Methods and systems for precise quantification of human sensory cortical areas
Inventors
Tu, Yanshuai • Wang, Yalin • Lu, Zhong-Lin
Assignees
New York University NYU • Arizona State University Downtown Phoenix campus
Publication Number
US-12303246-B2
Publication Date
2025-05-20
Expiration Date
2041-04-02
Interested in licensing this patent?
MTEC can help explore whether this patent might be available for licensing for your application.
Abstract
A sensory mapping method for a human brain is disclosed. The method includes the steps of flattening the cortical surface of the human brain, projecting functional imaging data onto the flattened surface, smoothing the functional imaging data, generating a sensory map, registering sensory maps across individuals and analyzing the maps in the common space. The flattening utilizes a conformal parametrization method. The smoothing utilizes a topological smoothing method that utilizes a diffeomorphic smoother. The registering is diffeomorphic. The sensory mapping method may further include a step of processing the functional imaging data to produce topological results.
Core Innovation
The invention discloses methods and systems for precise quantification of human sensory cortical areas using functional imaging data. The core process involves flattening the cortical surface of the human brain through conformal parametrization, projecting functional imaging data (such as fMRI) onto the flattened surface, and then applying specialized smoothing and topological processing techniques. The flattening utilizes harmonic mapping followed by iterative refinement to achieve a conformal mapping, ensuring the geometric structures of the cortical areas are well preserved.
A critical problem addressed by the invention is that conventional sensory mapping methods, particularly those based on fMRI, are hampered by low signal-to-noise ratio and spatial resolution, leading to noisy, non-topological maps that lack quantitative reliability. Current approaches typically produce qualitative, uncertain maps, which are inadequate for clinical and research purposes where robust quantitative descriptors are necessary for diagnosis, prognosis, and understanding of sensory processing.
To solve these shortcomings, the disclosed systems incorporate a diffeomorphic smoothing algorithm employing the Beltrami coefficient to ensure retinotopic maps are locally smooth, differentiable, and invertible, while also minimizing angle distortion. Additionally, topology-preserving segmentation and decoding steps are integrated into the mapping framework, allowing iterative refinement of boundaries between sensory areas based on both structural and functional data until guaranteed convergence is achieved. The methods also include diffeomorphic registration across subjects, enabling robust alignment and quantitative comparison of sensory maps in a common space.
Claims Coverage
The patent contains two principal independent methods covering six inventive features spanning diffeomorphic sensory map generation, smoothing algorithms, topological processing, conformal flattening, sensory map rendering, and registration across individuals.
Diffeomorphic smoothing algorithm for retinotopic coordinates
A sensory mapping method includes smoothing fMRI data by first determining retinotopic coordinates for each cortical area, then applying a diffeomorphic smoothing algorithm that: - Computes a parametric coordinate for each area. - Performs Laplacian smoothing on these coordinates. - Applies a 'chopping' step to maintain diffeomorphism. - Uses a Linear Beltrami Solver (LBS) to compute a sensory map per area. The smoothing process is repeated until the maximum absolute value of the Beltrami coefficient for retinotopic coordinates in all areas is less than one, resulting in a diffeomorphic sensory map.
Conformal flattening of the cortical surface
The method flattens the cortical surface onto a two-dimensional surface by applying harmonic mapping to a unit disk. This is followed by iterative refinement of the parametric coordinates for each point in the parameter domain to achieve a conformal mapping of the cortical area of interest.
Rendering of the sensory map with stimuli images
After generating the diffeomorphic sensory map, the method further comprises rendering this map with images representing stimuli that activate the corresponding cortical surface, facilitating intuitive visualization of perception centers.
Registration of sensory maps across individuals using diffeomorphic methods
The method provides for registering multiple sensory maps from different subjects. The registering step is diffeomorphic, ensuring the alignment preserves smooth, invertible transformations. It includes quantifying diffeomorphism using the Beltrami coefficient and performing the registration in an optimization framework.
Topology-preserving segmentation and decoding of fMRI data
A sensory mapping method processes fMRI data to produce topological results by: - Parameterizing the flattened cortical surface to a unit disk. - Computing a segmentation that preserves topology based on this parameterization. - Decoding fMRI data using this topology-preserving segmentation to create a topological retinotopic map. - Repeating segmentation and decoding until a theoretically-guaranteed convergence is achieved, resulting in a diffeomorphic sensory map.
Optimization-based modeling of diffeomorphic registration
The method registers a plurality of sensory maps across subjects by: - Applying the Beltrami coefficient to quantify diffeomorphism. - Modeling the registration in an optimization framework to ensure robust, accurate alignment and compliance with diffeomorphism constraints.
These inventive features together facilitate robust, topologically-accurate, and quantitative mapping of human sensory cortical areas, supporting high-fidelity alignment, visualization, and analysis within and across subjects.
Stated Advantages
Reduces noise and fixes topology violations in sensory maps generated from functional imaging data.
Produces diffeomorphic (smooth, differentiable, and invertible) maps that are biologically plausible and compatible with neurophysiological conclusions.
Enables precise quantification of sensory cortical area properties, suitable for research and clinical applications such as diagnosis and prognosis.
Improves alignment and analysis of sensory maps across time and individuals by providing robust, diffeomorphic registration methods.
Achieves superior accuracy and minimization of angle distortion and topology violations compared to conventional smoothing and registration methods.
Documented Applications
Quantitative analysis and visualization of individual human brain sensory maps, including visual retinotopic maps.
Measurement of properties of sensory areas associated with development, aging, or sensory processing diseases.
Clinical use in disease diagnosis and prognosis by quantifying sensory map features in patients.
Alignment and group analysis of sensory maps across multiple individuals to improve atlas accuracy.
Extension to other sensory cortical areas such as auditory, somatosensory, and olfactory cortex for mapping and quantification.
Interested in licensing this patent?