Medical imaging plays a crucial role in modern healthcare by providing doctors with a way to visualize internal structures of the body without invasive procedures. However, the images produced by these imaging techniques can be large, take up a lot of storage space as well as bandwidth when transferring these digital images over network. This is where compression comes in as an essential tool in medical imaging. HTJ2K compression codec reduces the size of an image file significantly without sacrificing image quality as it is lossless. By compressing medical images, healthcare providers can store, transmit, and analyze enormous amounts of data more efficiently, enabling faster diagnosis and treatment planning for patients. Compression techniques are used in various medical imaging modalities such as MRI (Magnetic Resonance Imaging), CT (Computed Tomography) scans, ultrasound, and X-ray imaging, making them an essential component of modern medical imaging.
With AVX512-based optimization contributed by Intel, here , the OpenJPH project provides the industry with even more reasons why HTJ2K is critical on the path towards next generation medical imaging powered by low latency compression/decompression. In testing, time to compress using OpenJPH saw up to 46% reduction with AVX512-based optimization enabled.
OpenJPH is an open-source software initiative aimed at developing a fully functional, high-performance implementation of the high-throughput JPEG2000 image compression standard that is free and accessible to all. Every millisecond counts when delivering medical imagery at scale and HTJ2K is perfectly positioned to deliver in this space. It is a highly efficient compression algorithm that can provide significant compression ratios while still preserving the high image quality “lossless” required for medical diagnosis. AVX-512 optimization provides significant performance improvements for demanding applications like HTJ2K encode by adding 512-bit vector instructions to the existing 256-bit SIMD (Single Instruction, Multiple Data) AVX2 (Advanced Vector Extensions 2) instruction set.
* “Performance varies by use, configuration and other factors. Learn more at www.Intel.com/PerformanceIndex.”
Faster radiology image viewing times can have a profound impact on care delivery. Keeping pace with the latest high-performance standards like HTJ2K and choosing implementations of these that fully utilize the underlying hardware they run on delivers tangible results that can power the next generation of care delivery.
Links and Resources
A Blazing-Fast JPEG 2000: What is HTJ2K?: https://htj2k.com/
OpenJPH project https://github.com/aous72/OpenJPH
Support for AVX512-based compression on the encode step https://github.com/aous72/OpenJPH/pull/103
Low Latency Lossless Compression of Monochrome Medical Images