Medical Imaging Implementation Using FPGAs

Introduction

Earlier prediction and treatment are driving the fusion of modalities such as positron emission tomography (PET)/computerized tomography (CT) and X-ray/CT equipment. The higher image resolutions that are needed require fine geometry micro-array detectors coupled with sophisticated software/hardware systems for the analysis of photonic and electronic signals. These systems must provide both highly accurate and extremely fast processing of large amounts of image data (up to 250 GMACS and 1 Gbps). Furthermore, to lower patient costs, each piece of equipment must be lower priced and possess a longer life utility. This calls for more flexible systems with the capability to continually update features and algorithms over the equipment’s lifetime. Together, flexible algorithm deployment and modality fusion compel the use of programmable system electronic components, such as high-powered CPUs and FPGAs.

Several factors should be considered in the efficient development of flexible medical imaging equipment:

Development of imaging algorithms requires high-level intuitive modeling tools for continual improvements in digital signal processing (DSP).
The performance needs for near-real-time analysis require system platforms that scale with both software (CPUs) and hardware (configurable logic). These processing
platforms must meet various performance price points and be capable of bridging the fusion of multiple imaging modalities.
System architects and design engineers need to quickly partition and debug algorithms on these platforms, using the latest tools and intellectual property (IP) libraries to speed their deployment and improve profitability.

With these factors in mind, Altera provides its modular Video and Image Processing (VIP) Suite, a blockset of key IP building blocks that can accelerate the development and implementation of sophisticated imaging algorithms into FPGAs. The VIP Suite blockset, along with other Altera® and partner IP modules and reference designs (including IQ modems, JPEG2000 compression, fast Fourier transform (FFT)/inverse fast Fourier transform (IFFT), edge detection, etc.), provide a broad range of tools designers can use to speed FPGA implementations of computationally intensive tasks.

Algorithm Developments in Medical Imaging

Some of the most critical pieces of equipment in today’s medical development environment include:

X-ray, magnetic resonance imaging (MRI), CT scanner, ultrasound, and 3D imaging systems
Measuring and analysis instruments
Optical manipulation and analysis
Surgical microscopes
Telemedicine systems

Designers are looking for rapid imaging solutions with applications including:

Image analysis and pattern recognition
Image enhancement and restoration
Image and data compression
Wavelet transform capabilities
Color space conversion

The following sections cover some of the trends and key developments driving the integration of programmable logic into medical imaging equipment.

Image-Guided Therapy

Intraoperative image processing for surgical guidance uses the registration (correlation) of preoperative (CT or MR) images with real-time 3D (ultrasound and X-ray) images to guide the surgical treatment of disease using non-invasive therapies (ultrasound, MR interventional, and X-ray treatments). Various algorithms have been developed to provide the best registration results for the specific fusion of modalities and therapy types.