The Covid-19 era was characterized by an increase in patients needing constant medical care because of the virus and post-covid symptoms. Patients were increasingly monitored 24 hours a day. Thus, remote monitoring of patients has became an established method of monitor- ing their vital parameters and keeping track of their activities. Using remote monitoring, healthcare workers can manage patients’ clinical ac- tivities without physically intervention. Using a dashboard, healthcare professionals can visualize the patient’s clinical course and intervene when necessary. Globally and in territorial healthcare, clinical data from patients are an increasingly valuable asset. Collection, analysis, and use of accurate patient information can help consolidate patient information. It is possible to analyze symptoms for a more accurate diagnosis, highlight demographics and geography to aid in disease di- agnosis, define scenarios, and plan for the necessary resources. Remote monitoring is once again defined by data. The amount of data collected is closely related to machine learning based analysis of data streams. Having access to real-time data enables convenient dosing of drugs and improves care quality. In this context, machine learning is a paradigm to perform patient monitoring remotely without requiring the physical presence of the healthcare provider. This thesis summarises our ef- forts to exploit the opportunities offered by machine learning in remote monitoring systems. The model we propose identifies possible data transmission tampering. Our model can also monitor whether patients properly follow the healing path suggested by the doctor. This work is motivated by our research on remote monitoring systems and su- pervised learning algorithms for clinical data management. It is based on an extensive literature review on remote monitoring systems and a survey of artificial intelligence paradigms and techniques capable of managing clinical data to provide a high level of data quality, which is essential for informed clinical care.

Approaches to Clinical Pathway Protection by means of Artificial Intelligence and Process Mining

Lofù, Domenico
2022-01-01

Abstract

The Covid-19 era was characterized by an increase in patients needing constant medical care because of the virus and post-covid symptoms. Patients were increasingly monitored 24 hours a day. Thus, remote monitoring of patients has became an established method of monitor- ing their vital parameters and keeping track of their activities. Using remote monitoring, healthcare workers can manage patients’ clinical ac- tivities without physically intervention. Using a dashboard, healthcare professionals can visualize the patient’s clinical course and intervene when necessary. Globally and in territorial healthcare, clinical data from patients are an increasingly valuable asset. Collection, analysis, and use of accurate patient information can help consolidate patient information. It is possible to analyze symptoms for a more accurate diagnosis, highlight demographics and geography to aid in disease di- agnosis, define scenarios, and plan for the necessary resources. Remote monitoring is once again defined by data. The amount of data collected is closely related to machine learning based analysis of data streams. Having access to real-time data enables convenient dosing of drugs and improves care quality. In this context, machine learning is a paradigm to perform patient monitoring remotely without requiring the physical presence of the healthcare provider. This thesis summarises our ef- forts to exploit the opportunities offered by machine learning in remote monitoring systems. The model we propose identifies possible data transmission tampering. Our model can also monitor whether patients properly follow the healing path suggested by the doctor. This work is motivated by our research on remote monitoring systems and su- pervised learning algorithms for clinical data management. It is based on an extensive literature review on remote monitoring systems and a survey of artificial intelligence paradigms and techniques capable of managing clinical data to provide a high level of data quality, which is essential for informed clinical care.
Artificial Intelligence; Process Mining; Information Security; Clinical Pathway Protection
File in questo prodotto:
File Dimensione Formato  
35 ciclo-LOFU Domenico.pdf

accesso aperto

Descrizione: Approaches to Clinical Pathway Protection by means of Artificial Intelligence and Process Mining
Tipologia: Tesi di dottorato
Licenza: Dominio pubblico
Dimensione 2.42 MB
Formato Adobe PDF
2.42 MB Adobe PDF Visualizza/Apri

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/11589/246060
Citazioni
  • Scopus ND
  • ???jsp.display-item.citation.isi??? ND
social impact