Prototype of Explainable Artificial Intelligence System for Assessing the Health Status of Geriatric Patient (ProtoArtIs)

Project no.: PP34/2102
Project website:

Project description:

Personalized (human centric) risk assessment and threat prevention has become a relevant challenge in health care recently. An aging society and various chronic and aggressive diseases such as dementia, cancer, and others are driving the development of new methods and systemic tools for early disease prevention as well as the provision of health care to the elderly. Measures are needed to assess the factors (symptoms) of everyone’s health risks and to help the physician to determine the likelihood of these risks quickly and objectively, i.e., advise what syndromes are specific to the patient. The use of artificial intelligence in geriatrics is very promising and relevant, as the diagnosis of a geriatric patient is a complex, experience-based, and time-consuming process that involves a variety of questionnaires and subjective and inaccurate patient responses.
The aim of the project is to create a prototype of a geriatric patient’s health condition and risk assessment system based on explainable artificial intelligence. The developed prototype will allow to assess the health status of patients according to the symptoms and factors used in medical practice, identify health risks, and present the most likely syndromes or diseases, will serve as a decision support tool for patient care staff.

Project funding:

KTU Research and Innovation Fund

Project results:

The innovative decision support system was developed, which can be used to provide guidance to geriatricians, family physicians, or other medical staff. The system helps to assess a patient’s health risks (syndromes) related to eating disorders, taking into account symptoms and factors identified and confirmed in medical practice.
The developed prototype allows the patient to be diagnosed objectively, quickly and without much medical experience, even to inexperienced medical staff (e.g. residents, family physicians and general practitioners), shortens the decision-making time by avoiding several days of patients monitoring (usually up to a week), which is common for inpatient geriatric examination.
The system is also able to detect additional relations between risks, which is difficult to an inexperienced specialist, and which is especially relevant when a person develops several syndromes (risks).

Period of project implementation: 2021-04-01 - 2021-12-31

Project partners: Lithuanian University of Health Sciences

Agnius Liutkevičius

2021 - 2021

Centre of Real Time Computer Systems, Faculty of Informatics