Showing papers by "Xingde Li published in 2019"
TL;DR: A novel artificial intelligence (AI)-assisted method for automated, real-time, in situ detection of glioma infiltration at high spatial resolution and excellent levels of sensitivity and specificity for detectingglioma-infiltrated brain tissue.
Abstract: Purpose: In glioma surgery, it is critical to maximize tumor resection without compromising adjacent non-cancerous brain tissue. Optical Coherence Tomography (OCT) is a non-invasive, label-free, real-time, high-resolution imaging modality that has been explored for glioma infiltration detection. Here we report a novel artificial intelligence (AI) assisted method for automated, real-time, in situ detection of glioma infiltration at high spatial resolution. Experimental Design: Volumetric OCT datasets were intraoperatively obtained from resected brain tissue specimens of 21 patients with glioma tumors of different stages and labeled as either non-cancerous or glioma-infiltrated based on histopathology evaluation of the tissue specimens (gold standard). Labeled OCT images from 12 patients were used as the training dataset to develop the AI assisted OCT-based method for automated detection of glioma-infiltrated brain tissue. Unlabeled OCT images from the other 9 patients were used as the validation dataset to quantify the method detection performance. Results: Our method achieved excellent levels of sensitivity (~100%) and specificity (~85%) for detecting glioma-infiltrated tissue with high spatial resolution (16 mm laterally) and processing speed (~100,020 OCT A-lines/second). Conclusions: Previous methods for OCT-based detection of glioma-infiltrated brain tissue rely on estimating the tissue optical attenuation coefficient from the OCT signal, which requires sacrificing spatial resolution to increase signal quality, and performing systematic calibration procedures using tissue phantoms. By overcoming these major challenges, our AI-assisted method will enable implementing practical OCT-guided surgical tools for continuous, real-time and accurate intra-operative detection of glioma-infiltrated brain tissue, facilitating maximal glioma resection and superior surgical outcomes for glioma patients.
28 citations
TL;DR: Parallel-trained deep neural networks for automated endoscopic OCT image segmentation feasible even with a limited training data set are reported and quantitative changes in the top esophageal layers' thickness in the EOE model are demonstrated.
Abstract: We report parallel-trained deep neural networks for automated endoscopic OCT image segmentation feasible even with a limited training data set. These U-Net-based deep neural networks were trained using a modified dice loss function and manual segmentations of ultrahigh-resolution cross-sectional images collected by an 800 nm OCT endoscopic system. The method was tested on in vivo guinea pig esophagus images. Results showed its robust layer segmentation capability with a boundary error of 1.4 µm insensitive to lay topology disorders. To further illustrate its clinical potential, the method was applied to differentiating in vivo OCT esophagus images from an eosinophilic esophagitis (EOE) model and its control group, and the results clearly demonstrated quantitative changes in the top esophageal layers’ thickness in the EOE model.
28 citations
TL;DR: A biopsy‐needle compatible rigid probe, capable of performing three‐dimensional (3D) two‐photon optical biopsy of an internal organ has been reported and has been successfully demonstrated on mouse kidney in vivo and in situ.
Abstract: In this work, we report a biopsy-needle compatible rigid probe, capable of performing three-dimensional (3D) two-photon optical biopsy. The probe has a small outer diameter of 1.75 mm and fits inside a gauge-14 biopsy needle to reach internal organs. A carefully designed focus scanning mechanism has been implemented in the rigid probe, which, along with a rapid two-dimensional MEMS scanner, enables 3D imaging. Fast image acquisition up to 10 frames per second is possible, dramatically reducing motion artifacts during in vivo imaging. Equipped with a high-numerical aperture micro-objective, the miniature rigid probe offers a high two-photon resolution (0.833 × 6.11 μm, lateral × axial), a lateral field of view of 120 μm, and an axial focus tuning range of 200 μm. In addition to imaging of mouse internal organs and subcutaneous tumor in vivo, first-of-its-kind depth-resolved two-photon optical biopsy of an internal organ has been successfully demonstrated on mouse kidney in vivo and in situ.
22 citations
TL;DR: The ultrahigh resolution and excellent imaging contrast at 800 nm of the tethered OCT capsule suggest the potential of the technology as an enabling tool for surveillance of early esophageal diseases on awake patients without the need for sedation.
Abstract: Endoscopic optical coherence tomography (OCT) is a noninvasive technology allowing for imaging of tissue microanatomies of luminal organs in real time. Conventional endoscopic OCT operates at 1300 nm wavelength region with a suboptimal axial resolution limited to 8-20 μm. In this paper, we present the first ultrahigh-resolution tethered OCT capsule operating at 800 nm and offering about 3- to 4-fold improvement of axial resolution (plus enhanced imaging contrast). The capsule uses diffractive optics to manage chromatic aberration over a full ~200 nm spectral bandwidth centering around 830 nm, enabling to achieve super-achromaticity and an axial resolution of ~2.6 μm in air. The performance of the OCT capsule is demonstrated by volumetric imaging of swine esophagus ex vivo and sheep esophagus in vivo, where fine anatomic structures including the sub-epithelial layers are clearly identified. The ultrahigh resolution and excellent imaging contrast at 800 nm of the tethered capsule suggest the potential of the technology as an enabling tool for surveillance of early esophageal diseases on awake patients without the need for sedation.
21 citations
TL;DR: An ultralow-voltage, electrothermal (ET) micro-electro-mechanical system (MEMS) based probe for forward-viewing endoscopic optical coherence tomography (OCT) imaging was reported and its performance was successfully demonstrated with cross-sectional imaging of biological tissues ex vivo and in vivo at a speed up to 200 frames per second.
Abstract: We report an ultralow-voltage, electrothermal (ET) micro-electro-mechanical system (MEMS) based probe for forward-viewing endoscopic optical coherence tomography (OCT) imaging. The fully assembled probe has a diameter of 5.5 mm and a length of 55 mm, including the imaging optics and a 40 mm long fiber-optic cantilever attached on a micro-platform of the bimorph ET MEMS actuator. The ET MEMS actuator provides a sufficient mechanical actuation force as well as a large vertical displacement, achieving up to a 3 mm optical scanning range with only a 3 VACp-p drive voltage with a 1.5 VDC offset. The imaging probe was integrated with a swept-source OCT system of a 100 kHz A-scan rate, and its performance was successfully demonstrated with cross-sectional imaging of biological tissues ex vivo and in vivo at a speed up to 200 frames per second.
12 citations
TL;DR: An unstructured review of published evidence of available pulmonary imaging technologies along with a demonstration of state-of-the-art OCT imaging technology of in vivo human and animal airways are presented.
8 citations