Magnetic Resonance Imaging (MRI)
NMRI (Nuclear Magnetic Resonance Imaging) is also called MRI for short. It is a medical imaging technology used to produce images of human bodies' detailed internal structure by utilizing phenomenon of nuclear magnetic resonance. This technology reconstructs images by implementing an accurate detection of the energy released by excited hydrogen atoms. It is currently one of a few medical imaging technologies that are not harmful to human bodies. MRI allows users select target organs by configuring multiple parameters, create images of cross sections at any depth, and perform excellently in soft tissue imaging. A typical MRI system consists of a magnetic field system, a RF system and an imaging reconstruction system.
Magnetic field system is built with a permanent magnet or a superconducting coil and three gradient coils. Permanent magnet or superconducting coil generates a static high field of up to 1.5 or 3 tesla. To build a well- distributed field, a shim coil could be added. Gradient coils are located on x, y, z axes of three-dimension coordinate to implement accurate locating of inspected parts of human bodies. Gradient field is commonly controlled by a high-precision DSP which utilizes a DAC to produce analog output for driving the coils.
RF system is made up of a RF transmitter and a receiver. RF transmitter has a waveform generator (typically composed of a FPGA and a direct digital synthesizer or a DAC) to generate RF signal, which is then amplified for eventual emission in a pattern of pulses. The emitted RF pulses results in NMR (nuclear magnetic resonance) phenomenon on hydrogen atoms in human bodies. RF receiver, which includes a low noise amplifier, a band-pass filter and a programmable amplifier, receives NMR signal and sends to image processing system after amplification and filtering.
Image reconstruction system converts the analog signal sent by RF receiver into digital signal, and then utilizes a DSP to process based on the signal and gradient field information. The output of DSP will be used by a host to reconstruct images with distinguishable grayscales that represent different body tissues according to a tissues-grayscales database. Eventually, images are display on screen or printed out for doctors' reference.