Search

INTRODUCTION Peritoneal Carcinomatosis (PC) is a late stage manifestation of several malignancies like gynecological, gastrointestinal or hepatobiliary cancers, among others. In PC, tumors spread on the peritoneal surface and often leave patients with only palliative treatment options. However, advanced EOC has better outcomes in terms of survival compared with other malignancies like colorectal neoplasms or mesotheliomas. The pathological study is essential to plan the suitable treatment and early intervention in EOC. In this work we analyzed the value of tumor markers CA 125, CEA, ratio CA 125/CEA and diameter of ovarian tumor (DOT) as a pre-surgical predictive tool of ovarian cancer.METHODS We recruited patients with suspicious CP studied in the Gynaecological Department of Complejo Hospitalario de Navarra, between 2005 and 2016. CA 125 and CEA were evaluated in preoperative venous blood samples and DOT was estimated by ultrasound or CT scan.RESULTS We identified 250 patients with suspicious CP. Median age was 64,5 years and most patients (86.4%) were diagnosed with EOC. Final diagnosis is described in Table 1. CA 125 and ratio CA 125/CEA was higher in those patients with EOC, while CEA was lower with significant differences (p

[ANGLÈS] Due to the overwhelming increment of information in multiple languages, Machine Translation has become an essential application in our lives. Proof of this are the continuous investments made by technology companies to develop new and improved translation systems. However, translation between distant language pairs such as Chinese and Spanish, which are commonly used in both business and daily life, have been seldom addressed from a research point of view. This project focuses on online translation between Chinese and Spanish. Initially, we present a brief introduction to the field of Machine Translation, and a quick overview to its history and its main approaches. After that, we introduce Statistical Machine Translation, which is the translation paradigm behind our online system. We explain the mathematical basis and the structure of the models. Likewise, we present the evaluation framework used to evaluate our system. Next, we describe how the online system has been built, which allows for translating either from a web-based interface or from two mobile applications (one for Android and one for iOS). Also, we explain how special methods for inputting Chinese and Latin characters are included in the web-based interface and the application. Then, we present the implementation details regarding our statistical translation system. We cover both the description of the corpora used to train the system and the quality assessment of the resulting translations. Finally, we briefly explore the paradigm of rule-based machine translation (also between Chinese and Spanish). We explain the theory of this translation system and we describe the construction of a toy system in order to illustrate how this kind of systems works. This last task is the basis for an open-source rule-based machine translation system that is being developed within the framework of the Google Summer of Code 2013.

The main goal in compiling this book was to highlight the situation in Africa in terms of AIDS and opportunistic diseases. Several chapters reveal great poverty, an apocalyptic situation in many parts of Africa. Global migration of people resulted in their exposure to pathogens from all over the world. This fact has to be acknowledged and accepted as African reality. New, unconventional hypotheses, not determined by established dogmas, have been incorporated into the book, although they have not yet been sufficiently validated experimentally. It still applies that any dogma in any area of science, and medicine in particular, has and always will hinder progress. According to some biologists, in the future, AIDS is very likely to occur in a number of variations, as a direct result of the ongoing processes in the global human society. Thus, we urgently need a comprehensive solution for AIDS, in order to be ready to fight other, much more dangerous intruders.

In this project, I proposed a set of methods to complete the task of multimodal machine translation, which is to generate a image caption in the target language given the image itself and corresponding image captions in the source language. I completed this task with deep learning techniques.