Lung cancer is the most prevalent type of cancer and the leading cause of cancer-related deaths worldwide. Coherent anti-Stokes Raman scattering (CARS) is capable of providing cellular-level images and resolving pathologically related features on human lung tissues. However, conventional means of analyzing CARS images requires extensive image processing, feature engineering, and human intervention. This study demonstrates the feasibility of applying a deep learning algorithm to automatically differentiate normal and cancerous lung tissue images acquired by CARS. We leverage the features learned by pretrained deep neural networks and retrain the model using CARS images as the input. We achieve 89.2% accuracy in classifying normal, small-cell carcinoma, adenocarcinoma, and squamous cell carcinoma lung images. This computational method is a step toward on-the-spot diagnosis of lung cancer and can be further strengthened by the efforts aimed at miniaturizing the CARS technique for fiber-based microendoscopic imaging.
A Coherent Anti-Stokes Raman Scattering (CARS) microendoscope probe for early stage label-free prostate cancer diagnosis at single cell resolution is presented. The handheld CARS microendoscope probe includes a customized micro-electromechanical systems (MEMS) scanning mirror as well as miniature optical and mechanical components. In our design, the excitation laser (pump and stokes beams) from the fiber is collimated, reflected by the reflecting mirror, and transmitted via a 2D MEMS scanning mirror and a micro-objective system onto the sample; emission in the epi-direction is returned through the micro-objective lens, MEMS and reflecting mirror, and collimation system, and finally the emission signal is collected by a photomultiplier tube (PMT). The exit pupil diameter of the collimator system is designed to match the diameter of the MEMS mirror and the entrance pupil diameter of the micro-objective system. The back aperture diameter of the micro-objective system is designed according to the largest MEMS scanning angle and the distance between the MEMS mirror and the back aperture. To increase the numerical aperture (NA) of the micro-objective system in order to enhance the signal collection efficiency, the back aperture diameter of the micro-objective system is enlarged with an upfront achromatic wide angle Keplerian telescope beam expander. The integration of a miniaturized micro-optics probe with optical fiber CARS microscopy opens up the possibility of in vivo molecular imaging for cancer diagnosis and surgical intervention.
Label-free multiphoton imaging is promising for replacing biopsy and could offer new strategies for intraoperative or surgical applications. Coherent anti-Stokes Raman scattering (CARS) imaging could provide lipid-band contrast, and second harmonic generation (SHG) imaging is useful for imaging collagen, tendon and muscle fibers. A combination of these two imaging modalities could provide rich information and this combination has been studied by researchers to investigate diseases through microscopy imaging. The combination of these two imaging modalities in endomicroscopy imaging has been rarely investigated. In this research, a fiber bundle consisted of one excitation fiber and 18 collection fibers was developed in our endomicroscopy prototype. The 18 collection fibers were divided into two collection channels with 9 fibers in each channel. These two channels could be used together as one channel for effective signal collection or used separately for simplifying detection part of the system. Differences of collection pattern of these two channels were investigated. Collection difference of central excitation fiber and surrounding 18 fibers was also investigated, which reveals the potential ability of this system to measure forward to backward (F/B) ratio in SHG imaging. CARS imaging of mouse adipocyte and SHG imaging of mouse tail tendon were performed to demonstrate the CARS and SHG tissue imaging performance of this system. Simultaneous CARS and SHG imaging ability of this system was demonstrated by mouse tail imaging. This fiber bundle based endomicroscopy imaging prototype, offers a promising platform for constructing efficient fiber-based CARS and SHG multimodal endomicroscopes for label free intraoperative imaging applications.
The development of real-time, label-free imaging techniques has recently attracted research interest for in situ differentiation of cancerous lesions from normal tissues. Molecule-specific intrinsic contrast can arise from label-free imaging techniques such as Coherent Anti-Stokes Raman Scattering (CARS), Two-Photon Excited AutoFluorescence (TPEAF), and Second Harmonic Generation (SHG), which, in combination, would hold the promise of a powerful label-free tool for cancer diagnosis. Among cancer-related deaths, lung carcinoma is the leading cause for both sexes. Although early treatment can increase the survival rate dramatically, lesion detection and precise diagnosis at an early stage is unusual due to its asymptomatic nature and limitations of current diagnostic techniques that make screening difficult. We investigated the potential of using multimodality nonlinear optical microscopy that incorporates CARS, TPEAF, and SHG techniques for differentiation of lung cancer from normal tissue. Cancerous and non-cancerous lung tissue samples from patients were imaged using CARS, TPEAF, and SHG techniques for comparison. These images showed good pathology correlation with hematoxylin and eosin (H and E) stained sections from the same tissue samples. Ongoing work includes imaging at various penetration depths to show three-dimensional morphologies of tumor cell nuclei using CARS, elastin using TPEAF, and collagen using SHG and developing classification algorithms for quantitative feature extraction to enable lung cancer diagnosis. Our results indicate that via real-time morphology analyses, a multimodality nonlinear optical imaging platform potentially offers a powerful minimally-invasive way to differentiate cancer lesions from surrounding non-tumor tissues in vivo for clinical applications.
The ability to visualize cellular structures and tissue molecular signatures in a live body could revolutionize the practice
of surgery. Specifically, such technology is promising for replacing tissue extraction biopsy and offering new strategies
for a broad range of intraoperative or surgical applications, including early cancer detection, tumor margin identification,
nerve damage avoidance, and surgical outcomes enhancement. Coherent anti-Stokes Raman scattering (CARS)
microendoscopy offers a way to achieve this with label-free imaging capability and sub-cellular resolution. However,
efficient collection of epi-CARS signals and reduction of nonlinear effects in fibers are two major challenges
encountered in the development of fiber-based CARS microendoscopy. To circumvent this problem, we designed and
developed a fiber bundle for a CARS microendoscopy prototype. The excitation lasers were delivered by a single
multimode fiber at the center of the bundle while the epi-CARS signals were collected by multiple MMFs surrounding
the central fiber. A polarization scheme was employed to suppress the four-wave mixing (FWM) effect in the excitation
fiber. Our experimental results suggest that, with this fiber bundle and the polarization FWM-suppressing scheme, the
signal-to-noise ratio of the CARS images was greatly enhanced through a combination of high collection efficiency of
epi-CARS signals, isolation of excitation lasers, and suppression of FWM. Tissue imaging capability of the
microendoscopy prototype was demonstrated by ex vivo imaging on mouse skin and lung tissues. This fiber bundle-based
CARS microendoscopy prototype, with the polarization FWM-suppressing scheme, offers a promising platform
for constructing efficient fiber-based CARS microendoscopes for label free intraoperative imaging applications.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.