Machine learning for classification of cutaneous sebaceous neoplasms: implementing decision tree model using cytological and architectural features.

BACKGROUND: This observational study aims to describe and compare histopathological, architectural, and nuclear characteristics of sebaceous lesions and utilized these characteristics to develop a predictive classification approach using machine learning algorithms. METHODS: This cross-sectional study was conducted on Iranian patients with sebaceous tumors from two hospitals between March 2015 and March 2019. Pathology slides were reviewed by two pathologists and the architectural and cytological attributes were recorded. Multiple decision tree models were trained using 5-fold cross validation to determine the most important predictor variables and to develop a simple prediction model. RESULTS: This study assessed the characteristics of 123 sebaceous tumors. Histopathological findings, including pagetoid appearance, neurovascular invasion, atypical mitosis, extensive necrotic area, poor cell differentiation, and non-lobular tumor growth pattern, as well as nuclear features, including highly irregular nuclear contour, and large nuclear size were exclusively observed in carcinomatous tumors. Among non-carcinomatous lesions, some sebaceoma and sebaceous adenoma cases had features like high mitotic activity, which can be misleading and complicate diagnosis. Based on multiple decision tree models, the five most critical variables for lesion categorization were identified as: basaloid cell count, peripheral basaloid cell layers, tumor margin, nuclear size, and chromatin. CONCLUSIONS: This study implemented a machine learning modeling approach to help optimally categorize sebaceous lesions based on architectural and nuclear features. However, studies of larger sample sizes are needed to ensure the accuracy of our suggested predictive model.

Clinicopathologic Assessment of Ocular Adnexal Lymphoproliferative Lesions at a Tertiary Eye Hospital in Iran.

BACKGROUND: The most common type of ocular lymphoma is non-Hodgkin lymphoma (NHL), categorized into two groups: indolent (slow growing) and aggressive (rapid growing). Differentiating benign reactive lymphoid hyperplasia (RLH) from malignant ocular adnexal lymphoma (OAL) is challenging. Histopathology, immunohistochemistry (IHC) and ow cytometry have been used as diagnostic tools in such cases. MATERIALS AND METHODS: In this retrospective case series, from 2002 to 2013 at Farabi Eye Center, 110 patients with ocular lymphoproliferative disease were enrolled. Prevalence, anatomical locations, mean age at diagnosis and the nal diagnosis of the disease with IHC were assessed. Comparison between previous pathologic diagnoses and results of IHC was made. Immunoglobulin light chains and B-cell and T-cell markers and other immuno-phenotyping markers including CD20, CD3, CD5, CD23, CD10, CYCLIND1 and BCL2 were evaluated to determine the most accurate diagnosis. The lymphomas were categorized based on revised European-American lymphoma (REAL) classi cation. RESULTS: Mean age±SD (years) of the patients was 55.6 ±19.3 and 61% were male. Patients with follicular lymphoma, large B-cell lymphoma or chronic lymphocytic leukemia/small cell lymphoma (CLL/SLL) tended to be older. Nine patients with previous diagnoses of low grade B-cell lymphoma were re-evaluated by IHC and the new diagnoses were as follows: extranodal marginal zone lymphoma(EMZL) (n=1), SLL(n=1), mantle cell lymphoma (MCL) (n=3), reactive lymphoid hyperplasia RLH (n=2). Two cases were excluded due to poor blocks. Flow cytometry reports in these seven patients revealed SLL with positive CD5 and CD23, MCL with positive CD5 and CyclinD1 and negative CD23, EMZL with negative CD5,CD23 and CD10. One RLH patient was negative for Kappa/Lambda and positive for CD3 and CD20 and the other was positive for all of the light chains, CD3 and CD20. Orbit (49.1%), conjunctiva (16.1%) and lacrimal glands (16.1%) were the most common sites of involvement. CONCLUSIONS: Accurate pathological classi cation of lesions is crucial to determine proper therapeutic approaches. This can be achieved through precise histologic and IHC analyses by expert pathologists.