Publicação
OPTIMIZING NON-MELANOMA SKIN CANCER DIAGNOSIS THROUGH DATA-DRIVEN PATIENT PROFILING
| Resumo: | Non-melanoma skin cancer encompasses various forms of skin malignancies, with the most common types of tumors being basal cell carcinoma and squamous cell carcinoma, and is currently the most frequent type of skin cancer worldwide. Since its incidence is expected to continue increasing, this cancer represents an important health problem in medical practice and it is becoming increasingly evident that there is a need to leverage the potential of biomedical engineering to improve the management of this disease. This work aims to provide an overview of these cancers in a region of Portugal, drawing insights on how organizational and informational processes function, based on real-world data collected from the Portuguese National Oncology Registry, which serves as the repository for oncology information systems. Using data analysis techniques, it was attempted to fit the diagnosed patients into well- defined groups that have similar waiting times until the diagnosis and identify co-occurring characteristics within those groups. The analysis looked both into the demographic and medical features of the patients aiming to uncover patterns that could allow a more efficient allocation of resources, streamline healthcare processes, and ultimately improve patient outcomes. Although distinct patterns were not often identified, it was clear, through the assessment of the registry performed, that the information systems have significant fragilities that affect the quality and accuracy of the data they provide. The identified limitations hamper a complete representation of the organizational and informational processes, consequently impacting their ability to fully support management needs. |
|---|---|
| Autores principais: | Pimenta, Inês Magalhães Capela |
| Assunto: | Non-melanoma skin cancer Clinical pathway Survival Analysis National Oncology Registry |
| Ano: | 2024 |
| País: | Portugal |
| Tipo de documento: | dissertação de mestrado |
| Tipo de acesso: | acesso aberto |
| Instituição associada: | Universidade Nova de Lisboa |
| Idioma: | inglês |
| Origem: | Repositório Institucional da UNL |
| Resumo: | Non-melanoma skin cancer encompasses various forms of skin malignancies, with the most common types of tumors being basal cell carcinoma and squamous cell carcinoma, and is currently the most frequent type of skin cancer worldwide. Since its incidence is expected to continue increasing, this cancer represents an important health problem in medical practice and it is becoming increasingly evident that there is a need to leverage the potential of biomedical engineering to improve the management of this disease. This work aims to provide an overview of these cancers in a region of Portugal, drawing insights on how organizational and informational processes function, based on real-world data collected from the Portuguese National Oncology Registry, which serves as the repository for oncology information systems. Using data analysis techniques, it was attempted to fit the diagnosed patients into well- defined groups that have similar waiting times until the diagnosis and identify co-occurring characteristics within those groups. The analysis looked both into the demographic and medical features of the patients aiming to uncover patterns that could allow a more efficient allocation of resources, streamline healthcare processes, and ultimately improve patient outcomes. Although distinct patterns were not often identified, it was clear, through the assessment of the registry performed, that the information systems have significant fragilities that affect the quality and accuracy of the data they provide. The identified limitations hamper a complete representation of the organizational and informational processes, consequently impacting their ability to fully support management needs. |
|---|