Banca de DEFESA: GELLY VIANA MOTA

Uma banca de DEFESA de MESTRADO foi cadastrada pelo programa.
STUDENT : GELLY VIANA MOTA
DATE: 21/12/2022
TIME: 14:00
LOCAL: IMD - Sala B206
TITLE:

UNSUPERVISED LEARNING FOR CLASSIFICATION IN THE SURVEY OF FORMATIVE DEMANDS FOR NURSING

KEY WORDS:

Continuing Education; Non-Supervised Learning; Neonatal; Nursing.

PAGES: 180
BIG AREA: Ciências Exatas e da Terra
AREA: Ciência da Computação
SUBÁREA: Metodologia e Técnicas da Computação
SUMMARY:

Inadequate or deficient training of healthcare professionals can lead to serious complications for patients. In particular, in the Neonatal Intensive Care Unit (NICU), the risk is evident in babies who have barely started their lives. Continuing Education (PE) in healthcare is based on the assumption of significant and problematizing learning, proposing strategies that enable collective construction, in addition to guiding paths for a dialogical relationship. The NICU presents its own intense dynamics in relation to the activities performed by the nursing staff, which hinders the PE in understanding the whole and acting preventively to possible sequelae. The production of changes in the work process of these teams, in the practices of management, care, and social control in the health area has been desired by the actors of this process through continuing education in line with the demands of the service itself. In the line of development of educational technologies, this work has as its main objective to help and contribute to hospital continuing education in the survey of formative demands for nursing at the NICU through Unsupervised Learning (UL). To this end, data from January to December 2021 of the system incorporated in the NICU sector, the Research Electronic Data Capture (REDCap), which contains information from discharge and death records, were analyzed, resulting in 392 records. The research has a quanti-qualitative approach and for the development process of this work the PDCA (Plan, Do, Check and Action) method was adopted, which aims at delivering continuous improvement in the processes it enhances.  Exploratory data analysis was performed using Colab, Python and R tools for this purpose . About 70.6% of the 2021 admissions were premature babies and 96% of these had hypothermia on admission. The difficulty in finding patterns due to the diversity of the data, made it possible to apply the UL to identify the formative possibilities for the NICU nursing team based on the data. Based on the data that were defined according to their specificities in REDCap, the algorithm was used to perform a clustering in order to explain the total variability of the data. When the algorithm was applied to the set of maternal antecedents, respiratory intercurrences and ophthalmologic intercurrences, these explained a variability from 56.6% to 47.3%. This demonstrates the explanatory variability of each set of data and that can guide continuing education on which training needs can be addressed by these factors that occur in the NICU. This study aims to help health professionals, especially nurses, in the indication of knowledge based on data from the unit's daily routine, which is necessary and appropriate for the exercise of their function in a safe and independent way, when promoting quality care to newborns in the neonatal unit. And finally, impact on the reduction of hospital costs that can be avoided through a qualification that is in line with the demand of the service.

COMMITTEE MEMBERS:
Presidente - 3229319 - APUENA VIEIRA GOMES
Interna - 2245086 - ISABEL DILLMANN NUNES
Interna - 1362181 - ISMENIA BLAVATSKY DE MAGALHÃES
Externa ao Programa - 4665456 - DANIELE VIEIRA DANTAS - UFRNExterna à Instituição - CINTIA CAPISTRANO TEIXEIRA ROCHA - EBSERH