Zaneta Swiderska-Chadaj

Content-based analysis of Ki-67 stained meningioma specimens for automatic hot-spot selection

BackgroundHot-spot based examination of immunohistochemically stained histological specimens is one of the most important procedures in pathomorphological practice. The development of image acquisition equipment and computational units allows for the automation of this process. Moreover, a lot of possible technical problems occur in everyday histological material, which increases the complexity of the problem. Thus, a full context-based analysis of histological specimens is also needed in the quantification of immunohistochemically stained specimens. One of the most important reactions is the Ki-67 proliferation marker in meningiomas, the most frequent intracranial tumour. The aim of our study is to propose a context-based analysis of Ki-67 stained specimens of meningiomas for automatic selection of hot-spots.MethodsThe proposed solution is based on textural analysis, mathematical morphology, feature ranking and classification, as well as on the proposed hot-spot gradual extinction algorithm to allow for the proper detection of a set of hot-spot fields. The designed whole slide image processing scheme eliminates such artifacts as hemorrhages, folds or stained vessels from the region of interest. To validate automatic results, a set of 104 meningioma specimens were selected and twenty hot-spots inside them were identified independently by two experts. The Spearman rho correlation coefficient was used to compare the results which were also analyzed with the help of a Bland-Altman plot.ResultsThe results show that most of the cases (84) were automatically examined properly with two fields of view with a technical problem at the very most. Next, 13 had three such fields, and only seven specimens did not meet the requirement for the automatic examination. Generally, the Automatic System identifies hot-spot areas, especially their maximum points, better. Analysis of the results confirms the very high concordance between an automatic Ki-67 examination and the expert’s results, with a Spearman rho higher than 0.95.ConclusionThe proposed hot-spot selection algorithm with an extended context-based analysis of whole slide images and hot-spot gradual extinction algorithm provides an efficient tool for simulation of a manual examination. The presented results have confirmed that the automatic examination of Ki-67 in meningiomas could be introduced in the near future.

Survey: interpolation methods for whole slide image processing

Evaluating whole slide images of histological and cytological samples is used in pathology for diagnostics, grading and prognosis . It is often necessary to rescale whole slide images of a very large size. Image resizing is one of the most common applications of interpolation. We collect the advantages and drawbacks of nine interpolation methods, and as a result of our analysis, we try to select one interpolation method as the preferred solution. To compare the performance of interpolation methods, test images were scaled and then rescaled to the original size using the same algorithm. The modified image was compared to the original image in various aspects. The time needed for calculations and results of quantification performance on modified images were also compared. For evaluation purposes, we used four general test images and 12 specialized biological immunohistochemically stained tissue sample images. The purpose of this survey is to determine which method of interpolation is the best to resize whole slide images, so they can be further processed using quantification methods. As a result, the interpolation method has to be selected depending on the task involving whole slide images.

Local Binary Patterns and Unser Texture Descriptions to the Fold Detection on the Whole Slide Images of Meningiomas and Oligodendrogliomas

The paper presents a method for an automatic folds detection in the whole slide images to support the pathomorphological diagnostic procedure. The studied slides represent the meningiomas and oligodendrogliomas tumour stained with the Ki-67/MIB-1 immunohistochemical reaction. The proposed method is based on texture analysis (local binary pattern and Unser), mathematical morphology and Support Vector Machine classification. The fold area detection is a necessary preprocessing step in the automatic examination of the histological specimens, such as hot-spot selection, quantitative evaluation etc. The results of the automatic fold detection were compared with the expert’s annotations. The achieved results confirm efficiency of the proposed solutions.

Texture Analysis to Trophoblast and Villi Detection in Placenta Histological Images

The paper presents two methods of an automatic tropho-blasts and villi detection in histological images to support the pathomorphological diagnostic procedure. The studied slides represent the placenta villi from spontaneous miscarriage stained with the Hematoxylin and Eosin. The proposed methods are based on texture analyses, as Local Binary Pattern and Unser, and mathematical morphology operations. The research on placenta villi detection and the evaluation on the histological images is needed to support clinical studies. The results of the automatic trophoblasts and villi detection were compared with the expert’s annotations. The average coverage of the detected trophoblast areas is 93.65 % for the Unser- method and 77.06 % for the LBP method. The obtained results confirm efficiency of the proposed solutions.

Semantic Segmentation of Colon Glands in Inflammatory Bowel Disease Biopsies

Robust delineation of tissue components in hematoxylin and eosin (H&E) stained slides is a critical step in quantifying tissue morphology. Fully convolutional neural networks (FCN) are ideally suited for automatic and efficient segmentation of tissue components in H&E slides. However, their performance relies on the network architecture, quality and depth of training. Here we introduce a set of 802 image tiles of colon biopsies from 2 subjects with inflammatory bowel disease (IBD) annotated for glandular epithelium (EP), gland lumen together with goblet cells (LG), and stroma (ST). We either trained the FCN-8s de-novo on our images (DN-FCN-8s) or pre-trained on the ImageNet dataset and fine-tuned on our images (FT-FCN-8s). For comparison, we used the U-Net trained de-novo. The training involved 700/802 images, leaving 102 images as a testing set. Ultimately, each model was validated in an independent digital biopsy slide. We also determined how the number of images used for training affects the performance of the model and observed a plateau in trainability at 700 images. In the testing set, U-Net and FT-FCN-8s achieved accuracies of 92.30% and 92.26% respectively. In the independent biopsy slide, U-Net demonstrated a segmentation accuracy of 88.64%, with F1-scores of 0.74 (EP), 0.92 (LG), and 0.93 (ST). The performance of the FT-FCN-8s was slightly worse, but the model required fewer images to reach a high classification performance. Our data demonstrate that all 3 FCNs are appropriate for segmentation of glands in biopsies from patients with IBD and open the door for quantification of IBD associated pathologies.

Image processing methods for the structural detection and gradation of placental villi

The context-based examination of stained tissue specimens is one of the most important procedures in histopathological practice. The development of image processing methods allows for the automation of this process. We propose a method of automatic segmentation of placental structures and assessment of edema present in placental structures from a spontaneous miscarriage. The presented method is based on texture analysis, mathematical morphology, and region growing operations that are applicable to the heterogeneous microscopic images representing histological slides of the placenta. The results presented in this study were obtained using a set of 50 images of single villi originating from 13 histological slides and was compared with the manual evaluation of the pathologist. In the presented experiments, various structures, such as villi, villous mesenchyme, trophoblast, collagen, and vessels have been recognized. Moreover, the gradation of villous edema for three classes (no villous edema, moderate villous edema, and massive villous edema) has been conducted. Villi images were correctly identified in 98.21%, villous mesenchyme was correctly identified in 83.95%, and the villi evaluation was correct in 74% for the edema degree and 86% for the number of vessels. The presented segmentation method may serve as a support for current manual diagnosis methods and reduce the bias related to individual, subjective assessment of experts.

A Deep Learning Pipeline to Delineate Proliferative Areas of Intracranial Tumors in Digital Slides

Separating tumor cells from other tissues such as meninges and blood is one of the vital steps towards automated quantification of the proliferative index in digital slides of brain tumors. In this paper, we present a deep learning based pipeline to delineate areas of tumor in meningioma and oligodendroglioma specimens stained with Ki-67 marker. A pre-trained convolutional neural network (CNN) was fine-tuned with 7057 image tiles to classify whole slide images (n = 15) in a tile-by-tile mode. The performance of the model was evaluated on slides manually annotated by the pathologist. The CNN model detected tumor areas with 89.4% accuracy. Areas with blood and meninges were respectively classified with 98.2% and 89.8% accuracy. The overall classification accuracy was 88.7%, and the Cohen’s kappa coefficient reached 0.748, indicating a very good concordance with the manual ground truth. Our pipeline can process digital slides at full resolution, and has the potential to objectively pre-process slides for proliferative index quantification.

Automatic quantification of vessels in hemorrhoids whole slide images

The paper presents the automatic method to examine the histological Whole Slide Images of hemorrhoids. The specimen were stained by Hematoxiline and Eosine, and evaluated by number of vessels, occurrence blood clots and hemorrhage. The proposed method is based on colour deconvolution, mathematical morphology, decision tree classification, region growing operation and region reconstruction. The results of algorithm were compared with medical expert results, what confirmed efficiency of presented algorithm. In 80% of cases, vessels were correctly detected. In 85% cases correctly evaluated occurrence of blood clots and hemorrhages were correctly evaluated in 75% of cases.