achieve better performance on the processors it runs. Intel? oneAPI Math Core Function Library (oneMKL) and Intel? oneAPI Deep Neural Network Library (oneDNN) provide basic support for commonly used deep learning frameworks and many self-developed AI applications.
Intel Software Protection Extension is an instruction set that can improve the security of application code and data, and better prevent them from being leaked or modified. Developers can divide sensitive information into hardware-based Trusted Execution Environment (TEE) or secure areas, that is, areas with higher levels of security protection in memory. This technology helps to ensure that the source of trust is limited to a small part of the central processing unit’s hardware, thereby better protecting the confidentiality and integrity of code and data.