RESUMEN
OBJECTIVE: The aim of this retrospective study was to evaluate the programmed cell death 1 (PD‑1) expression in esophageal squamous cell carcinoma (ESCC) and association with clinical characteristics. MATERIALS AND METHODS: From January 2009 to December 2014, 88 patients with ESCC were retrospectively included in this study. Eighty‑eight cancer tissues, 35 paraneoplastic atypical hyperplasia tissues (PAHTs), and 30 relative normal esophageal tissues (RNETs) were collected and tested for expression of PD‑1 by immunohistochemistry assay. The PD‑1 expression and clinical characteristics of the ESCC patients were evaluated. The prognosis of the ESCC patients was compared between the PD‑1 positive and negative patients. RESULTS: The PD‑1 positive rate was 51.2% (45/88), 22.9% (8/35), and 6.7% (2/30) for the cancer tissue, PAHT, and RNET, respectively, with statistical difference (P < 0.05); The PD‑1 expression was significantly associated with lymph node metastasis (P < 0.05) and pathology grade (P < 0.05). The median overall survival was 29.8 months and 32.1 months for the PD‑1 positive and negative groups without statistical difference (hazard ratio = 1.00, 95% confidence interval = 0.58–1.71, P < 0.05). CONCLUSION: PD‑1 may play a key role in the process of carcinogenesis of ESCC but not associated with prognosis and overall survival.
RESUMEN
OBJECTIVE: The present contribution concentrates on the application of support vector machines (SVM) for coronary heart disease and non-coronary heart disease classification. METHODS: We conducted many experiments with support vector machine and different variables of low-density lipoprotein cholesterol (LDLC), high-density lipoprotein cholesterol (HDLC), total cholesterol (TC), triglycerides (TG), glucose and age (dataset 346 patients with completed diagnostic procedures). Linear and non-linear classifiers were compared: linear discriminant analysis (LDA) and SVM with a radial basis function (RBF) kernel as a non-linear technique. RESULTS: The prediction accuracy of training and test sets of SVM were 96.86% and 78.18% respectively, while the prediction accuracy of training and test sets of LDA were 90.57% and 72.73% respectively. The cross-validated prediction accuracy of SVM and LDA were 92.67% and 85.4%. CONCLUSION: Support vector machine can be used as a valid way for assisting diagnosis of coronary heart disease.
OBJETIVO: El presente trabajo trata de la utilización de las máquinas de vector de apoyo a la hora de clasificar cardiopatías coronarias y cardiopatías no coronarias. MÉTODOS: Llevamos a cabo numerosos experimentos con máquinas de vector de apoyo y diferentes variables de colesterol de lipoproteínas de baja densidad (CLBD), colesterol de lipoproteínas de alta densidad (CLAD), colesterol total (TC), triglicéridos (TG), glucosa y edad de nuestro conjunto de datos (346 pacientes con procedimientos de diagnóstico completos). Se compararon los clasificadores lineales y no lineales: el análisis lineal discriminante (ALD) y las máquinas de vector de apoyo (SVM) con un kernel de función de base radial (FBR) como técnica no lineal. RESULTADO: La exactitud de predicción del conjunto de pruebas y de entrenamientos de SVM fue 96.86% y 78.18% respectivamente, mientras que la exactitud de prediccin de los conjuntos de prueba y entrenamientos de ALD fue 90.57% y 72.73% respectivamente. La exactitud de predicción de SVM y ALD tras la validación cruzada fue 92.67% y 85.4%. CONCLUSIÓN: La máquina de vector de apoyo puede usarse como una forma válida de ayuda a la hora de realizar el diagnóstico de la cardiopatía coronaria.