Magnetic Resonance Imageless Diagnosis MRID

he advents of sophisticated data processing Al algorithms on one hand, and low-cost magnetic resonance MR imaging hardware on the other, are creating a new paradigm for clinical diagnosis based on the phenomenon of MR. The rather undemanding hardware requirements associated to low-field MR scanners has opened the door to a hitherto unfathomable degree of System availability and portability. It is foreseeable that countless low-cost MR systems will be sited in the near future in small clinics, outpatient departments, rural regions, low-income areas, etc. As the use of these systems expands, the data processing and user interface modules must be simplified with respect to today¿s standards, to avoid the need for intervention of highly skilled and specialized personnel. Al tools are already heavily used for image interpretation and clinical diagnostic aid. Our hypothesis is that, as these algorithms evolve and people without formal training require useful output from highly available systems, the degree of human intervention in the process will decrease and become altogether unnecessary for numerous applications. This patent discloses a classification method for clinical diagnosis i.e. a method that decides in favour of a particular hypothesis from a finite hypothesis space based on direct interpretation of the MR data collected by any MR system, which need not be designed to, or even capable of, producing medical images. We have called this Magnetic Resonance Imageless Diagnostic MRID and it is a method to answer clinical questions based on the analysis of data collected from an MR System regardless of its configuration field strength, homogeneity, gradient capabilities, etc..