scholarly journals Multinomial Classification of Patterns in Lung Cancer Biopsy Slides Using Customized Convolutional Neural Network

2021 ◽  
Author(s):  
Jung Wook Yang ◽  
Dae Hyun Song ◽  
Hyo Jung An ◽  
Sat Byul Seo
Keyword(s):  
Cell Carcinoma ◽  
Lung Carcinoma ◽  
Treatment Plan ◽  
Area Under The Curve ◽  
Large Cell ◽  
Workflow Systems ◽  
Suitable Treatment ◽  
Cancer Biopsy ◽  
Pathology Image

Abstract Identifying the lung carcinoma subtype in small biopsy specimens is an important part of determining a suitable treatment plan but is often challenging without the help of special and/or immunohistochemical stains. Pathology image analysis that tackles this issue would be helpful for diagnoses and subtyping of lung carcinoma. In this study, we developed AI models to classify multinomial patterns of lung carcinoma (adenocarcinoma, squamous cell carcinoma, small cell carcinoma, large cell neuroendocrine carcinoma) and non-neoplastic lung tissue based on convolutional neural networks (CNN or ConvNet). Four CNNs that were pre-trained using transfer learning and one CNN built from scratch were used to classify patch images from pathology whole-slide images (WSIs). We evaluated the diagnostic performance of each model in the test sets. The Xception model achieved the highest performance among pre-trained CNNs with an accuracy of 0.86 and an area under the curve (AUC) of 0.97. The built from scratch CNN model obtained an accuracy of 0.92 and an AUC ranging from 0.99 to 1.00 for subtyping lung carcinoma tasks. These results demonstrate how promising CNN models are for developing improved diagnostic workflow systems for diagnosis and subtyping of lung carcinoma. Of particular note is the fact that the built from scratch CNN described in this paper achieves prompt and consistent results so has the potential to be applied in working hospitals for pathological diagnoses.

Histological Typing of Lung Tumours

1981 ◽  
Vol 67 (4) ◽  
pp. 253-272 ◽  
Author(s):  
◽  
E. Chaves ◽  
Toão Pessoa ◽  
O. Campobasso ◽  
J. Chrétien ◽  
...  
Keyword(s):  
Cell Carcinoma ◽  
Large Cell Carcinoma ◽  
Large Cell ◽  
Intermediate Cell ◽  
Cell Type ◽  
Histological Differentiation ◽  
Lung Tumours ◽  
Oat Cell Carcinoma

The WHO Histological Classification of Lung Tumours, published in 1967, has been revised. The main features are as follows: Squamous cell carcinoma (epidermoid carcinoma) has the same definition as in the original version, i.e., the identification of keratin and/or intercellular bridges by light microscopy. Three degrees of histological differentiation are described. Dysplasia and carcinoma in situ are discussed. Small cell carcinoma is divided into oat-cell carcinoma, an intermediate cell type and a category for oat-cell carcinomas combined with other major types. Adenocarcinoma includes the acinar, papillary and bronchiolo-alveolar forms and the solid carcinomas with mucus formation (previously part of the large cell carcinoma group). Mesothelial tumours are divided into fibrous, epithelial and biphasic subtypes. A number of less common tumours and tumour-like lesions are defined.

α-Methylacyl CoA Racemase in Pulmonary Adenocarcinoma, Squamous Cell Carcinoma, and Neuroendocrine Tumors: Expression and Survival Analysis

2007 ◽  
Vol 131 (10) ◽  
pp. 1555-1560
Author(s):  
Konstantin Shilo ◽  
Tatiana Dracheva ◽  
Haresh Mani ◽  
Junya Fukuoka ◽  
Isabell A. Sesterhenn ◽  
...  
Keyword(s):  
Squamous Cell Carcinoma ◽  
Cell Carcinoma ◽  
Squamous Cell ◽  
Lung Carcinoma ◽  
Small Cell Lung Carcinoma ◽  
Large Cell ◽  
Small Cell ◽  
Stage I ◽  
Small Cell Lung ◽  
Cell Lung Carcinoma

Abstract Context.—α-Methylacyl CoA racemase (AMACR) is an oxidative enzyme involved in isomeric transformation of fatty acids entering the beta-oxidation pathway. AMACR serves as a useful marker in establishing a diagnosis of prostatic malignancy; however, limited information is available in regard to its presence in pulmonary neoplasms. Objective.—To investigate AMACR expression within a spectrum of lung carcinomas and its correlation with patients' survival. Design.—Four hundred seventy-seven pulmonary carcinomas, including 150 squamous cell carcinomas, 150 adenocarcinomas, 46 typical carcinoids, 31 atypical carcinoids, 28 large cell neuroendocrine carcinomas, and 72 small cell carcinomas, were studied immunohistochemically using tissue microarray–based samples. Results.—Overall, pulmonary tumors were positive for AMACR in a significant percentage (47%) of cases. Among tumor types, 22% of squamous cell carcinoma, 56% of adenocarcinoma, 72% of typical carcinoid, 52% of atypical carcinoid, 70% of large cell neuroendocrine carcinoma, and 51% of small cell lung carcinoma were positive for AMACR. Furthermore, the Kaplan-Meier analysis revealed that the patients with AMACR-positive small cell carcinoma had better survival (19% vs 5% after 5 years, P = .04) than patients with AMACR-negative tumors. Such survival advantage was seen for patients with stage I–II (P = .01) but not stage III–IV small cell carcinomas (P = .58). Conclusions.—These results indicate that, similar to prostate cancer, the overexpression of AMACR frequently occurs in pulmonary carcinomas. Additionally, its positive correlation with outcome of stage I–II small cell lung carcinoma warrants further investigation of the AMACR role in the prognosis of this tumor.

scholarly journals Das anaplastische Lungenkarzinom des Hundes

1973 ◽  
Vol 10 (2) ◽  
pp. 102-113 ◽  
Author(s):  
H. Stünzi
Keyword(s):  
Cell Carcinoma ◽  
Large Cell Carcinoma ◽  
Large Cell ◽  
Domestic Animals ◽  
Small Cell ◽  
Round Cell ◽  
Histological Criteria ◽  
Pulmonary Tumors ◽  
Pulmonary Neoplasia

Of 86 canine pulmonary tumors five had the histological criteria of undifferentiated small-cell carcinoma and two of undifferentiated large-cell carcinoma. The five small-cell anaplastic careinomas could be divided, as is the case in man, according to strict histological criteria into fusiform, lymphocyte-like, and polygonal subgroups. In the fusiform and round-cell subgroups there were focal sites of equivocal differentiation towards epidermoid careinoma or adenocarcinoma. These local changes must be considered in making a diagnosis, but it was not possible to positively determine the behavior of anaplastic, epidermoid or glandular carcinomas. The classification of pulmonary carcinomas of domestic animals has significance not only for experimental oncology but also for clarification of the cause of pulmonary neoplasia. Anaplastic pulmonary carcinomas have not been found in the cat.

scholarly journals Classification of Lung Carcinomas in the Dog and Cat

1981 ◽  
Vol 18 (4) ◽  
pp. 513-528 ◽  
Author(s):  
J. E. Moulton ◽  
C. Von Tscharner ◽  
R. Schneider
Keyword(s):  
Squamous Cell Carcinoma ◽  
Cell Carcinoma ◽  
Large Cell ◽  
Cell Types ◽  
Small Cell ◽  
Alveolar Cell ◽  
Lung Carcinomas ◽  
Gland Carcinoma ◽  
Bronchial Gland

A total of 218 lung carcinomas from dogs and cats were examined histologically. The tumors were classified into adenocarcinoma, squamous-cell carcinoma, bronchial gland carcinoma, and alveolar-cell carcinoma. We believe that adenocarcinoma should be subdivided into differentiated and undifferentiated types because the two are distinct histologically and vary in frequency in the cat and dog. It is also important to recognize bronchial gland carcinoma, a distinct histological type, and to subdivide alveolar-cell carcinoma into three separate types: anaplastic small-cell and large-cell types, and adenomatosis type.

Prognostic Significance of the Cyto-Histological Classification of Infiltrating Cervical Epidermoid Carcinoma

1980 ◽  
Vol 66 (2) ◽  
pp. 183-190 ◽  
Author(s):  
Paolo Dalla Palma ◽  
Carla Arslan Pagnini ◽  
Giuseppe de Laurentiis
Keyword(s):  
Cell Carcinoma ◽  
Classification System ◽  
Survival Rates ◽  
Prognostic Significance ◽  
Large Cell Carcinoma ◽  
Large Cell ◽  
Small Cell ◽  
Histological Classification ◽  
Cervical Carcinomas

One hundred-fifty-five cervical carcinomas were classified according to the Wentz and Reagan (21) cyto-histological system, and correlated to 5 year survival rates. This classification subdivides epidermoid carcinomas into three sub-groups: keratinizing carcinoma, large cell carcinoma, and small cell carcinoma. No significant differences between one group and another were observed. The importance of the type of therapy in conditioning the significance of this classification system is discussed.

Sign in / Sign up

Export Citation Format

Share Document