Work

Platform for AI-driven medical data analysis to support clinical decision

Year: 2022
Type: dissertation
Abstract: Cancer is one of the leading causes of death on the world and surviving its treatment does not mean that the process is over. Several patients that have undergone cancer treatment, feel insecure in relation to their health, due to the stress and anxiety of cancer reappearance and post-treatment symptoms such as: sleeping disorders, fatigue and memory problems, pain, anxiety, and stress. Patients that undergone cancer treatment are followed periodically by a clinician, that evaluates its clinical situation, but also, his Quality of Life. This information is vital to understand the patient well-being, since cancer as a huge impact on all aspects of the patient’s life. Nevertheless, clinicians lack on tools capable of measuring objectively the patient’s Quality of Life, nor tools that enable more data visualization that could improve the clinician’s decision-making. So, the purposed aim of this dissertation is to provide a Clinical Decision Support System Platform with visualization tools capable of giving information from patients, gathered from a wearable device and a smart scale, and using Fuzzy Logic, an Artificial Intelligence subset, to give new insights about patient well-being. The designed CDSS Platform was able to integrate commercially used smart device, with minimal human intervention required. Also, the data gathered from those devices was used to create a continuous monitoring system, associated with visualization tools that enhanced the clinician knowledge of the patient. Furthermore, an indicator denominated as Patient Progression Indicator was developed with the use of the Fuzzy Logic algorithm, that provides an indirect but objective measurement of the patient well-being. Although the results seem promising, more in-depth research is required such as a trial study capable of validating the results obtained.
Author: Dias, Álison Válter
Language: English
Open Access status: green
OpenAlex ID: W7033074752
Cites: 42
Cited by:
Related to: 10