Expand this Topic clickable element to expand a topic
Skip to content
Optica Publishing Group

Chemically-selective imaging of brain structures with CARS microscopy

Open Access Open Access

Abstract

We demonstrate the use of coherent anti-Stokes Raman scattering (CARS) microscopy to image brain structure and pathology ex vivo. Although non-invasive clinical brain imaging with CT, MRI and PET has transformed the diagnosis of neurologic disease, definitive pre-operative distinction of neoplastic and benign pathologies remains elusive. Definitive diagnosis still requires brain biopsy in a significant number of cases. CARS microscopy, a nonlinear, vibrationally-sensitive technique, is capable of high-sensitivity chemically-selective three-dimensional imaging without exogenous labeling agents. Like MRI, CARS can be tuned to provide a wide variety of possible tissue contrasts, but with sub-cellular spatial resolution and near real time temporal resolution. These attributes make CARS an ideal technique for fast, minimally invasive, non-destructive, molecularly specific intraoperative optical diagnosis of brain lesions. This promises significant clinical benefit to neurosurgical patients by providing definitive diagnosis of neoplasia prior to tissue biopsy or resection. CARS imaging can augment the diagnostic accuracy of traditional frozen section histopathology in needle biopsy and dynamically define the margins of tumor resection during brain surgery. This report illustrates the feasibility of in vivo CARS vibrational histology as a clinical tool for neuropathological diagnosis by demonstrating the use of CARS microscopy in identifying normal brain structures and primary glioma in fresh unfixed and unstained ex vivo brain tissue.

©2007 Optical Society of America

Full Article  |  PDF Article
More Like This
Ex vivo and in vivo imaging of myelin fibers in mouse brain by coherent anti-Stokes Raman scattering microscopy

Yan Fu, T. Brandon Huff, Han-Wei Wang, Haifeng Wang, and Ji-Xin Cheng
Opt. Express 16(24) 19396-19409 (2008)

Brain tumor delineation enhanced by moxifloxacin-based two-photon/CARS combined microscopy

Viet-Hoan Le, Su Woong Yoo, Yeoreum Yoon, Taejun Wang, Bumju Kim, Seunghun Lee, Kyung-Hwa Lee, Ki Hean Kim, and Euiheon Chung
Biomed. Opt. Express 8(4) 2148-2161 (2017)

Liquid-crystal tunable filter spectral imaging for brain tumor demarcation

Steven C. Gebhart, Reid C. Thompson, and Anita Mahadevan-Jansen
Appl. Opt. 46(10) 1896-1910 (2007)

Supplementary Material (2)

Media 1: MOV (14654 KB)     
Media 2: MOV (3513 KB)     

Cited By

Optica participates in Crossref's Cited-By Linking service. Citing articles from Optica Publishing Group journals and other participating publishers are listed here.

Alert me when this article is cited.


Figures (9)

Fig. 1.
Fig. 1. Diagrams of the CARS microscopy system. (a) Laser configuration used to maximize the CARS imaging depth. The signal and idler pulse trains of the OPO exit collinearly and overlapped in time. (b) Laser configuration used to maximize the detected epi-CARS signal. The signal of the OPO is combined with the synchronized 1064 nm output of the Nd:VYO4 on a dichroic mirror (DM). A delay stage is used to overlap the two pulse trains in time. (c) Schematic of the CARS microscope. An inverted microscope was modified to use a nondescanned epi-detector to maximize the epi-CARS signal from the tissue.
Fig. 2.
Fig. 2. Mosaic CARS microscopy image of a healthy APP and PS1AD mouse brain. Each single image is 700×700 µm in size. The brain slice was sectioned 2.8 mm posterior to the bregma. The single image enclosed by the white square is shown in Fig. 3. The pump and Stokes wavelengths were 816.7 and 1064 nm, respectively.
Fig. 3.
Fig. 3. (a). Single CARS image from Fig. 2. The pump and Stokes wavelengths were 816.7 and 1064 nm, respectively. (b) H&E image of the same region of the same mouse brain. Note the structures observable in both images, from upper left corner to the bottom right are the cortex, corpus callosum, oriens layer, and pyramidal layer.
Fig. 4.
Fig. 4. (3.4 MB) Animation and three-dimensional reconstruction of a 50 µm depth stack in brain tissue taken with CARS microscopy. The first segment of the animation shows an uncorrected depth stack taken at 20x magnification in a coronal section containing the cortex (top) and white matter tracks streaming from the corpus collosum (bottom, not shown). At a depth of ~50 µm in the tissue, scattering from the white matter tracks has significantly decreased the CARS signal, while the CARS signal persists in the less scattering grey matter of the cortex. The second segment of the animation shows the three-dimensional reconstruction of the depth stack. [Media 1]
Fig. 5.
Fig. 5. Imaging the brain with chemical selectivity. All images were taken from the same brain used in the Fig. 2 mosaic. A nonresonant image was subtracted from (a) and (b). Regions of positive contrast (signal greater than the nonresonant background) are color coded green, while regions of negative contrast are color coded blue. (a) Positive lipid contrast image of CARS microscopy at 2845 cm-1. (b) Negative lipid contrast image of CARS microscopy at 2955 cm-1 showing the cell bodies with positive contrast, (c) Corresponding H&E image. Neuronal cell bodies appear blue due to the haemotaxylin stain, while axons are dyed pink by eosin. The Stokes wavelength was 1064 nm.
Fig. 6.
Fig. 6. Raman spectra acquired from fresh coronal slices of SCID mouse brain tissue immersed in PBS solution. The red line represents the Raman spectrum from the corpus collosum, while the blue line is the Raman spectrum from the nuclei-rich pyramidal layer of the hippocampus. Raman spectra were acquired in 100s and averaged twice.
Fig. 7.
Fig. 7. CARS images of astrocytoma in a SCID mouse sacrificed 4 weeks after inoculation of tumor cells. The pump and Stokes wavelengths were 924.0 and 1254.2 nm, respectively. (a) A low resolution, large field of view mosaic CARS microscopy image provides chemically-selective anatomical information. Part (b) illustrates the ability of CARS to produce higher resolution images, in this case corresponding to the area enclosed by the rectangle in (a). This 80X, 175×175 µm image demonstrates the microscopic infiltration at the boundary between the tumor and normal tissue with a precision comparable to the conventional fixed tissue H&E histology images in Fig. 8.
Fig. 8.
Fig. 8. H&E images of a brain tumor harvested from a different SCID mouse in the same group as the mouse imaged in Fig. 7. The brain slice was taken at approximately the same location as that of Fig. 7. (a) The boundary between the tumor and normal tissue at 4X, (b) 10X, and (c) 20X magnification.
Fig. 9.
Fig. 9. Raman spectrum acquired from the center of a large astrocytoma in a fresh coronal slice of a SCID mouse brain tissue immersed in PBS. There is minimal Raman intensity at 2852 cm-1 and 2880 cm-1, indicating the tissue is lipid deficient relative to normal brain tissue. The peaks at 2920 cm-1 and 2960 cm-1 arise due to the CH3 symmetric and anti-symmetric stretching vibrations. The large signal increase after 3000 cm-1 is due to the water stretching vibration of PBS solution. The spectrum was integrated for 200s and averaged twice.
Select as filters


Select Topics Cancel
© Copyright 2024 | Optica Publishing Group. All rights reserved, including rights for text and data mining and training of artificial technologies or similar technologies.