Spectral imaging methods and systems

Details Spectral imaging methods and systems Spectral imaging methods and systems

2002-06-25

2004-06-15

Smith, Zandra V. (Department: 2877)

Optics: measuring and testing

By dispersed light spectroscopy

Utilizing a spectrometer

Reexamination Certificate

active

06750964

ABSTRACT:

BACKGROUND
Advances in the molecular understanding of many diseases and conditions depend, to a large extent, on microscopic evaluation of tissues. For example, malignant cells are identified by evaluating tissue samples. It is one of the most arduous and time-consuming tasks in pathology. To facilitate the evaluation, tissue samples can be treated with stains to provide feature contrast.
Human color vision is a form of spectral imaging by which we determine the intensity and proportion of wavelengths present in our environment, and their spatial distribution. However, unlike the eye, which breaks up the light content of an image into red, green, and blue, instrument-assisted spectral imaging can use an arbitrarily large number of wavelength-classes. Furthermore, it can extend the range to include the ultraviolet and infrared regions of the spectrum invisible to the unaided eye. The result of spectral imaging is a data set (known as a data cube) in which spectral information is present at every picture-element (pixel) of a digitally acquired image.
SUMMARY
The invention features a method that forms one or more images of an unknown sample by illuminating the sample with a weighted spectral distribution for each image. The method analyzes the one or more resulting images and identifies target features. The identifying can include classifying regions of the images based on their spectral or/and spatial properties. In addition to classifying, the method can quantify the amount of some biomarkers either in combination or apart from the classifying. The identification of particular target features can guide the automation of subsequent processes.
In general, in one aspect, the invention features a method for imaging. The method includes: obtaining a spectral weighting function indicative of an attribute of the reference sample; illuminating a target sample with light whose spectral flux distribution corresponds to the spectral weighting function to produce a corresponding target image, wherein the target image is indicative of a response of the target sample to the corresponding illumination at multiple spatial locations of the target sample; and identifying one or more target features in the target sample based on the target image.
Embodiments of the method may include any of the following features.
The spectral weighting function may be obtained from a set of reference images. The spectral weighting function may be determined based on at least one of principal component analysis, projection pursuit, independent component analysis, convex-hull analysis, and machine learning The method may also include illuminating a reference sample at each of a plurality of pure spectral bands to produce the set of reference images, wherein each reference image is indicative of a response of the reference sample to the corresponding illumination at multiple spatial locations of the reference sample.
The method may also include determining one or more additional spectral weighting functions indicative of additional attributes of the reference sample based on the set of reference images; and illuminating a target sample with light whose spectral flux distribution corresponds to each of the additional spectral weighting functions to produce additional corresponding target images, wherein each target image is indicative of a response of the target sample to the corresponding illumination at the multiple spatial locations of the target sample. The response of the reference sample may include transmission, reflectance, or fluorescence. The response of the target sample may include transmission, reflectance, or fluorescence. The spectral weighting function may include multiple ones of the spectral bands. The spectral weighting function is a single one of the spectral bands.
The method may include preparing the reference and target samples with markers suitable for chromogenic in-situ hybridization. The response of the target sample to the spectral weighting function may correlate with the presence of the in-situ hybridization marker in the target sample. The method may include preparing the reference and target samples with a marker suitable for color immunohistochemistry. The response of the target sample to the spectral weighting function correlates with the presence of the color immunohistochemistry marker in the target sample. The method may include preparing the first reference and first target with a general stain.
The identification may include assigning the one or more target features to one or more classes. The identification may include quantifying the amount of a chromogen in a target feature.
The method may further include automating a subsequent process based on the identification of the one or more target features. The subsequent process may include laser capture microdissection. The laser capture microdissection may include directing laser energy to the identified target features in the target sample to remove corresponding portions of the target sample. The method may include performing mass spectroscopy on the portions of the target sample removed by laser capture microdissection process. The mass spectroscopy may be protein mass spectroscopy. The may also include performing protein purification on the portions of the target sample removed by laser capture microdissection process. The subsequent process may include determining the extent of a condition or disease in a target sample. The condition may be fibrosis. The condition may be chronic organ rejection
The method may also include repetitively illuminating the target sample with light whose spectral flux distribution corresponds to the spectral weighting function to record the target image as a function of time. The identifying one or more target features may be based on the time dependence of the target image.
This application incorporates other documents by reference. In case of conflict, this document controls.
The details of one or more embodiments of the invention are set forth in the accompanying drawings and the description below. Other features, objects, and advantages of the invention will be apparent from the description and drawings, and from the claims.


REFERENCES:
patent: 4379233 (1983-04-01), Rosenthal
patent: 4519707 (1985-05-01), Kuffer
patent: 4669878 (1987-06-01), Meier
patent: 4800279 (1989-01-01), Hieftje et al.
patent: 5029245 (1991-07-01), Keranen et al.
patent: 5042893 (1991-08-01), Ong
patent: 5066124 (1991-11-01), Wulf
patent: 5137364 (1992-08-01), McCarthy
patent: 5424545 (1995-06-01), Block et al.
patent: 5433197 (1995-07-01), Stark
patent: 5539517 (1996-07-01), Cabib et al.
patent: 5567937 (1996-10-01), Pinkus
patent: 5608213 (1997-03-01), Pinkus et al.
patent: 5719024 (1998-02-01), Cabib et al.
patent: 5760407 (1998-06-01), Margosiak et al.
patent: 5838451 (1998-11-01), McCarthy
patent: 5912165 (1999-06-01), Cabib et al.
patent: 5991028 (1999-11-01), Cabib et al.
patent: 5995645 (1999-11-01), Soenksen et al.
patent: 6007996 (1999-12-01), McNamara et al.
patent: 6075595 (2000-06-01), Malinen
patent: 6142629 (2000-11-01), Adel et al.
patent: 6160618 (2000-12-01), Garner
patent: 6373568 (2002-04-01), Miller et al.
patent: WO 01/11343 (2001-02-01), None
W.C. Sweatt et al., “ISIS; An Information-Efficient Spectral Imaging System,” Imaging Spectrometry IV, Proc. SPIE, vol. 3438, pp. 98-106, San Diego, 1998.
B.R. Stallard, Construction of Filter Vectors for the Information-Efficient Spectral Imaging Sensor, Imaging Spectroscopy IV, Proc. SPIE, vol. 3438, pp. 172-182, San Diego, 1998.
L.O. Jimenez et al., “High Dimensional Feature Reduction via Projection Pursuit,” TR-ECE 96-5, School of Electrical Engineering, Purdue University, West Lafayette, IN 47907-1285, Apr., 1995.
Hyvarien et al., “Novel Spectroscopic Techniques for Biomedical Applications,” Optoelectronics Laboratory, Finland, SPIE vol. 2084, pp. 224-230.
Keraanen et al., “Thirty-two Channel LED Array Spectrometer Module with Compact Optomechanical Construction,” Technical Research Centre of Finland, Electronics Laboratory, Finland, SPIE vol. 1533 Opt

Affiliated with

Also associated with

Rating

Spectral imaging methods and systems does not yet have a rating. At this time, there are no reviews or comments for this patent.

If you have personal experience with Spectral imaging methods and systems, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Spectral imaging methods and systems will most certainly appreciate the feedback.

Rate now

Comments { 0 }

Profile ID: LFUS-PAI-O-3325462