Research Progress of Deep Learning in Bladder Cancer Pathology / 肿瘤防治研究

The incidence of bladder cancer is increasing annually, and the gold standard for its diagnosis relies on histopathological biopsy. Whole-slide digitization technology can produce thousands of high-resolution captured pathological images and has greatly promoted the development of digital pathology. Deep learning, as a new method of artificial intelligence, has achieved remarkable results in the analysis of pathological images for tumor diagnosis, molecular typing, and prediction of prognosis and recurrence of bladder cancer. Traditional pathology relies heavily on the professional level and experience of pathologists; as such, it is highly subjective and has poor reproducibility. Deep learning can automatically extract image features. It can also improve diagnostic efficiency and repeatability and reduce missed and misdiagnosed rates when used to assist pathologists in making decisions. This technology cannot only alleviate the pressure of the current shortage of skilled workforce and uneven medical resources but also promote the development of precision medicine. This article reviews the latest research progress and prospects of deep learning in pathological image analysis of bladder cancer.

Ethical concerns and recommendations for sharing anatomic pathology images in online social media networks

@#Anatomic pathology is a field that relies on visual examination to provide diagnosis. Photos of specimens and microscopic slides play an important role in pathology education. With the internet, sharing and seeing images from different patient cases has become efficient and accessible. However, ethical concerns may be raised since patient images are used for academic purposes in a public setting. Proper de-identification, informed consent and setting professional guidelines for sharing pathology images are suggested.

Metodología para la evaluación automatizada de la expresión del receptor de vitamina D mediante técnica inmunocitoquímica / Methodology of automated Vitamin D receptor immunocytochemical expression análisis

Introducción: la citología permite examinar células de un tejido de manera mínimamente invasiva, sin embargo, la capacidad de realizar técnicas complementarias como la inmunocitoquímica (ICQ) no está exenta de dificultades. Es el objetivo de nuestro trabajo presentar una metodología que permita la utilización de ICQ automatizada asociada a un análisis automatizado mediante técnica de patología digital. Métodos: se incluyeron 5 sujetos sanos y se obtuvieron muestras de superficie ocular utilizando un citocepillo. La muestra fue procesada de manera automatizada mediante citología en fase líquida. Posteriormente se realizó ICQ automatizada para detectar la positividad nuclear del receptor de vitamina D. Para la evaluación, se utilizaron dos métodos: cuantificación directa bajo microscopio de luz y análisis automatizado usando analizador de imágenes en las diapositivas digitales obtenidas con un Scanner. El porcentaje de positividad encontrado con ambos métodos fueron comparados utilizando la prueba de Kappa. Resultados: todas las muestras presentaron una celularidad adecuada. En todos los casos fue posible realizar ICQ automatizada, más aún, todas las muestras presentaron una calidad óptima. Al comparar ambos métodos (manual versus automatizado) se observó un nivel de acuerdo sustancial (Kappa=0,69). Conclusiones: la metodología presentada en este manuscrito permite la evaluación automatizada de marcadores inmunohistoquímicos de la superficie ocular de manera mínimamente invasiva, siendo similar al conteo manual, pero más objetivo y reproducible. Esta técnica podría ser útil para el estudio proteómico en patologías como la enfermedad por ojo seco.

Introduction: Cytology tests use small amounts of tissue samples for diagnosis as a minimally invasive technique; however, the ability to perform complementary methods such as immunocytochemistry (ICC) is not without difficulties. The aim of our work is to present a method that allows the use of automated ICC associated with an automated image analysis using digital pathology. Methods: Five healthy subjects were included, and ocular surface samples were obtained using a cytobrush. The sample was processed as liquid-based cytology. Automated ICC was subsequently performed to detect vitamin D receptor nuclear positivity. Two methods were used for evaluation: manual counting under a light microscope and automated analysis using an image analyzer on digitized slides. The percentage of positivity found in both methods was compared using the Kappa test. Results: All samples presented adequate cellularity. In all cases, it was possible to perform automated ICC; moreover, all samples presented optimal quality. When comparing both methods (manual versus automated), a substantial level of agreement was seen (Kappa=0.69). Conclusions. The method presented in this manuscript allows the minimally invasive automated evaluation of ocular surface ICC markers, being like manual counting but more objective and reproducible. This technique could be useful for proteomic study in pathologies such as dry eye disease.

Research Progress and Application of Digital Pathology Technology in Bone Marrow Morphology / 中国医疗器械杂志

Advances in digital pathology technology have enabled pathologists and laboratory physicians to perform quick, easy, accurate and reproducible analysis of digital images of tissues and cells with the aid of electronic screens and software tools, rather than relying solely on traditional optical microscopy observations. The conventional clinical cytology testing practice is to be replaced by a digital workflow, which includes both digital imaging and image analysis. This article provides an overview of the basic principles of digital pathology techniques, the advances of development of device in cytology digital pathology, and their clinical applications in bone marrow morphology, and existing problems and prospects of digital pathology application in hematology.

Feasibility of fully automated classification of whole slide images based on deep learning

Although microscopic analysis of tissue slides has been the basis for disease diagnosis for decades, intra- and inter-observer variabilities remain issues to be resolved. The recent introduction of digital scanners has allowed for using deep learning in the analysis of tissue images because many whole slide images (WSIs) are accessible to researchers. In the present study, we investigated the possibility of a deep learning-based, fully automated, computer-aided diagnosis system with WSIs from a stomach adenocarcinoma dataset. Three different convolutional neural network architectures were tested to determine the better architecture for tissue classifier. Each network was trained to classify small tissue patches into normal or tumor. Based on the patch-level classification, tumor probability heatmaps can be overlaid on tissue images. We observed three different tissue patterns, including clear normal, clear tumor and ambiguous cases. We suggest that longer inspection time can be assigned to ambiguous cases compared to clear normal cases, increasing the accuracy and efficiency of histopathologic diagnosis by pre-evaluating the status of the WSIs. When the classifier was tested with completely different WSI dataset, the performance was not optimal because of the different tissue preparation quality. By including a small amount of data from the new dataset for training, the performance for the new dataset was much enhanced. These results indicated that WSI dataset should include tissues prepared from many different preparation conditions to construct a generalized tissue classifier. Thus, multi-national/multi-center dataset should be built for the application of deep learning in the real world medical practice.

PTEN loss in Gleason grade 7 prostate tumors exhibits intratumoral heterogeneity and is associated with unfavorable pathological features

Background: PTEN loss is observed in 20­30% of prostate cancers and is associated with a poor outcome, but clinical details of the impact of this biomarker are unclear for intermediate grade tumors. Methods: We investigated 43 radical prostatectomy-derived grade 7 prostate tumors from the Clinics Hospital of Ribeirão Preto. Tissue microarray (TMA) blocks were constructed and PTEN copy number status was determined for all patients through fluorescence in situ hybridization (FISH). To determine the presence of PTEN protein loss in our study cohort, we performed immunohistochemistry (IHC) in TMA sections. We then developed an automated algorithm in HALO™ to identify regions of PTEN protein loss in whole prostate scanned sections from ten patients with known PTEN deletion status by FISH. Clinical analyses were conducted to determine the associations between PTEN loss and patient outcome. All statistical analyses were conducted in R v3.4.3 with P-values below 0.05 being considered statistically significant. Results: In this study of 43 grade 7 tumors, we found PTEN deletions by FISH in 18.9% of tumors, and PTEN protein loss by IHC in 16.3% of tumors. Both techniques were highly concordant and complementary. Clinical analysis demonstrated that PTEN deletion by FISH was significantly associated with positive margin invasion (P = 0.04) and Gleason score upgrade (P = 0.001). Digital image analysis of ten representative tumors demonstrated distinct intratumoral heterogeneity for PTEN protein loss in four tumors. Conclusions: This study shows that PTEN loss in Gleason grade 7 tumors can be heterogeneous and that a systematic analysis of this biomarker using a combination of FISH, IHC, and digital imaging may identify patients with a greater risk of poor outcome (AU)

Application perspective of large pathological section coupled with artificial intelligence in the diagnosis of pancreatic tumors / 中华胰腺病杂志

Large pathological sections can display tumors and the para-tumor tissues holistically and completely on one slice, which is beneficial to the whole observation and evaluation of tumors and their surrounding microenvironments, thus gaining a comprehensive understanding about the disease. With the gradual realization of digital pathology and advancement in computational pathology, artificial intelligence has made it possible to achieve accurate and individualized diagnosis and treatment of pancreatic tumors by linking up morphologies of different tumor cells and surrounding microenvironments with various types of data including image omics, gene proteomics and clinical data, which is both an opportunity and a challenge for Chinese researchers.

Introduction of Artificial Intelligence in Pathology / 한양의대학술지

Pathology has a long history of artificial intelligence (AI) as much as any other field of medicine, and has used AI algorithms continuously. However, in Korea, pathology AI is unfamiliar even to the pathologists. In this article, I will summarize the terms and definitions, the basic elements of pathology AI, and the future direction. Digital pathology is a system or environment that digitizes glass slides into binary files, observes them through a monitor or any digital devices, interprets it, analyzes it, and maintains it. Computational pathology is a comprehensive concept of diagnosis support or research system that deals with image, text and omics data. Virtual microscopy is a method or technology that allows pathologists to view and share glass slides images from whole slide scanners. Image analysis is a technique or method that processes various digital images and quantifies features. The basic elements of pathology AI are as follows: environmental factors called digital pathology and technical elements such as AI, machine learning, and deep learning. Digital pathology workflow consists of three elements; acquisition or collection of data, data processing and data storage. The basic process of image analysis consists of preprocessing of image, identification of region of interest, and feature extraction. There is enormous potential for improvement of patient care through digital pathology and/or AI, and a harmonized discussion about activation of Korean digital pathology among government, academia and industry will be mandatory for future medicine and healthcare in Korea.