The purpose of this study was to develop a computer-aided diagnosis (CAD) system for automatic classification of histopathological images of lung tissues. Two datasets (private and public datasets) were obtained and used for developing and validating CAD. The private dataset consists of 94 histopathological images that were obtained for the following five categories: normal, emphysema, atypical adenomatous hyperplasia, lepidic pattern of adenocarcinoma, and invasive adenocarcinoma. The public dataset consists of 15,000 histopathological images that were obtained for the following three categories: lung adenocarcinoma, lung squamous cell carcinoma, and benign lung tissue. These images were automatically classified using machine learning and two types of image feature extraction: conventional texture analysis (TA) and homology-based image processing (HI). Multiscale analysis was used in the image feature extraction, after which automatic classification was performed using the image features and eight machine learning algorithms. The multicategory accuracy of our CAD system was evaluated in the two datasets. In both the public and private datasets, the CAD system with HI was better than that with TA. It was possible to build an accurate CAD system for lung tissues. HI was more useful for the CAD systems than TA.
Malignant pleural mesothelioma (MPM), associated with unfavorable outcomes, is closely associated with asbestos exposure. Early detection and treatment are critical to prolong survival of patients with MPM because of the rapid progression and resistance to treatment. The recently defined malignant mesothelioma in situ (MIS) has been gaining increasing attention with advances in genome-based methods including fluorescence in situ hybridization (FISH) as well as immunohistochemistry. We herein report the case of a MIS in a 73-year-old male with a history of asbestos exposure presenting with massive pleural effusion in the right thoracic cavity. Video-assisted thoracoscopic surgery with pleural biopsy of the right side revealed a single layer of atypical mesothelial cells without invasive lesions by hematoxylin and eosin staining.However, these mesothelial cells exhibited a loss of methylthioadenosine phosphorylase (MTAP) by immunohistochemistry and homozygous deletion of CDKN2A (p16) by FISH, leading to the diagnosis of MIS.
DNA methylation is considered the primary epigenetic mechanism underlying the development of malignant melanoma. Since DNA methylation can be influenced by environmental factors, it is preferable to compare cancer and normal cells from the same patient. In order to compare the methylation status in melanoma tissues and melanocytes from the same individuals, we employed a novel epidermal sheet cultivation technique to isolate normal melanocytes from unaffected sites of melanoma patients. We also analyzed primary and metastatic melanoma samples, three commercially available melanocytes, and four melanoma cell lines. Cluster analysis of DNA methylation data classified freshly isolated melanomas and melanocytes into the same group, whereas the four melanoma cell lines were clustered together in a distant clade. Moreover, our analysis discovered methylation at several novel loci (KRTCAP3, AGAP2, ZNF490), in addition to those identified in previous studies (COL1A2, GPX3); however, the latter two were not observed in fresh melanoma samples. Subsequent studies revealed that NPM2 was hypermethylated and downregulated in melanomas, which was consistent with previous reports. In many normal melanocytes, NPM2 showed distinct immunohistochemical staining, while its expression was lost in malignant melanoma cells. In particular, intraepithelial lesions of malignant melanoma, an important challenge in clinical practice, could be distinguished from benign nevi. The present findings indicate the importance of using fresh melanoma samples, not melanoma cell lines and melanocytes in epigenetic studies. In addition, NPM2 immunoreactivity could be used to differentiate melanomas from normal melanocytes or benign disease.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.