Image analysis for classification of dysplasia in Barrett's esophagus using endoscopic optical coherence tomography

Multimodal Raman spectroscopy and optical coherence tomography for biomedical analysis

Optical techniques hold great potential to detect and monitor disease states as they are a fast, non-invasive toolkit. Raman spectroscopy (RS) in particular is a powerful label-free method capable of quantifying the biomolecular content of tissues. Still, spontaneous Raman scattering lacks information about tissue morphology due to its inability to rapidly assess a large field of view. Optical Coherence Tomography (OCT) is an interferometric optical method capable of fast, depth-resolved imaging of tissue morphology, but lacks detailed molecular contrast. In many cases, pairing label-free techniques into multimodal systems allows for a more diverse field of applications. Integrating RS and OCT into a single instrument allows for both structural imaging and biochemical interrogation of tissues and therefore offers a more comprehensive means for clinical diagnosis. This review summarizes the efforts made to date toward combining spontaneous RS-OCT instrumentation for biomedical analysis, including insights into primary design considerations and data interpretation.

Feasibility of the soft attention-based models for automatic segmentation of OCT kidney images

Clinically, optical coherence tomography (OCT) has been utilized to obtain the images of the kidney's proximal convoluted tubules (PCTs), which can be used to quantify the morphometric parameters such as tubular density and diameter. Such parameters are useful for evaluating the status of the donor kidney for transplant. Quantifying PCTs from OCT images by human readers is a time-consuming and tedious process. Despite the fact that conventional deep learning models such as conventional neural networks (CNNs) have achieved great success in the automatic segmentation of kidney OCT images, gaps remain regarding the segmentation accuracy and reliability. Attention-based deep learning model has benefits over regular CNNs as it is intended to focus on the relevant part of the image and extract features for those regions. This paper aims at developing an Attention-based UNET model for automatic image analysis, pattern recognition, and segmentation of kidney OCT images. We evaluated five methods including the Residual-Attention-UNET, Attention-UNET, standard UNET, Residual UNET, and fully convolutional neural network using 14403 OCT images from 169 transplant kidneys for training and testing. Our results show that Residual-Attention-UNET outperformed the other four methods in segmentation by showing the highest values of all the six metrics including dice score (0.81 ± 0.01), intersection over union (IOU, 0.83 ± 0.02), specificity (0.84 ± 0.02), recall (0.82 ± 0.03), precision (0.81 ± 0.01), and accuracy (0.98 ± 0.08). Our results also show that the performance of the Residual-Attention-UNET is equivalent to the human manual segmentation (dice score = 0.84 ± 0.05). Residual-Attention-UNET and Attention-UNET also demonstrated good performance when trained on a small dataset (3456 images) whereas the performance of the other three methods dropped dramatically. In conclusion, our results suggested that the soft Attention-based models and specifically Residual-Attention-UNET are powerful and reliable methods for tubule lumen identification and segmentation and can help clinical evaluation of transplant kidney viability as fast and accurate as possible.

Tissue self-attention network for the segmentation of optical coherence tomography images on the esophagus

Automatic segmentation of layered tissue is the key to esophageal optical coherence tomography (OCT) image processing. With the advent of deep learning techniques, frameworks based on a fully convolutional network are proved to be effective in classifying pixels on images. However, due to speckle noise and unfavorable imaging conditions, the esophageal tissue relevant to the diagnosis is not always easy to identify. An effective approach to address this problem is extracting more powerful feature maps, which have similar expressions for pixels in the same tissue and show discriminability from those from different tissues. In this study, we proposed a novel framework, called the tissue self-attention network (TSA-Net), which introduces the self-attention mechanism for esophageal OCT image segmentation. The self-attention module in the network is able to capture long-range context dependencies from the image and analyzes the input image in a global view, which helps to cluster pixels in the same tissue and reveal differences of different layers, thus achieving more powerful feature maps for segmentation. Experiments have visually illustrated the effectiveness of the self-attention map, and its advantages over other deep networks were also discussed.

Artificial intelligence technique in detection of early esophageal cancer

Due to the rapid progression and poor prognosis of esophageal cancer (EC), the early detection and diagnosis of early EC are of great value for the prognosis improvement of patients. However, the endoscopic detection of early EC, especially Barrett's dysplasia or squamous epithelial dysplasia, is difficult. Therefore, the requirement for more efficient methods of detection and characterization of early EC has led to intensive research in the field of artificial intelligence (AI). Deep learning (DL) has brought about breakthroughs in processing images, videos, and other aspects, whereas convolutional neural networks (CNNs) have shone lights on detection of endoscopic images and videos. Many studies on CNNs in endoscopic analysis of early EC demonstrate excellent performance including sensitivity and specificity and progress gradually from in vitro image analysis for classification to real-time detection of early esophageal neoplasia. When AI technique comes to the pathological diagnosis, borderline lesions that are difficult to determine may become easier than before. In gene diagnosis, due to the lack of tissue specificity of gene diagnostic markers, they can only be used as supplementary measures at present. In predicting the risk of cancer, there is still a lack of prospective clinical research to confirm the accuracy of the risk stratification model.

Adversarial convolutional network for esophageal tissue segmentation on OCT images

Automatic segmentation is important for esophageal OCT image processing, which is able to provide tissue characteristics such as shape and thickness for disease diagnosis. Existing automatical segmentation methods based on deep convolutional networks may not generate accurate segmentation results due to limited training set and various layer shapes. This study proposed a novel adversarial convolutional network (ACN) to segment esophageal OCT images using a convolutional network trained by adversarial learning. The proposed framework includes a generator and a discriminator, both with U-Net alike fully convolutional architecture. The discriminator is a hybrid network that discriminates whether the generated results are real and implements pixel classification at the same time. Leveraging on the adversarial training, the discriminator becomes more powerful. In addition, the adversarial loss is able to encode high order relationships of pixels, thus eliminating the requirements of post-processing. Experiments on segmenting esophageal OCT images from guinea pigs confirmed that the ACN outperforms several deep learning frameworks in pixel classification accuracy and improves the segmentation result. The potential clinical application of ACN for detecting eosinophilic esophagitis (EoE), an esophageal disease, is also presented in the experiment.

Adversarial convolutional network for esophageal tissue segmentation on OCT images

Automatic segmentation is important for esophageal OCT image processing, which is able to provide tissue characteristics such as shape and thickness for disease diagnosis. Existing automatical segmentation methods based on deep convolutional networks may not generate accurate segmentation results due to limited training set and various layer shapes. This study proposed a novel adversarial convolutional network (ACN) to segment esophageal OCT images using a convolutional network trained by adversarial learning. The proposed framework includes a generator and a discriminator, both with U-Net alike fully convolutional architecture. The discriminator is a hybrid network that discriminates whether the generated results are real and implements pixel classification at the same time. Leveraging on the adversarial training, the discriminator becomes more powerful. In addition, the adversarial loss is able to encode high order relationships of pixels, thus eliminating the requirements of post-processing. Experiments on segmenting esophageal OCT images from guinea pigs confirmed that the ACN outperforms several deep learning frameworks in pixel classification accuracy and improves the segmentation result. The potential clinical application of ACN for detecting eosinophilic esophagitis (EoE), an esophageal disease, is also presented in the experiment.

Ensemble of Deep Convolutional Neural Networks for Classification of Early Barrett’s Neoplasia Using Volumetric Laser Endomicroscopy

Barrett’s esopaghagus (BE) is a known precursor of esophageal adenocarcinoma (EAC). Patients with BE undergo regular surveillance to early detect stages of EAC. Volumetric laser endomicroscopy (VLE) is a novel technology incorporating a second-generation form of optical coherence tomography and is capable of imaging the inner tissue layers of the esophagus over a 6 cm length scan. However, interpretation of full VLE scans is still a challenge for human observers. In this work, we train an ensemble of deep convolutional neural networks to detect neoplasia in 45 BE patients, using a dataset of images acquired with VLE in a multi-center study. We achieve an area under the receiver operating characteristic curve (AUC) of 0.96 on the unseen test dataset and we compare our results with previous work done with VLE analysis, where only AUC of 0.90 was achieved via cross-validation on 18 BE patients. Our method for detecting neoplasia in BE patients facilitates future advances on patient treatment and provides clinicians with new assisting solutions to process and better understand VLE data.

Computer-Aided Analysis of Gland-Like Subsurface Hyposcattering Structures in Barrett’s Esophagus Using Optical Coherence Tomography

(1) Background: Barrett’s esophagus (BE) is a complication of chronic gastroesophageal reflux disease and is a precursor to esophageal adenocarcinoma. The clinical implication of subsurface glandular structures of Barrett’s esophagus is not well understood. Optical coherence tomography (OCT), also known as volumetric laser endomicroscopy (VLE), can assess subsurface glandular structures, which appear as subsurface hyposcattering structures (SHSs). The aim of this study is to develop a computer-aided algorithm and apply it to investigate the characteristics of SHSs in BE using clinical VLE data; (2) Methods: SHSs were identified with an initial detection followed by machine learning. Comprehensive SHS characteristics including the number, volume, depth, size and shape were quantified. Clinical VLE datasets collected from 35 patients with a history of dysplasia undergoing BE surveillance were analyzed to study the general SHS distribution and characteristics in BE. A subset of radiofrequency ablation (RFA) patient data were further analyzed to investigate the pre-RFA SHS characteristics and post-RFA treatment response; (3) Results: SHSs in the BE region were significantly shallower, more vertical, less eccentric, and more regular, as compared with squamous SHSs. SHSs in the BE region which became neosquamous epithelium after RFA were shallower than those in the regions that remained BE. Pre-ablation squamous SHSs with higher eccentricity correlated strongly with larger reduction of post-ablation BE length for less elderly patients; (4) Conclusions: The computer algorithm is potentially a valuable tool for studying the roles of SHSs in BE.

Advanced Imaging for Barrett's Esophagus and Early Neoplasia: Surface and Subsurface Imaging for Diagnosis and Management

PURPOSE OF REVIEW

Esophageal adenocarcinoma bears one of the fastest rising incidence of any cancers and generally arises in the setting of gastroesophageal reflux and Barrett's esophagus. However, early detection of neoplasia can be challenging since most patients are asymptomatic until they progress to more advanced and less curable stages, and early dysplastic lesions can be small, multifocal, and difficult to detect. Clearly, new imaging tools are needed in light of sampling error associated with random biopsies, the current standard of practice.

RECENT FINDINGS

Advances in endoscopic imaging including virtual chromoendoscopy, confocal laser endomicroscopy, and subsurface imaging with optical coherence tomography have ushered in a new era for detecting subtle neoplastic lesions. Moreover, in light of esophagus-sparing treatments for neoplastic lesions, such tools are likely to guide ablation and follow-up management. While there is no ideal single imaging modality to facilitate improved detection, staging, ablation, and follow-up of patients with dysplastic Barrett's esophagus, new advances in available technology, the potential for multimodal imaging, and the use of computer-aided diagnosis and biomarkers all hold great promise for improving detection and treatment.

Robust layer segmentation of esophageal OCT images based on graph search using edge-enhanced weights

Automatic segmentation of esophageal layers in OCT images is crucial for studying esophageal diseases and computer-assisted diagnosis. This work aims to improve the current techniques to increase the accuracy and robustness for esophageal OCT image segmentation. A two-step edge-enhanced graph search (EEGS) framework is proposed in this study. Firstly, a preprocessing scheme is applied to suppress speckle noise and remove the disturbance in the esophageal structure. Secondly, the image is formulated into a graph and layer boundaries are located by graph search. In this process, we propose an edge-enhanced weight matrix for the graph by combining the vertical gradients with a Canny edge map. Experiments on esophageal OCT images from guinea pigs demonstrate that the EEGS framework is more robust and more accurate than the current segmentation method. It can be potentially useful for the early detection of esophageal diseases.

Predictive features for early cancer detection in Barrett's esophagus using Volumetric Laser Endomicroscopy

The incidence of Barrett cancer is increasing rapidly and current screening protocols often miss the disease at an early, treatable stage. Volumetric Laser Endomicroscopy (VLE) is a promising new tool for finding this type of cancer early, capturing a full circumferential scan of Barrett's Esophagus (BE), up to 3-mm depth. However, the interpretation of these VLE scans can be complicated, due to the large amount of cross-sectional images and the subtle grayscale variations. Therefore, algorithms for automated analysis of VLE data can offer a valuable contribution to its overall interpretation. In this study, we broadly investigate the potential of Computer-Aided Detection (CADe) for the identification of early Barrett's cancer using VLE. We employ a histopathologically validated set of ex-vivo VLE images for evaluating and comparing a considerable set of widely-used image features and machine learning algorithms. In addition, we show that incorporating clinical knowledge in feature design, leads to a superior classification performance and additional benefits, such as low complexity and fast computation time. Furthermore, we identify an optimal tissue depth for classification of 0.5-1.0 mm, and propose an extension to the evaluated features that exploits this phenomenon, improving their predictive properties for cancer detection in VLE data. Finally, we compare the performance of the CADe methods with the classification accuracy of two VLE experts. With a maximum Area Under the Curve (AUC) in the range of 0.90-0.93 for the evaluated features and machine learning methods versus an AUC of 0.81 for the medical experts, our experiments show that computer-aided methods can achieve a considerably better performance than trained human observers in the analysis of VLE data.

Computer-aided detection of early Barrett's neoplasia using volumetric laser endomicroscopy

BACKGROUND AND AIMS

Volumetric laser endomicroscopy (VLE) is an advanced imaging system that provides a near-microscopic resolution scan of the esophageal wall layers up to 3-mm deep. VLE has the potential to improve detection of early neoplasia in Barrett's esophagus (BE). However, interpretation of VLE images is complex because of the large amount of data that need to be interpreted in real time. The aim of this study was to investigate the feasibility of a computer algorithm to identify early BE neoplasia on ex vivo VLE images.

METHODS

We used 60 VLE images from a database of high-quality ex vivo VLE-histology correlations, obtained from BE patients ± neoplasia (30 nondysplastic BE [NDBE] and 30 high-grade dysplasia/early adenocarcinoma images). VLE features from a recently developed clinical VLE prediction score for BE neoplasia served as input for the algorithm: (1) higher VLE surface than subsurface signal and (2) lack of layering. With this input, novel clinically inspired algorithm features were developed, based on signal intensity statistics and grayscale correlations. For comparison, generic image analysis methods were examined for their performance to detect neoplasia. For classification of the images in the NDBE or neoplastic group, several machine learning methods were evaluated. Leave-1-out cross-validation was used for algorithm validation.

RESULTS

Three novel clinically inspired algorithm features were developed. The feature "layering and signal decay statistics" showed the optimal performance compared with the other clinically features ("layering" and "signal intensity distribution") and generic image analyses methods, with an area under the receiver operating characteristic curve (AUC) of .95. Corresponding sensitivity and specificity were 90% and 93%, respectively. In addition, the algorithm showed a better performance than the clinical VLE prediction score (AUC .81).

CONCLUSIONS

This is the first study in which a computer algorithm for BE neoplasia was developed based on VLE images with direct histologic correlates. The algorithm showed good performance to detect BE neoplasia in ex vivo VLE images compared with the performance of a recently developed clinical VLE prediction score. This study suggests that an automatic detection algorithm has the potential to assist endoscopists in detecting early neoplasia on VLE. Future studies on in vivo VLE scans are needed to further validate the algorithm.

Quantitative attenuation analysis for identification of early Barrett's neoplasia in volumetric laser endomicroscopy

Early neoplasia in Barrett’s esophagus (BE) is difficult to detect. Volumetric laser endomicroscopy (VLE) incorporates optical coherence tomography, providing a circumferential scan of the esophageal wall layers. The attenuation coefficient (μVLE) quantifies decay of detected backscattered light versus depth, and could potentially improve BE neoplasia detection. The aim is to investigate feasibility of μVLE for identification of early BE neoplasia. In vivo and ex vivo VLE scans with histological correlation from BE patients ± neoplasia were used. Quantification by μVLE was performed manually on areas of interest (AoIs) to differentiate neoplasia from nondysplastic (ND)BE. From ex vivo VLE scans from 16 patients (13 with neoplasia), 68 AoIs were analyzed. Median μVLE values (mm−1) were 3.7 [2.1 to 4.4 interquartile range (IQR)] for NDBE and 4.0 (2.5 to 4.9 IQR) for neoplasia, not statistically different (p=0.82). Fourteen in vivo scans were used: nine from neoplastic and five from NDBE patients. Median μVLE values were 1.8 (1.5 to 2.6 IQR) for NDBE and 2.1 (1.9 to 2.6 IQR) for neoplasia, with no statistically significant difference (p=0.37). In conclusion, there was no significant difference in μVLE values in VLE scans from early neoplasia versus NDBE. Future studies with a larger sample size should explore other quantitative methods for detection of neoplasia during BE surveillance.

Feature Extraction and Classification on Esophageal X-Ray Images of Xinjiang Kazak Nationality

Esophageal cancer is one of the fastest rising types of cancers in China. The Kazak nationality is the highest-risk group in Xinjiang. In this work, an effective computer-aided diagnostic system is developed to assist physicians in interpreting digital X-ray image features and improving the quality of diagnosis. The modules of the proposed system include image preprocessing, feature extraction, feature selection, image classification, and performance evaluation. 300 original esophageal X-ray images were resized to a region of interest and then enhanced by the median filter and histogram equalization method. 37 features from textural, frequency, and complexity domains were extracted. Both sequential forward selection and principal component analysis methods were employed to select the discriminative features for classification. Then, support vector machine and K-nearest neighbors were applied to classify the esophageal cancer images with respect to their specific types. The classification performance was evaluated in terms of the area under the receiver operating characteristic curve, accuracy, precision, and recall, respectively. Experimental results show that the classification performance of the proposed system outperforms the conventional visual inspection approaches in terms of diagnostic quality and processing time. Therefore, the proposed computer-aided diagnostic system is promising for the diagnostics of esophageal cancer.

Staging Early Esophageal Cancer

Staging esophageal cancer provides a standardized measure of the extent of disease that can be used to inform decisions about therapy and guide prognosis. For esophageal cancer, the treatment pathways vary greatly depending on stage of disease, and accurate staging is therefore crucial in ensuring the optimal therapy for each patient. For early esophageal cancer (T1 lesions), endoscopic resection can be curative and simultaneously gives accurate staging of depth of invasion. For tumors invading the submucosa or more advanced disease, comprehensive investigation is required to accurately stage the tumor and assess suitability for curative resection. A combined imaging approach of computed tomography (CT), positron emission tomography (PET), and endoscopic ultrasound (EUS) offers complementary diagnostic information and gives the greatest chance of accurate staging. Staging laparoscopy can identify peritoneal disease and small superficial liver lesions that could be missed on CT or PET, and alters management in up to 20 % of patients. Optical diagnostic techniques offer the prospect of further extending the possibilities of endoscopic staging in real time. Optical coherence tomography can image superficial lesions and could provide information on depth of invasion for these lesions. Real-time lymph node analysis using optical diagnostics such as Raman spectroscopy could be used to support immediate endoscopic therapy without waiting for results of cytology or further investigations.

Classification of basal cell carcinoma in human skin using machine learning and quantitative features captured by polarization sensitive optical coherence tomography

We report the first fully automated detection of basal cell carcinoma (BCC), the most commonly occurring type of skin cancer, in human skin using polarization-sensitive optical coherence tomography (PS-OCT). Our proposed automated procedure entails building a machine-learning based classifier by extracting image features from the two complementary image contrasts offered by PS-OCT, intensity and phase retardation (PR), and selecting a subset of features that yields a classifier with the highest accuracy. Our classifier achieved 95.4% sensitivity and specificity, validated by leave-one-patient-out cross validation (LOPOCV), in detecting BCC in human skin samples collected from 42 patients. Moreover, we show the superiority of our classifier over the best possible classifier based on features extracted from intensity-only data, which demonstrates the significance of PR data in detecting BCC.

Computer-aided texture analysis combined with experts' knowledge: Improving endoscopic celiac disease diagnosis

AIM: To further improve the endoscopic detection of intestinal mucosa alterations due to celiac disease (CD).

METHODS: We assessed a hybrid approach based on the integration of expert knowledge into the computer-based classification pipeline. A total of 2835 endoscopic images from the duodenum were recorded in 290 children using the modified immersion technique (MIT). These children underwent routine upper endoscopy for suspected CD or non-celiac upper abdominal symptoms between August 2008 and December 2014. Blinded to the clinical data and biopsy results, three medical experts visually classified each image as normal mucosa (Marsh-0) or villous atrophy (Marsh-3). The experts’ decisions were further integrated into state-of-the-art texture recognition systems. Using the biopsy results as the reference standard, the classification accuracies of this hybrid approach were compared to the experts’ diagnoses in 27 different settings.

RESULTS: Compared to the experts’ diagnoses, in 24 of 27 classification settings (consisting of three imaging modalities, three endoscopists and three classification approaches), the best overall classification accuracies were obtained with the new hybrid approach. In 17 of 24 classification settings, the improvements achieved with the hybrid approach were statistically significant (P < 0.05). Using the hybrid approach classification accuracies between 94% and 100% were obtained. Whereas the improvements are only moderate in the case of the most experienced expert, the results of the less experienced expert could be improved significantly in 17 out of 18 classification settings. Furthermore, the lowest classification accuracy, based on the combination of one database and one specific expert, could be improved from 80% to 95% (P < 0.001).

CONCLUSION: The overall classification performance of medical experts, especially less experienced experts, can be boosted significantly by integrating expert knowledge into computer-aided diagnosis systems.

Automated segmentation and characterization of esophageal wall in vivo by tethered capsule optical coherence tomography endomicroscopy

Optical coherence tomography (OCT) is an optical diagnostic modality that can acquire cross-sectional images of the microscopic structure of the esophagus, including Barrett's esophagus (BE) and associated dysplasia. We developed a swallowable tethered capsule OCT endomicroscopy (TCE) device that acquires high-resolution images of entire gastrointestinal (GI) tract luminal organs. This device has a potential to become a screening method that identifies patients with an abnormal esophagus that should be further referred for upper endoscopy. Currently, the characterization of the OCT-TCE esophageal wall data set is performed manually, which is time-consuming and inefficient. Additionally, since the capsule optics optimally focus light approximately 500 µm outside the capsule wall and the best quality images are obtained when the tissue is in full contact with the capsule, it is crucial to provide feedback for the operator about tissue contact during the imaging procedure. In this study, we developed a fully automated algorithm for the segmentation of in vivo OCT-TCE data sets and characterization of the esophageal wall. The algorithm provides a two-dimensional representation of both the contact map from the data collected in human clinical studies as well as a tissue map depicting areas of BE with or without dysplasia. Results suggest that these techniques can potentially improve the current TCE data acquisition procedure and provide an efficient characterization of the diseased esophageal wall.

In vivo endoscopic tissue identification by rapid evaporative ionization mass spectrometry (REIMS)

Gastrointestinal cancers are a leading cause of mortality, accounting for 23 % of cancer-related deaths worldwide. In order to improve outcomes from these cancers, novel tissue characterization methods are needed to facilitate accurate diagnosis. Rapid evaporative ionization mass spectrometry (REIMS) is a technique developed for the in vivo classification of human tissue through mass spectrometric analysis of aerosols released during electrosurgical dissection. This ionization technique was further developed by utilizing surface induced dissociation and was integrated with an endoscopic polypectomy snare to allow in vivo analysis of the gastrointestinal tract. We tested the classification performance of this novel endoscopic REIMS method in vivo. It was shown to be capable of differentiating between healthy layers of the intestinal wall, cancer, and adenomatous polyps based on the REIMS fingerprint of each tissue type in vivo.

Endoscopic Optical Coherence Tomography (OCT): Advances in Gastrointestinal Imaging

In the rapidly evolving field of endoscopic gastrointestinal imaging, Optical Coherence Tomography (OCT) has found many diverse applications. We present the current status of OCT and its practical applications in imaging normal and abnormal mucosa in the esophagus, stomach, small and large intestines, and biliary and pancreatic ducts. We highlight technical aspects and principles of imaging, assess published data, and suggest future directions for OCT-guided evaluation and therapy.

I-SCAN targeted versus random biopsies in Barrett's oesophagus

BACKGROUND

The accuracy and effectiveness of targeted oesophageal biopsies in Barrett's oesophagus to detect dysplasia using new magnification techniques are unknown. Aim of this study was to investigate whether the combined use of acetic acid, magnification and electronic filters allows the same accuracy as the four-quadrant random biopsies pattern; pathologist interobserver agreement both in low grade and high grade dysplasia was also assessed.

METHODS

Fifty-four consecutive patients newly diagnosed with Barrett's oesophagus were enrolled in a prospective study from a single endoscopy unit. Biopsies were evaluated by the local pathologist and by an expert pathologist from another pathology unit.

MAIN OUTCOME MEASUREMENT

Dysplasia detection rate and interobserver agreement for the histologic diagnosis of dysplasia.

RESULTS

The use of acetic acid, magnification and electronic filters showed an unacceptably low dysplasia detection rate by the two pathologists (9.2% and 5.5% for targeted biopsies, respectively). The interobserver agreement for low grade dysplasia between pathologists was low (Cohen's K weighted=0.45).

CONCLUSIONS

In an average setting, the standard four-quadrant method should still be preferred, along with the implementation of a routine second evaluation by an expert pathologist.

Toward the guidance of transbronchial biopsy: identifying pulmonary nodules with optical coherence tomography

BACKGROUND

Solitary pulmonary nodules (SPNs) frequently require transbronchial needle aspiration (TBNA) or biopsy to determine malignant potential, but have variable diagnostic yields. Confirming needle placement within SPNs during TBNA could significantly increase diagnostic yield. Optical coherence tomography (OCT) provides nondestructive, high-resolution, microstructural imaging with potential to distinguish SPN from parenchyma. We have developed needle-based OCT probes compatible with TBNA. Before OCT can play any significant role in guiding clinical TBNA, OCT interpretation criteria for differentiating SPN from lung parenchyma must be developed and validated.

METHODS

OCT of SPN and parenchyma was performed on 111 ex vivo resection specimens. OCT criteria for parenchyma and SPN were developed and validated in a blinded assessment. Six blinded readers (two pulmonologists, two pathologists, and two OCT experts) were trained on imaging criteria in a 15-min training session prior to interpreting the validation data set.

RESULTS

OCT of lung parenchyma displayed evenly spaced signal-void alveolar spaces, signal-intense backreflections at tissue-air interfaces, or both. SPNs lacked both of these imaging features. Independent validation of OCT criteria by the six blinded readers demonstrated sensitivity and specificity of 95.4% and 98.2%, respectively.

CONCLUSIONS

We have developed and validated OCT criteria for lung parenchyma and SPN with sensitivity and specificity &gt; 95% in this ex vivo study. We anticipate that OCT could be a useful complementary imaging modality to confirm needle placement during TBNA to potentially increase diagnostic yield.

Volumetric optical frequency domain imaging of pulmonary pathology with precise correlation to histopathology

BACKGROUND

Lung cancer is the leading cause of cancer-related mortality. Radiology and bronchoscopy techniques do not have the necessary resolution to evaluate lung lesions on the microscopic scale, which is critical for diagnosis. Bronchial biopsy specimens can be limited by sampling error and small size. Optical frequency domain imaging (OFDI) provides volumetric views of tissue microstructure at near-histologic resolution and may be useful for evaluating pulmonary lesions to increase diagnostic accuracy. Bronchoscopic OFDI has been evaluated in vivo, but a lack of correlated histopathology has limited the ability to develop accurate image interpretation criteria.

METHODS

We performed OFDI through two approaches (airway-centered and parenchymal imaging) in 22 ex vivo lung specimens, using tissue dye to precisely correlate imaging and histology.

RESULTS

OFDI of normal airway allowed visualization of epithelium, lamina propria, cartilage, and alveolar attachments. Carcinomas exhibited architectural disarray, loss of normal airway and alveolar structure, and rapid light attenuation. Squamous cell carcinomas showed nested architecture. Atypical glandular formation was appreciated in adenocarcinomas, and uniform trabecular gland formation was seen in salivary gland carcinomas. Mucinous adenocarcinomas showed alveolar wall thickening with intraalveolar mucin. Interstitial fibrosis was visualized as signal-dense tissue, with an interstitial distribution in mild interstitial fibrotic disease and a diffuse subpleural pattern with cystic space formation in usual interstitial pneumonitis.

CONCLUSIONS

To our knowledge, this study is the first demonstration of volumetric OFDI with precise correlation to histopathology in lung pathology. We anticipate that OFDI may play a role in assessing airway and parenchymal pathology, providing fresh insights into the volumetric features of pulmonary disease.

Screening for precancerous lesions of upper gastrointestinal tract: from the endoscopists' viewpoint

Upper gastrointestinal tract cancers are one of the most important leading causes of cancer death worldwide. Diagnosis at late stages always brings about poor outcome of these malignancies. The early detection of precancerous or early cancerous lesions of gastrointestinal tract is therefore of utmost importance to improve the overall outcome and maintain a good quality of life of patients. The desire of endoscopists to visualize the invisibles under conventional white-light endoscopy has accelerated the advancements in endoscopy technologies. Nowadays, image-enhanced endoscopy which utilizes optical- or dye-based contrasting techniques has been widely applied in endoscopic screening program of gastrointestinal tract malignancies. These contrasting endoscopic technologies not only improve the visualization of early foci missed by conventional endoscopy, but also gain the insight of histopathology and tumor invasiveness, that is so-called optical biopsy. Here, we will review the application of advanced endoscopy technique in screening program of upper gastrointestinal tract cancers.

In vivo optical coherence tomography: the role of the pathologist

Optical coherence tomography (OCT) is a nondestructive, high-resolution imaging modality, providing cross-sectional, architectural images at near histologic resolutions, with penetration depths up to a few millimeters. Optical frequency domain imaging is a second-generation OCT technology that has equally high resolution with significantly increased image acquisition speeds and allows for large area, high-resolution tissue assessments. These features make OCT and optical frequency domain imaging ideal imaging techniques for surface and endoscopic imaging, specifically when tissue is unsafe to obtain and/or suffers from biopsy sampling error. This review focuses on the clinical impact of OCT in coronary, esophageal, and pulmonary imaging and the role of the pathologist in interpreting high-resolution OCT images as a complement to standard tissue pathology.

Motion artifacts associated with in vivo endoscopic OCT images of the esophagus

3-D optical coherence tomography (OCT) has been extensively investigated as a potential screening and/or surveillance tool for Barrett's esophagus (BE). Understanding and correcting motion artifact may improve image interpretation. In this work, the motion trace was analyzed to show the physiological origin (respiration and heart beat) of the artifacts. Results showed that increasing balloon pressure did not sufficiently suppress the physiological motion artifact. An automated registration algorithm was designed to correct such artifacts. The performance of the algorithm was evaluated in images of normal porcine esophagus and demonstrated in images of BE in human patients.

Morphological analysis of optical coherence tomography images for automated classification of gastrointestinal tissues

The impact of digestive diseases, which include disorders affecting the oropharynx and alimentary canal, ranges from the inconvenience of a transient diarrhoea to dreaded conditions such as pancreatic cancer, which are usually fatal. Currently, the major limitation for the diagnosis of such diseases is sampling error because, even in the cases of rigorous adherence to biopsy protocols, only a tiny fraction of the surface of the involved gastrointestinal tract is sampled. Optical coherence tomography (OCT), which is an interferometric imaging technique for the minimally invasive measurement of biological samples, could decrease sampling error, increase yield, and even eliminate the need for tissue sampling provided that an automated, quick and reproducible tissue classification system is developed. Segmentation and quantification of ophthalmologic pathologies using OCT traditionally rely on the extraction of thickness and size measures from the OCT images, but layers are often not observed in nonopthalmic OCT imaging. Distinct mathematical methods, namely Principal Component Analysis (PCA) and textural analyses including both spatial textural analysis derived from the two-dimensional discrete Fourier transform (DFT) and statistical texture analysis obtained independently from center-symmetric autocorrelation (CSAC) and spatial grey-level dependency matrices (SGLDM), have been previously reported to overcome this problem. We propose an alternative approach consisting of a region segmentation according to the intensity variation along the vertical axis and a pure statistical technique for feature quantification, i.e. morphological analysis. Qualitative and quantitative comparisons with traditional approaches are accomplished in the discrimination of freshly-excised specimens of gastrointestinal tissues to exhibit the feasibility of the proposed method for computer-aided diagnosis (CAD) in the clinical setting.

Development of angle-resolved low coherence interferometry for clinical detection of dysplasia

This review covers the development of angle-resolved low coherence interferometry (a/LCI) from initial development through clinical application. In the first applications, the approach used a time-domain interferometry scheme and was validated using animal models of carcinogenesis to assess the feasibility of detecting dysplasia in situ. Further development of the approach led to Fourier-domain interferometry schemes with higher throughput and endoscope-compatible probes to enable clinical application. These later implementations have been applied to clinical studies of dysplasia in Barrett's esophagus tissues, a metaplastic tissue type that is associated with an increased risk of esophageal adenocarcinoma. As an alternative to systematic biopsy, the a/LCI approach offers high sensitivity and specificity for detecting dysplasia in these tissues while avoiding the need for tissue removal or exogenous contrast agents. Here, the various implementations of a/LCI are discussed and the results of the preliminary animal experiments and ex vivo human tissue studies are reviewed. A review of a recent in vivo clinical study is also presented.

Defect detection and property evaluation of indium tin oxide conducting glass using optical coherence tomography

This study demonstrates a new approach for evaluating the properties of indium tin oxide (ITO) conducting glass and identifying defects using optical coherence tomography (OCT). A swept-source OCT system was implemented to scan the ITO conducting glass to enable two-dimensional or three-dimensional imaging. With OCT scanning, the defects can be clearly identified at various depths. Several parameters in addition to morphological information can be estimated simultaneously, including the thickness of the glass substrate, the refractive index, reflection coefficient, and transmission coefficient, all of which can be used to evaluate the quality of ITO conducting glass. This study developed a modified method for evaluating the refractive index of glass substrates without having to perform multiple scans as well as a segmentation algorithm to separate the interfaces. The results show the potential of OCT as an imaging tool for the inspection of defects in ITO conducting glass.