SAR Calculation Processor for MRI Patient Safety

The patent creates a digital twin (virtual model) of the patient's anatomy that replicates the physical patient's electromagnetic properties. This virtual copy allows SAR calculations to be performed on the digital model rather than relying on bio-mesh simulations, thereby avoiding the systematic errors inherent in bio-mesh approaches while maintaining measurement accuracy.

The patent replaces the mechanical/bio-mesh simulation approach with a physics-based electromagnetic field calculation method using measured B1 maps. Instead of relying on simplified bio-mesh models that introduce systematic errors, the system uses actual measured field data to compute SAR, substituting the simulation mechanism with a measurement-based calculation mechanism.

The patent applies local quality by tailoring the SAR calculation to each individual patient's specific anatomy and electromagnetic properties rather than using a uniform worst-case scenario for all patients. By measuring B1 maps specific to each patient and computing their personalized SAR distribution, the system ensures safety for that specific patient while avoiding the overly conservative RF duty cycle limitations imposed by blanket worst-case estimates.

The patent implements dynamic SAR calculation that adapts to each patient's specific characteristics through measured B1 maps. Rather than using a static worst-case scenario, the system dynamically computes SAR based on actual patient-specific field measurements, allowing the RF duty cycle to be optimized for each individual patient's safety profile rather than being constrained by a one-size-fits-all approach.

The patent performs preliminary measurement of B1 maps for each transmit channel before the actual SAR-critical scan. These pre-acquired B1 maps capture the electromagnetic field distribution for each channel and are stored for later use. When SAR calculation is needed, the system retrieves these pre-measured B1 maps rather than performing new measurements, thereby simplifying the SAR calculation process while maintaining per-channel accuracy.

The patent uses B1 maps as an intermediary data structure that bridges the gap between raw RF transmission parameters and SAR calculations. By measuring and storing B1 maps for each transmit channel, the system creates an intermediate representation of the electromagnetic fields that can be directly used in SAR computations, simplifying the overall process while maintaining accuracy.

The patent implements feedback by using measured B1 maps to inform SAR calculations for each patient. The system measures the actual electromagnetic field distribution (B1 maps) for each patient, uses this feedback information to compute patient-specific SAR distributions, and then adjusts RF pulse parameters accordingly. This closed-loop approach ensures both known and unknown hotspots are identified and controlled based on actual patient-specific measurements rather than assumptions.

The patent changes the parameters used for SAR calculation from fixed, population-based assumptions to dynamic, patient-specific measured B1 map data. By adjusting the input parameters (actual measured fields rather than theoretical models) to match each patient's unique electromagnetic properties, the system can identify and control both known and unknown hotspots specific to that patient's anatomy.