Aplicación para extracción de predicciones a partir de datos de cáncer de próstata
Loading...
Official URL
Full text at PDC
Publication date
2024
Authors
Advisors (or tutors)
Editors
Journal Title
Journal ISSN
Volume Title
Publisher
Citation
Abstract
Según la OMS [8], el cáncer es una de las principales causas de muerte. Sólo en 2022, hubo casi 10 millones de muertes relacionadas con esta enfermedad en todo el mundo; y se espera que para 2040 esta cifra aumente a más de 15 millones. Para las mujeres, el tipo de cáncer más común es el cáncer de mama, mientras que para los hombres son el cáncer de pulmón y el cáncer de próstata. Uno de los principales casos de riesgo en este último es que en la gran mayoría de casos es indoloro y asintomático. Por tanto, es muy importante para la sociedad desarrollar métodos para detectar el cáncer de próstata y diagnosticarlo a tiempo. Este es el propósito de este trabajo de Fin de Grado: a partir de una base de datos reales de tumores de próstata, desarrollar y evaluar varios modelos de aprendizaje automático y aprendizaje profundo para crear una aplicación que a partir de mediciones de tumores sea capaz de dar un diagnóstico sobre los mismos.
The World Health Organization (WHO) identifies cancer as one of the leading causes of mortality globally [8]. In 2022 alone, cancer accounted for nearly 10 million deaths worldwide, with projections estimating that this number could exceed 15 million by 2040. Breast cancer is the most prevalent form of cancer among women, while lung and prostate cancer are most common in men. A key challenge with prostate cancer is that, in the majority of cases, it is both painless and asymptomatic. This highlights the critical need for early detection and timely diagnosis. The objective of this undergraduate thesis is to address this need by utilizing a real-world database of prostate tumors to develop and assess various machine learning and deep learning models. The ultimate goal is to create an application capable of providing diagnostic insights based on tumor measurements.
The World Health Organization (WHO) identifies cancer as one of the leading causes of mortality globally [8]. In 2022 alone, cancer accounted for nearly 10 million deaths worldwide, with projections estimating that this number could exceed 15 million by 2040. Breast cancer is the most prevalent form of cancer among women, while lung and prostate cancer are most common in men. A key challenge with prostate cancer is that, in the majority of cases, it is both painless and asymptomatic. This highlights the critical need for early detection and timely diagnosis. The objective of this undergraduate thesis is to address this need by utilizing a real-world database of prostate tumors to develop and assess various machine learning and deep learning models. The ultimate goal is to create an application capable of providing diagnostic insights based on tumor measurements.