Method and system for identifying tissue type based on peak-fitted parameters in high-wavenumber Raman spectrum
Inventors
Pandey, Rishikesh • Kersey, Alan
Assignees
Interested in licensing this patent?
MTEC can help explore whether this patent might be available for licensing for your application.
Publication Number
US-12038383-B2
Publication Date
2024-07-16
Expiration Date
Abstract
A method and system for identifying tissue types is provided that utilizes Raman scattered light in the high wavenumber region (HWN) of Raman spectrum. The method includes: a) interrogating a tissue specimen with at least one wavelength of light, the wavelength of light operable to produce Raman scattered light in a HWN region of Raman spectrum from the tissue specimen; b) producing signal data representative of the Raman scattering, the signal data relating signal intensity as a function of wavenumber within the HWN region; c) fitting a curve to a portion of the signal data attributable to a HWN peak within the signal data, for a plurality of different HWN peaks; d) determining a plurality of characteristics of each fitted curve; and e) determining a tissue type using the determined characteristics and stored collective data representative of fitted peaks in the HWN region of a plurality of different tissue types.
Core Innovation
The invention relates to identifying tissue types using Raman scattered light in a high wavenumber (HWN) region of a Raman spectrum. A tissue specimen is interrogated with at least one wavelength of light operable to produce Raman scattered light in the HWN region, and signal data is produced representative of Raman scattering, where the signal data relates signal intensity as a function of wavenumber within the HWN region.
The signal data includes a plurality of signal intensity contributions associated with respective biomolecules present within the tissue. A curve is fitted to a portion of the signal data attributable to the signal intensity contribution associated with a respective biomolecule, for each of a plurality of the respective biomolecules, wherein each curve includes a fitted peak in the HWN region, and a plurality of characteristics of each fitted curve is determined.
A tissue type is then determined using the determined characteristics of the fitted curves and stored collective data representative of the fitted peaks in the HWN region of a plurality of different tissue types. The stored collective data is representative of fitted peaks in the HWN region across different tissue types, including adipose tissue, benign tissue, and cancer tissue. In an alternative method, curves are fitted to a portion of the signal data attributable to HWN peaks for a plurality of different HWN peaks, and the determined characteristics include determining an area plot for each of a plurality of different tissue types at one or more respective wavenumbers within the HWN region.
Claims Coverage
The independent claims cover three inventive scopes: a method using HWN-region peak fitting tied to biomolecule-associated signal intensity contributions to determine tissue type from stored collective fitted-peak data; a corresponding system including light sources, spectrometer or monochromator, detector, and controller configured to perform the same HWN-region fitting and tissue-type determination; and an alternative method that fits curves to multiple HWN peaks and uses area plots across tissue types and wavenumbers together with stored collective fitted-peak data.
High wavenumber Raman intensity-versus-wavenumber peak fitting with biomolecule-associated contributions
Interrogating a tissue specimen with at least one wavelength of light to produce Raman scattered light in an HWN region, producing signal data relating signal intensity as a function of wavenumber within the HWN region including a plurality of signal intensity contributions associated with respective biomolecules, fitting a curve to portions of the signal data attributable to each biomolecule contribution such that each curve includes a fitted peak in the HWN region, determining characteristics of each fitted curve, and determining a tissue type using the determined characteristics and stored collective data representative of fitted peaks in the HWN region of a plurality of different tissue types.
Raman HWN region system producing signal data, fitting curves to biomolecule-associated HWN peaks, and tissue-type determination using stored collective fitted-peak data
Providing at least one light source, a spectrometer or monochromator, at least one light detector, and a system controller with non-transitory memory storing instructions that cause the controller to control the at least one light source to interrogate a tissue specimen at one or more predetermined wavelengths; control the at least one light detector to sense scattered light processed by the spectrometer or monochromator and produce signal data representative of a Raman spectrum in the HWN region; process the signal data; fit a curve to a portion of the processed signal data attributable to signal intensity contributions associated with respective biomolecules such that each curve includes a fitted peak in the HWN region; determine characteristics of each fitted curve; and determine a tissue type using the determined characteristics and stored collective data representative of fitted peaks in the HWN region of a plurality of different tissue types.
HWN peak curve fitting with area plots across tissue types and stored collective fitted-peak data
Interrogating a tissue specimen with at least one wavelength of light to produce Raman scattered light in an HWN region; producing signal data representative of Raman scattering relating signal intensity as a function of wavenumber within the HWN region; fitting a curve to a portion of the signal data attributable to an HWN peak for a plurality of different HWN peaks; determining characteristics of each fitted curve including determining an area plot for each of a plurality of different tissue types at one or more respective wavenumbers within the HWN region where each area plot is based on the signal data; and determining a tissue type using the determined characteristics and stored collective data representative of fitted peaks in the HWN region of a plurality of different tissue types.
Across the independent claims, the core coverage is HWN-region Raman signal data and curve fitting to HWN peaks, extracting fitted-curve characteristics, and determining tissue type using those characteristics together with stored collective fitted-peak data for multiple tissue types.
Stated Advantages
Interpretable/explainable classification based on fitted peaks and extracted characteristics.
Reduced input dimensionality and less overfitting.
Lower cost via non-high-resolution spectrometers.
High throughput.
Reduced susceptibility to artifacts.
Documented Applications
Automated tissue-type identification, exemplified for adipose tissue, benign tissue, and cancer tissue including malignant breast tissue.
Interested in licensing this patent?