AALISABETH: lifestyle tutoring through ambient intelligence G.Vespasiani 1, F. Brancaccio 2, G. Matrella 3, I. De Munari 4 and P. Ciampolini 5, on behalf of the AALISABETH partnership 6 Abstract In this paper, the AALISABETH (Ambient-Aware LIfeStyle tutor, Aiming at a BETter Health) project is discussed. It was recently started in the framework of the applied research program Casa intelligente per una longevità attiva ed indipendente dell'anziano, funded by INRCA and Regione Marche. The project aims at tackling prevention issues, with reference to some among most diffused pathologies among elderly people, by means of innovative solutions, featuring low invasiveness and ease of use. 1 Introduction AALISABETH (Ambient-Aware LIfeStyle tutor, Aiming at a BETter Health) focuses at elderly (65+) people, not suffering from major chronic diseases or disabilities, but suffering from (or being at risk of) metabolic of circulatory diseases (such as diabetes of hypertension) or with mild cognitive impairments. 1 Giacomo Vespasiani, direttore dell Unità Operativa di Diabetologia e Malattie del Ricambio, ASUR Marche, Ascoli Piceno e S. Benedetto del Tronto, giacomo@vespasiani.com 2 Fausta Brancaccio, MeTeDa Srl, S. Benedetto del Tronto (AP), fausta.brancaccio@meteda.it 3 Ilaria De Munari, Università degli Studi di Parma, ilaria.demunari@unipr.it Name of the Second Author Name, Address of Institute, e-mail: name@email.address 4 Guido Matrella, Università degli Studi di Parma, guido.matrella@unipr.it 5 Paolo Ciampolini, Università degli Studi di Parma, paolo.ciampolini@unipr.it 6 Consorzio Progetto AALISABETH: Me.Te.Da. S.r.l. S. Benedetto del Tronto (AP); MAC Srl, Recanati (MC); Università degli Studi di Parma; Università degli Studi di Camerino (MC); I.T.C. Srl, Recanati (MC); Sassomeccanica Srl, S. Benedetto del Tronto (AP); Coop. Soc. Nuova Ricerca Agenzia Res, P.to S. Elpidio (FM); Valdichienti Srl, Tolentino (MC); Ralò Srl, S. Severino Marche (MC); ERS Elettronica Srl, Ancona, On-Off Srl, S. Benedetto del Tronto (AP); Dienpi Srl, S. Benedetto del Tronto (AP);Eusebi Srl, S. Benedetto del Tronto (AP).
2 G. Vespasiani et al. A vast majority of elderly population actually fits such a picture: for instance, more than 35% of the overall Italian population is overweight, with 10% being obese, this being a major risk factor for cardiocirculatory diseases. Similarly, 13% of Italian elderly population (+65) suffers from diabetes, this percentage raising up to 20% for over-75 people. Onset and progression of such pathologies are notoriously fostered by improper lifestyles: nutrition, physical exercise and strict compliance to medical therapies are the main keys for prevention or control of diseases. Effective monitoring of health status, however, implies regular checking of physiological parameters and frequent interaction with physicians and caregivers: this is frequently overlooked, due to boredom or complexity of checks, to carelessness or to lack of motivation. The main AALISABETH goal is to develop an innovative strategy exploiting ambient intelligence to make health monitoring less complex and demanding, this possibly incentivizing the adoption of health lifestyles. 2 AALISABETH technological vision AALISABETH project is rooted in several technology areas: from the one hand, smart home technologies, mostly aimed at safety, comfort, entertainment purposes (and recently being exploited for energy awareness and saving as well); on the other hand, telemedicine, mainly focused at remote monitoring and for implementing home care strategies, beneficial to both the user and the health services. Even though both areas share some technical issues (for instance, networking and communication), they are considered as being completely independent, having quite different market and application visions. Our project, instead, aims at exploiting and emphasize synergies: in particular, a common environment sharing both functionalities will be devised: data coming from such environment will be fused and analyzed in order to infer behavioral information, recognizing and measuring parameters related to daily living activities (frequencies and intensities). AALISABETH system is conceived for supporting independent life of elderly people, mixing different technologies in a distributed-intelligence approach. Different components are gathered in a modular and hierarchical infrastructure [1], cooperating in an holistic approach to support health and wellness.
AALISABETH: intelligenza ambientale per la prevenzione 3 At the peripheral level, the system accounts for smart objects, i.e., objects having autonomous functionalities and capable of communicating and interacting with each other. At this level, environmental medical and personal sensors are found. Environmental sensors enable home automation and environmental safety features, as well as with energy management; medical sensors deal self-checking of clinical and physiological parameters such as body weight, blood pressure, heart and breath rates, glycaemia, oxygen saturation level, etc.; personal sensors (i.e, wearable [2]) allows for monitoring and measuring motion, evaluating posture, checking for falls and for user s identification. Further, unconventional, objects having different primary functions will be provided with network-connectivity features and will enter the system communication infrastructure [3]. Either wired and wireless sensors will be exploited, all of them contributing to build a behavioral profile [4], especially oriented, in particular, to most relevant aspects such as physical exercise and nutrition. A first control layer implements simpler and straightforward home functions (e.g., lighting, heating, security) as well as tasks related to logging and transmission of clinical data. Information gathered by the system is also accessed by a second, higher processing layer, in which data are fused and behavioral patterns are sought for. The project vision accounts for an hybrid interpretative model, in which the clinical knowledge of the project research staff is compared and integrated with the inference capability of the data analysis system, in order to train and improve its perspicacity. To this purpose, the project also include a short pilot phase, managed by participating caregiving structure and providing the project with feedbacks on the actual impact and acceptance of the system. User s interaction is of the utmost importance: the system, once qualitative and quantitative behavioral assessment will have been carried out, will provide the user with information, advice and motivation toward the target lifestyle. Similarly, caregivers (formal as well as informal) will be provided with simple and clear view of the relevant information, making them available an additional tool to evaluate health status. At a third upper layer, data coming from different households will converge, allowing for working out statistical data providing medical research with additional, unconventional data, potentially useful to improve knowledge of specific issues regarding such a fast-growing demographic class. We expect major challenges to come from the need of accounting for unprecictable interaction and relationships among different smart objects and the users themselves. Due to such unpredictability (which may become especially critical if cognitive impairments are taken into account) a plain, statistical approach is scarcely effective, and more sophisticated mathematical and computational models have to be involved to build a reliable behavioral model [6,7]. Inferred behavioral knowledge is then fed back
4 G. Vespasiani et al. to the field : it can be exploited to provide environmental control functions with adaptivity and awareness, whereas supplementing physiological data with context information may significantly enhance their meaningfulness and reliability. Moreover, data fusion will possibly enable innovative functions, based on the correlation among clinical data and inferred behavioral profiles; interaction with environmental sensor may provide a way for integrating health-related information coming from dedicated sensors with complementary, indirect information about user s activity, obtained in an inexpensive and non-invasive way. 3 Conclusions AALISABETH project, funded by INRCA and Regione Marche in the framework of the applied research program Casa intelligente per una longevità attiva ed indipendente dell'anziano, foresee the exploitation of ambient intelligence technologies to provide monitoring, guidance and motivation toward healthy lifestyles, suitable for preserving as long as possible autonomy and independence in daily living. It is worth remarking, however, the inherent prevention aim of the AALISABETH approach: with this respect, it addresses needs of a much ampler class of users than just elderly people: market perspective are hence consequently wider. More generally speaking, main project innovation comes from the behavioural inference capability, mostly grounded on already available and stable base technologies. I.e., added value will mostly come from big data analysis techniques; by accounting for suitable abstraction of physical data, such techniques will not be tightly bounded to the actual technological base but will have a more general scope, allowing for future extension toward different scenarios, preserving their value well beyond the rapid obsolescence of supporting hardware technologies. References 1. A. Losardo, F. Grossi, G. Matrella, V. Bianchi, A. Ricci, I. De Munari, P. Ciampolini (2010). Remote control and monitoring of the home environment. GERONTECHNOLOGY, vol. 9, No. 2, ISSN: 1569-1101, doi: 10.4017/gt.2010.09.02.238.00 2. V. Bianchi, F. Grossi, I. De Munari, P. Ciampolini (2012). Multi Sensor Assistant: A Multisensor Wearable Device for Ambient Assisted Living. JOURNAL OF MEDICAL IMAGING AND HEALTH INFORMATICS, vol. 2, Number 1, p. 70-75, ISSN: 2156-7018, doi: 10.1166/jmihi.2012.1058 3. A. Ricci, E. Smargiassi, D. Mancini, I. De Munari, V. Aisa, P. Ciampolini (2011). Wr@p: the last meter technology for Energy-aware networked smart appliances. In: -. Proceedings of the 2011 IEEE International Symposium on Power Line Communications and its Applications
AALISABETH: intelligenza ambientale per la prevenzione 5 (ISPLC). Udine, Italy, 3-6 aprile, p. 193-198, ISBN: 9781424477494 4. A. Losardo, G. Matrella, F. Grossi, I. De Munari, P. Ciampolini (2012). Indirect wellness monitoring through AAL environments. GERONTECHNOLOGY, vol. 11:2, ISSN: 1569-1101, doi: 10.4017/gt.2012.11.02.492.00 5. EU-FP7-FET (Future and Emerging Technologies) programme, TOPDRIM: Topology-Driven Methods for Multi-level Complex Systems, number FP7-ICT-318121-2012-2015. 6. E. Merelli, M. Rasetti: The Immune System as a Metaphor for Topology Driven Patterns Formation in Complex Systems. ICARIS 2012: 289-291 7. Flavio Corradini, Emanuela Merelli, Diletta Romana Cacciagrano, Rosario Culmone, Luca Tesei, Leonardo Vito: ACTIVAge: proactive and Self-Adaptive Social Sensor Network for Ageing People. ERCIM News 2011(87): (2011)