PERCISTz: A Method for the Quantitative Assessment of Disease Response using Medical Image Data.

C11731: Medical Image Analysis and Diagnosis Improvement Tools

Novelty:

This technology bundle is a set of processing tools for medical image analysis.

Value Proposition:

The current invention is a suite of tools for improving diagnoses from medical images, involving three separate technologies. Currently, there is a need for precise, quantitative medical image assessment. Methods currently used leave much to be desired in terms of assessment over time, precision, and sensitivity. Current methods are not well-personalized for each patient, and may lead to incorrect diagnosis. The disclosed inventions seek to address these issues and provide other benefits to the user. This bundle of technologies includes many statistical measures implemented to provide the following advantages:

Technical Details:

Johns Hopkins researchers have developed tools to address insufficiencies in medical image analysis. The first is called PERCISTz, and is a method for the assessment of disease response using medical image data (e.g. Positron Emission Tomography or FDG-PET) to quantify and track disease over time. This method is unique in that it is image-driven and self-normalizing (i.e. personalized), to quantitatively describe a disease trajectory when tracked longitudinally over multiple imaging encounters. The second technology is a 'Hyper-Precise' Spherical Volumetric Sampler of Volumetric Data. This method allows precise spherical Volume of Interest (VOI) measurements of PET and other quantitative volumetric data for greater precision and consistency when comparing quantitative measurements taken from disparate digital datasets. The method allows the precise sampling of volumetric digital data using spherical Volumes of Interest (VOIs) in a manner which is independent of the underlying sampling matrix. The third technology in this bundle is called 'PAX' Sampler of Digital Data, which provides increased sensitivity of signal detection over customary sampling in a manner which can be either user- or auto-regulated to balance between sensitivity and specificity of measurement. The method provides a graduated and controlled measurement tool which can transition from a simple averaging statistic over a defined interval (sometimes referred to as 'PEAK') to a single element measurement within that same defined interval (sometimes referred to as 'MAX'). This technique provides sampling statistic which self-transitions from 'PEAK' (with high specificity) to 'MAX' (with high sensitivity), thus the hybrid name 'PAX'.

Looking for Partners:

To develop and commercialize the technology as a set of tools for medical image evaluation.

Stage of Development:

Proof of Concept

Data Availability:

Under CDA/NDA

Publications/Associated Cases:

Associated Cases: C11732, C11733