Comprehensive confocal endomicroscopy of the esophagus in vivo

Scattering-Based Light-Sheet Microscopy Imaging of Human Papillomavirus-Associated Squamous Lesions of the Anal Canal: A Proof-of-Principle Study

Demand for anal cancer screening is expected to rise following the recent publication of the Anal Cancer-HSIL Outcomes Research trial, which showed that treatment of high-grade squamous intraepithelial lesions significantly reduces the rate of progression to anal cancer. While screening for human papillomavirus-associated squamous lesions in the cervix is well established and effective, this is less true for other sites in the lower anogenital tract. Current anal cancer screening and prevention rely on high-resolution anoscopy with biopsies. This procedure has a steep learning curve for providers and may cause patient discomfort. Scattering-based light-sheet microscopy (sLSM) is a novel imaging modality with the potential to mitigate these challenges through real-time, microscopic visualization of disease-susceptible tissue. Here, we report a proof-of-principle study that establishes feasibility of dysplasia detection using an sLSM device. We imaged 110 anal biopsy specimens collected prospectively at our institution's dysplasia clinic (including 30 nondysplastic, 40 low-grade squamous intraepithelial lesion, and 40 high-grade squamous intraepithelial lesion specimens) and found that these optical images are highly interpretable and accurately recapitulate histopathologic features traditionally used for the diagnosis of human papillomavirus-associated squamous dysplasia. A reader study to assess diagnostic accuracy suggests that sLSM images are noninferior to hematoxylin and eosin images for the detection of anal dysplasia (sLSM accuracy = 0.87; hematoxylin and eosin accuracy = 0.80; P = .066). Given these results, we believe that sLSM technology holds great potential to enhance the efficacy of anal cancer screening by allowing accurate sampling of diagnostic tissue at the time of anoscopy. While the current imaging study was performed on ex vivo biopsy specimens, we are currently developing a handheld device for in vivo imaging that will provide immediate microscopic guidance to high-resolution anoscopy providers.

CAESNet: Convolutional AutoEncoder based Semi-supervised Network for improving multiclass classification of endomicroscopic images

OBJECTIVE

This article presents a novel method of semisupervised learning using convolutional autoencoders for optical endomicroscopic images. Optical endomicroscopy (OE) is a newly emerged biomedical imaging modality that can support real-time clinical decisions for the grade of dysplasia. To enable real-time decision making, computer-aided diagnosis (CAD) is essential for its high speed and objectivity. However, traditional supervised CAD requires a large amount of training data. Compared with the limited number of labeled images, we can collect a larger number of unlabeled images. To utilize these unlabeled images, we have developed a Convolutional AutoEncoder based Semi-supervised Network (CAESNet) for improving the classification performance.

MATERIALS AND METHODS

We applied our method to an OE dataset collected from patients undergoing endoscope-based confocal laser endomicroscopy procedures for Barrett's esophagus at Emory Hospital, which consists of 429 labeled images and 2826 unlabeled images. Our CAESNet consists of an encoder with 5 convolutional layers, a decoder with 5 transposed convolutional layers, and a classification network with 2 fully connected layers and a softmax layer. In the unsupervised stage, we first update the encoder and decoder with both labeled and unlabeled images to learn an efficient feature representation. In the supervised stage, we further update the encoder and the classification network with only labeled images for multiclass classification of the OE images.

RESULTS

Our proposed semisupervised method CAESNet achieves the best average performance for multiclass classification of OE images, which surpasses the performance of supervised methods including standard convolutional networks and convolutional autoencoder network.

CONCLUSIONS

Our semisupervised CAESNet can efficiently utilize the unlabeled OE images, which improves the diagnosis and decision making for patients with Barrett's esophagus.

Immediate Label-Free Ex Vivo Evaluation of Human Brain Tumor Biopsies With Confocal Reflectance Microscopy

Confocal microscopy utilizing fluorescent dyes is widely gaining use in the clinical setting as a diagnostic tool. Reflectance confocal microscopy is a method of visualizing tissue specimens without fluorescent dyes while relying on the natural refractile properties of cellular and subcellular structures. We prospectively evaluated 76 CNS lesions with confocal reflectance microscopy (CRM) to determine cellularity, architecture, and morphological characteristics. A neuropathologist found that all cases showed similar histopathological features when compared to matched hematoxylin and eosin-stained sections. RNA isolated from 7 tissues following CRM imaging retained high RNA integrity, suggesting that CRM does not alter tissue properties for molecular studies. A neuropathologist and surgical pathologist masked to the imaging results independently evaluated a subset of CRM images. In these evaluations, 100% of images reviewed by the neuropathologist and 95.7% of images reviewed by the surgical pathologist were correctly diagnosed as lesional or nonlesional. Furthermore, 97.9% and 91.5% of cases were correctly diagnosed as tumor or not tumor by the neuropathologist and surgical pathologist, respectively, while 95.8% and 85.1% were identified with the correct diagnosis. Our data indicate that CRM is a useful tool for rapidly screening patient biopsies for diagnostic adequacy, molecular studies, and biobanking.

Large-area spectrally encoded confocal endomicroscopy of the human esophagus in vivo

BACKGROUND AND OBJECTIVE

Diagnosis of esophageal diseases is often hampered by sampling errors that are inherent in endoscopic biopsy, the standard of care. Spectrally encoded confocal microscopy (SECM) is a high-speed reflectance confocal endomicroscopy technology that has the potential to visualize cellular features from large regions of the esophagus, greatly decreasing the likelihood of sampling error. In this paper, we report results from a pilot clinical study imaging the human esophagus in vivo with a prototype SECM endoscopic probe.

MATERIALS AND METHODS

In this pilot clinical study, six patients undergoing esophagogastroduodenoscopy (EGD) for surveillance of Barrett's esophagus (BE) were imaged with the SECM endoscopic probe. The device had a diameter of 7 mm, a length of 2 m, and a rapid-exchange guide wire provision for esophageal placement. During EGD, the distal portion of the esophagus of each patient was sprayed with 2.5% acetic acid to enhance nuclear contrast. The SECM endoscopic probe was then introduced over the guide wire to the distal esophagus and large-area confocal images were obtained by helically scanning the optics within the SECM probe.

RESULTS

Large area confocal images of the distal esophagus (image length = 4.3-10 cm; image width = 2.2 cm) were rapidly acquired at a rate of ∼9 mm² /second, resulting in short procedural times (1.8-4 minutes). SECM enabled the visualization of clinically relevant architectural and cellular features of the proximal stomach and normal and diseased esophagus, including squamous cell nuclei, BE glands, and goblet cells.

CONCLUSIONS

This study demonstrates that comprehensive spectrally encoded confocal endomicroscopy is feasible and can be used to visualize architectural and cellular microscopic features from large segments of the distal esophagus at the gastroesophageal junction. By providing microscopic images that are less subject to sampling error, this technology may find utility in guiding biopsy and planning and assessing endoscopic therapy. Lasers Surg. Med. 49:233-239, 2017. © 2016 Wiley Periodicals, Inc.

High-resolution handheld rigid endomicroscope based on full-field optical coherence tomography

Full-field optical coherence tomography (FF-OCT) is a powerful tool for nondestructive assessment of biological tissue, i.e., for the structural examination of tissue in depth at a cellular resolution. Mostly known as a microscopy device for ex vivo analysis, FF-OCT has also been adapted to endoscopy setups since it shows good potential for in situ cancer diagnosis and biopsy guidance. Nevertheless, all the attempts to perform endoscopic FF-OCT imaging did not go beyond lab setups. We describe here, to the best of our knowledge, the first handheld FF-OCT endoscope based on a tandem interferometry assembly using incoherent illumination. A common-path passive imaging interferometer at the tip of an optical probe makes it robust and insensitive to environmental perturbations, and a low finesse Fabry-Perot processing interferometer guarantees a compact system. A good resolution (2.7 μm transverse and 6 μm axial) is maintained through the long distance, small diameter relay optics of the probe, and a good signal-to-noise ratio is achieved in a limited 100 ms acquisition time. High-resolution images and a movie of a rat brain slice have been recorded by moving the contact endoscope over the surface of the sample, allowing for tissue microscopic exploration at 20 m under the surface. These promising ex vivo results open new perspectives for in vivo imaging of biological tissue, in particular, in the field of cancer and surgical margin assessment.

Miniature objective lens with variable focus for confocal endomicroscopy

Spectrally encoded confocal microscopy (SECM) is a reflectance confocal microscopy technology that can rapidly image large areas of luminal organs at microscopic resolution. One of the main challenges for large-area SECM imaging in vivo is maintaining the same imaging depth within the tissue when patient motion and tissue surface irregularity are present. In this paper, we report the development of a miniature vari-focal objective lens that can be used in an SECM endoscopic probe to conduct adaptive focusing and to maintain the same imaging depth during in vivo imaging. The vari-focal objective lens is composed of an aspheric singlet with an NA of 0.5, a miniature water chamber, and a thin elastic membrane. The water volume within the chamber was changed to control curvature of the elastic membrane, which subsequently altered the position of the SECM focus. The vari-focal objective lens has a diameter of 5 mm and thickness of 4 mm. A vari-focal range of 240 μm was achieved while maintaining lateral resolution better than 2.6 μm and axial resolution better than 26 μm. Volumetric SECM images of swine esophageal tissues were obtained over the vari-focal range of 260 μm. SECM images clearly visualized cellular features of the swine esophagus at all focal depths, including basal cell nuclei, papillae, and lamina propria.