Classifying breast cancer in ultrahigh-resolution optical coherence tomography images using convolutional neural networks.

Optical coherence tomography (OCT) is being investigated in breast cancer diagnostics as a real-time histology evaluation tool. We present a customized deep convolutional neural network (CNN) for classification of breast tissues in OCT B-scans. Images of human breast samples from mastectomies and breast reductions were acquired using a custom ultrahigh-resolution OCT system with 2.72 µm axial resolution and 5.52 µm lateral resolution. The network achieved 96.7% accuracy, 92% sensitivity, and 99.7% specificity on a dataset of 23 patients. The usage of deep learning will be important for the practical integration of OCT into clinical practice.

Particle-Tracking Microrheology Using Micro-Optical Coherence Tomography.

Clinical manifestations of cystic fibrosis (CF) result from an increase in the viscosity of the mucus secreted by epithelial cells that line the airways. Particle-tracking microrheology (PTM) is a widely accepted means of determining the viscoelastic properties of CF mucus, providing an improved understanding of this disease as well as an avenue to assess the efficacies of pharmacologic therapies aimed at decreasing mucus viscosity. Among its advantages, PTM allows the measurement of small volumes, which was recently utilized for an in situ study of CF mucus formed by airway cell cultures. Typically, particle tracks are obtained from fluorescence microscopy video images, although this limits one's ability to distinguish particles by depth in a heterogeneous environment. Here, by performing PTM with high-resolution micro-optical coherence tomography (µOCT), we were able to characterize the viscoelastic properties of mucus, which enables simultaneous measurement of rheology with mucociliary transport parameters that we previously determined using µOCT. We obtained an accurate characterization of dextran solutions and observed a statistically significant difference in the viscosities of mucus secreted by normal and CF human airway cell cultures. We further characterized the effects of noise and imaging parameters on the sensitivity of µOCT-PTM by performing theoretical and numerical analyses, which show that our system can accurately quantify viscosities over the range that is characteristic of CF mucus. As a sensitive rheometry technique that requires very small fluid quantities, µOCT-PTM could also be generally applied to interrogate the viscosity of biological media such as blood or the vitreous humor of the eye in situ.

Optical coherence tomography holds promise to transform the diagnostic anatomic pathology gross evaluation process.

SIGNIFICANCE: Real-time histology can close a variety of gaps in tissue diagnostics. Currently, gross pathology analysis of excised tissue is dependent upon visual inspection and palpation to identify regions of interest for histopathological processing. Such analysis is limited by the variable correlation between macroscopic and microscopic findings. The current standard of care is costly, burdensome, and inefficient. AIM: We are the first to address this gap by introducing optical coherence tomography (OCT) to be integrated in real-time during the pathology grossing process. APPROACH: This is achieved by our high-resolution, ultrahigh-speed, large field-of-view OCT device designed for this clinical application. RESULTS: We demonstrate the feasibility of imaging tissue sections from multiple human organs (breast, prostate, lung, and pancreas) in a clinical gross pathology setting without interrupting standard workflows. CONCLUSIONS: OCT-based real-time histology evaluation holds promise for addressing a gap that has been present for >100 years.

Millimeter-scale chip-based supercontinuum generation for optical coherence tomography.

Supercontinuum sources for optical coherence tomography (OCT) have raised great interest as they provide broad bandwidth to enable high resolution and high power to improve imaging sensitivity. Commercial fiber-based supercontinuum systems require high pump powers to generate broad bandwidth and customized optical filters to shape/attenuate the spectra. They also have limited sensitivity and depth performance. We introduce a supercontinuum platform based on a 1-mm2 Si3N4 photonic chip for OCT. We directly pump and efficiently generate supercontinuum near 1300 nm without any postfiltering. With a 25-pJ pump pulse, we generate a broadband spectrum with a flat 3-dB bandwidth of 105 nm. Integrating the chip into a spectral domain OCT system, we achieve 105-dB sensitivity and 1.81-mm 6-dB sensitivity roll-off with 300-µW optical power on sample. We image breast tissue to demonstrate strong imaging performance. Our chip will pave the way toward portable OCT and incorporating integrated photonics into optical imaging technologies.

Fully Automated Postlumpectomy Breast Margin Assessment Utilizing Convolutional Neural Network Based Optical Coherence Tomography Image Classification Method.

BACKGROUND: The purpose of this study was to develop a deep learning classification approach to distinguish cancerous from noncancerous regions within optical coherence tomography (OCT) images of breast tissue for potential use in an intraoperative setting for margin assessment. METHODS: A custom ultrahigh-resolution OCT (UHR-OCT) system with an axial resolution of 2.7 µm and a lateral resolution of 5.5 µm was used in this study. The algorithm used an A-scan-based classification scheme and the convolutional neural network (CNN) was implemented using an 11-layer architecture consisting of serial 3 × 3 convolution kernels. Four tissue types were classified, including adipose, stroma, ductal carcinoma in situ, and invasive ductal carcinoma. RESULTS: The binary classification of cancer versus noncancer with the proposed CNN achieved 94% accuracy, 96% sensitivity, and 92% specificity. The mean five-fold validation F1 score was highest for invasive ductal carcinoma (mean standard deviation, 0.89 ± 0.09) and adipose (0.79 ± 0.17), followed by stroma (0.74 ± 0.18), and ductal carcinoma in situ (0.65 ± 0.15). CONCLUSION: It is feasible to use CNN based algorithm to accurately distinguish cancerous regions in OCT images. This fully automated method can overcome limitations of manual interpretation including interobserver variability and speed of interpretation and may enable real-time intraoperative margin assessment.