Epithelium Segmentation Research Articles

An Immunofluorescence-Guided Segmentation Model in Hematoxylin and Eosin Images Is Enabled by Tissue Artifact Correction Using a Cycle-Consistent Generative Adversarial Network

Despite recent advances, the adoption of computer vision methods into clinical and commercial applications has been hampered by the limited availability of accurate ground truth tissue annotations required to train robust supervised models. Generating such ground truth can be accelerated by annotating tissue molecularly using immunofluorescence (IF) staining and mapping these annotations to a post-IF hematoxylin and eosin (H&E) (terminal H&E) stain. Mapping the annotations between IF and terminal H&E increases both the scale and accuracy by which ground truth could be generated. However, discrepancies between terminal H&E and conventional H&E caused by IF tissue processing have limited this implementation. We sought to overcome this challenge and achieve compatibility between these parallel modalities using synthetic image generation, in which a cycle-consistent generative adversarial network was applied to transfer the appearance of conventional H&E such that it emulates terminal H&E. These synthetic emulations allowed us to train a deep learning model for the segmentation of epithelium in terminal H&E that could be validated against the IF staining of epithelial-based cytokeratins. The combination of this segmentation model with the cycle-consistent generative adversarial network stain transfer model enabled performative epithelium segmentation in conventional H&E images. The approach demonstrates that the training of accurate segmentation models for the breadth of conventional H&E data can be executed free of human expert annotations by leveraging molecular annotation strategies such as IF, so long as the tissue impacts of the molecular annotation protocol are captured by generative models that can be deployed prior to the segmentation process.

Automatic quantification of tumor-stroma ratio as a prognostic marker for pancreatic cancer.

This study aims to introduce an innovative multi-step pipeline for automatic tumor-stroma ratio (TSR) quantification as a potential prognostic marker for pancreatic cancer, addressing the limitations of existing staging systems and the lack of commonly used prognostic biomarkers. The proposed approach involves a deep-learning-based method for the automatic segmentation of tumor epithelial cells, tumor bulk, and stroma from whole-slide images (WSIs). Models were trained using five-fold cross-validation and evaluated on an independent external test set. TSR was computed based on the segmented components. Additionally, TSR's predictive value for six-month survival on the independent external dataset was assessed. Median Dice (inter-quartile range (IQR)) of 0.751(0.15) and 0.726(0.25) for tumor epithelium segmentation on internal and external test sets, respectively. Median Dice of 0.76(0.11) and 0.863(0.17) for tumor bulk segmentation on internal and external test sets, respectively. TSR was evaluated as an independent prognostic marker, demonstrating a cross-validation AUC of 0.61±0.12 for predicting six-month survival on the external dataset. Our pipeline for automatic TSR quantification offers promising potential as a prognostic marker for pancreatic cancer. The results underscore the feasibility of computational biomarker discovery in enhancing patient outcome prediction, thus contributing to personalized patient management.

Diagnosis of posterior staphyloma using the radius of steepest curvature among retinal pigment epithelium segmentation line measured by optic coherent tomography

PurposeTo investigate a novel marker to diagnose posterior staphylomas by measuring the radius of the steepest curvature on the retinal pigment epithelium (RPE) segmentation line using optical coherence tomography (OCT).Study DesignRetrospective Cross-sectional Study.MethodsThe authors developed a prototype software to measure the radius of curvature on the RPE segmentation line of OCT. Twelve images of 9-mm radial OCT scans were used. The radius of curvature was measured at the steepest area of the RPE segmentation line, and the macular curvature (MC) index was calculated based on its reciprocal. Based on the wide-field fundus findings, the study sample was divided into three groups: definite posterior staphyloma, no posterior staphyloma, and undetermined. The differences of MC index among the groups and the correlation between the MC index, age, and axial length were analyzed.ResultsThe present study analyzed 268 eyes, with 54 (20.1%) with definite posterior staphyloma, 202 (75.4%) with no posterior staphyloma, and 12 (4.5%) with undetermined disease status. A maximum MC index of 37.5 was observed in the group with no posterior staphyloma, which was less than the minimum MC index of 42.7 observed in the group with definite posterior staphyloma. The MC index had strong correlations with the axial length and age in eyes with high myopia.ConclusionsEyes with posterior staphyloma have a steeper curvature than those with radius 8.44 mm, while eyes without posterior staphyloma do not. MC index 40 (radius 8.44 mm) might act as a reference to distinguish between those with and those without posterior staphyloma.

Invasion depth estimation of carcinoma cells using adaptive stain normalization to improve epidermis segmentation accuracy.

Submucosal invasion depth is a significant prognostic factor when assessing lymph node metastasis and cancer itself to plan proper treatment for the patient. Conventionally, oncologists measure the invasion depth by hand which is a laborious, subjective, and time-consuming process. The manual pathological examination by measuring accurate carcinoma cell invasion with considerable inter-observer and intra-observer variations is still challenging. The increasing use of medical imaging and artificial intelligence reveals a significant role in clinical medicine and pathology. In this paper, we propose an approach to study invasive behavior and measure the invasion depth of carcinoma from stained histopathology images. Specifically, our model includes adaptive stain normalization, color decomposition, and morphological reconstruction with adaptive thresholding to separate the epithelium with blue ratio image. Our method splits the image into multiple non-overlapping meaningful segments and successfully finds the homogeneous segments to measure accurate invasion depth. The invasion depths are measured from the inner epithelium edge to outermost pixels of the deepest part of particles in image. We conduct our experiments on skin melanoma tissue samples as well as on organotypic invasion model utilizing myoma tissue and oral squamous cell carcinoma. The performance is experimentally compared to three closely related reference methods and our method provides a superior result in measuring invasion depth. This computational technique will be beneficial for the segmentation of epithelium and other particles for the development of novel computer-aided diagnostic tools in biobank applications.

CNN stability training improves robustness to scanner and IHC-based image variability for epithelium segmentation in cervical histology.

In digital pathology, image properties such as color, brightness, contrast and blurriness may vary based on the scanner and sample preparation. Convolutional Neural Networks (CNNs) are sensitive to these variations and may underperform on images from a different domain than the one used for training. Robustness to these image property variations is required to enable the use of deep learning in clinical practice and large scale clinical research. CNN Stability Training (CST) is proposed and evaluated as a method to increase CNN robustness to scanner and Immunohistochemistry (IHC)-based image variability. CST was applied to segment epithelium in immunohistological cervical Whole Slide Images (WSIs). CST randomly distorts input tiles and factors the difference between the CNN prediction for the original and distorted inputs within the loss function. CNNs were trained using 114 p16-stained WSIs from the same scanner, and evaluated on 6 WSI test sets, each with 23 to 24 WSIs of the same tissue but different scanner/IHC combinations. Relative robustness (rAUC) was measured as the difference between the AUC on the training domain test set (i.e., baseline test set) and the remaining test sets. Across all test sets, The AUC of CST models outperformed "No CST" models (AUC: 0.940-0.989 vs. 0.905-0.986, p < 1e - 8), and obtained an improved robustness (rAUC: [-0.038, -0.003] vs. [-0.081, -0.002]). At a WSI level, CST models showed an increase in performance in 124 of the 142 WSIs. CST models also outperformed models trained with random on-the-fly data augmentation (DA) in all test sets ([0.002, 0.021], p < 1e-6). CST offers a path to improve CNN performance without the need for more data and allows customizing distortions to specific use cases. A python implementation of CST is publicly available at https://github.com/TIGACenter/CST_v1.

Deep Learning-Based Recognition of Cervical Squamous Interepithelial Lesions.

Cervical squamous intraepithelial lesions (SILs) are precursor lesions of cervical cancer, and their accurate diagnosis enables patients to be treated before malignancy manifests. However, the identification of SILs is usually laborious and has low diagnostic consistency due to the high similarity of pathological SIL images. Although artificial intelligence (AI), especially deep learning algorithms, has drawn a lot of attention for its good performance in cervical cytology tasks, the use of AI for cervical histology is still in its early stages. The feature extraction, representation capabilities, and use of p16 immunohistochemistry (IHC) among existing models are inadequate. Therefore, in this study, we first designed a squamous epithelium segmentation algorithm and assigned the corresponding labels. Second, p16-positive area of IHC slides were extracted with Whole Image Net (WI-Net), followed by mapping the p16-positive area back to the H&E slides and generating a p16-positive mask for training. Finally, the p16-positive areas were inputted into Swin-B and ResNet-50 to classify the SILs. The dataset comprised 6171 patches from 111 patients; patches from 80% of the 90 patients were used for the training set. The accuracy of the Swin-B method for high-grade squamous intraepithelial lesion (HSIL) that we propose was 0.914 [0.889-0.928]. The ResNet-50 model for HSIL achieved an area under the receiver operating characteristic curve (AUC) of 0.935 [0.921-0.946] at the patch level, and the accuracy, sensitivity, and specificity were 0.845, 0.922, and 0.829, respectively. Therefore, our model can accurately identify HSIL, assisting the pathologist in solving actual diagnostic issues and even directing the follow-up treatment of patients.

Deep learning-based method for segmenting epithelial layer of tubules in histopathological images of testicular tissue.

There is growing concern that male reproduction is affected by environmental chemicals. One way to determine the adverse effect of environmental pollutants is to use wild animals as monitors and evaluate testicular toxicity using histopathology. We propose an automated method to process histology images of testicular tissue. Testicular tissue consists of seminiferous tubules. Segmenting the epithelial layer of the seminiferous tubule is a prerequisite for developing automated methods to detect abnormalities in tissue. We suggest an encoder-decoder fully connected convolutional neural network model to segment the epithelial layer of the seminiferous tubules in histological images. The ResNet-34 is used in the feature encoder module, and the squeeze and excitation attention block is integrated into the encoding module improving the segmentation and localization of epithelium. We applied the proposed method for the two-class problem, where the epithelial layer of the tubule is the target class. The -score and Intersection over Union of the proposed method are 0.85 and 0.92. Although the proposed method is trained on a limited training set, it performs well on an independent dataset and outperforms other state-of-the-art methods. The pretrained ResNet-34 in the encoder and attention block suggested in the decoder result in better segmentation and generalization. The proposed method can be applied to testicular tissue images from any mammalian species and can be used as the first part of a fully automated testicular tissue processing pipeline. The dataset and codes are publicly available on GitHub.

A CAD system for automatic dysplasia grading on H&E cervical whole-slide images

Cervical cancer is the fourth most common female cancer worldwide and the fourth leading cause of cancer-related death in women. Nonetheless, it is also among the most successfully preventable and treatable types of cancer, provided it is early identified and properly managed. As such, the detection of pre-cancerous lesions is crucial. These lesions are detected in the squamous epithelium of the uterine cervix and are graded as low- or high-grade intraepithelial squamous lesions, known as LSIL and HSIL, respectively. Due to their complex nature, this classification can become very subjective. Therefore, the development of machine learning models, particularly directly on whole-slide images (WSI), can assist pathologists in this task. In this work, we propose a weakly-supervised methodology for grading cervical dysplasia, using different levels of training supervision, in an effort to gather a bigger dataset without the need of having all samples fully annotated. The framework comprises an epithelium segmentation step followed by a dysplasia classifier (non-neoplastic, LSIL, HSIL), making the slide assessment completely automatic, without the need for manual identification of epithelial areas. The proposed classification approach achieved a balanced accuracy of 71.07% and sensitivity of 72.18%, at the slide-level testing on 600 independent samples, which are publicly available upon reasonable request.

A machine learning model for separating epithelial and stromal regions in oral cavity squamous cell carcinomas using H&E-stained histology images: A multi-center, retrospective study

ObjectiveTissue slides from Oral cavity squamous cell carcinoma (OC-SCC), particularly the epithelial regions, hold morphologic features that are both diagnostic and prognostic. Yet, previously developed approaches for automated epithelium segmentation in OC-SCC have not been independently tested in a multi-center setting. In this study, we aimed to investigate the effectiveness and applicability of a convolutional neural network (CNN) model to perform epithelial segmentation using digitized H&E-stained diagnostic slides from OC-SCC patients in a multi-center setting. MethodsA CNN model was developed to segment the epithelial regions of digitized slides (n = 810), retrospectively collected from five different centers. Deep learning models were trained and validated using well-annotated tissue microarray (TMA) images (n = 212) at various magnifications. The best performing model was locked down and used for independent testing with a total of 478 whole-slide images (WSIs). Manually annotated epithelial regions were used as the reference standard for evaluation. We also compared the model generated results with IHC-stained epithelium (n = 120) as the reference. ResultsThe locked-down CNN model trained on the TMA image training cohorts with 10x magnification achieved the best segmentation performance. The locked-down model performed consistently and yielded Pixel Accuracy, Recall Rate, Precision Rate, and Dice Coefficient that ranged from 95.8% to 96.6%, 79.1% to 93.8%, 85.7% to 89.3%, and 82.3% to 89.0%, respectively for the three independent testing WSI cohorts. ConclusionThe automated model achieved a consistently accurate performance for automated epithelial region segmentation compared to manual annotations. This model could be integrated into a computer-aided diagnosis or prognosis system.

Code-Free Development and Deployment of Deep Segmentation Models for Digital Pathology.

Application of deep learning on histopathological whole slide images (WSIs) holds promise of improving diagnostic efficiency and reproducibility but is largely dependent on the ability to write computer code or purchase commercial solutions. We present a code-free pipeline utilizing free-to-use, open-source software (QuPath, DeepMIB, and FastPathology) for creating and deploying deep learning-based segmentation models for computational pathology. We demonstrate the pipeline on a use case of separating epithelium from stroma in colonic mucosa. A dataset of 251 annotated WSIs, comprising 140 hematoxylin-eosin (HE)-stained and 111 CD3 immunostained colon biopsy WSIs, were developed through active learning using the pipeline. On a hold-out test set of 36 HE and 21 CD3-stained WSIs a mean intersection over union score of 95.5 and 95.3% was achieved on epithelium segmentation. We demonstrate pathologist-level segmentation accuracy and clinical acceptable runtime performance and show that pathologists without programming experience can create near state-of-the-art segmentation solutions for histopathological WSIs using only free-to-use software. The study further demonstrates the strength of open-source solutions in its ability to create generalizable, open pipelines, of which trained models and predictions can seamlessly be exported in open formats and thereby used in external solutions. All scripts, trained models, a video tutorial, and the full dataset of 251 WSIs with ~31 k epithelium annotations are made openly available at https://github.com/andreped/NoCodeSeg to accelerate research in the field.

A Classification-Guided Segmentation Algorithm Based on Deep Learning for Epithelium Segmentation in Histopathological Images of Radicular Cysts.

In histopathological analysis of radicular cysts (RCs), lesions in epithelium can provide pathologists with rich information on pathologic degree, which is helpful to determine the type of periapical lesions and make precise treatment planning. Automatic segmentation and localization of epithelium from whole slide images (WSIs) can assist pathologists to complete pathological diagnosis more quickly. However, the class imbalance problem caused by the small proportion of fragmented epithelium in RCs imposes challenge on the typical automatic one-stage segmentation method. In this paper, we proposed a classification-guided segmentation algorithm (CGSA) for accurate segmentation. Our method was a two-stage model, including a classification network for region of interest (ROI) location and a segmentation network guided by classification. The classification stage eliminated most irrelevant areas and alleviated the class imbalance problem faced by the segmentation model. The results of 5-fold cross validation demonstrated that CGSA outperformed the one-stage segmentation method which was lacking in prior epithelium localization information. The epithelium segmentation achieved an overall Dice's coefficient of 0.722, and intersection over union (IoU) of 0.593, which improved by 5.5% and 5.9% respectively compared with the one-stage segmentation method using UNet.Clinical Relevance- This work presents a framework for automatic epithelium segmentation in histopathological images of RCs. It can be applied to make up for the shortcomings of manual annotation which is labor-intensive, time-consuming and objective.

Abstract 3: Deep learning-based integration of esophageal cancer morphology with genomics

Abstract Esophageal cancer is a major cause of cancer mortality world-wide. Even with advanced treatment options available, patients with esophageal cancer have a poor 5-year survival rate of 16%. Genomic studies of esophageal cancer have revealed frequently mutated genes involved in different oncogenic pathways. In this study, we attempted to predict genomic alterations, pathological and clinical features directly from H&amp;E-stained whole slide images (WSIs) of individual esophageal cancer cases by utilizing modern deep learning methods. We used WSIs with esophageal cancer obtained from TCGA and annotated cancerous epithelial and stromal components separately for esophageal adenocarcinoma (EAC) and squamous cell carcinoma (ESCC) (n= 25 cases for each) to create a training set. Subsequently, patches of size 256×256 pixels were extracted, and a U-Net convolutional neural network (CNN) was trained for semantic segmentation of tumor epithelium and stromal pixels. Tumor epithelium regions identified by the first U-Net CNN were fed into a Squeezenet model to learn to distinguish between EAC and ESCC in the training phase. Finally, for each subtype (EAC and ESCC), separate Densenet models were trained to predict gene mutation status. We also trained independent Densenet models to predict tumor stage, grade, and survival outcome from the tumor epithelial patches. After training, we tested these models on an independent validation set obtained from TCGA (n = 99) and UCSF (n = 25) patients, ensuring patient level separation between the training and the validation sets. We obtained an average AUC of 0.90 and 0.87 for epithelium vs. stromal and EAC vs. ESCC classification, respectively, on the validation datasets. These models enabled us to predict tumor stage, grade, and survival outcome with an accuracy of &amp;gt; 0.9. TP53 gene alterations were detected in over 50% of esophageal cancer cases, and our model was able to predict TP53 gene mutation status on WSIs with an AUC of 0.91. Overall, this study demonstrates how deep learning algorithms can aid in better understanding and predicting key biological and clinical features of aggressive malignancies at both population and individual patient level. This integration of artificial intelligence with molecular and histopathologic features could also help to accelerate precision cancer diagnosis and treatment in the clinical setting. Citation Format: Ruchika Verma, Wei Wu, Neeraj Kumar, Elizabeth Yu, Won-Tak Choi, Sarah Umetsu, Trever Bivona. Deep learning-based integration of esophageal cancer morphology with genomics [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2021; 2021 Apr 10-15 and May 17-21. Philadelphia (PA): AACR; Cancer Res 2021;81(13_Suppl):Abstract nr 3.

Artificial intelligence (AI)–powered pathologic response (PathR) assessment of resection specimens after neoadjuvant atezolizumab in patients with non-small cell lung cancer: Results from the LCMC3 study.

106 Background: PathR is an efficacy endpoint in Phase II and III neoadjuvant trials and is proposed as a surrogate for disease-free survival (DFS) and overall survival. Machine learning (ML)–based, automated approaches standardize quantification of areas of tumor bed and residual viable tumor. Here we show that automation may provide a scalable alternative to or complementary tool for manual assessment. Methods: We determined inter-reader variability for PathR among pathologists in the LCMC3 (NCT02927301) study and developed an AI-powered digital PathR assessment tool in line with manual consensus recommendations. Study cases were reviewed for PathR by a local site pathologist and 3 central expert pathologists (n = 127). When determined manually, major PathR (MPR) was defined as ≤10% viable tumor averaged per case. ML models were trained and validated by the PathAI research platform using digitized H&amp;E-stained tumor sections. The digital PathR model predicted percent viable tumor for each case as the sum of the cancer epithelium area from each slide divided by the sum of tumor bed area for each slide. DFS (clinical cutoff: Oct 23, 2020) was reported for patients with manual and digital PathR assessment (n = 135). For digital MPR, we used a prevalence-matched cutoff that maintained the same proportion of patients as manual MPR. Results: Inter-reader agreement among 1 local and 3 central pathologists for manual PathR was good (n = 127; ICC = 0.87; 95% CI: 0.84-0.90). Agreement was 91% (κ = 0.82) on manual MPR and 98% (κ = 0.88) on pathologic complete response (pCR). 6 patients had unanimous pCR. Digital and manual PathR were strongly correlated (n = 135, Pearson r = 0.78) and digital PathR demonstrated an outstanding predictability for manual MPR (AUROC = 0.975). The range was 0%-60% for digital PathR and 0%-100% for manual PathR with a regression line slope &lt; 1.0 (m = 0.303) indicating systematic differences between the methods, consistent with digital PathR using a high-resolution segmentation of cancer epithelium from stroma across each slide. Longer DFS was observed for MPR yes vs no with both digital and manual assessment (Table). Conclusions: This analysis showed good inter-reader agreement for manual and strong correlation of AI-powered digital and manual PathR. Comparable DFS rates for manual MPR and digital MPR are encouraging in the preliminary data. These data support further studies of digital PathR as a standardized and scalable tool to determine PathR. Clinical trial information: NCT02927301. [Table: see text]

Automated Cervical Digitized Histology Whole-Slide Image Analysis Toolbox

Background: Cervical intraepithelial neoplasia (CIN) is regarded as a potential precancerous state of the uterine cervix. Timely and appropriate early treatment of CIN can help reduce cervical cancer mortality. Accurate estimation of CIN grade correlated with human papillomavirus type, which is the primary cause of the disease, helps determine the patient's risk for developing the disease. Colposcopy is used to select women for biopsy. Expert pathologists examine the biopsied cervical epithelial tissue under a microscope. The examination can take a long time and is prone to error and often results in high inter-and intra-observer variability in outcomes. Methodology: We propose a novel image analysis toolbox that can automate CIN diagnosis using whole slide image (digitized biopsies) of cervical tissue samples. The toolbox is built as a four-step deep learning model that detects the epithelium regions, segments the detected epithelial portions, analyzes local vertical segment regions, and finally classifies each epithelium block with localized attention. We propose an epithelium detection network in this study and make use of our earlier research on epithelium segmentation and CIN classification to complete the design of the end-to-end CIN diagnosis toolbox. Results: The results show that automated epithelium detection and segmentation for CIN classification yields comparable results to manually segmented epithelium CIN classification. Conclusion: This highlights the potential as a tool for automated digitized histology slide image analysis to assist expert pathologists.

Accurate segmentation of prostate cancer histomorphometric features using a weakly supervised convolutional neural network.

.Purpose: Prostate cancer primarily arises from the glandular epithelium. Histomophometric techniques have been used to assess the glandular epithelium in automated detection and classification pipelines; however, they are often rigid in their implementation, and their performance suffers on large datasets where variation in staining, imaging, and preparation is difficult to control. The purpose of this study is to quantify performance of a pixelwise segmentation algorithm that was trained using different combinations of weak and strong stroma, epithelium, and lumen labels in a prostate histology dataset.Approach: We have combined weakly labeled datasets generated using simple morphometric techniques and high-quality labeled datasets from human observers in prostate biopsy cores to train a convolutional neural network for use in whole mount prostate labeling pipelines. With trained networks, we characterize pixelwise segmentation of stromal, epithelium, and lumen (SEL) regions on both biopsy core and whole-mount H&E-stained tissue.Results: We provide evidence that by simply training a deep learning algorithm on weakly labeled data generated from rigid morphometric methods, we can improve the robustness of classification over the morphometric methods used to train the classifier.Conclusions: We show that not only does our approach of combining weak and strong labels for training the CNN improve qualitative SEL labeling within tissue but also the deep learning generated labels are superior for cancer classification in a higher-order algorithm over the morphometrically derived labels it was trained on.

Automatic carcinoma identification from breast epithelial tissue WSI through U-Net deep learning network

Epithelium tissue covers and lines all internal organs of the body. Breast cancer carcinomas arise from the epithelial cells of the breast. The epithelial component lines tubules, ducts, etc. and segmenting out this region from other tissues is important to detect breast cancer. This paper proposes application of deep learning technique U-Net, to segment epithelium tissue automatically from a whole slide image (WSI) image. It also implements U-Net, an image segmentation algorithm to automatically learn the features of epithlium components from the experts-annotated WSI image dataset and tested the results. Various image processing techniques such as thresholding are applied to improve the quality of dataset before and after the training of images by U-Net. The performance of the U-Net is measured by statistic parameters Sorensen–Dice coefficient and F1 score. The automatic system generated an epithelium segmentation of accuracy of 0.932. Context: Automatic, faster and accurate computational technique implementation for breast cancer diagnostic pathology. Aims: Automatic identification of carcinoma in WSI of Breast Epithelium Tissue by using deep learning network, U-Net. Settings and Design: Epithelium tissue covers and lines all internal organs of the body. Breast cancer carcinomas arise from the epithelial cells of the breast. The epithelial component lines tubules, ducts, etc. and segmenting out this region from other tissues is important to detect breast cancer. Methods and Material: This paper proposes application of deep learning technique U-Net, to segment epithelium tissue automatically from a whole slide image (WSI) image. It also implements U-Net, an image segmentation algorithm to automatically learn the features of epithlium components from the experts-annotated WSI image dataset and tested the results. Various image processing techniques such as thresholding are applied to improve the quality of dataset before and after the training of images by U-Net

EpithNet: Deep Regression for Epithelium Segmentation in Cervical Histology Images

Background: Automated pathology techniques for detecting cervical cancer at the premalignant stage have advantages for women in areas with limited medical resources. Methods: This article presents EpithNet, a deep learning approach for the critical step of automated epithelium segmentation in digitized cervical histology images. EpithNet employs three regression networks of varying dimensions of image input blocks (patches) surrounding a given pixel, with all blocks at a fixed resolution, using varying network depth. Results: The proposed model was evaluated on 311 digitized histology epithelial images and the results indicate that the technique maximizes region-based information to improve pixel-wise probability estimates. EpithNet-mc model, formed by intermediate concatenation of the convolutional layers of the three models, was observed to achieve 94% Jaccard index (intersection over union) which is 26.4% higher than the benchmark model. Conclusions: EpithNet yields better epithelial segmentation results than state-of-the-art benchmark methods.

Epithelium segmentation and automated Gleason grading of prostate cancer via deep learning in label-free multiphoton microscopic images.

In the current clinical care practice, Gleason grading system is one of the most powerful prognostic predictors for prostate cancer (PCa). The grading system is based on the architectural pattern of cancerous epithelium in histological images. However, the standard procedure of histological examination often involves complicated tissue fixation and staining, which are time-consuming and may delay the diagnosis and surgery. In this study, label-free multiphoton microscopy (MPM) was used to acquire subcellular-resolution images of unstained prostate tissues. Then, a deep learning architecture (U-net) was introduced for epithelium segmentation of prostate tissues in MPM images. The obtained segmentation results were then merged with the original MPM images to train a classification network (AlexNet) for automated Gleason grading. The developed method achieved an overall pixel accuracy of 92.3% with a mean F1 score of 0.839 for epithelium segmentation. By merging the segmentation results with the MPM images, the accuracy of Gleason grading was improved from 72.42% to 81.13% in hold-out test set. Our results suggest that MPM in combination with deep learning holds the potential to be used as a fast and powerful clinical tool for PCa diagnosis.

Tumour parcellation and quantification (TuPaQ): a tool for refining biomarker analysis through rapid and automated segmentation of tumour epithelium.

Immunohistochemistry (IHC) is an essential component of biomarker research in cancer. Automated biomarker quantification is hampered by the failure of computational algorithms to discriminate 'negative' tumour cells from 'negative' stromal cells. We sought to develop an algorithm for segmentation of tumour epithelium in colorectal cancer (CRC), irrespective of the biomarker expression in the cells. We developed tumour parcellation and quantification (TuPaQ) to segment tumour epithelium and parcellate sections into 'epithelium' and 'non-epithelium'. TuPaQ comprises image pre-processing, extraction of regions of interest (ROIs) and quantification of tumour epithelium (total area occupied by epithelium and number of nuclei in the occupied area). A total of 286 TMA cores from CRC were manually annotated and analysed using the commercial halo software to provide ground truth. The performance of TuPaQ was evaluated against the ground truth using a variety of metrics. The image size of each core was 7000 × 7000 pixels and each core was analysed in a matter of seconds. Pixel × pixel analysis showed a sensitivity of 84% and specificity of 95% in detecting epithelium. The mean tumour area obtained by TuPaQ was very close to the area quantified after manual annotation (r = 0.956, P < 0.001). Moreover, quantification of tumour nuclei by TuPaQ correlated very strongly with that of halo (r = 0.891, P < 0.001). TuPaQ is a very rapid and accurate method of separating the epithelial and stromal compartments of colorectal tumours. This will allow more accurate and objective analysis of immunohistochemistry.

A dense multi-path decoder for tissue segmentation in histopathology images

Background and ObjectiveSegmenting different tissue components in histopathological images is of great importance for analyzing tissues and tumor environments. In recent years, an encoder–decoder family of convolutional neural networks has increasingly adopted to develop automated segmentation tools. While an encoder has been the main focus of most investigations, the role of a decoder so far has not been well studied and understood. Herein, we proposed an improved design of a decoder for the segmentation of epithelium and stroma components in histopathology images. MethodsThe proposed decoder is built upon a multi-path layout and dense shortcut connections between layers to maximize the learning and inference capability. Equipped with the proposed decoder, neural networks are built using three types of encoders (VGG, ResNet and preactived ResNet). To assess the proposed method, breast and prostate tissue datasets are utilized, including 108 and 52 hematoxylin and eosin (H&E) breast tissues images and 224 H&E prostate tissue images. ResultsCombining the pre-activated ResNet encoder and the proposed decoder, we achieved a pixel wise accuracy (ACC) of 0.9122, a rand index (RAND) score of 0.8398, an area under receiver operating characteristic curve (AUC) of 0.9716, Dice coefficient for stroma (DICE_STR) of 0.9092 and Dice coefficient for epithelium (DICE_EPI) of 0.9150 on the breast tissue dataset. The same network obtained 0.9074 ACC, 0.8320 Rand index, 0.9719 AUC, 0.9021 DICE_EPI and 0.9121 DICE_STR on the prostate dataset. ConclusionsIn general, the experimental results confirmed that the proposed network is superior to the networks combined with the conventional decoder. Therefore, the proposed decoder could aid in improving tissue analysis in histopathology images.