Medical hyperspectral imaging for evaluation of tissue and tumor
Apparatus and methods for hyperspectral imaging analysis that assists in real and near-real time assessment of biological tissue condition, viability, and type, and monitoring the above over time. Embodiments of the invention are particularly useful in surgery, clinical procedures, tissue assessment, diagnostic procedures, health monitoring, and medical evaluations, especially in the detection and treatment of cancer.
1. A medical instrument for detecting a tumor, the instrument comprising:
a first stage optic configured to receive light from in-vivo tissue of a subject in a region of interest;
a spectral separator configured to filter the received light;
an imaging sensor for recording an image of the filtered light;
a diagnostic protocol module adapted for tumor detection in in-vivo tissue;
a diagnostic processor configured to:
instruct the spectral separator to filter the received light in a plurality of spectral bands based on the diagnostic protocol module,
instruct the imaging sensor to record a plurality of images of the region of interest based on the diagnostic protocol module, each image of the plurality of images corresponding to a spectral band of the plurality of spectral bands,
obtain a hyperspectrally resolved image based on the plurality of images, and
obtain a pseudo-color image based on the hyperspectrally resolved image, the pseudo-color image enhancing the visibility of any tumor tissue present in the region of interest; and
a projection subsystem configured to project the pseudo-color image onto the region of interest, wherein the projection subsystem comprises an image projector with a field-of-view co-aligned with a field of view of the first stage optic.
2. The medical instrument of claim 1 , wherein the medical instrument weighs less than 25 pounds.
3. The medical instrument of claim 1 , wherein the diagnostic processor comprises an image acquisition interface having an input responsive to the imaging sensor and an output provided to the general-purpose operating module.
4. The medical instrument of claim 1 , wherein the diagnostic processor comprises a filter control interface configured to instruct the spectral separator to filter the light received by the first stage optic independently of an illumination of the region of interest.
5. The medical instrument of claim 1 , further comprising a second-stage optic configured to focus the filtered light onto the imaging sensor.
6. The medical instrument of claim 1 , wherein the filtered light includes wavelengths in one or more of UV, visible, NIR, and IR spectral regions.
7. The medical instrument of claim 1 , wherein the pseudo-color image includes information about at least one of the presence of a tumor, the presence of a residual tumor at a margin of a surgical excision bed, and the progression of a tumor.
8. The medical instrument of claim 7 , wherein the tumor comprises breast cancer or lymphoma.
9. The medical instrument of claim 7 , wherein the information about the progression of the tumor includes at least one of tumor stage grading and microvascular changes in a vascular tissue.
10. The medical instrument of 1 , further comprising a light emitting diode configured to illuminate the region of interest.
11. The medical instrument of claim 1 , wherein the diagnostic processor is further configured to preprocess the images of the plurality of images by removing background radiation.
12. The medical instrument of claim 11 , wherein the diagnostic processor is further configured to:
convert intensities in the hyperspectrally resolved image into units of optical density by taking a negative logarithm of each decimal base to obtain a converted image;
decompose a spectrum for each pixel of the converted image into several independent components;
determine three planes for a first RGB image based on the decomposed spectrum for each pixel;
determine a sharpness factor plane based on the converted image;
convert the first RGB image to a hue-saturation-value/intensity image having a plane;
scale the hue-saturation-value/intensity image plane based on the sharpness factor plane;
convert the scaled hue-saturation-value/intensity image into a second RGB image; and
remove outliers in color intensity beyond a standard deviation in the second RGB image and stretch the resulting image to fill an entire color intensity range between 0 and 1 to obtain the pseudo-color image.
13. The medical instrument of claim 12 , wherein the diagnostic processor is configured to determine the three planes for the first RGB image based on one or more characteristic features of the spectrum of each pixel.
14. The medical instrument of claim 12 , wherein the diagnostic processor is configured to determine the sharpness factor plane based on a combination of images of the plurality of images obtained in different spectral regions.
15. The medical instrument of claim 12 , wherein the diagnostic processor is configured to determine the sharpness factor plane based on a ratio of a yellow plane in the range of about 550-580 nm to a green plane in the range of about 495-525 nm.
16. The medical instrument of claim 12 , wherein the diagnostic processor is configured to determine the sharpness factor plane based on a combination of oxyhemoglobin and deoxyhemoglobin spectral components.
17. The medical instrument of claim 12 , wherein the diagnostic processor is configured to determine the sharpness factor plane based on a ratio between a wavelength in the red region in the range 615-710 nm and a wavelength in the yellow region in the range of about 550-580 nm or in the orange region in the range of about 580-615 nm.
18. The medical instrument of claim 12 , wherein the diagnostic processor is configured to remove outliers beyond three standard deviations.
19. The medical instrument of claim 1 , wherein the filtered light includes wavelengths in the visible spectral region, and wherein the pseudo-color image projected onto the region of interest includes wavelengths in the visible spectral region.
20. A method for detecting a tumor, the method comprising:
receiving light from in-vivo tissue of a subject in a region of interest;
filtering the received light in a plurality of spectral bands based on a diagnostic protocol module adapted for tumor detection in in-vivo tissue;
recording a plurality of images of the region of interest based on the diagnostic protocol module, each image of the plurality of images corresponding to a spectral band of the plurality of spectral bands;
obtaining a hyperspectrally resolved image based on the plurality of images;
obtaining a pseudo-color image based on the hyperspectrally resolved image, the pseudo-color image enhancing the visibility of any tumor tissue present in the region of interest; and
projecting the pseudo-color image onto the region of interest.
21. The method of claim 20 , further comprising the steps of:
preprocessing the images of the plurality of images by removing background radiation;
converting intensities in the hyperspectrally resolved image into units of optical density by taking a negative logarithm of each decimal base to obtain a converted image;
decomposing a spectrum for each pixel of the converted image into several independent components;
determining three planes for a first RGB image based on the decomposed spectrum for each pixel;
determining a sharpness factor plane based on the converted image;
converting the first RGB image to a hue-saturation-value/intensity image having a plane;
scaling the hue-saturation-value/intensity image plane based on the sharpness factor plane;
converting the hue-saturation-value/intensity image into a second RGB image;
removing outliers in color intensity beyond a standard deviation in the second RGB image and stretching the resulting image to fill an entire color intensity range between 0 and 1 to obtain the pseudo-color image.
22. The method of claim 21 , wherein determining the three planes for the first RGB image is based on one or more characteristic features of the spectrum of each pixel.
23. The method of claim 21 , wherein determining the sharpness factor plane is based on a combination of images of the plurality of images obtained in different spectral regions.
24. The method of claim 21 , wherein the sharpness factor plane is determined based on a ratio of a yellow plane in the range of about 550-580 nm to a green plane in the range of about 495-525 nm.
25. The method of claim 21 , wherein the sharpness factor plane is determined based on a combination of oxyhemoglobin and deoxyhemoglobin spectral components.
26. The method of claim 21 , wherein the sharpness factor plane is determined based on a ratio between a wavelength in the red region in the range 615-710 nm and a wavelength in the yellow region in the range of about 550-580 nm or in the orange region in the range of about 580-615 nm.
27. The method for detecting cancer in tissue of claim 21 , wherein removing outliers beyond a standard deviation comprises removing outliers beyond three standard deviations.
28. The method of claim 20 , wherein the filtered light includes wavelengths in the visible spectral region, and wherein the pseudo-color image projected onto the region of interest includes wavelengths in the visible spectral region.
29. The method of claim 20 , wherein the pseudo-color image includes information about at least one of the presence of a tumor, the presence of a residual tumor at a margin of a surgical excision bed, and the progression of a tumor.
30. The method of claim 29 , medical instrument of claim 7 , wherein the tumor comprises breast cancer or lymphoma.
31. The method of claim 20 , further comprising collecting the received light endoscopically, laparoscopically, thoracoscopically, cystoscopically, hysteroscopically, bronchoscopically, or mediastinoscopically.
32. The method of claim 29 , wherein the information about the progression of the tumor includes at least one of tumor stage grading and microvascular changes in a vascular tissue.