ABSTRACT
Objetivos: Crear un sistema experto que sea abierto y cuya base de conocimiento esté representada por imágenes y pueda ser utilizada como vehículo de información relevante. Material y métodos: Uno de los autores actuó como ingeniero del conocimiento. Con un ordenador Pentium 133 MHz y en Visual Basic se ha creado SARA (Sistema Abierto de Razonamiento Artificial). Para su elaboración se ha construido la aplicación SEADLINH (Sistema Experto de Ayuda al Diagnóstico de Linfomas No Hodgkin). Resultados: El sistema ofrece imágenes con las que comparar la arquitectura microscópica y el aspecto citológico del caso problema. La selección de distintas imágenes (patrones) lleva a un diagnóstico indicando la probabilidad de certeza. El sistema permite acceder a su base de conocimiento y modificarla. Admite información complementaria en forma de imágenes o texto. Conclusiones: Se ha creado un sistema experto (SARA) de tipo abierto, que está disponible para llenar de contenidos su base de datos y su base de conocimientos. Su calidad ulterior dependerá de los recursos que se empleen en su confección, abriendo el campo de acción de los patólogos (AU)
Subject(s)
Artificial Intelligence , Lymphoma/diagnosis , Microscopy/methods , Information Systems , Information Systems/supply & distribution , Electronic Data Processing/methods , Anatomy/education , Anatomy/methods , Anatomy/trends , Pathology/methods , Pathology/trends , Pathology/education , Medical Informatics/instrumentation , Medical Informatics/trends , Medical Informatics/education , 28640/methods , Bias , Classification/methods , Decision Making, Computer-Assisted , Artificial Intelligence , Diagnostic Services , Diagnostic Services/trends , National Health Programs/trends , Expert Systems , Telemedicine/methods , Telemedicine/trends , Health Knowledge, Attitudes, PracticeABSTRACT
PURPOSE: The aim of this study was to investigate a possible influence of lymphocytic infiltration in Chronic Lymphocytic Leukemia (CLL) on bone mass, according to the infiltration patterns. MATERIAL AND METHODS: We have studied histological preparations from decalcified and paraffin included bone biopsies and used a histomorphometric study through a mixed (interactive-automatic) image analysis, in order to quantify bone trabecular (BT/TV) and cellular volume (CV). Preparations were visualized via a 2.5 objective; a video camera was adapted and connected to a computer with analogical-digital converter incorporated. After the image was captured, a binarization was made by trabecular and medullary spaces segmentation; automatically, we calculated trabecular area (TA) and cellular area (CA). Expressions indicated above were used to calculate BV/TV and CV, expressed in percentage: BV/TV (%) = [TA/(TA + CA)] x 100, and CV (%) = [CA/(TA + CA)] x 100. RESULTS: 92 normal biopsies and 79 CLL biopsies were studied (Nodular 12, Diffuse 22, Interstitial 23, Mixed 22). BT/TV in normal subjects age over 45, was 18.40%; it was lower in CLL subjects, 17.25%. CV was 33.62% in normal cases and significantly higher in CLL with 42.95%. According to the infiltration patterns, Nodular pattern showed similar value to normal biopsies, versus Diffuse pattern with BT/TV of 16.19% and CV of 51.49%. Significantly, a high CV value was accompanied of a low B/TV value, in normal and CLL subjects. CONCLUSIONS: We analyzed histomorphometrically trabecular volume as expression of bone mass in undecalcified biopsies. CLL infiltration tended to show a lower trabecular volume. Diffuse pattern presented the higher cellular infiltration and bone mass destruction.
Subject(s)
Bone and Bones/pathology , Image Processing, Computer-Assisted , Leukemia, Lymphocytic, Chronic, B-Cell/pathology , Leukemic Infiltration , Aged , Aged, 80 and over , Analog-Digital Conversion , Biopsy, Needle , Bone Density , Bone Marrow/pathology , Female , Humans , Male , Middle Aged , Retrospective StudiesABSTRACT
A cytomorphodensitometric study was made by analysis of the computerised image of the germinative and intermediate epithelial layers in a sample of five specimens of cholesteatoma and the corresponding skin from the external auditory canal. A morphometric variable corresponding to the nuclear area was determined and densitometric variables equivalent to the relative quantity of nuclear DNA were determined. Within the germinative layers of the cholesteatoma epithelium, we observed larger nuclei and more relative DNA per nucleus. The average densitometric values were in the range of 2C-4C, which is indicative of a greater cellular population in S phase in the cholesteatoma than in the skin.
Subject(s)
Cholesteatoma/diagnosis , Cholesteatoma/physiopathology , Densitometry/methods , Ear, Middle/cytology , Ear, Middle/physiopathology , Otitis Media/physiopathology , Adult , Cell Count , Cholesteatoma/complications , Culture Techniques , DNA/analysis , Ear, External/cytology , Epithelial Cells , Epithelium/physiopathology , Female , Humans , Image Interpretation, Computer-Assisted , Male , Middle Aged , Otitis Media/complications , S Phase , Skin/cytologyABSTRACT
Epithelial proliferation in colesteatoma and its influence on the subepithelial inflammatory reaction was studied using morphometry and nuclear photometry of specimens. Twenty specimens of colesteatoma and 15 specimens of skin from the external auditory canal were examined. Nuclear content was greater in the basal cells of colesteatoma epithelium than in the basal cells of external auditory canal epithelium, suggesting increased proliferative activity. This activity was found to be reted to the inflammatory infiltrate of the conjunctive tissue and varied in different specimens.
Subject(s)
Cell Movement , Cholesteatoma/physiopathology , Ear, Middle/physiopathology , Adolescent , Adult , Child , Connective Tissue , Ear, External/physiopathology , Epithelium/physiopathology , Female , Humans , Lymphocytes , MaleABSTRACT
We report a rare case of non-Hodgkin lymphoma located in the vulva. The diagnosis was supported by the cytological anomalies in material obtained by fine needle aspiration. The tumor was removed with wide excision, and the patient was alive and in good health ten months after surgery.
