1. Introduction
Human communities and medical systems worldwide have been facing the new reality characterized by many consequences of an ageing population. During the next few decades, by 2050, almost one-quarter of the world’s population will be 60 years of age and older [
1]. In the European Union, the disability-free life expectancy (healthy life years) at the age of 65 is approximately 10 years for both women and men [
2]. The goal of medical–social interventions is not only to increase longevity but also to maintain independent functioning, active contribution to society, and a good health-related quality of life in older age. Cerebrovascular, heart diseases, and cancer are the main causes of death and they all have in common a series of modifiable lifestyle risk factors such as poor diet, tobacco use, alcohol consumption, and physical inactivity [
3]. These factors along with social isolation, depression, cognitive impairment, physical disabilities, sensory impairments, and lack of access to specialized medical care have a significant negative impact on well-being and quality of life in older age. Present medical and social care practices are based on a passive approach where services are provided when requested. Arguably, this archetype suggests issues that can raise some debate. Even though scientifically supported and overall cost-effective, preventive medicine, a more proactive approach, is not a consistent practice and plays only a minor role within healthcare systems [
4,
5]. The negative effects of risk factors are dose- and exposure-duration-dependent across the lifespan. It has been established that especially older people can greatly benefit from risk reduction; however, behavior modification is difficult to achieve and it is a complex process [
4]. When successfully implemented, preventive interventions require a great deal of time and human resources and, in the specific context of the older population, more specialized trained healthcare professionals [
6,
7,
8,
9]. Caring for seniors with various degrees of disabilities, either physical or cognitive, represents a constant challenge for families and communities. When displaced from their homes in order to be placed in long-term care facilities, their well-being and health status often significantly decline. The desirable approach is to provide assistance and care at home, thus enabling older people to live independently and, of course, focusing specialized health care on maintaining functional independence for the longest possible time.
Nowadays there are opportunities for meeting the care needs of older adults. In this respect, more and more research efforts and successful results have led to reshaping the technology landscape with increasingly advanced technological solutions such as biosensors, ambient sensors, and artificial intelligence [
10]. Widespread availability of wireless devices and high-speed internet access technologies has also contributed to augment the care and assistance available for seniors [
11]. Any technological solution for older people has to be adapted to their characteristics, such as visual or hearing impairments, cognitive function status, level of education, computer literacy, and financial resources. The vINCI technology represents an innovative instrument developed, specifically but not exclusively, for seniors, by computer sciences researchers together with a medical team specialized in geriatrics and gerontology. It was designed to be independently and effortlessly used by seniors in the comfort of their own environment. It is a modular and flexible platform that can integrate a large array of various sensors and therefore easily adapts to specific care needs. The pilot study tested a system consisting of sensors and standardized instruments capable of evaluating several parameters and also of generating a personalized feedback for the user dedicated to optimizing physical activity level, social interaction, and health-related quality of life. Moreover, the system was able to detect and signal events and health-related aspects that would require medical assistance [
12,
13].
In this paper, we present the results of the pilot vINCI technology on health-related quality of life (QoL), which was tested on an adequate sample of older adults. The system integrates smart devices with monitoring sensors for older adults to collect physical activity, psychological, and social parameters, as well as health and context information describing each subject.
1.1. Health-Related Quality of Life
Many authors agree that when defining QoL there are two main domains: objective domains (such as health status, economic status, social functioning and status, housing, etc.) and subjective domains (emotional and psychological well-being, personal fulfillment, morale, self-esteem, etc.); all these should be accounted for [
14,
15]. Another discussion of these concepts entails the argument of imposed values of life as opposed to self-reported dimensions of quality of life. The World Health Organization (WHO) defines QoL as “an individual’s perception of their position in life in the context of the culture and value systems in which they live and in relation to their goals, expectations, standards and concerns” [
16].
The concept of health-related quality of life, as defined by the WHO, introduces a link between two relevant terms: quality of life and health. In 1948, the WHO formulated the definition of health as follows: “a state of complete physical, mental and social well-being and not merely the absence of disease or infirmity”. There are two aspects that need to be taken into account when evaluating the health-related quality of life: the approach should be multidimensional, which means that is important to know the physical, social, and psychological characteristics of the individual and it should be monitored from both perspectives (objective and subjective) for each domain [
17,
18,
19,
20,
21,
22].
In older age, health inevitably becomes the main concern of an individual and has the most important load on a person’s QoL and well-being [
23,
24]. Physical activity is a major determinant of health-related self-perceived quality of life, especially in older age compared to younger adults [
25,
26].
The World Health Organization Quality of Life Questionnaire—Short Form (WHOQOL-BREF) is a “generic” instrument, centered on the self-reported impact of health status on social activities, psychological well-being, and autonomy. The WHOQOL-BREF instrument has been extensively used in medical and social research and public policy-making, and has been successfully applied in clinical studies on older populations [
27].
1.2. vINCI Technology
The vINCI project (Clinically-validated INtegrated Support for Assistive Care and Lifestyle Improvement: the Human Link) has developed an assistive monitoring technology for the elderly. vINCI, an integrated and validated framework, using the internet of things (IoT), provides monitoring services and nonintrusive assistance to the elderly, to support them but also their families, who do not have the time or opportunity to care for them permanently. The vINCI system is also aimed at people who provide care and medical assistance to the elderly, as well as various medical organizations. The challenge for vINCI technologies is to meet the needs of older adults, adapt to cognitive and perceptual decline, and support older adults in daily activities while protecting privacy, independence, and information security.
The monitored person downloads a free application on their mobile phone. Using the application on the smart phone, or the web version of vINCI technology, the user creates an account and gives the application access to personal data. From the mobile application, the person can periodically fill in a series of medical questionnaires to which otherwise he/she would have had access only after complex medical consultations: a questionnaire to identify the perception of quality of life, a questionnaire to identify the perception of physical activity, and another to rate their level of physical activity and mental comfort. The application also provides the scores obtained, even in time, for the respective questionnaires. From the dashboard (web version), the family can track in real time the monitored parameters of the person.
vINCI is a clinically validated ambient intelligence framework, where multiple wearable and ambient devices work together to create an aggregated solution able to capture the various facets of events leading to the decrease in the perceived quality of life as associated with old age.
The vINCI application connects several innovative user-oriented devices: a smart watch (Fitbit and CMD watch, provided by the partner Connected Medical Devices, Bucharest, Romania), smart insoles, and tablets on which users can answer surveys and questions about their mood. All data is aggregated and analyzed by a phone application (vINCI app, which the user uses) and a web platform (vINCI dashboard, used by the owners). This application intends to create a much clearer picture of the patient’s health, life, and activities, and also provide intelligent care for the elderly in clinics. The final goal is to use the data collected from devices, analyze it, and use the results to improve the quality of life of the monitored patients and increase the active ageing rate. The technology designed for user interaction was developed with the active involvement of the elderly [
13,
28,
29].
vINCI technology was tested in the pilots proposed in the project. The most relevant was the pilot developed at the Ana Aslan National Institute of Gerontology and Geriatrics (NIGG), Bucharest, Romania, were there was a salon dedicated to the project and medical staff to validate the technology with patients [
12].
In parallel, a second medical pilot was deployed in Cyprus, in an institute working with the elderly belonging to the University of Nicosia (medical partner).
In previously published papers, important research was presented on technology components in various stages of development. Preliminary versions of the vINCI platform, as well as the involvement of patients and their caregivers in the design of the technology, are demonstrated in the work [
29]. This consists of investigating their needs and requirements from information collected using the specific questionnaires.
All collected data are integrated into a model that ontologically models the medico-geriatric profile of a patient. This innovative model was developed following an extensive study that followed both medical records in the field and consulted specialists in various relevant fields. Finally, the model was tested and then implemented via a single service on the vINCI platform, where all data are aggregated unitarily for a patient. Two things were accomplished in this patient model: the data obtained in the tests were used for medical validation of the evidence obtained by monitoring with the help of the technology, and an algorithm was proposed to interpret the monitoring information (providing the score that the user receives via the app) [
30]. Preliminary study of the establishment of the quality of life model proposed in the project and analysis methods are presented in detail in [
31,
32]. An original analysis of the data collected through the WHOQoL-BREF questionnaires from the elderly, which identifies groups with distinct QOL profiles, is presented in [
33].
1.3. vINCI Architecture
The vINCI system, an application that offers non-intrusive monitoring and care to the elderly, contributes to improving quality of life (QoL) by offering modern and efficient solutions that meet their expectations [
12].
The vINCI architecture is based on the JHipster microservices architecture, which provides preconfigured tools and components that offer the management of microservices applications: gateway, JHipster Registry, and JHipster Console. Each of these microservices is responsible for collecting data from the devices and sending the data further to the IO-server microservice [
29].
Figure 1 shows the microservices architecture of the vINCI application.
To simplify the schema, the communication links between the JHipster Registry and the rest of the microservices, or between the JHipster console and them, have been removed. However, the gateway and all the newly created microservices still use the JHipster Registry for discovery and the JHipster Console to store logs and metrics.
In the first development phase, the vINCI platform architecture was constituted mainly of the basic JHipster components and the microservices for the CMD watch, Fitbit, insole, and survey microservice.
The architecture was designed so that each microservice had its own PostgreSQL database, where the records from its corresponding device should be stored. The reason was that each device had its own data format and each database would have its specific schema. For example, the watch sends JSON objects that contain coordinates, while the survey application sends JSON objects with fields such as “surveyType” or “scoring Result”.
The gateway is a Java Spring application, like the other microservices generated with JHipster, but with a front-end side written in ReactJS and much more logic of its own. In the case of vINCI, the client uses only one entry for requests: the gateway. It acts as a proxy from client applications to back-end services and vice versa. It exposes a single end point to client applications and maps requests internally to the microservices that should handle the business from that end point. This happens with most microservices architectures, where functionality needs to be used by many services. The API gateway model also allows the developer to more easily add new microservices to the architecture, reduces the overhead of complicated requests, means less duplicate logic and code, and improves security. This task that the gateway performs is called HTTP request routing. It automatically submits each newly added microservice, which can be accessed using its name “/microservice-name”, which is of course secure.
The dashboard is the client side of the gateway microservice. It is an application developed with the latest libraries: ReactJS, bootstrap, HTML5, and CSS3. Since it is written in React, it is very well organized in components, meaning that adding new custom features is very easy. It is built and runs with the npm or yarn JavaScript package managers, which download and install the dependencies needed by the dashboard. The dashboard comes with an administration section for the admin user where one can see information about the microservices that are running in the vINCI platform and their configuration, manage the users, explore application metrics, check the logs, or see the swagger API documentation.
Furthermore, the data saved in IO-Server is fetched and displayed for each device type separately in the entities table. New data can be introduced by the admin manually as well. These are the out-of-the-box features of the gateway. However, more logic has been added to the gateway to satisfy the requirements and objectives of the vINCI project.
When receiving device data, the IO-Server identifies the device from which the information came by a UUID field. However, the information about the device is stored in the tables as an ID, not a string UUID. Therefore, the IO-Server needs to map the received UUID to its corresponding ID from the gateway. The communication between the IO-Server and the gateway is done through REST API calls using a feign client.
The business regarding the watch device is handled by the watch microservice. This is a Spring Boot application. The watches used in the vINCI platform are provided by the Connected Medical Devices company, which developed them to send data periodically to one of its servers (the external server shown in the above figure). For vINCI, the company provided us with an API for importing the data. The watch service makes API calls every 30 s to check if any new data are available. The imported data is then sent to the IO-Server through a feign client, then saved in the specific watch table.
Another important side of the architecture consists of the insole microservice and the insole Android application. The latter allows one or more smart insole devices to connect to it simultaneously and send data to it. The mobile application processes the received data and translates it into more relevant information (i.e., the type of the recorded activity: standing, walking). Then, the information is delivered to the shoe microservice, which forwards it to the IO-Server.
The survey Android application and the survey microservice, written in Spring, handle the business regarding the questionnaires completed by the patients. Unlike the smart shoes or the watches business, the starting point here is not a smart device, but the actual mobile application installed on the patient’s tablet or smart phone. The users complete the surveys, then the answers and the calculated scoring results are sent to the survey microservice as JSON objects. As said before, the specific survey data were stored in a specific PostgreSQL database.
1.4. vINCI Kits
In our study, we used four monitoring kits, but the number can be extended according to the availability and preferences of end users and the monitoring team, and this reflects the flexibility of vINCI technology. The four monitoring kits used in this study were:
The smart watch (CMD One smart watch) is an existing technology provided by the partner Connected Medical Devices (CMD). The information provided by the smart watch is GPS location/time, the number of steps, how many times the watch has left a defined area, and time intervals when the watch has been removed from the wrist.
The Fitbit Ionic watch has been integrated into vINCI technology, and is a watch that contains a series of biometric sensors useful in the project for assistive monitoring of the elderly. Fitbit offers users a wide range of physical activity monitoring devices. In the current stage, numerous analyses have been carried out on the validity of these devices in measuring the different indices for adults and the elderly. The Adidas Edition device was selected to offer the main features needed to monitor the elderly. The Fitbit microservice was implemented in Spring Boot (the technology behind the vINCI platform). The tracker communicates with the Fitbit mobile application using BLU technology, which interacts with Fitbit servers using HTTPS.
Smart insoles, technology developed during the project, which can be used indoors, features a wireless communication interface (BLE or LoRA). The insoles identify different patient activity conditions, namely: standing rest, walking, running, and non-contact (in the sense that shoes are not worn or the foot is not placed on a walking surface). The history of physical activity performed by the subject can be tracked by collecting the data packets transmitted in a given time interval and checking the time stamps applied by the server.
Questionnaires: WHOQOL-BREF (World Health Organization Quality of Life Questionnaire—Short Form) and IPAQ-SF (International Physical Activity Questionnaire—Short Form). The WHOQOL-BREF was used in the project to measure the quality of life of selected individuals recruited to be part of the study; this questionnaire is available in several languages and for the project, under a legal agreement between us and the World Health Organization, we were granted a license to use the WHOQOL-BREF in accordance with the WHO’s terms and conditions. The WHOQOL-BREF version is a questionnaire with 26 items that assess the quality of life in the physical, psychological, social, and environmental domains. The IPAQ-SF is used to assess the level of physical activity of people recruited to be part of the study; this questionnaire assesses physical activity in many domains, leisure time physical activity, household activities, work-related physical activity, and transport-related physical activity.
The collected data, related to the physical and mental health conditions of the older people, composed a model of the monitored data for each person, which was validated with the support of medical professionals (at the Ana Aslan National Institute of Gerontology and Geriatrics, Bucharest, Romania).
Each vINCI application kit was represented by a microservice. The data collected from patients was medically validated by specialists and then the patient was shown an account of their general condition, receiving recommendations specific to the identified health condition.
All data from surveys and IoT devices was stored in a secure manner to keep data private. This simplified interaction can help detect early symptoms of old age or trigger alerts for special cases such as an older adult’s fall, stroke, etc. This information was provided to the elderly person, their carers, and their family. The patient profile model was used to evaluate the impact of the vINCI technology on the level of quality of life perceived by the elderly adult, allowing an appropriate adjustment of the intervention assistance provided by the family and caregivers [
30].
The data from the devices, derived from information based on a conceptual model, enabled possible alerts based on a detected deterioration in conditions of advanced age, and is shown in
Figure 2.
1.5. Security and Privacy through Blockchain in vINCI Platform
One of the major problems in health care is the responsibility for data preservation and authenticity. In health care, a variety of personal information is generated by clinics, hospitals, and health-related applications, data that ends up being stored in large databases. During their life, people interact with a large number of medical specialists, for example, pediatricians, general practitioners, nurses, dentists, specialists in medical insurance, etc. Each of them stores data in their IT system, leading to a fragmented system and databases that are not shared.
In our case, the vINCI platform has been developed considering both security and privacy considerations. This means that data being generated for the patient needs to protect their personal data, and access needs to be restricted and fully controlled by the patient alone. Here, some data might be used by clinics, by doctors, or health companies even, but when all these actors need to be involved, who will administer the system? We say that the security of the communicated data needs to be ensured through the use of classic mechanisms, authentication based on certificates, and through the use of blockchain.
In the vINCI platform, the blockchain manages access and permissions to collect health information and use the personal data stored in the platform. For this purpose, the blockchain collaborates with the vINCI platform, the edge nodes, and the Digital Caregiver application. For a starting point in our description, we refer the reader also to our previous paper.
The vINCI blockchain-based data access control system has been developed using the Hyperledger Fabric framework [
34], which manages the access and permissions on the information collected, respectively, and the personal data stored in the vINCI platform.
Blockchain is a digital ledger where all the executed transactions are stored. It uses a distributed peer-to-peer network to make a continuously growing list of ordered records called blocks. Every block contains a set of signed transactions and is validated by the network itself, by means of a consensus mechanism. Copies of the blockchain are distributed on each participating node in the network. Blockchain can be considered as a permanent database because the implemented algorithms prevent alteration of the already stored information. Blockchain provides unified, secure, and user-controlled access to patient health data. It allows users to easily grant, modify, or revoke access to their data.
Blockchain has been integrated as an intermediary in the collaboration between edge nodes and the Digital Caregiver application. The Digital Caregiver application interacts directly with the platform and provides data from patient clinical questionnaire completions and data from medical investigations performed by clinical staff. Entities with the correct permissions can deliver the data to the platform or access health records only when their own identities and cryptographic keys used are verified by the blockchain.
The blockchain structure uses public key cryptography to create an immutable chain of accounts. Each node participating in the network has a copy of the blockchain (tablets and edge nodes, or the vINCI platform). Due to the large amount of data generated, the content of the nodes consists of information links, permissions, and other additional information. The data are either on the participant platform or on the vINCI platform.
Clinics, hospitals, research institutes, insurance companies, healthcare applications and other entities that generate medical data, which have the role of nodes, make up the distributed network. Users interact with a particular node and, as a result, data is generated. This stored information is encrypted and digitally signed to ensure the privacy and authenticity of the information. The data must have a certain format accepted by all the other nodes. After this interaction, the node sends a request to the patient asking him/her if the information will be published in the blockchain or not. This communication flow is presented in the
Figure 3:
In our implementation, we consider that the proof of work concept is not suitable in healthcare blockchain. Instead, proof of interoperability is an alternative method that eliminates some disadvantages, including powerful and costly hardware required for high computations. Proof of interoperability implies that transactions and stored data are interoperable with regard to a known set of structural and semantic constraints. Structural constraints imply attributes such as type and cardinality, and semantic constraints imply using an agreed value set. Proof of interoperability assumes that all miners reach consensus regarding the set of data templates advised by specialists in medical terminology. The miners, the nodes within a while list, are in fact the medical institutions that check the data format.
Therefore, considering the example of the older adult being consulted at a clinic, the institution generates a set of analysis information. This stored information is encrypted and digitally signed to ensure the privacy and authenticity of the information. The institution sends a link with this information and, if the patient agrees to publish it, the clinic broadcasts the transaction to the network. The miners verify the data format and, if the majority agrees, then the clinic generates a new block and appends it to the blockchain in chronological order. All the miners must update it, as shown in the
Figure 4.
For vINCI in particular, the blockchain platform implementation, based on open source software—HyperLedger Fabric version 1.4 [
35]—includes three modules: microservice blockchain, fabric proxy, and blockchain network. The interaction of the vINCI system with the blockchain network is based on the blockchain microservice and its submodule fabric proxy.
The microservice defines two main entities whose data is saved as transactions in the blockchain network: storage and policy. The storage entity specifies information about a single medical analysis for a patient. A policy entity is defined for each storage transaction and specifies data access policy by calling the appropriate REST API methods defined in the microservice.
The fabric-proxy module is responsible for the integration of the blockchain vINCI microservice with the blockchain network components.
The blockchain microservice is a component of both the vINCI platform and the blockchain platform. Thus, the data saved on the blockchain platform will be reached from the gateway front-end. Validation of both platforms covers the processes of user registration, creation of data access policies, and data requests by the user.
Storage transactions are recorded periodically through REST API communication between analytics modules and the microservice blockchain. These are performed automatically by the analysis module running on the vINCI platform.
The patient defines the list of authorized persons to access personal information through the REST API functions of the blockchain microservice. After this, the policy transaction associated with the storage transaction is saved. At this moment, the person who wants to access the patient data is verified based on the corresponding IDs. The individual queries the appropriate entity for the policy transaction to retrieve the keys to access the patient data.
Based on the security access key received, the patient results saved in the blockchain database can be obtained through the REST API function
4. Discussion
During the past decade, technological assisted living for healthy aging has become a distinct research field with many initiatives. Using monitoring sensors, other projects developed intelligent home care systems, step counters, location, and identification systems, signal algorithms for analyzing heart arrhythmias and cognitive function promotion systems [
50,
51,
52,
53].
There are several original aspects of our project that we would like to highlight.
Firstly, the vINCI technology is a tripod system with a user-friendly interface in the form of a smart phone application specifically designed for older people, an integrative platform that allows networking as many sensors as necessary to provide different models adapted to the care needs of older people and dedicated sensors for physical activity and location monitoring.
Secondly, the physical activity level is evaluated both subjectively and objectively, providing an accurate estimation, while the feedback received by the user is based on a medical backed algorithm.
Another original feature of our project is that testing of the vINCI technology was in clinical settings, in a controlled environment on actual older patients with various chronic diseases.
As presented, the results of the clinical study are promising. The patients admitted to the Ana Aslan NIGG clinics showed a high interest in the use of smart devices and in completing the experiment and the attrition rate was 0%.
Participants were open and motivated to wear these devices to obtain specific information about a range of biophysiological measurements. The use of Fitbit raised particular interest by providing easily accessible additional information.
According to the satisfaction questionnaire, the study’s participants considered the vINCI devices easy to use and understand, without a negative interference on their daily activities. The older patients were generally satisfied with the use of vINCI devices for intelligent continuous monitoring of clinical status and behavior.
In the experimental group, the results showed that there were statistically significant differences between day 1 and day 8 regarding quality of life (QoL) in each of the WHOQOL-BREF domains: physical, psychological, social, and environmental. Finally, in the control group, the results showed that there were no statistically significant differences in scores for participants between day 1 and day 8 regarding QoL in each of the WHOQOL-BREF domains, except for the social domain.
In the experimental group (day 8), of the 30 patients, 66.7% had a high level of physical activity (PA) and the median value of total PA (MET-minutes/week) was 3304.50, the high category describing high levels of PA participation. In the control group (day 8), of the 30 patients, 50.0% had a moderate level of PA and the median value of total PA (MET-minutes/week) was 2029.50.
We also identified several limitations of our clinical pilot study.
The sample size was small and the duration of the intervention was short, although these features were characteristic for a pilot study.
Secondly, we can speculate that the significant increase in the physical activity levels and health-related quality of life might have been partly due to a confounding motivation of the participants generated by the attention they received from our supporting staff during the experiment.