The objective of the project is to develop and validate a Virtual Assistant – “Coach” that supports the process of healthy and active ageing of elderly people that are still active but in risk of disconnecting from the Society. SAAM aims at monitoring and supporting (coaching) elderly in four basic (main) domains: mobility, activity, sleep and social activity; and three advanced domains: dietary, cognition and emotion, and cardio-vascular health. The main challenge in designing the overall SAAM system architecture is that the employed technology needs to remain hidden or unobtrusive to the largest extent possible, yet in service of the user to help her/him gain the sense of safety and independence while benefitting from the social inclusion provided by its social circles and caregivers.

SAAM poster
SAAM poster

From the architecture point of view, the SAAM system consists of two main parts – a centralized cloud-based server part, and a user-side data-collection infrastructure with a personal client running within the user’s home. The whole concept is based on a cloud-based approach connecting all components via broadband connections and provides centralized services through APIs and interfaces. In the user side segment, the SAAM coaching system relies on the use of unobtrusive personal (mobile phone, wearable sensors) and smart home sensors (electricity smart meter, simple IoT devices) to monitor and analyse the activity of individual users. In the server-side segment, a predefined subset of data from all end-user locations is collected, cleared, conditioned and stored, i.e. prepared for subsequent processing and reasoning. Coaching actions are presented to end users through applications supporting multitude of different levels of details and interactions required, efficiently distinguishing between the access right levels, and privacy and preference settings of social circles, caregivers, family members and elderly interested in their data themselves.

The involvement of our laboratory in the SAAM project is mainly in the development and testing of the cardio-vascular health monitoring module. The goal of the cardio module is to establish personal monitoring (coaching) based on long-term ECG and blood pressure measurements. A medical-grade ECG body sensor, developed at JSI and commercially available as Savvy ECG, is used for the assessment of eventual cardio-vascular problems.

More information about the SAAM project can be found at


Jozef Stefan Institute is the leading Slovenian scientific research institute, covering a broad spectrum of basic and applied research. The staff consists of more than 960 specializes in natural sciences, life sciences and engineering.

SAAM partners
SAAM partners

JSI team

Roman Trobec
Aleksandra Rashkovska-Koceva
Viktor Avbelj
Miha Mohorčič
Matjaž Depolli



E. Merdjanovska, A. Rashkovska; Respiration extraction from single-channel ecgusing signal-processing methods and deep learning, MIPRO 2020 : 43nd International Convention, September 28 - October 2, 2020, Opatija, Croatia, 2020 [DOI: 10.23919/MIPRO48935.2020.9245219][COBISS: 36548099]
I. Tomašić, N. Petrović, M. Linden, A. Rashkovska; Comparison of publicly available beat detection algorithms performances on the ECGs obtained by a patch ECG device, MIPRO 2019 : 42nd International Convention, May 20 -24, 2019, Opatija, Croatia, MIPRO ..., 2019 [DOI: 10.23919/MIPRO.2019.8756769][COBISS: 32388903]
J. Bogatinovski, D. Kocev, A. Rashkovska; Feature Extraction for Heartbeat Classification inSingle-Lead ECG, MIPRO 2019 : 42nd International Convention, May 20 -24, 2019, Opatija, Croatia, MIPRO ..., 2019 [COBISS: 32389415]
I. Grubišić, D. Davidović, K. Skala, M. Depolli, M. Mohorčič, R. Trobec; Enriching heart monitoring with accelerometer data, MIPRO 2019 : 42nd International Convention, May 20 -24, 2019, Opatija, Croatia, MIPRO ..., 2019 [COBISS: 32375847]