Improved Image Analysis Using a Biomechanical Framework
Medical imaging and the ability to assess morphological changes in tissue volume is important for many clinical applications. However, current image processing techniques are one-dimensional, limiting the ability to accurately track changes over time. Our algorithm and associated software measures changes in organ morphometry considering volumetric, directional and sheer deformations to create a three-dimensional analysis, which can improve the sensitivity and specificity of disease prognosis and diagnosis.
Technology Overview
Scientists at Wake Forest Baptist Medical Center have developed a longitudinal/cross-sectional image registration algorithm for three-dimensional deformation analysis of medical images. By applying a finite strain theory of continuum mechanics, the algorithm produces an accurate 3D description of tissue deformation. This approach has higher detection sensitivity of morphological changes than conventional approaches. For example, in addition to detecting changes in volume, this novel image analysis method is also able to detect directional changes, shear deformation, connection with neighboring voxels and orientation-specific changes.
This algorithm has a broad application range. It can be applied to every organ system to track changes associated with disease progression and pathologic processes. It can also identify more sensitive and reliable MRI biomarkers, which may increase the sensitivity and specificity of disease diagnosis. The algorithm may provide prognostic information regarding outcomes, intermediate biomarkers or response to therapy.
Medical imaging techniques such as magnetic resonance imaging (MRI) and computerized tomography (CT) are commonly used to quantify patients’ structural changes over time. This includes tracking changes associated with disease progression (e.g., hippocampal structure in Alzheimer’s Disease progression), as well as pathological processes (e.g., changes in malignant tumors).
Current imaging techniques such as Jacobian determinant (JD) and cortical thickness (CTHK) only characterize morphological changes in one-dimension (volumetric). However, certain changes, such as those associated with neural atrophy in adults with mild cognitive impairment, may not be detected using these techniques, or with any other techniques currently available.
This algorithm has been fully developed and tested in a longitudinal MRI study of 33 infants and 35 adults with mild cognitive impairment.
- Jeongchul Kim, PhD, Department of Radiology
- Youngkyoo Jung, PhD, Department of Radiology
Patent Pending (WO2018098213A1)