System and method for managing the healthcare processes

Abstract

The present invention relates to a system and a method for managing the 'healthcare processes' capable of improving the 'adherence' of a patient to one or more steps of a 'healthcare process'. The system comprises a portable device whose position, path and 'contacts' and relative times inside and outside a 'healthcare facility' are tracked. The tracking within the 'healthcare facility' is also operated by means of the communication with fixed devices. The system uses 'log files' relative to 'contacts' between patients and / or 'caregiver' and / or 'healthcare workers' and / or 'healthcare facilities', comprises data analysis algorithms, and 'predictors' which are computer variables relative to data on the actors of the system. An AI forecasts, promotes, measures, evaluates and optimizes the 'adherence' of a patient, in a continuous updating process, based on the 'predictors', of the processing of the algorithms, of the tracking data, of the 'log files' and of the 'contacts'.

Description

[0001] LEIBO. / 67e2025

[0002] “System and method for managing the healthcare processes”

[0003] Description

[0004] Field of the invention

[0005] The present invention relates to the management of the ‘healthcare processes’ (see Definitions section) and, in particular, to the management of the ‘adherence’ (see Definitions section) of the patient to the steps and times of the ‘healthcare processes’, i.e. the level of actual coincidence between the individual behavior of the patient and the prescriptions received from the ‘healthcare workers’ (see Definitions section). The present invention relates to the system and to the method for managing aforesaid ‘healthcare processes’ in a manner such to identify the actual steps of the ‘healthcare process’ and allow the verification of the ‘adherence’ thereof with the pre-established steps.

[0006] Prior art

[0007] In the healthcare field, research historically focused its attention on ‘adherence’ in the context of the pharmacological treatment, i.e. of a ‘patient governing procedure’ (see ‘Healthcare procedure’, Definitions section).

[0008] The ‘adherence’ of the patient to the pharmacological therapy in fact represents one of the most important problems of clinical practice, since the success of any ‘therapeutic process’ (see ‘Healthcare process’, Definitions section) depends on the actual adhesion of the patient to the therapy and that only an accurate measurement and evaluation thereof can actually allow ascribing the outcomes observed during the prescribed treatment.

[0009] Typically there are two measurement approaches used: one is generally objective, the other is more subjective.

[0010] Within the objective approach, one solution provides for counting, during the control LEIBO. / 67e2025 visits, the doses of drugs that remained in the packages dispensed to the patient, but even the possible accuracy of the count still neglects the guantities and times of the doses taken or forgotten. Additionally, the electronic monitoring system that records date and time at which the package of the drug was opened, in addition to the high cost that limits the use thereof to the clinical research phase, if it can provide some description of the modes of therapy execution, does not indicate anything regarding the doses that are actually taken. Likewise, the control of the times and of the packages purchased by the patients through the use of the databases of the pharmacies is not necessarily indicative of the level of ‘adherence’, given that buying drugs does not signify taking them. Finally, the detection of several biochemical parameters, in the blood or urine, as test of having taken the drug in guestion, to which non-toxic biological markers have been added, does not lack drawbacks; indeed, moreover, its organizational complexity limits the use to only the research phase, though the data collected can at any rate be misleading, since they are affected by a wide variety of individual factors, such as diet, absorption of the drug and relative excretion.

[0011] Within the objective approach, an example of a patent that involves providing a solution to the problem of ‘adherence’ of the patients to a therapy is the object of the patent US9147163B1 by R.F. NEASE, D.A. TOMALA, S. FOLLIS and T.L. BAUMAN. The invention describes a system and a method for improving the ‘adherence’ to the therapeutic programs in particular to pharmacological therapies and / or healthcare therapies. The methods presented can also be used for improving the compliance and / or the ‘adherence’ to other programs of wellbeing and / or healthcare assistance. The methods and the systems of the invention can include methods for identifying the patients who are at risk of non-‘adherence’, non-conformity or probability that a therapeutic program will terminate, predicting a base for such non-conformity and aiming for direct interventions with the patients who have been identified as probably not compliant. The invention by R.F. Nease et al. Provides for determining LEIBO. / 67e2025 the ‘adherence’ of a patient to a medical / pharmacological therapy on the basis of two elements: the MPR (Medication Possess Ratio), i.e. a ratio that indicates the stockpile of drugs available for the patient with respect to those which the patient should have by following a therapy, and the so-called “Claims Data”, i.e. clinical data of the patients. The problems tied to the determination of the MPR are obvious, since over the years a series of devices have been implemented for determining if a patient takes specific drugs; in any case their utility is limited, the collected data might not be reliable and their use for all the patients is often complex. Additionally, in the event in which there is now “Claims Data” regarding a patient, the described invention is not very effective, not employing ‘machine learning’ models (see section ‘Definitions’), Artificial Intelligence nor neural networks, and hence the evaluation of the probability of non-‘adherence’ being limited to direct comparisons or simple statistical methods.

[0012] The approach of subjective type comprises the direct reguest to the patients regarding the times, guantities and modes of taking the drugs, the direct reguest to ‘healthcare workers’ (see Definitions section) and patients of their personal evaluation regarding that which they deem to be a good ‘adherence’ to the treatments, the use of standardized self-administrable ‘guestionnaires’ (see Definitions section) or even the analysis of the general characteristics of the patient or of several aspects of his / her personality. Nevertheless, the ‘healthcare workers’ as with the patients tend to overestimate the level of ‘adherence’, only several ‘guestionnaires’ which evaluate specific behaviors correlated to specific medical indications have shown to provide valid prediction elements for ‘adherence’ to the treatments and said prediction elements were not up to now correlated with the identification of any stable factor or specific personality trait.

[0013] In conclusion, presently there is no gold-standard for measuring the ‘adherence’ to the pharmacological treatment nor is there a gold-standard for defining, based on evidence data, a “good” or “bad” ‘adherence’. At the same time, the attention placed LEIBO. / 67e2025 by research mainly on the factors correlated to the patients has in fact negatively affected the possibility of identifying a gold-standard for the process or method, valid for all patients, all situations and all conditions, capable of improving the level of ‘adherence’.

[0014] On the other hand, many other patient behaviors have proven to be significant, including personal hygiene, smoking, alcohol, contraception, sexual behaviors, diet and physical activity, considered by all to rightly be “therapeutic behaviors” and with reference to said “therapeutic behaviors” it is clear that the above-observed problems of measurement, evaluation and forecast of the level of ‘adherence’ to the pharmacological treatments become even more evident.

[0015] The necessary extension of the concept of ‘adherence’ also to the steps of prophylaxis and of diagnosis, given the partiality of a definition limited to only the therapeutic phase, further enlarges the number of behaviors and ‘procedures’ (see “Definitions” section) to be measured, evaluated and predicted such as for example the response to the vaccination campaigns and to the screening programs, as well as the times and modes in reguesting the intervention of a ‘healthcare worker’, executing the prescriptions in terms of visits and diagnostic investigations and respecting the relative appointments. In this context, the problems of ‘adherence’ are generally expressed in the lack of reguest of intervention of the same ‘healthcare worker’ or in the lack of attendance of the relative appointment, if reserved. But it is only in the latter case, when the patient - in this case defined ‘no-show’ - is absent without notice, is significantly late or cancels the appointment with a delay such to not allow the reassignment of the previously allocated time slot, that the level of ‘adherence’ affects the efficiency of the ‘healthcare facility’. It is natural therefore that the ‘healthcare facilities’ have focused investments and research in the development of systems and solutions adapted to minimize said problems, i.e. to ensure the efficiency in the management of all the phases of the ‘processes’ connected with a healthcare service; it is also natural that said solutions in this LEIBO. / 67e2025 manner limit their objective to the reduction of the no-show, only one of the possible expressions of the non-‘adherence’ of the steps of prophylaxis and diagnosis. Typically, the aforesaid solutions - in connection with a wide range of resource allocation and reservation systems - use, also in combination, various methods for planning and / or reminding; limited by an approach that mainly accounts for the drivers tied to the needs of the same ‘structures’ and the relative ‘operators’ (also because more easily known) and, only marginally, to that of the patients, said solutions do not resolve the problem of the no-show nor do they significantly reduce it. The planning methods, attainable by means of free access systems or reservation systems with use of overbooking, indeed have limits due to a high variability in completing the ‘procedure’ or in the possibility to reserve. In particular, the programming of a free access ‘procedure’ eliminates the problem of the no-show but shifts the possible inefficiency to the times in the waiting room and / or to the actual capacity to carry out the same ‘procedure’. On the contrary, the overbooking aims to reduce the no-show per se, since only the relative effects of inefficiency on the ‘structure’ and the relative affected ‘operators’, also in this case shifting the problem to the waiting room times and / or to the actual capacity to carry out the ‘procedure’ in case of presentation of the reserved patients to a greater extent than the no-show forecast. The reminder methods, on the other hand, also attainable by means of automated systems capable of carrying out calls or sending messages to remind the patient of the reserved appointment, are unable to ensure a high rate of reduction of the no-shows, since they act on only one determining factor, i.e. that the patient could forget or have forgotten the appointment, while there are in fact various determining factors, relative to a plurality of causes, for example personal, financial, logistic, economic, emotional and qualitative. For example, the sentiment of the ‘healthcare workers’ is often neglected, which strongly affects the ‘adherence’ of a patient. Preferably, it would be necessary to analyze how the ‘healthcare workers’ interact with specific patients, if there is a certain affinity between them (also of LEIBO. / 67e2025 character type) and if, and how, this changes over time. The effect of the sentiment of a ‘healthcare worker’ on a patient is widely discussed in various documents of the prior art such as, by way of a non-limiting or non-binding example, by Pensieri C. et al [Pensieri, C., et al. Doctor-patient communication tricks. Oncological study at Campus Bio-Medico University of Rome. La Clinica Terapeutica 169.5 (2018): e224- e204].

[0016] In conclusion, contrary to the evidence, in the use of solutions for measuring, evaluating and predicting the ‘adherence’, there remains the tendency to use ineffective instruments, with a static and single-dimensional approach normally limited to single ‘procedures’ and steps of the ‘healthcare path’ (see Definitions section); in the identification of the problems relative to the ‘adherence’, to be focused on the factors correlated to the patient with respect to those which instead more greatly and more freguently influence him / her, by way of a non-limiting example linked to the characteristics of the disease and of the therapy and to the representation that the patient has of these as we well as to the following factors: environmental setting, social-economic, instrumental, technological, organizational and professional of the ‘healthcare facilities’ (see Definitions section) involved and their relation with the characteristics of the patient.

[0017] It is desirable to arrange a system and a method for managing the ‘healthcare processes’ capable of minimizing the above-described drawbacks. In particular, it would be desirable to arrange a system and a method for managing the ‘healthcare processes’ that is capable of ensuring an optimal ‘management of the adherence’ (see Definitions section) of the actual ‘healthcare process’ phases with respect to the programmed ‘healthcare process’ phases.

[0018] Definitions

[0019] In the body of the text, reference is made to elements, phenomena and characteristics for which a definition within the scope of the invention is provided LEIBO. / 67e2025 hereinbelow. Each reference to elements, phenomena and characteristics defined hereinbelow is reported in the present description between single quote marks (e.g. ‘adherence’).

[0020] - ‘Adherence’: a parameter / variable, a plurality of parameters / variables, or even a parameter resulting from a combination of values of a plurality of parameters / variables, which, also on the basis of the dynamic ‘adherence’ model with 6 phases proposed by Gearing et al. [Gearing, R. E., Townsend, L., MacKenzie, M., & Charach, A. (2011). Reconceptualizing medication adherence: Six phases of dynamic adherence. Harvard Review of Psychiatry, 19(4), 177-189], indicates the level or the different levels with which a patient observes good practices in the response behavior to symptoms, to a state of discomfort and / or pain and in the decision to turn to a ‘healthcare worker’ and / or to a ‘healthcare facility’, correctly following the recommendations received therefrom, typically at the start and within a ‘healthcare path’ and with respect to the relative ‘healthcare processes’ and ‘healthcare procedures’.

[0021] - ‘Management of the adherence’ or ‘Managing the adherence’: the entirety of the actions, structured in macro-steps (Forecast: predict, forecast, plan; Promotion: educate, promote; Measurement: monitor, measure, verify; Evaluation: evaluate, analyze; Optimization: intervene, correct, adapt), and object of continuous updating, adapted to ensure, through a cyclic management of ‘contacts’, the improvement of the ‘adherence’ of the patients and, consequently, of the overall effectiveness of the healthcare treatment.

[0022] - ‘Healthcare worker’ or ‘Healthcare workers’: a subject (i.e. physician, pharmacist, dentist, psychologist, nurse, obstetrician, dental hygienist, dietician, physiotherapist, speech therapist) who performs activities of prevention, diagnosis, assistance, treatment and / or rehabilitation, typically within a ‘healthcare facility’ and / or at the home of a patient.

[0023] - ‘Healthcare facility’ or ‘Healthcare facilities’: the specific seat (o plurality of seats) LEIBO. / 67e2025 and / or the organization (intended as the set of the subjects who are a part of this, whether ‘healthcare workers’ or otherwise) which (i) delivers basic assistance services and / or specialistic services and / or surgery and / or instrumental diagnostics and / or laboratory services and / or rehabilitation services and / or socio-healthcare clinical services and / or hospitalization services and / or emergency services and / or residential services, with continuous or daily cycle, (i.e. general physician office, poly-specialistic hospital, emergency room, analysis laboratory), (ii) prepares and dispenses pharmaceutical products, with or with prescription, and / or healthcare devices and / or healthcare aids (i.e. pharmacy, drugstore, healthcare facility), (iii) provide first aid on the territory and provides pre-hospital emergency care, stabilization for grave diseases and traumas and transport services to the ‘healthcare facilities’ (i.e. emergency room, ambulance service).

[0024] - ‘Caregiver’: a subject who, as a professional and / or family member and / or friend and / or for solidarity, conducts treatment, assistance and support, continuously or occasionally, to a patient, even one who is self-sufficient.

[0025] - ‘Guidelines’: the entirety of the clinical behavior recommendations produced, through a systematic process, for the purpose of assisting ‘healthcare workers’ and patients in the decisions on the most appropriate assistance modes in specific clinical circumstances.

[0026] - ‘Healthcare path’: the instrument aimed for implementing the ‘guidelines’ as result of the integration of the clinical recommendations of the reference ‘guidelines’ and the local context elements capable of conditioning the application thereof. It can be separated into a ‘healthcare path’ for prevention or prophylaxis, and assistance or treatment path.

[0027] - ‘Healthcare process’ or ‘healthcare processes’: the elementary unit of the ‘healthcare path’, i.e. the set of the activities that create value for the patient, transforming the input resources of the ‘healthcare process’ into a final output LEIBO. / 67e2025 product of the ‘healthcare process’ (i.e. diagnostic process, therapeutic process, rehabilitation process).

[0028] - ‘Healthcare procedures’ or ‘Healthcare procedures’ or ‘Procedure’ or ‘Procedures’: the elementary unit of the ‘healthcare process’, constituted by a variable number of ‘procedures’, whose governing and execution modes, typically (except in the cases of incapacity and / or life-threatening cases) in accordance with the patient, determine the following types: a) ‘healthcare governing procedures’, in which the ‘procedure’ is governed and / or executed by ‘healthcare workers’ on the patient (i.e. visit session, diagnostic exam, blood test, surgical procedure, physiotherapy session, functional rehabilitation etc.); said procedure is, in all respects, a ‘contact’; b) ‘patient self-governing procedures’, in which the ‘procedure’ is governed and / or executed by the patient on himself / herself i.e. self-managed by the patient, upon indication of ‘healthcare workers’ but without their presence and / or their direct intervention, typically while at home (i.e. taking of drugs, following a diet, performing physical activities).

[0029] - ‘Appropriateness’ (of a ‘procedure’): the indication of the correct use, based on the evidence (‘guidelines’) and / or clinical experience and / or good practices, of effective ‘healthcare procedures’ in a patient who can actually benefit therefrom due to his / her clinical conditions, due to his / her preferences (e.g. not limiting of place and time) and due to his / her expectations.

[0030] - ‘Log files’: a digital / register file that documents, in a structured and chronological manner, operations carried out by a computer system (server, storage, client, applications or any other computerized device or program) or even operations where a patient and / or a ‘caregiver’ and / or a ‘healthcare worker’ and / or a ‘healthcare facility’ is involved. In addition, by ‘log files’ it is also intended a register that documents the ‘contacts’ in a structured and chronological manner. Includes information of the available attributes of a ‘contact framework’. LEIBO. / 67e2025

[0031] - ‘Contact’ or ‘Contacts’: the act or the result of the interaction, between at least two ‘participants’, which can occur both in person (face-to-face or in presence) and through other means (paper, electronic or digital). In both contexts, one of the ‘participants’ can be an automated system, Artificial Intelligence or a robot, which enlarges the concept of ‘contact’ in order to also include interactions with nonhuman entities.

[0032] - ‘Contact framework’ or ‘Contact frameworks’: the set, partial or total, of the attributes which characterize each ‘contact’. Said attributes are: o ‘Macro-step’ or ‘Macro-steps’ (types and actions of the ‘Management of the adherence’ cycle: Forecast, i.e. predict, forecast, plan; Promotion, i.e. educate, promote; Measurement, i.e. monitor, measure, verify; Evaluation, i.e. evaluate, analyze; Optimization, i.e. intervene, correct, adapt); o ‘Participant’ or ‘Participants’ (types and subjects and / or entities involved: human, i.e. patient, ‘caregiver’, ‘healthcare worker’ and otherwise; non-human, i.e. automated system, Artificial Intelligence, robots, and others); o ‘Objective’ or ‘Objectives’ (desired types and aims and / or results, specific and measurable: activities pertaining to a ‘healthcare governing procedure’, i.e. visit reservation, visit memo, clinical verification, check-up visit, therapy re- evaluation; activities pertaining to a ‘patient self-governing procedures’, i.e. patient state verification, ‘adherence’ verification, ‘guestionnaire’ administration); o ‘Rule’ or ‘Rules’ (specific types and norms: internal, i.e. company policies, standard operating procedures; external, i.e. sector standards, legislative regulations, privacy norms; relations between attributes of the ‘contact’, i.e. ‘participants’-‘objective’-‘channel’; interaction with other ‘contacts’, i.e. followup within 24 hours of initial ‘contact’); o ‘Communication’ or ‘Communications’ (types and contents of the interaction: factual information, i.e. data, facts, exchanged objective information; opinions LEIBO. / 67e2025 and viewpoints, i.e. beliefs, expressed evaluations; personal emotions and experiences, i.e. shared concerns, joys; social and courteous exchanges, i.e. salutations, expressions of thanks, asking forgiveness; exchanges of guestions and responses, i.e. guestions of information, clarification, opinion, evaluation, reflection, survey, evaluation scales, tests, interviews, open or closed guestions); o ‘Communication modes’ (types and forms; verbal, i.e. spoken, written; extra verbal, non-verbal, i.e. body language, gestures, postures, facial expressions, physical distance between ‘participants’, visual ‘contact’, and para-verbal, i.e. voice tone, hesitations; hybrid, like a combination of the different types and forms); o ‘Channel’ or ‘Channels’ (type and medium; in person, i.e. face-to-face; paper, i.e. conventional mailing; electronic; i.e. fax; digital, i.e. telephone, sms, e-mail, instant messaging, social media, app, videoconference, virtual reality and augmented reality); o ‘Relationship context’ or ‘Relationship contexts’ (types and levels between ‘participants’: personal, i.e. informal, formal, nonexistent, consolidated; social, i.e. low, high; cultural, i.e. acceptable, incompatible); o ‘Dynamics’ (type and interactions between ‘participants’: cooperation, i.e. collaboration, mutual support; conflict, i.e. disagreements, tensions, contrasts, differences of opinions, contrasting interests; negotiation, i.e. seeking an agreement or compromise; empathy, i.e. active listening, comprehension, emotional support between ‘participants’; socialization, i.e. informal conversation, expression of common interests; learning and teaching, i.e. knowledge exchange, competencies, abilities); o ‘Intensity’ (types and level of involvement of the ‘participants’: physical, i.e. low, medium, high, variable; emotional, i.e. low, medium, high, variable; cognitive, i.e. low, medium, high, variable); LEIBO. / 67e2025 o ‘Duration’ or ‘Durations’ (types and level of extension over time: ‘contact’ general information, by way of a non-limiting or non-binding example a duration is brief when it is on the order of 30s, 1m; by way of a non-limiting or non -binding example, in the case of a telephone contact, a specific duration for ‘contact’ phase can be measured as: 2 rings, 10s, 2m and / or in another way); o ‘Location context’ or ‘Location contexts’ or ‘Environmental contexts’ (types and places, also virtual: for appointment for hospitalization, i.e. hospital, clinic; for appointment for clinical procedure, i.e. cardiological clinic, Rome clinic; for ‘contact’ for health state verification, i.e. at the home of the patient, not in movement, non-noisy environment, non-crowded environment); o ‘Time context’ or ‘Time contexts’ (types and time slots: for appointment, i.e. today, in one hour, next Thursday at 17:00; for ‘contact’, i.e. in 5 minutes, between 9:30 and 10:00).

[0033] - ‘Questionnaire’ or ‘Questionnaires’: a structured set of guestions, possibly including possible responses, and / or one or more linguistic statements that can be the object of assertion, doubt, interrogation, desire, adapted to collect data and / or reguest information and / or actions and / or to measure the size or consistency of an attitude or of an individual capacity.

[0034] - ‘Predictor’ or ‘Predictors’: variables or indicators used for making forecasts or estimates on a future result or on the verification of an event.

[0035] - ‘Characterization’ (of a patient, ‘caregiver’ and / or ‘healthcare worker’) or ‘Characterizations’: one or more of the dimensions of the personality and of the behavior between ‘Personality states’ and / or ‘Profiles’ and / or ‘States of change’: o ‘Personality state’ or ‘Personality states’: definition based on the Whole Trait

[0036] Theory [Fleeson, W., & Jayawickreme, E. (2015). Whole trait theory. Journal of research in personality, 56, 82-92], as an alternative or supplement of the model of the conventional personality traits [Goldberg, L. R. (1990). An alternative “description of personality”: The big-five factor structure. Journal LEIBO. / 67e2025 of Personality and Social Psychology, 59 (6), 1216-1229] and / or of the model “six-factor HEXACO model” [Feher, A., & Vernon, P. A. (2021). Looking beyond the Big Five: A selective review of alternatives to the Big Five model of personality. Personality and Individual Differences, 169, 110002] which tend to evaluate the traits such as the separate, distinct personality dimensions that are often associated with a certain stability over time. The ‘personality states’, instead, are not considered isolated but part of an interconnected set of behaviors, thoughts and sentiments. Within the domain of personality traits, the ‘personality states’ consider the variability in the behavior of an individual and imply that each individual can have, at different times and situations, a distribution of levels of a given personality trait. The measurement and the evaluation of the traits and of the ‘personality states’ is carried out through the use of psychometric instruments, such as ‘guestionnaires’, tests or interviews. o ‘Profile’ or ‘Profiles’: the set and the integration of the different dimensions of the behavior and of the personality of an individual analyzed, by way of a nonlimiting or non-binding example, through the different methods of the behavioral analysis, psycho-behavioral analysis, cognitive-behavioral and psychographic analysis. As a function of the data available and of the different methods of analysis conducted on the data, the ‘profile’ can be, by way of a nonlimiting or non-binding example, of behavioral, psycho-behavioral, cognitive- behavioral, psychographic and / or integrated type, as result of the integration of one or more of the same ‘profiles’.

[0037] The analysis behavioral is focused on the actions observable and measurable of an individual, registered and analyzed objectively, beforehand eliminating - since they are deemed unknowable and to a certain extent irrelevant - the internal mental processes (i.e. the thoughts, cognition, emotions and motivations). According to the “behaviorist” [Watson John B. (1913), Psychology as the Behaviorist Views It. Psychological Review, 20, 158-177], LEIBO. / 67e2025 indeed, the behavior of an individual is the resultant of an environmental stimulus, immediately discernable. This approach has led to the development of behavior modification techniques based on the “conditioning” [Pavlov Ivan (1903); Skinner B. (1938), Behaviour of Organisms. New York, Macmillan]. The use of instruments oriented on the direct observation of the behavior and the data collection on the actions and the reactions of the individual include direct observations in specific contexts, forms of monitoring the daily activities and behavioral diaries, the latter introducing however a level of subjectivity that is only partially mitigatable through strategies such as the use of cross validation procedures for comparing the recordings with other data sources when possible.

[0038] The psycho-behavioral analysis has, with the theory of social learning, an essential point of reference [Bandura, A. (1986). Social Foundations of Thought and Action: A Social Cognitive Theory. Englewood Cliffs, N.J.: Prentice-Hall.; (1997). Self-efficacy: the exercise of control. New York: W.H. Freeman.]. This theory is concentrated on the importance of observation and imitation of the behaviors of others, on the influence of reinforcement and of punishment in behavior modeling, and on the importance of the self-effectiveness in determining the capacity of an individual in confronting new challenges and tasks.

[0039] The cognitive-behavioral analysis seeks to comprise how the cognitive processes influence the behavior of an individual, for example how the perceptions influence his / her response to specific stimuli, or how his / her decision processes are influenced by his / her beliefs or expectations.

[0040] Aaron Beck is considered the father of cognitive-behavioral analysis and therapy (TCC) and has developed theoretical therapeutic models and practices based on the idea that the distorted negative thoughts and beliefs influence the behavior and the emotions of an individual. The TCC is concentrated on the LEIBO. / 67e2025 recognition and modification of the dysfunctional thoughts in order to promote behavioral change and emotional wellbeing.

[0041] The psychographic analysis, with the work by Ernest Dichter, has profoundly influenced the development of market research and psychological marketing. Dichter was one of the first to apply psychological principles to advertising and marketing, exploring the unconscious motivations of consumers and the influence of the emotions and desires on the purchase behavior.

[0042] The measurement and the evaluation of the integrated ‘profile’ uses the integration of various information sources, like observed behavior, cognitive styles, preferences and values, involving various instruments including psychological tests, interviews, behavioral and psychographic evaluations. o ‘State of change’ or ‘States of change’: seguential steps of the behavioral change through which individuals progress while they seek to modify an undesired behavior, described by the transtheoretical model by Prochaska and Di Clemente [Prochaska, J. 0., Di Clemente, C. C., & Norcross, J. C. (1997). In search of how people change: applications to addictive behaviors].

[0043] - ‘Machine learning’: the approach that uses statistical methods in order to improve the performance of an algorithm in the identification of data pattern. of the invention

[0044] Object of the present invention is to provide a system and a method for managing the ‘healthcare processes’ capable of forecasting, promoting, measuring, evaluating, optimizing and, thus, with a transverse updating of said actions, ‘managing the adherence’ of a patient to one or more steps of a ‘healthcare process’. Where by “transverse updating” it is intended the fact that the updating of the ‘adherence’ occurs in all the abovementioned steps of forecast, promotion, measurement, evaluation and optimization.

[0045] The previously mentioned objects are reached by a system and by a method for LEIBO. / 67e2025 managing the ‘healthcare processes’ in accordance with the enclosed claims.

[0046] The ‘healthcare process’ management system, with at least a ‘healthcare worker’ and at least a patient, is adapted for ‘managing the adherence’ of said patient to one or more ‘healthcare processes’ and comprises:

[0047] - at least a computerized apparatus suitable of hosting one or more databases, one or more Al and one or more agendas of the ‘procedures’; said computerized apparatus being managed and accessible with a back-end structure and a frontend application; said Al being trained on the basis of historical data present on said databases and on the basis of data present in scientific literature; the system also comprises:

[0048] - at least a portable device associated with said patient and / or with a ‘caregiver’ and / or with said ‘healthcare worker’ in order to electronically track their position, the path and the ‘contacts’ inside and outside a ‘healthcare facility’ and the relative times; where, with regard to the face-to-face ‘contacts’, the tracking occurs through the mutual interaction between portable devices with wireless technology, e.g. BLE, UWB, Wi-Fi, with the support of the geolocation unit, e.g. GPS, comprised in the portable devices. More correctly, the use of devices with wireless technology, even without the need for additional infrastructures such as beacons or access points, allows detecting the presence and the vicinity of other similar devices; monitoring the duration and the freguency of the vicinity between devices, it is possible to obtain interaction data relative to inferences on the verification of a face-to-face ‘contact’ and, therefore, of an interaction or a conversation of said patient, ‘caregiver’ and / or said ‘healthcare worker’, above all if the vicinity is at a distance of less than 2 meters and the duration suitable (e.g. at least a minute). In the measurement of the vicinity, the technology BLE offers a discrete accuracy at brief distance, which can be further improved with calibration algorithms, while that UWB excels in precision, render it ideal for safely detecting when two devices are found within the typical distance of a conversation LEIBO. / 67e2025

[0049] (about 1-2 meters). These inferences, when associated with agendas of the ‘procedures’ and / or a ‘log file’ of ‘contacts’ of said patient, ‘caregiver’, ‘healthcare worker’ and / or ‘healthcare facility’, and preferably supported by geolocation data, acguire significant truth and are considered objective data;

[0050] - in several embodiments of the present invention, the system further comprises a plurality of fixed devices positioned in said ‘healthcare facility’ and adapted to detect the passage and the stay time, within one or more areas, of the portable devices with wireless technology, and to communicate said passage and said stay time to said computerized apparatus; the interaction between said portable devices and said fixed devices suitable of providing data indicating the position, the path and the stay time of said portable devices in the areas monitored by the fixed devices; the mutual interaction between said portable devices with wireless technology and said fixed devices suitable of providing data indicating said face- to-face ‘contacts’ within said ‘healthcare facility’;

[0051] - ‘log files’ relative to face-to-face ‘contacts’ and / or paper and / or electronic and / or digital ‘contacts’ of said patient and / or a ‘caregiver’ thereof and / or a ‘healthcare worker’ and / or a ‘healthcare facility’, and hence also relative to ‘contacts’ among the abovementioned ‘participants’;

[0052] - at least a first algorithm for analysis of the geolocation data, of the data generated by the mutual interaction between portable devices, of the data generated by the interaction, mutual or otherwise, between portable and fixed devices (when the latter devices are present) and data extracted from said ‘log files’;

[0053] - ‘predictors’ contained in said database; said ‘predictors’ being computer variables indicative of data on a patient and / or a ‘caregiver’ thereof and / or a ‘healthcare worker’ and / or a ‘healthcare facility’; said patient, ‘caregiver’ and / or ‘healthcare worker’ being associated with one or more ‘personality states’ and / or ‘profiles’ and / or ‘states of change’ based on the ‘predictors’, whose values depend on objective data. Said objective data are the LEIBO. / 67e2025 geolocation data generated by geolocation units comprised in the portable devices, said mutual interaction data between portable devices, said interaction data, mutual or otherwise, between portable and fixed devices (when the latter devices are present), said data extracted from said ‘log files’, the processing of said first algorithm and other data present in database; the ‘management of the adherence’ of said patient being conducted by said Al based on the ‘predictors’ and on the abovementioned data.

[0054] The method for managing the ‘healthcare processes’ according to the present invention comprises the following steps:

[0055] - one or more steps of acguisition and analysis of objective data and / or selfreported data (as a non-limiting example, the data relative to electronic tracking and to ‘contacts’); said objective and / or self-referred data being analyzed, by algorithms and / or Al, in order to automatically attribute to a patient, a ‘caregiver’ and / or a ‘healthcare worker’ and / or a ‘healthcare facility’ one or more attributes of a ‘contact framework’ and to automatically attribute ‘personality states’ and / or ‘profiles’ and / or ‘states of change’ for said patient, said ‘caregiver’ and / or said ‘healthcare worker’;

[0056] - one or more steps of selection of a ‘contact framework’, by means of ‘machine learning’, in which ‘macro-steps’ and / or ‘participants’ and / or ‘objectives’ and / or ‘rules’ and / or ‘communications’ and / or ‘communication modes’ and / or ‘channels’ and / or ‘relationship contexts’ and / or ‘dynamics’ and / or ‘intensities’ and / or ‘durations’ and / or ‘location contexts’ and / or ‘time contexts’ (e.g. ‘contact’ time slots) adapted for the ‘contact’ to be activated with the patient, the ‘caregiver’ and / or the ‘healthcare worker’ and / or the ‘healthcare facility’, are selected from said databases based on the objective and / or self-referred data and / or on the values of the ‘predictors’ and / or of the types of ‘personality states’ and / or ‘profiles’ and / or ‘states of change’, attributed, by means of ‘machine learning’, to said patient, ‘caregiver’ and / or ‘healthcare worker’ on the basis of said objective LEIBO. / 67e2025 data and / or self-referred and / or of said values of the ‘predictors’ present in said database; said selection step adapted to forecast a first identification of a ‘contact framework’ by means of statistical data and a then forecast said identification on the basis of the objective and / or self-referred data acquired for said patient, ‘caregiver’ and / or ‘healthcare worker’ and / or ‘healthcare facility’ in the ‘contacts’ carried out with said patient, ‘caregiver’ and / or ‘healthcare worker’ and / or ‘healthcare facility’;

[0057] - one or more adequacy prediction steps of the selected ‘contact framework’, in which a probability of adequacy is calculated by employing said objective and / or self-referred data and / or said ‘predictors’ and / or said ‘personality states’ and / or ‘profiles’ and / or ‘states of change’; said adequacy prediction steps of the ‘contact framework’ being attained by employing said Artificial Intelligence (Al) engine; said probability of adequacy being assigned to each attribute of the ‘contact framework’;

[0058] - one or more steps of adequacy verification of the ‘contact framework’, by means of ‘machine learning’; said steps of adequacy verification provided for the confirmation of said probability of adequacy and / or their updating for each attribute of the ‘contact framework’;

[0059] - one or more steps of verifying the “outcome” of the ‘contact’ and of the “actions” following the “outcome”. Where by “outcome” it is intended the evaluation and measurement of the correspondence with respect to pre-established ‘objectives’ (see ‘Contact framework’, Definitions section), i.e. positive outcome, negative outcome, outcome which is assigned a certain score. Where by “actions” it is intended activities subsequent to the “outcome”, i.e. no further action, definition of a new therapy, follow-up, programming of a new ‘contact’. Said steps of verifying “outcome” and “actions” are attained on the basis of objective and selfreferred data and by employing said Artificial Intelligence (Al) engine and / or algorithms of the system. The “outcome” is, by way of a non-limiting or non- LEIBO. / 67e2025 binding example, determined by patient, ‘caregiver’, ‘healthcare worker’, ‘healthcare facility’ and / or Al. In several embodiments of the invention, indeed, the Al determines the “outcome” on the basis of the evaluations “outcome” of patient patient, ‘caregiver’, ‘healthcare worker’ and / or ‘healthcare facility’;

[0060] - one or more steps of outcome forecast of an identified ‘procedure’, in which a probability of positive or negative outcome of the identified ‘procedure’ is calculated by employing said ‘predictors’; said probability of positive or negative outcome is a function of the ‘adherence’ of the patient to the ‘procedure’ and of the ‘appropriateness’ of the ‘procedure’; said steps of outcome forecast being attained by employing said Artificial Intelligence (Al) engine;

[0061] - one or more steps of outcome verification of the identified ‘procedure’, carried out and / or being completed, by means of ‘machine learning’; said verification steps provide for the assignment of a cumulative score or of distinct scores to urgency indicators and ‘appropriateness’ of said ‘procedure’ identified for said patient; said verification steps provide for the assignment of one or more values of ‘adherence’ of the patient to the ‘procedure’ on the basis of the ‘predictors’, of the objective data and of the self-reported data and of the ‘personality states’, ‘profiles’ and ‘states of change’;

[0062] - preferably the method comprises one or more further steps of verification of the adeguacy of the ‘contact framework’;

[0063] - one or more steps of updating of said database with types of ‘personality states’, types of ‘profiles’, types of ‘states of change’ and / or ‘macro-steps’ and / or ‘participants’ (comprising for example patients, ‘caregiver’, ‘healthcare workers’) and / or ‘objectives’ and / or ‘rules’ and / or ‘communications’ and / or ‘communication modes’ and / or ‘channels’ and / or ‘relationship contexts’ and / or ‘dynamics’ and / or ‘intensities’ and / or ‘durations’ and / or ‘location contexts’ and / or ‘time contexts’ (i.e. communication time slots and / or appointment time slots) pairable with said patient, ‘caregiver’ and / or ‘healthcare worker’ as a LEIBO. / 67e2025 function of the ‘predictors’ and of the acquired objective and self-referred data, in which each type of ‘personality states’, type of ‘profiles’, type of ‘states of change’ and / or ‘macro-steps’ and / or ‘participants’ and / or ‘objectives’ and / or ‘rules’ and / or ‘communications’ and / or ‘communication modes’ and / or ‘channels’ and / or ‘relationship contexts’ and / or ‘dynamics’ and / or ‘intensities’ and / or ‘durations’ and / or ’location contexts’ and / or ‘time contexts’ (i.e. ‘contact’ time slot and / or appointment time slot) is assigned with a probability of suitability to said patient, ‘caregiver’ and / or ‘healthcare worker’; types and values of the ‘contact framework’ can be paired also with ‘healthcare facility’; said steps of updating being by employing said Artificial Intelligence (Al) engine.

[0064] The identification, given the identification of the ‘characterization’ (see Definitions section) of the patient, of the ‘caregiver’ and / or of the ‘healthcare worker’, of the attributes of the ‘contact framework’, and the conformity to said attributes, increases the probability of conveniently communicating with the patient, the ‘caregiver’ and / or the ‘healthcare worker’ and / or the ‘healthcare facility’, so as to exchange data and information of each step of the ‘healthcare process’, facilitating the ‘management of the adherence’ of the patient to the ‘healthcare procedures’ and to the relative ‘healthcare process’, within the logic of the ‘healthcare path’.

[0065] In a first step, in partial or absolute absence of objective and / or self-referred data on patient, ‘caregiver’ and / or ‘healthcare worker’ and / or ‘healthcare facility’, the dentification of the suitable attributes of the ‘contact framework’ for said patient, ‘caregiver’ and / or ‘healthcare worker’ and / or ‘healthcare facility’ is attained on the basis of statistical data, just like, on the basis of statistical data, the adequate first identification is attained of the belonging of said patient, ‘caregiver’ and / or ‘healthcare worker’ to one of the types of ‘personality states’ and / or ‘profiles’ and / or ‘states of change’ of a database.

[0066] Preferably, the steps of updating are attained by employing an Artificial Intelligence (Al) engine, adapted to identify or modify the probabilities of suitability and to add LEIBO. / 67e2025 or eliminate one or more types of ‘personality states’ and / or ‘profiles’ and / or ‘states of change’ and / or types and / or values of the attributes of the ‘contact framework’.

[0067] Still more preferably, the Artificial Intelligence (Al) engine employs algorithms of automatic learning, recognition and / or processing of the natural language and of the speech (verbal language) and of the “extra-verbal” language, in order to extract and process the information adapted to identify or modify the probabilities of suitability and to add or eliminate one or more types of ‘personality states’ and / or ‘profiles’ and / or ‘states of change’ and / or types and / or values of the attributes of the ‘contact framework’.

[0068] The "extra-verbal" data represent information communicate outside of the verbal language. These data include gestures, facial expressions, posture of the body, visual ‘contact’, tones of voice and other signals which contribute to the ‘communication’ (see ‘Contact framework’, Definitions section) but are not expressed by means of words or verbal language. The use of the term "extra-verbal" underlines that this data is found outside of the domain of the verbal language and can be egually significant in the comprehension of the human ‘communication’.

[0069] With processing of the extra-verbal language it is intended the processing of the nonverbal and para-verbal language. Specifically:

[0070] - non-verbal data: are gestures, facial expressions, body posture, visual ‘contact’, movements, etc. The non-verbal data is used in a mainly unconscious manner for communicating emotions, intentions or moods.

[0071] - para-verbal data: are aspects of the language which surround the spoken content of a ‘contact’. Para-verbal are the voice tone, the word speed, the intonation and other vocal characteristics which can affect the significance of the ‘communication’. The para-verbal data adds nuances and additional significance to the words.

[0072] A behavioral analysis of a person, a psycho-behavioral analysis, a cognitive- behavioral analysis, a psychographic analysis, as well as an analysis of the LEIBO. / 67e2025

[0073] ‘personality states’ and of the ‘states of change’ can be investigated by using a combination of verbal and extra-verbal data acguired by means of digital ‘guestionnaires’ and audio and / or video recordings.

[0074] The collected verbal data can be analyzed by using NLP algorithms (Natural Language Processing) in order to extract information on the emotions, the sentiments, the opinions, the tone of the responses. This can assist in better understanding the emotional and cognitive state and the person’s preferences. The analysis of the para-verbal language can be used for enriching the understanding of the responses during an interview and / or an audio ‘contact’. This analysis can be conducted manually by experts or it can be automated by using specific technigues of recognition and processing of the natural and para-verbal language (NLP). The accurate interpretation of the paralinguistic language can contribute to identifying emotions, emotional states or significance nuances which might not emerge from the verbal text on its own.

[0075] The videotaping can also be used for the analysis of non-verbal data, as a nonlimiting example monitoring the posture and the movements of the body. The analysis of the posture and of the body movements is concentrated on the observation and on the interpretation of the body positions and gestures of the people. This type of analysis seeks to understand what the body posture of a person can indicate, with regarding to his / her emotional state, to the level of comfort, to attitude or intention. For example, a person who is erect and looks directly in the eyes of someone during a conversation can be interpreted as self-confident or interested in the discussion, while a person who is closed in himself / herself and crosses his / her arms could indicate defense, discomfort or insecurity. Also the analysis of the body movements can assist in better understanding the non-verbal ‘communication’. The analysis also regards the facial expressions, the gestures of the hands or other visible signals which communicate non-verbal message. This type of analysis seeks to interpret the visible signs that can reveal emotions, intentions or LEIBO. / 67e2025 moods of a person. For example, a smile can indicate joy or happiness, while a wrinkle on the forehead can suggest worry or confusion. For example, a person could cross his / her arms as sign of defense or shake the head to express disapproval. The videorecordings can be analyzed by using the technology of detection of the facial expressions for identifying changes in the facial expression, such as smiles, smirks or signs of stress. This information can provide clues on the emotions and the emotional reactions of the person. The use of ‘machine learning’ algorithms and Artificial Intelligence can assist in creating predictive models based on the collected data. These models can be instructed to forecast the future behavior of the person based on the collected data, both verbal and extra-verbal. This can assist in identifying behavior models and social relations. The analysis of the social interactions can involve the observation of how a person behaves in group situations or during a conversation.

[0076] The extra-verbal language is therefore the set of prosodic traits (such as intonation, rhythm, duration, accent), of traits such as the type of voice, the timber used, the resonance, as well as the facial and body gesturing, which can be acguired by means of systems of audio detection, video cameras, webcam and / or other audio or viewing systems comprised in the system and used in the ‘contacts’ with the patient and / or the ‘caregiver’ and / or the ‘healthcare worker’ and / or the ‘healthcare facility’.

[0077] Preferably, the method for managing the ‘healthcare processes’ provides for, starting from the step of completing the ‘procedure’ and up to the end of the ‘healthcare process’, one or more steps of verification (or of measurement and verification) of the outcome of the identified ‘procedure’, carried out and / or being completed, by means of ‘machine learning’. In particular, still as a function of the detection of the conditions of the patient (also self-referred), the verification step provides for the assignment of scores to the urgency indicators and ‘appropriateness’ of the ‘procedure’ and their updating as well as the assignment of a score to the level of ‘adherence’ of the patient to the ‘procedure’. In such mode, it LEIBO. / 67e2025 is possible to measure, predict, verify and correlate the outcome of the ‘procedure’ with the actual level of ‘adherence’ of the patient.

[0078] The advantages offered by the present invention are evident in light of the description described up to now and will be even clearer due to the enclosed figures and to the relative detailed description.

[0079] Description of the fiqures

[0080] The invention will be described hereinbelow in at least a preferred embodiment by way of a non-limiting example with the aid of the enclosed figures, in which:

[0081] - FIGURE 1 shows a schematic representation of a system 100 according to the present invention;

[0082] - FIGURE 2 shows a schematic representation of a system 100 according to the present invention;

[0083] - FIGURE 3 shows a block diagram representative of a method 200 according to the present invention;

[0084] - FIGURE 4 shows a block diagram representative of a method 200 according to the present invention;

[0085] - FIGURE 5 shows a patient 2 who compiles a ‘questionnaire’ 106 and is monitored by means of the system 100;

[0086] - FIGURE 6 shows a diagram representative of a process for generating a ‘questionnaire’ 106 in an automated manner;

[0087] - FIGURE 7 shows a diagram representative of a process for generating an input 610 in an automated manner.

[0088] In the enclosed figures, equivalent elements correspond with equal numbers.

[0089] Detailed description of the invention

[0090] According to the present invention, a system is attained together with the relative method of use for managing ‘healthcare processes’, capable, continuously over time, of forecasting, promoting, measuring, evaluating, optimizing and, thus, managing LEIBO. / 67e2025 and improving the ‘adherence’ of the patients through the use of a series of technologies and processes illustrated in detail hereinbelow. The synergistic operation of said technologies and processes allows predicting, measuring and evaluating the outcome of the ‘healthcare governing procedures’ and of the ‘patient self-governing procedures’ as a function of the ‘adherence’ of the patient to said ‘procedure’ and promoting, correcting and adapting ‘healthcare procedures’, ‘healthcare processes’ and / or ‘healthcare paths’ so as to improve the ‘adherence’ of the patient to said ‘healthcare procedures’, ‘healthcare processes’ and ‘healthcare paths’.

[0091] In order to measure the ‘adherence’, several non-limiting, non-binding options are the following:

[0092] - analysis of the main components (PCA): useful for identifying the main dimensions of the ‘adherence’ and reduce the complexity of the dataset, maintaining the most significant information. This allows identifying the main factors that affect the ‘adherence’ and to evaluate their overall impact;

[0093] - structural eguation models (SEM): useful for examining the complex relations between the different dimensions of the ‘adherence’ and evaluating the theoretical models that explain the adherent behavior. This allows testing the effect of the different variables on the overall ‘adherence’ and of identifying possible key factors which affect the behavior of the patient;

[0094] - network analysis: useful for examining the relations between the different variables which compose the construct of the ‘adherence’ and identifying possible patterns or subgroups within the phenomenon. This allows better understanding the complex interactions between the different factors that influence the ‘adherence’ and identifying the main determining factors of the adherent behavior.

[0095] Said ‘adherence’ can be described by employing one or more parameters / variables, each of which measures a specific dimension of the ‘adherence’ and is measured in a LEIBO. / 67e2025 distinct manner. In several embodiments, for the measurement of the ‘adherence’ a single parameter / index is created that combines the different dimensions of the ‘adherence’ in a single value. For this purpose, different approaches could be used, including:

[0096] - weighted approach: assigning a weigh to each dimension of the ‘adherence’ based on its relative importance and then combining the weighted values in a single score. For example, for a specific therapy it could be more important to respect the time of the doses with respect to the guantity taken and thus assign a greater weight to this aspect;

[0097] - sum approach: by simply summing the scores or the measurements of the different dimensions of the ‘adherence’. For example, assigning a point for each time that the patient respects the time of the doses, a point for each time the prescribed freguency is respected and a point for each time that the prescribed guantity is respected;

[0098] - mean approach: calculating the mean of the different dimensions of the ‘adherence’. This approach is particularly advantageous when it is deemed that all the dimensions have the same weight in the overall ‘adherence’.

[0099] The object of the present invention is to provide a system 100 and a method 200 for managing the ‘healthcare processes’ capable of improving the ‘adherence’ of a patient to a ‘healthcare governing procedure’ that is reserved, to be reserved or to be executed (e.g. free access), and to a ‘patient self-governing procedure’. In particular, the method 200 for managing aforesaid ‘healthcare processes’ is such to identify the actual steps of the ‘healthcare process’ and allow the verification of the ‘adherence’ with the pre-established steps of the same ‘healthcare process’.

[0100] In order to be able to evaluate the ‘adherence’ to a ‘healthcare path’, ‘healthcare process’ and / or ‘healthcare procedures’ by a patient, it is essential to understand the behavioral, psychological and cognitive processes thereof and, thus, carry out a ‘characterization’ thereof (see Definitions section). LEIBO. / 67e2025

[0101] The present invention sets the objective of resolving the following technical problems:

[0102] - ensuring a forecast, promotion, measurement, evaluation and optimization of the ‘adherence’ of a patient to a ‘healthcare path’ on the basis of observations or objective data, self-reported data and on the basis of a ‘characterization’ of said patient;

[0103] - ensuring, in each ‘contact’ and in every useful occasion, the collection of said objective and self-referred data for said ‘characterization’ of said patient through the use of various technologies and processes, including the administration of suitable ‘guestionnaires’ in ‘location contexts’ and ‘time contexts’ suitable for the optimization of said ‘characterization’;

[0104] - carrying out said ‘characterization’ of said patient (for the evaluation of his / her ‘adherence’) being based on the integration and on a comparison between said objective data (observed and measured with various technologies) and selfreferred data (perceived by the patient and / or ‘caregiver’ and / or ‘healthcare worker’ and, therefore, subjective);

[0105] - ensuring a forecast and a verification of the outcome of ‘healthcare governing procedures’ and of ‘patient self-governing procedures’ in a ‘healthcare path’.

[0106] The present invention will not be illustrated, merely by way of a non-limiting or nonbinding example, with reference to the figures which illustrate several embodiments relative to the present inventive concept.

[0107] With reference to FIG. 1, a schematic representation is shown of a system 100 for managing ‘healthcare processes’ according to the present invention.

[0108] Said system 100 for managing ‘healthcare processes’, with at least a ‘healthcare worker’ 1 and at least a patient 2, is adapted to forecast, promote, measure, evaluate, optimize and generally ‘manage the adherence’ of said patient 2 to one or more ‘healthcare processes’ (see Definitions section).

[0109] The system 100 comprises at least a computerized apparatus 101 suitable of hosting LEIBO. / 67e2025 one or more databases 110, algorithms and software architectures necessary for performing the steps of a method 200. The computerized apparatus 101 comprises said database 110, one or more Al 120 and one or more agendas 130 of the ‘procedures’. In several embodiments, the computerized apparatus 101 has a distributed hardware and / or software architecture. Said computerized apparatus 101 is managed and accessible by means of a back-end structure 102 adapted to be sed, by way of a non-limiting or non-binding example, by said ‘healthcare workers’, for the modification of one or more available time slots viewable in the agendas 130 and of the completable ‘healthcare procedures’. A front-end application 103 is used, by way of a non-limiting or non-binding example, by said patient, for the reservation and management of said ‘healthcare procedures’. Said front-end application 103, by way of a non-limiting or non-binding example, can be installed on a device of the patient such as a smartphone, a computer, a tablet. Said Al 120 is trained on the basis of historical data present in the database 110 and / or of data acguired and derived by scientific literature and / or by manual validation operated initially by ‘healthcare workers’, and by other sources.

[0110] More specifically, the data present in said database 110 in a non-limiting manner include:

[0111] - various types of ‘personality states’ (see ‘Characterization’, Definitions section);

[0112] - various types of ‘profiles’ (see ‘Characterization’, Definitions section);

[0113] - various types of ‘states of change’ (see ‘Characterization’, Definitions section);

[0114] - various ‘characterizations’ (see ‘Characterization’, Definitions section);

[0115] - various types of ‘contact frameworks’; and, as a function of the attributes that characterize the ‘contact framework’, the data in said database 110 further includes: various ‘macro-steps’ (see ‘Contact framework’, Definitions section); various ‘participants’ (see ‘Contact framework’, Definitions section); various ‘objectives’ (see ‘Contact framework’, Definitions section); LEIBO. / 67e2025

[0116] - various ‘rules’ (see ‘Contact framework’, Definitions section);

[0117] - various ‘communications’ (see ‘Contact framework’, Definitions section);

[0118] - various ‘communication modes’ (see ‘Contact framework’, Definitions section);

[0119] - various ‘channels’ (see ‘Contact framework’, Definitions section);

[0120] - various ‘relationship contexts’ (see ‘Contact framework’, Definitions section);

[0121] - various ‘dynamics’ (see ‘Contact framework’, Definitions section);

[0122] - various ‘intensities’ (see ‘Contact framework’, Definitions section);

[0123] - various ‘durations’ (see ‘Contact framework’, Definitions section);

[0124] - various ‘location contexts’ (see ‘Contact framework’, Definitions section);

[0125] - various ‘time contexts’ ( ‘contact’ and appointment time slots) (see ‘Contact framework’, Definitions section);

[0126] - specific data relative to various ‘participants’, by way of a non-limiting or nonbinding example, to ‘healthcare workers’ such as, by way of a non-limiting or nonbinding example, the availability to appointments and ‘contacts’ with the patients, said availability being visible in said agendas 130;

[0127] - ‘predictors’ 50 which are computer variables indicative of data on a patient and / or a ‘caregiver’ thereof and / or a ‘healthcare worker’ and / or a ‘healthcare facility’; said ‘predictors’ 50, in several embodiments of the invention, are computer variables relative to the relationship between a patient and / or a ‘caregiver’ and / or a ‘healthcare worker’ and / or a ‘healthcare facility’ with a nonhuman operator see ‘Participants’, ‘Contact framework’, Definitions section);

[0128] - data 602 relative to the patient and / or ‘caregiver’ and / or ‘healthcare worker’ and / or ‘healthcare facility’ in several embodiments of the invention; said data 602 can by way of example include data on time availability, personal data, data relative to clinical files and / or other files, ‘predictors’ 50, ‘personality states’ and / or ‘profiles’ and / or ‘states of change’.

[0129] Specifically, one type of ‘profile’, by way of a non-limiting or non-binding example of integrated ‘profile’ of a patient is, by way of a non-limiting or non-binding example, LEIBO. / 67e2025 digitally associated with a vector comprised in the database 110, preferably with a number of inputs equal to four. Said vector inputs can be referred to the values of four variables representative of the behavioral ‘profile’, of the psycho-behavioral ‘profile’, of the cognitive-behavioral ‘profile’ and of the psychographic ‘profile’ in which the patient is framed. The four or more variables can, by way of a non-limiting or non-binding example, take on a value comprised between 0 and 1, representative of the percentage of correspondence of the patient with a behavioral, psycho- behavioral, cognitive-behavioral and psychographic ‘profile’.

[0130] The embodiment just illustrated, which provides for the use of vectors, is not the only option available. Generally, the multidimensional constructs are measured through the use of multiple variables or indicators which reflect the different dimensions of the phenomenon under examination. These indicators can be quantified and organized in a vector, where each characteristic of the vector represents one of the dimensions of the construct. In the scope of the ‘adherence’ to the medical treatment, the different dimensions can include (as a non-limiting example) the frequency with which the patient takes the prescribed drugs, his / her involvement in the programmed medical visits, his / her adaptation to the modifications of the recommended lifestyle, and so forth. Each of these dimensions can be measured through specific variables, such as for example the number of doses of drugs taken correctly, the number of programmed medical visits at which the patient participates or the use of recommended healthy behaviors. These variables can then be organized in a multidimensional vector in which each characteristic represents a dimension of the ‘adherence’ construct. This vector can be used for quantifying and evaluating the overall level of ‘adherence’ of the patient to the treatment.

[0131] By way of a non-limiting or non-binding example, the database 110 comprises ten predefined behavioral ‘profiles’, ten predefined psycho-behavioral ‘profiles’, ten predefined cognitive-behavioral ‘profiles’ and ten predefined psychographic ‘profiles’. From the analysis conducted with the system 100, it results that a patient LEIBO. / 67e2025 has a maximum percentage of correspondence equal to 71% (0.71) at the second behavioral ‘profile’, a maximum percentage of correspondence equal to 67% (0.67) at the fourth psycho-behavioral ‘profile’, a maximum percentage of correspondence equal to 75% (0.75) at the first cognitive-behavioral ‘profile’ and a maximum percentage of correspondence equal to 54% (0.54) at the seventh psychographic ‘profile’. In this manner, the inputs of the vector of the integrated ‘profile’ can be described, by way of a non-limiting or non-binding example, in the following manner: [0.71 B, 0.67D, 0.75A, 0.54G] where the letters B, D, A and G are respectively tied to the second behavioral ‘profile’, to the fourth psycho-behavioral ‘profile’, to the first cognitive-behavioral ‘profile’ and to the seventh psychographic ‘profile’. In the example in which ten behavioral ‘profiles’, ten psycho-behavioral ‘profiles’, ten cognitive-behavioral ‘profiles’ and ten predefined psychographic ‘profiles’ are defined, for each patient forty variables (ten for each type of ‘profile’) will be present in the database 110, with decimal values comprised between 0 and 1. Multiple variables can be associated values that are non-zero, framing a patient not in a distinct ‘profile’ but in a more flexible way, accounting for various nuances.

[0132] The integrated ‘profile’ of the patient, in several embodiments of the present invention, can be represented by means of a matrix with four columns and ten lines in which all the values of said forty variables are present. The Artificial Intelligence 120 evaluates, case by case, for each patient, the “weight” that the percentage of correspondence to an integrated ‘profile’ (or percentage of framing or identification in an integrated integrated ‘profile’) has and, consequently, over time and for each ‘contact’, the ‘contact framework’ most suitable for each patient, having registered in the database 110 a series of historical data (also including continuously updated data) that shows how some integrated ‘profiles’ are more similar to several types and values of the attributes of a ‘contact framework’.

[0133] In addition to the representation of the multidimensional construct of the ‘profile’, by way of a non-limiting or non-binding example of the integrated ‘profile’, through a LEIBO. / 67e2025 vector or a matrix with a number of inputs equal to number of variables, the present invention can employ, in various embodiments, various other options for measuring and analyzing these types of complex constructs. Said different options can be, by way of a non-limiting or non-binding example:

[0134] - correlation matrices: used for evaluating the relations between the different variables that compose the multidimensional construct. Each cell of the matrix represents the level of correlation between two variables, allowing the identification of patterns and associations therebetween;

[0135] - analysis of the main components (PCA): a size reduction technique that allows identifying the patterns underlying the multivariate data. By means of PCA it is possible to identify the main dimensions (or components) that explain the variation in the data and reduce the complexity of the dataset, maintaining the most number of possible significant information pieces;

[0136] - analysis of the clusters: used for identifying homogeneous groups within a multidimensional dataset. The data are grouped together based on the similarities between the observations, allowing the identification of patterns or subgroups within the multidimensional construct;

[0137] - structural equation models (SEM): a statistical technique that allows examining the relations between observed variables and latent variables, i.e. variables not observable which represent complex constructs. By using SEM, it is possible to test complex theory models and evaluate the relations between the different dimensions of the multidimensional construct;

[0138] - network analysis (graphs): technique used for examining the relations between the variables through the representation of a graph, where the nodes represent the variables, and the connections between the nodes represent the relations or the correlations. The network analysis allows viewing and understanding the complex interactions between the different dimensions of the multidimensional construct. LEIBO. / 67e2025

[0139] The analyses of multivariate data, available, by way of a non-limiting or non-binding example, on the basis of the behavioral monitoring and of the ‘guestionnaires’, is a crucial option for confronting the complexity of a multidimensional and multifactor construct such as the ‘profile’ and, in particular, the integrated ‘profile’. The analyses preferably employed are the factorial analysis, the analysis of the main components (PCA) and the cluster analysis:

[0140] - factorial analysis and analysis of the main components (PCA): both of these methods are used for identifying the patterns of association between the variables and reducing the dimension of the data, simplifying the complexity of the ‘profile’ and / or integrated ‘profile’. The factorial analysis is particularly useful when it is desired to identify the underlying factors that can explain the variation in the observed variables. On the other hand, the PCA is a technigue for reducing the dimensionality that identifies the main patters of variation in the data;

[0141] - cluster analysis: for identifying homogeneous groups of individuals based on their characteristics. It is particularly useful when it is desired to explore if there are distinct patterns or types of ‘profiles’ within a sample. In the scope of the present invention, a first example of ‘profile’, by way of a non-limiting or non-binding example of integrated type, is that of a person who is generally on time, who has the tendency to respect the times and the rules, who tends to remember the appointments, the visits to execute, the drugs to take and who trusts in the ‘healthcare process’ that he / she is following. A second example of integrated ‘profile’ is that of a person who is generally disorganized and not on time but who still trusts in the ‘healthcare process’ that he / she is following. The person of the second example will tend to forget the times when he / she took the medicine, to take them at different times from those recommended or even sometimes forget to take them.

[0142] Once the type of ‘profile’ is defined that best corresponds to the patient (as in the vector of the reported example), the system 100 can provide instruments to relate LEIBO. / 67e2025 and treat the patient himself / herself with ad hoc techniques, as will be clarified hereinbelow.

[0143] For each attribute of the ‘contact framework’, by way of a non-limiting or non-binding example, is set of variables is arranged, which can each take on a decimal value comprised between 0 and 1 as a function of the percentage of suitability of said attribute to a patient, ‘caregiver’ and / or ‘healthcare worker’ and / or ‘healthcare facility’.

[0144] An example of ‘macro-steps’ (see ‘Contact framework’, Definitions section) is that of a set of numerical digital and / or alphanumeric identifications and / or of another nature, each associated with a ‘macro-step’ and / or with an ‘action’ already defined within the database 110.

[0145] An example of ‘participants’ (see ‘Contact framework’, Definitions section) is that of numerical digital and / or alphanumeric and / or of another nature, which are referred to patients, ‘caregivers’ and / or ‘healthcare workers’ (and / or other ‘participants’) and / or a ‘profiles’ of said patients, ‘caregivers’ and / or ‘healthcare workers’ already registered within the system 100.

[0146] An example of ‘objectives’ (see ‘Contact framework’, Definitions section) is that of a set of numerical digital and / or alphanumeric identifications and / or of another nature, each associated with an ‘objective’ already defined within the database 110, and / or with a new ‘objective’ inserted in the system 100, to which the identification is automatically and instantaneously attributed by algorithms of said system 100.

[0147] An example of ‘rules’ (see ‘Contact framework’, Definitions section) is that of a ‘contact’ that occurs only once, for example to confirm an appointment, and which does not provide for another subsequent ‘contact’. Another ‘rule’ is, by way of a nonlimiting or non-binding example, that which instead provides for a second ‘contact’, with an interval (e.g. equal to 48 hours) from the first. Another ‘rule’ is, still for example, that for which a ‘contact’ (physical and / or digital) is periodically repeated (e.g. daily), for the administration of the drugs to a patient hospitalized in a LEIBO. / 67e2025

[0148] ‘healthcare facility’ (physical ‘contact’, ‘healthcare governing procedures’), and / or to remind said patient, once dismissed, to continue to take the drugs (digital ‘contact’, ‘patient self-governing procedures’).

[0149] An example of ‘communication’ (see ‘Contact framework’, Definitions section) is that of a text file within which information is reported, by said Al 120, relative to the content, interpreted and understood, of a ‘contact’ (physical and / or digital) between a patient and the ‘healthcare worker’ involved in the treatment thereof.

[0150] An example of ‘communication modes’ (see ‘Contact framework’, Definitions section) is that of a set of numerical digital identifications and / or alphanumeric identifications and / or of another nature, each associated with the modes and / or with the single forms used (verbal modes in spoken form, and extra-verbal modes, tone of voice, posture and physical distance) by a patient and by a ‘caregiver’ during a ‘contact’.

[0151] An example of ‘channel’ (see ‘Contact framework’, Definitions section) is the telephone, as means used, still as an example, for a ‘contact’ in verbal mode. Specifically, said ‘channel’ is associated, by way of a non-limiting or non-binding example, with a parameter (variable over time) of Boolean type, comprised in the database 110, which takes on a value 0 when it does not result suitable for a specific subject, and a value 1 when it results suitable, on the basis, by way of a non-limiting or non-binding example, of the type of ‘profile’, of the ‘state of change’ and of the data acguired by the preceding ‘contacts’ with the same subject. For example, in the event in which a patient, from ‘contact’ with a ‘healthcare worker’, should be without cell phone, the following will surely be defined as not employable (and hence, in the example made, with Boolean value egual to 0); the sms ‘channels’, WhatsApp® and / or others, of a cell phone or smartphone. In another example, in the event in which the behavioral ‘profile’ of the patient should be indicative of a subject who is generally disorganized, bidirectional synchronous ‘channel’ types will be selected such as, for example, telephone or video calls, with respect to asynchronous monodirectional LEIBO. / 67e2025 types such as the e-mail ‘channel’.

[0152] An example of ‘relationship context’ (see ‘Contact framework’, Definitions section) is a set of computer variables relative to the relation, e.g. of personal type, existing between a patient and / or a ‘caregiver’ or between a ‘healthcare worker’ and a nonhuman operator (see ‘Participants’, ‘Contact framework’, Definitions section).

[0153] An example of ‘dynamics’ (see ‘Contact framework’, Definitions section) is that of one or more variables, with value comprised between 0 and 1 (indicative of a percentage value), in which the value depends on the level of cooperation, detected by the algorithms of the system 100 and / or of said Al 120, between the ‘participants’ of a ‘contact’. The variables, by way of a non-limiting or non-binding example, are continuously updated in an automatic manner by the, based on the sentiment (mood deduced from the set of the verbal and extra-verbal analyses) detected.

[0154] An example of ‘intensity’ (see ‘Contact framework’, Definitions section) is that of a set of variables that are related to each other and / or to the variables of the attribute ‘dynamics’ and / or of other attributes. By way of a non-limiting or non-binding example, an ‘intensity’ variable is a variable adapted to take on a percentage value (preferably comprised between 0 and 1) that is automatically compiled by the algorithms of the system 100 and / or of said Al 120 based on the “positivity” of the facial expressions of a patient recognized during a ‘contact’.

[0155] An example of ‘durations’ (see ‘Contact framework’, Definitions section) is that of a variable adapted to record the total time of a telephone ‘contact’ and the partial times associated with the different contents (courtesy exchange, exchanged information and salutations) of the same ‘contact’.

[0156] An example of ‘location context’ (see ‘Contact framework’, Definitions section) is that of variables compiled, by algorithms of the system 100 and / or by the Al 120, based on data extracted from geolocation devices of a patient during a ‘contact’.

[0157] An example of ‘time context’ ( ‘contact’ time slot) (see ‘Contact framework’, Definitions section) is that of one or more variables that take on a value comprised LEIBO. / 67e2025 between 0 and 24 and which indicate a precise time in which it is preferable to contact a certain patient. Specifically, said ‘contact’ time slots, by way of a nonlimiting or non-binding example, can be associated, in the database 110, with Boolean parameters corresponding to specific times / time intervals of the day, which acguire value 1 or 0 if they can be used or not used with a given patient. In another example of ‘time context’, the appointment time slots, by way of a non-limiting or non-binding example, can be Boolean parameters corresponding to times / time intervals of the day, which acguire value value 1 or 0 if they can be used or not used with a given patient and a given ‘healthcare worker’. Said appointment time slots adapted to be usable or not usable for a patient and a ‘healthcare worker’ on the basis of their availability and agendas 130.

[0158] An example of a ‘frame of a contact’ is the set, partial or total, of the types and / or values of the attributes of a ‘contact’. In the case, by way of a non-limiting and nonbinding example, of two different ‘contacts’ between a ‘healthcare facility’ and a patient, and in particular, of a telephone call following a SMS message, both originated from the ‘healthcare facility’, each of the two said ‘contacts’ can, according to the present invention, be characterized by different attributes. Here are, by way of a non-limiting and non-binding example, several values of the attributes of the ‘contact framework’ relative to said first and second ‘contact’, according to the present invention:

[0159] - ‘macro-step’: both ‘contacts’ are associated with the “promotion” of the adherence;

[0160] - ‘objectives’: the first ‘contact’ wishes to sensitize the patient with respect to the opportunity of a check-up visit; the second ‘contact’ is for the planning of the same visit;

[0161] - ‘rules’: the selection of the ‘channel’ (SMS message) for the first ‘contact’ is determined by an internal protocol of the ‘healthcare facility’; the selection of the ‘channel’ (telephone call) and of the time of the second ‘contact’ is a function of LEIBO. / 67e2025 the preferences of the patient, confirmed by the position (‘location context’) of the patient at the time of the ‘contact’;

[0162] - ‘communication’: the first ‘contact’ offers preliminary information; the second in- depth information for patient support;

[0163] - ‘communication modes’: the first ‘contact’ (SMS message) uses that verbal, in written form; the second (telephone call), that verbal, in spoken form, and that para-verbal, showing the variety of the ‘communication modes’ used;

[0164] - ‘channels’: the first ‘contact’ uses the SMS ‘channel’ and the second ‘contact’ the telephone channel;

[0165] - ‘relationship contexts’: the decision to use two ‘contacts’ is influenced by the context where the interaction occurs, in particular regarding communication efficiency, where the sender selects to send a preliminary message to plan a more detailed telephone conversation.

[0166] Said ‘predictors’ 50 are variables which, by way of a non-limiting or non-binding example, can be vectors with two inputs of decimal numerical type comprised between 0 and 1, one indicating the percentage of relative probability (intended as the probability that said ‘predictor’ 50 has a certain relevance for a certain patient) and, the other, a weight (intended as incidence, i.e. the value that the single ‘predictor’ 50 has with respect to the other ‘predictors’ 50).

[0167] The system 100 of the present invention comprises:

[0168] - at least a portable device 140 associated with said patient, with one or more ‘caregivers’ and / or with said ‘healthcare worker’ in order to electronically track their position, the path and the face-to-face ‘contacts’ inside and outside a ‘healthcare facility’ and the relative times. With electronic trackability it is intended the interaction between technological devices with which it is possible to know the position of one of these within an area and see such position and its variation over time on electronic apparatuses like said back-end structures 102. Said portable device 140 comprises a geolocation unit, by way of a non-limiting or LEIBO. / 67e2025 non-binding example of GPS type, suitable of providing geolocation data and / or a wireless communication unit of BLE, UWB or Wi-Fi type, suitable of providing mutual data interaction between multiple portable devices 140. By “tracking of the face-to-face ‘contacts’” it is intended the electronic trackability of ‘contacts’ within an area of physical ‘contact’ and for a minimum ‘contact’ time, through the measurement of the presence and of the vicinity, and of the relative times, of at least two portable devices 140, associated with said patient, ‘caregiver’ and / or ‘healthcare worker’, and provided with wireless technology, by way of a nonlimiting or non-binding example of Bluetooth, Bluetooth Low Energy (BLE), UWB, Wi-Fi type or through the correlation between said measurement of said portable devices 140 and the geolocation data of said geolocation unit. In several embodiments of the invention the face-to-face ‘contacts’ can be tracked, by way of a non-limiting or non-binding example, of a patient, also with respect to subjects, provided for example with devices BLE, not associated with ‘caregiver’ and / or ‘healthcare worker’ and / or another ‘patient’. This results useful in the step preceding the establishment of a ‘contact’ and / or during the same ‘contact’, for the verification, by Al, of the fact that the patient is alone or in presence of other people, and for the conseguent decision to proceed or not proceed with the ‘contact’. Said portable device 140, when provided to a patient, is uniguely associated with the identify of said patient and analogously associated with the identity of a ‘healthcare worker’ when provided with said ‘healthcare worker’. Said portable device 140, in several preferred embodiments of the present invention, is a PDS, a tablet, smartphone, a smartwatch or another wearable device such as, by way of a non-limiting example, smart glasses, a badge, a bracelet, a ring and / or another item. In several embodiments of the present invention said portable device 140 is a personal device of said patient such as a smartphone. The mutual interaction between said portable devices 140 with wireless technology provides data of ‘contacts’ of face-to-face type inside and outside a ‘healthcare facility’. LEIBO. / 67e2025

[0169] In order to infer that two people have had a meeting, intended as ‘contact’ of face- to-face type, for example that they spoke for about 5 minutes within a park, the data BLE (or UWB) and GPS can be used, combining the spatial analysis (coming from the GPS data) with the time and proximity analysis (coming from the BLE or UWB data). Hereinbelow, by way of a non-limiting or non-binding example, a possible mode:

[0170] - Proximity analysis (BLE or UWB data)

[0171] - Distance between the devices: the BLE data indicates that the two devices were about 1 meter distance apart for 5 minutes. This vicinity is compatible with the social interaction between people, such as a conversation face-to- face. The BLE signal decreases in intensity with the distance, hence a constant distance of about 1 meter for an extended period is a good indicator of a potential interaction.

[0172] - Stability of the connection: is the BLE signal remains stable (without significant fluctuations) for the entire period of 5 minutes, it is probable that the two people did not move away from each other, which reinforces the assumption that a conversation or extended interaction took place.

[0173] - Analysis of the Location (GPS data)

[0174] - Verification of the position in the park: the GPS geolocation data GPS confirms that both people were situated within the park during the period in which the vicinity was detected by means of BLE. This environmental context is crucial, since it makes it more plausible that the people remained stopped or if they moved slow, which is compatible with a conversation or a meeting.

[0175] - Spatial coherence: the correlation between the geolocation data GPS of both devices 140 shows that the two people are situated in the same area of the park. If the GPS shows that both people stopped or moved together, this further supports the idea of an interaction, such as walking alongside each other and talking. LEIBO. / 67e2025

[0176] Inference of the social interaction

[0177] - Combining the information, one can reasonably infer that the two people had a meeting and spoke for about 5 minutes in the park.

[0178] - Constant proximity: the distance of about 1 meter detected from the BLE data for 5 minutes suggests an extended physical vicinity, typical of a social interaction.

[0179] - Appropriate environmental context: being in a park, a place where people often stop to converse, reinforces the assumption of a social encounter.

[0180] - Coherent movement: if the GPS data show that the people were stopped or walk slowly together, it can be deduced that they were engaged in a conversation;

[0181] - in several embodiments of the present invention, the system 100 further comprises a plurality of fixed devices 150 positioned in said ‘healthcare facility’ and adapted to detect the passage and the stay time, within one or more areas, of the portable devices 140 with wireless technology, (by way of a non-limiting or non-binding example, of Bluetooth Low Energy, UWB, Wi-Fi and / or another type) and to communicate said passage and said stay time to said computerized apparatus 101. The communication with said computerized apparatus 101 is carried out, in various embodiments, with various technologies such as, by way of a non-limiting or non-binding example, wire connections, technology of connection to a wireless network and / or another type. Each of said fixed devices 150 defines a distinct position within said ‘healthcare facility’. The interaction between said portable devices 140 and said fixed devices 150 suitable of providing data 145’ indicating the position, the path and the stay times of said portable devices 140 in the areas monitored by the fixed devices 150. The mutual interaction between said portable devices 140 with wireless technology and said fixed devices 150 provides data of ‘contacts’ of face-to-face type within a ‘healthcare facility’ based on the verification of the position and of the spatial LEIBO. / 67e2025 coherence of said ‘contacts’ within said ‘healthcare facility’;

[0182] - ‘log files’ relative to ‘contacts’ of said patient and / or one or more ‘caregivers’ and / or a ‘healthcare worker’ and / or a ‘healthcare facility’ intended as organization (the logs, for example, can be relative to ‘contacts’ between a ‘caregiver’ and the secretary of the ‘healthcare facility’). Said ‘log files’ comprising all the types of ‘contacts’, face-to-face, paper, electronic and digital both inside and outside the ‘healthcare facility’. Said ‘log files’ contain, by way of a nonlimiting or non-binding example:

[0183] - types and values of the attributes of the ‘contact framework’, when available;

[0184] - general ‘contact’ data: distinct ID (identification), start and end timestamp, log generation timestamp, type (human or non-human, system, bot, robot) and distinct ID (identification) of ‘participants’, direction (entering and / or exiting), duration (expressed in seconds), ‘channel’ (face-to-face, telephone call, SMS, e- mail, chat message);

[0185] - data specific per type of ‘channel’:

[0186] - telephone call: response type (connected, free but no response, busy, voicemail, call refused), wait time before the response;

[0187] - SMS, e-mail: timestamp of reception and / or reading of the message;

[0188] - live chat: timestamp of the last visit on the chat, timestamp of reading of single chat messages, number of chat messages, duration of an exchange of chat messages.

[0189] Said ‘log files’, in several embodiments of the present invention, by way of a nonlimiting or non-binding example, are relative to the tracking within web sites and / or applications that use cookies of first parties or tracking pixels and / or cookies of third parties;

[0190] - at least a first algorithm; said first algorithm adapted to analyze data 145’, geolocation data, interaction data and data extracted from said ‘log files’; said first algorithm then processes said data, allowing, by way of a non-limiting or non- LEIBO. / 67e2025 binding example, the evaluation of when and with whom the phone calls occurred, how many took place and their duration, the phone calls that a patient made and received during his / her stay time in a ‘healthcare facility’; said Al 120 being further instructed by said first algorithm;

[0191] - ‘predictors’ 50 contained in said database 110; said ‘predictors’ 50 being computer variables indicative of data on a patient and / or a ‘caregiver’ thereof and / or a ‘healthcare worker’ and / or a ‘healthcare facility’; said patient, ‘caregiver’ and / or ‘healthcare worker’ being associated with one or more ‘characterizations’ (see Definitions section) based on the ‘predictors’ 50, whose values depend on objective data. Said objective data are: geolocation data, interaction data, data extracted from the ‘log files’, data 145’ and the processing of said first algorithm; the measurement, the forecast and the continuous updating of the ‘adherence’ of said patient being conducted by said Al 120 based on the ‘predictors’ 50 and on said objective data.

[0192] In several embodiments of the present invention, the system 100 comprises said ‘guestionnaires’ 106, which are administered to the patient, to the ‘caregiver’ and / or to the ‘healthcare worker’ by ‘healthcare workers’ and / or by the ‘healthcare facility’ (intended as organization and, hence, also by operators of non-sanitary type, by way of non-limiting example, administrative / management / secretary) personnel and / or automatically by said Al 120 and / or semi-automatically when said indications of position, path, time and face-to-face ‘contacts’, provided by said objective data (geolocation data, interaction data, data extracted from the ‘log files’, data 145’ and the processing of said first algorithm), are employed by said Al 120 to suggest to a ‘healthcare worker’ or to another operator of the ‘healthcare facility’ (always intended as an organization and hence, as seen above, “to suggest” for example to secretary personnel) to proceed with the administration of a ‘guestionnaire’ 106; said ‘guestionnaires’ 106 providing self-reported data 106’. By administration of a ‘guestionnaire’ 106 it is intended, as a non-limiting example, a digital sending over a LEIBO. / 67e2025 portable device of a patient, a telephone call made by a human or non-human interlocutor (robot), or even an a verbal interlocution during a face-to-face ‘contact’. Said ‘questionnaires’ 106 being adapted to be sent to the computerized apparatus 101. Said ‘questionnaires’ 106, in several embodiments of the present invention, are fully present in said database 110 with sets of questions and / or language statements predefined by human operators by means of said back-end structure 102. Said ‘questionnaires’ 106 are selected by said Al 120 from among ‘questionnaires’ 106, with relative scale, present in the literature, validated scientifically and included in one of the databases 110 of the system and / or are dynamically and automatically composed of said Al 120 through questions and / or linguistic statements selected from single questions and / or predefined linguistic statements present in said database 110. Said ‘questionnaires’ 106 are adapted to request, by way of a nonlimiting or non-binding example, personal data information, information on lifestyles, on time / day availability of the patient, on his / her clinical history, his / her health state, his / her satisfaction with regard to a ‘healthcare procedure’, to a ‘healthcare process’ or to the ‘healthcare path’, his / her state of agitation, his / her ‘adherence’ to the ‘healthcare path’ and / or other information. Said ‘questionnaires’ 106 comprise one or more questions on said objective data in order to obtain data relative to the attention and / or to the memory and / or to the perception of the patient and / or of the ‘caregiver’ and / or of the ‘healthcare worker’. Said ‘log files’ comprise the updating of one or more types and / or values of one or more attributes of a ‘contact framework’ associated with said patient and / or ‘caregiver’ and / or ‘healthcare worker’ and / or ‘healthcare facility’ on the basis of said data 106’. Said patient and / or ‘caregiver’ and / or ‘healthcare worker’ is associated with one or more ‘personality states’ and / or ‘profiles’ and / or to one or more ‘states of change’ based on the ‘predictors’ 50 whose values further depend on said self-reported data 106’ collected from said ‘questionnaires’ 106. Said ‘adherence’ of the patient is continuously predicted, promoted, measured, evaluated and optimized by said Al 120 based on the LEIBO. / 67e2025 comparison and / or integration between objective data and said self-reported data 106’.

[0193] In several embodiments of the present invention, like that shown in FIG. 1, the system

[0194] 100 further comprises one or more video acquisition units 104, suitable for acquiring images and audio of a patient, of one or more ‘caregiver’ and / or of a ‘healthcare worker’, by way of a non-limiting or non-binding example, within a ‘healthcare facility’ and / or during a video call by means of Web. Said video acquisition unit 104 adapted to acquire audio and video data 104’ and to send it (partly or entirely) to said computerized apparatus 101. The sending of the data to said computerized apparatus

[0195] 101 is carried out, in various embodiments of the invention, through various technologies, by way of a non-limiting or non-binding example, by means of wired connection to a back-end structure 102, by means of technologies wireless such as Bluetooth or UWB, by means of Wi-Fi network connection, since suitable communication units (like Wi-Fi antennas) are integrated within the video acquisition unit 104. Said video acquisition unit 104 is, by way of a non-limiting or non-binding example, a video camera, a webcam or other. The video acquisition unit 104 acquires said data 104’ on the basis of different inputs which, by way of a non-limiting or nonbinding example, comprise: commands, in natural language, for consensus or nonconsensus for the acquisition and recording, for sending, interruption and / or turning off, pronounced by the subjects (patients, ‘caregivers’, ‘healthcare workers’ and / or other personnel of the ‘healthcare facility’) affected (the commands are recorded and sent to the system 100 as proof of the management of the consensus and, together, as ‘characterization’ data of said affected subjects); interactions (automatic and / or commanded) with a portable device 140 and / or with applications for its management. Said objective data comprise said data 104’.

[0196] Said ‘log files’ comprise the updating of one or more types and / or values of one or more attributes of a ‘contact framework’ associated with said patient and / or ‘caregiver’ and / or ‘healthcare worker’ and / or ‘healthcare facility’ on the basis of said LEIBO. / 67e2025 data 104’.

[0197] In several embodiments of the present invention, like that shown in FIG. 1, the system 100 further comprises:

[0198] - one or more audio acquisition units 105 suitable for acquiring the audio of a ‘contact’ between patient and / or one or more ‘caregivers’ and / or a ‘healthcare worker’ and / or a ‘healthcare facility’. Said audio acquisition unit 105 adapted to acquire audio data 105’ and to send it (partly or entirely) to said computerized apparatus 101. The sending of the data to said computerized apparatus 101 is carried out, in various embodiments of the invention, through various technologies, by way of a non-limiting or non-binding example, by means of wired connection to a back-end structure 102, by means of technologies wireless such as Bluetooth or UWB, by means of Wi-Fi network connection since suitable communication units (such as Wi-Fi antennas) are integrated within the audio acquisition unit 105. By way of a non-limiting or non-binding example, said audio acquisition unit 105 is a call recorder microphone that has a connection unit which uses Wi-Fi protocols for sending data 105’ to said computerized apparatus 101. Said objective data comprise said data 105’. Said ‘log files’ comprise the updating of one or more types and / or values of one or more attributes of a ‘contact framework’ associated with said patient and / or ‘caregiver’ and / or ‘healthcare worker’ and / or ‘healthcare facility’ on the basis of said data 105’.

[0199] - at least an algorithm 160, comprised in said computerized apparatus 101, for analysis of the verbal language and / or extra-verbal adapted to analyze said data 105’. In several preferred embodiments of the present invention, said algorithm 160 carries out both the analysis of the extra-verbal language (specifically, for the data 105’, of the para-verbal language) and the analysis of the verbal language based on the data 105’ and, in the embodiments that provide for the use of the video acquisition unit 104, also of the data 104’. Said Al 120 acquires the processing conducted by said algorithm 160 and is further instructed thereby. LEIBO. / 67e2025

[0200] Said algorithm 160 uses, by way of a non-limiting or non-binding example, the free software Praat® for the prosodic analysis of the data 105’ and the software FaceReader by Noldus® for the analysis of the expressions and micro-facial expressions of the data 104’. Said algorithm 160 allows, by way of a non-limiting or non-binding example, detecting in a time of about 0.2 s, through the data 105’, the sex of interlocutor (allowing the distinction between man and woman) and allows having a probable age range for said interlocutor in a time of about 2 s.

[0201] In several embodiments of the present invention, the computerized apparatus 101, the audio acguisition unit 105, the video acguisition unit 106 and the algorithm 160 for language analysis are all integrated in the portable device 140.

[0202] By way of a non-limiting or non-binding example, said algorithm 160 detects a tone of voice, talking speed, moving speed, freguency of movement of the hands, legs, feet, eyes, assumed posture (such as for example crossed arms, chin down), facial expressions and other personal characteristics of the patient, of the ‘caregiver’ and / or of the ‘healthcare worker’. Said algorithm 160 recognizes when one is potentially lying through the analysis of the expressions and micro-facial expressions, of the posture, of the movements, and / or of the tone of voice.

[0203] By way of a non-limiting or non-binding example, said algorithm 160 makes use of the following resources and technologies:

[0204] - Facial Action Coding System (FACS): a system used for analyzing the facial expressions. FACS is a system that does not reguire Al per se, but often the software implementations based on FACS use algorithms for analysis of the images or facial recognition which can use Al 120 or ‘machine learning’;

[0205] - OpenFace: an open-source toolkit for the analysis of the body language and of the facial expression. It offers functions for detecting the facial expressions, the movement of the eyes, and more. OpenFace uses Al, in particular deep learning, for the detection of the facial expressions, of the eye expressions and of the movements of the face. It is based on deep neural networks for analyzing the facial LEIBO. / 67e2025 characteristics;

[0206] - OpenPose: an open-source framework for detecting human posture. OpenPose uses deep learning technigues for detecting the human posture and the body movement. It is based on convolutional neural networks (CNN) for identifying the articulations of the body in the images or in videos;

[0207] - Deep Learning Framework: by way of a non-limiting or non-binding example, TensorFlow, PyTorch and Keras, used for developing personalized models for the analysis of the extra-verbal data. For example, it is possible to train neural networks for the detection of the facial expressions or the recognition of the body language;

[0208] - Tobii Eye Tracker®: devices capable of attaining a tracking of the eye movements by monitoring glance direction and freguency.

[0209] In other embodiments of the present invention, said algorithm 160 uses other solutions, such as those provided by several cloud service providers, for example Microsoft Azure® and Amazon Rekognition®, which offer API for the analysis of images and videos and which include emotion and facial recognition functions, or additionally commercial solutions, such as Smart Eye®, which offer packages, based on ‘machine learning’, for analysis of the facial expressions and body language.

[0210] Said algorithm 160 adapted to analyze the extra-verbal language both for said patient 2 and for said ‘healthcare worker’ 1, allowing an evaluation of the sentiment of both as well as an evaluation of the influence that the sentiment of the ‘healthcare worker’ 1 can have on the sentiment of the patient 2 and / or the influence of the presence and / or of the sentiment of a ‘caregiver’ on the sentiment of the patient 2. As a non-limiting example, the sentiment of the patient 2 can be positive, neutral or negative as a function of a certain sentiment of the ‘healthcare worker’ 1.

[0211] The analysis of the verbal and extra-verbal language, in several embodiments of the present invention, is conducted by employing software and said algorithm 160, which can use, by way of a non-limiting or non-binding example, the following resources: LEIBO. / 67e2025

[0212] - Transcription of the speech into text: Google Cloud Speech-to-Text®, IBM Watson Speech to Text® and Microsoft Azure Speech Service®;

[0213] - Processing of the natural language (NLP): spaCy, NLTK® (Natural Language Toolkit) and Stanford NLP® can be used for analyzing the transcribed text and extracting linguistic information, sentiment and tones;

[0214] - Detection of moods: for example Microsoft Azure Cognitive Services® provides an API for analysis of the emotions based on audio acguisitions;

[0215] - Programming languages for the analysis of the natural language: Python® is a programming language widely used for the analysis of the natural language and offers numerous libraries and frameworks, such as TextBlob®, spaCy® and NLTK® which can be used for processing voice data;

[0216] - Acoustic analysis software used for extracting information on the tone, speed, volume and other voice characteristics: Praat® and Audacity®;

[0217] - Acoustic analysis software, for emotions and for the scene: OpenSmile®;

[0218] - Platforms for analysis of the natural language and sentiment analysis: Lexalytics®, RapidMiner® and MonkeyLearn®;

[0219] - Libraries for ‘machine learning’: TensorFlow® or PyTorch® in order to train personalized models for analysis of the extra-verbal language.

[0220] Said patient, ‘caregiver’ and / or ‘healthcare worker’ is characterized (see ‘Characterization’, Definitions section) on the basis of said ‘predictors’ 50 and, thus, on the basis of objective data and of self-reported data 106’, collected from said ‘guestionnaires’ 106 administered to the patient, to the ‘caregiver’ and / or to the ‘healthcare worker’ by ‘healthcare workers’ and / or automatically and / or semi- automatically from said Al 120. Said ‘characterization’ also depending on the type and step of the ‘healthcare process’ underway (e.g. management appointment or therapy).

[0221] In several embodiments of the present invention, said video acguisition unit 104, as mentioned, is comprised in wearable smart glasses. Said wearable smart glasses are LEIBO. / 67e2025 worn, by way of a non-limiting or non-binding example, by said patient, ‘caregiver’ and / or ‘healthcare worker’ (e.g. in a ‘healthcare facility’) and provide the viewing area of the wearer (the patient, the ‘caregiver’ and / or the ‘healthcare worker’). Advantageously, said Al 120 analyses the data 104’ of said video acguisition unit 104 in order to determine when an object / document / screen / point (e.g. within a visit room of a ‘healthcare facility’) remains in said viewing area for a defined time. Said “defined time” can, by way of a non-limiting or non-binding example, be manually defined by means of back-end structures 102 or automatically by the Al 120. This analysis allows obtaining an objective data point, tied to the fact that one said object / document / screen / point is a room, within the area of the glance of the user who wears the smart glasses for a “defined time”, considered sufficient so that said object / document / screen / point of a room has actually been seen. As a non-limiting example, the Al 120, knowing the viewing area of the patient, can understand if he / she has effectively observed a document or a screen indicated by the ‘healthcare worker’. Such objective data can be employed for carrying out a verification of the attention and / or of the memory and / or of the perception of the patient by means of ‘guestionnaires’ 106.

[0222] In several embodiments of the present invention, said indication of position, path, time and ‘contacts’ provided by the objective and / or self-referred data is employed by said Al 120 in order to suggest, by means of said back-end structure 102, to a ‘healthcare worker’ to administer, or automatically administer to a patient, a ‘caregiver and / or a ‘healthcare worker’, a ‘guestionnaire’ 106 containing one or more guestions on objective data measured with said portable devices 140, said fixed devices 150, said video acguisition unit 104 and / or said audio acguisition unit 105, in order to obtain data related to the attention and / or to the memory and / or to the perception of the patient and / or of the ‘caregiver’ and / or of the ‘healthcare worker’ (i.e. relative to the subjected perception of the patient and / or ‘caregiver’ and / or ‘healthcare worker’) with regard to different dimensions of the ‘characterization’ and LEIBO. / 67e2025 of the level of ‘adherence’ of the patient and with an estimated compilation time, for example for said patient, lower than the wait time estimated in a waiting room where one is found. The compilation time of the ‘questionnaire’ 106 more generally must be less than the wait time for an event, regarding which the compilation of said ‘questionnaire’ 106 is deemed useful.

[0223] The tracking of the position of the patient, carried out by the portable device 140 outside a ‘healthcare facility’ is possible through the use, by said portable device 140, of one unit, by way of example GPS, for the geolocation. Said geolocation allows obtaining the position of a patient who has said portable device 140 therewith. Said geolocated position is sent to said computerized apparatus 101 in order to allow suitable analysis algorithms and / or said Al 120 to understand if the patient is situated, by way of a non-limiting or non-binding example, at his / her residence, if he / she is outside, if in a ‘healthcare facility’. Said analysis of the geolocated position of the patient is exploited by said Al 120 in order to determine whether or not to administer a ‘questionnaire’ 106 to said patient. By way of a non-limiting or nonbinding example, the Al 120 administers the ‘questionnaire’ 106 to the patient who is situated at his / her residence and avoids administering it if situated outside and / or if moving. Said Al 120 is also adapted to analyze the moment in which a patient has left, for example from his own residence, to go to a ‘healthcare facility’ for an appointment. In this case, the Al 120 can comprise if the patient is potentially going to be late or early, evaluating the average travel times of a road section that connects the position of the patient with the position of the ‘healthcare facility’ where the appointment has been set. The evaluation of such path can be carried out by the Al 120 through known instruments, such as, by way of a non-limiting or non-binding example, the Google® Maps service. In several embodiments of the invention, the Al 120 determines that said patient is surely late or surely will not arrive for said appointment when the geolocated position of the patient is identified at a point quite far from the ‘healthcare facility’ a short time before his / her appointment, or even LEIBO. / 67e2025 when the position of the patient is in a geographic place completely different and considerably distant from that of the appointment. In ambivalent cases, the Al 120, based on the geolocated position, by way of a non-limiting or non-binding example, can administer to a patient a ‘guestionnaire’ 106 in order to resolve said ambivalence. In the cases of tardiness or missed appointment, the Al 120 can send a notification to a ‘healthcare worker’ and / or automatically update said reservation agendas 130. By way of a non-limiting or non-binding example, said ‘guestionnaire’ 106 is employed for having a measurement of the actual memory of the patient, of the ‘caregiver’ or of the ‘healthcare worker’ regarding what was said / what happened during a ‘contact’. In several embodiments of the invention, said data 104’ and 105’, transcribed by suitable transcription algorithms, can be analyzed by the Al 120, organized in “summaries” of the main arguments treated during a ‘contact’ with a patient, and made available, by way of a non-limiting or non-binding example, to patients and ‘healthcare workers’ by means of said back-end structure 102 and / or front-end application 103.

[0224] The continuous forecast, promotion, measurement, evaluation, optimization and updating of the ‘adherence’ of said patient is automatically conducted by said Al 120 based on the ‘predictors’ 50 and on the objective and self-referred data. Said ‘adherence’ of the patient is predicted, promoted, measured, evaluated and optimized by said Al 120 based on the integration and on the automatic comparison between objective data provided by said video acguisition unit 104 (data 104’), said audio acguisition unit 105 (date 105’), said portable device 140, said geolocation unit, said fixed devices 150 (data 145’) and self-reported data of said patient, ‘caregiver’ and / or ‘healthcare worker’, acguired by means of said ‘guestionnaires’ 106 (date 106’) and relative to facts and / or perceptions of subjected type. Said value of ‘adherence’ is then automatically updated each time new objective and / or selfreferred data is obtained. Said Al 120 associates with the patient the attributes of the ‘contact framework’, based on said ‘predictors’ 50 and said objective and self- LEIBO. / 67e2025 referred data in order to increase the value of said ‘adherence’ for said patient.

[0225] By way of a non-limiting or non-binding example, the type of ‘communication’ with the patient is a function of the ‘healthcare path’. The Al 120, based on the therapy followed by a patient can define the guestions to be inserted in said ‘guestionnaires’ 106, or select suitable pre-set ‘guestionnaires’ 106 relative to one or more drugs or to a therapy. For example, if by scientific literature a specific side effect is known, associated with the intake of a drug, and it is known that such side effect appears after a certain period of time after intake, said ‘guestionnaires’ 106 can include guestions aimed to understand if the patient has said side effect, to which extent, and for how long he / she has had it, thus obtaining also a response relative to the ‘adherence’ of the patient to the therapy (a patient who has a side effect is probably indicative of a patient who is taking the prescribed drug(s).

[0226] By way of a non-limiting or non-binding example, the ‘adherence’ is registered in the database 110 by means of a variable with value comprised between 0 and 100 and representative of the percentage probability of the patient to follow the indications provided by the ‘healthcare workers’ for his / her ‘healthcare process’. By way of a non-limiting or non-binding example, a greater ‘adherence’ (higher percentage probability) is assigned to the patients who are mainly identified in a ‘profile’ of a person who is generally punctual, who has the tendency to respect the times and rules, who tends to remember the appointments, the visits to be completed, the drugs to take and who trusts in the ‘healthcare process’ that he / she is following. By way of example, a greater ‘adherence’ (higher percentage probability) is assigned to the patients whose data 104’ and 105’, analyzed by the algorithm 160, are indicative of a person who tends to have calm behavior and who is at ease in discussing with a ‘healthcare worker’. Still by way of example, a lower ‘adherence’ (lower percentage probability) is assigned to the patients who, notwithstanding the fact that they belong to a ‘profile’ of a person who tends to be punctual and respectful of times and rules, from the analyses carried out by the algorithm 160 have replied to specific LEIBO. / 67e2025 questions, relative to their tendency to follow a recommended pharmacological therapy, in a manner contrary to that detected. Still by way of example, a low ‘adherence’ is assigned to the patients who have been mainly identified in a behavioral ‘profile’ of a person who is generally disorganized and not on time.

[0227] In several preferred embodiments, like that shown in FIG. 1, said portable device 140 comprises an accelerometer 141 adapted to carry out movement measurements and, through a communication unit (which can be of Bluetooth type, with network connection 4G, 5G, Wi-Fi and / or other), to transmit the detected data 141’ to said computerized apparatus 101 which comprises movement algorithms. Said objective data comprise the data 141’. Said movement algorithms provide further data through which said Al 120 manages said ‘adherence’ of said patient.

[0228] In several embodiments of the present invention, said accelerometer 141 communicates with a gyroscope, allowing the identification and interpretation of subtle and distinct gestures like those associated with the intake of drugs. The advantages of the communication between accelerometer 141 and gyroscope are reported hereinbelow:

[0229] - improved precision in the movement detection: the accelerometer 141 measures the linear acceleration, while the gyroscope detects the angular speed and orientation. Combined, this information provides a more complete measurement of the movement;

[0230] - identification of the movement direction: the gyroscope can assist in determining the direction and orientation of the movement and, therefore, distinguish, by way of a non-limiting or non-binding example, the lifting of the glass towards the mouth, through the orientation of the arm during lifting, from the tilt of said glass during drinking, through the rotation and tilt of the wrist. This can allow greater specificity in detecting the actions connected to the intake of drugs;

[0231] - reduction of the detection errors: integrating more sensors can reduce the errors that can occur by using only one sensor; LEIBO. / 67e2025 greater context for the analysis: by combining the data, it is possible to obtain a more detailed vision of the movement.

[0232] In several embodiments of the invention in which the computerized apparatus 101 is comprised within the portable device 140, the movement algorithms can also be directly implemented within the portable device 140 (which is for example wearable). This allows the accelerometer 141 to process the collected data 141’ in real time and determine the position, the movement and the state of the user without having to rely on an external connection or on remote processing resources.

[0233] The accelerometer 141 can distinguish if one is standing, seated or lying down and thus provide information on the distance from the ground during a movement. This occurs through the analysis of the changes in a gravitational acceleration along the axes x, y and z. When standing, the gravitational acceleration along the axis z (vertical, perpendicular to the walking surface) will be predominant, while when seated or lying down this acceleration will be reduced. By using this information together with movement algorithms, the accelerometer can determine position and state, for example if one is standing, seated or lying down.

[0234] Said computerized apparatus 101 processes the data 141’ acguired by said accelerometer 141 in order to determine a state of agitation of a patient. The movement algorithms preserve, in the databases 110, ranges of analyzed values, such as for example the freguency of movement of the hands, characterizing the patient and his / her normal level of agitation. In the event in which, for example, the portable device 140 is a smartwatch, or a bracelet, the accelerometer 141 registers the movements of the hand / wrist of the patient; when, for example, the registered data 141’ is indicative of a movement value of said hand / wrist of the patient that lies beyond the ranges provided by the movement algorithms, the patient is considered “agitated” and his / her state of agitation is considered greater with the increase of the difference between the hands movement value of data 141’ and the maximum value in the range provided by the movement algorithms. Said Al 120 acguires LEIBO. / 67e2025 information on said state of agitation for the ‘management of the adherence’ of the patient.

[0235] Still in another non-limiting example, the data 104’ can be used for identifying, on average, the state of agitation of a patient when filmed with said video acguisition unit 104. The algorithm 160 preserves, in the databases 110, ranges of analyzed values, such as for example the freguency of movement of the hands and legs, characterizing the patient and his / her usual level of agitation. The agitation level could be enhanced also a function of the environment context (waiting room, visit room) and of the relative step (anamnesis, diagnosis, therapy) of the ‘healthcare path’. Still in another non-limiting example, the determination by the movement algorithms of the state of agitation of a patient, through the processing and the data analysis 141’ registered by the accelerometer 141 of a portable device 140 worn by said patient in the form of a smartwatch, can be compared, by said Al 120, with the determination of said state of agitation of said patient operated by the algorithm 160 on the data 104’. When, for example, the data 141’ is indicative of a movement value of the hands of the patient, which is beyond the ranges provided by the algorithm 160, the patient is considered “agitated” and his / her state of agitation is considered greater with the increase of the difference between the hands movement value of data 141’ and the maximum value in the range provided by the algorithm 160. By way of a non-limiting or non-binding example, the state of agitation of the patient is detected by said accelerometer 141 before his / her entrance in a clinic and thus before the patient has been filmed by a video acguisition unit 104 placed in the same clinic. The Al 120 thus obtains, on the basis of two distinct modes (accelerometer 141 and video unit 104), measurements of the patent’s agitation level before, during and after a visit with a ‘healthcare worker’. In this manner the Al 120 can compare the data in order to understand, for example, that it is preferable to administer a ‘guestionnaire’ 106 to a patient after the visit, said patient usually being less agitated, in the historical record of his / her medical visits, and hence more inclined to reply in LEIBO. / 67e2025 an attentive and truthful manner after the visit itself. In this manner the Al 120, still in another example, can use the data on the state of agitation in order to update the ‘characterization’ (see Definitions) of the patient, improving the prediction on the ‘adherence’ of the patient itself and identifying preferable actions for the relative optimization.

[0236] The video unit 104 and / or the audio unit 105 therefore allow verifying the dynamics of the state of agitation in association with the different ‘location contexts’ and the different steps of the ‘healthcare path’. By way of a non-limiting or non-binding example, said state of agitation is evaluated by identifying the ‘profile’ associated with the patient. By way of a non-limiting or non-binding example, each ‘profile’ of a patient is associated with typical ranges of data 141’ present in the scientific literature and / or collected from all the patients and accelerometers 141 employed in the system 100. A patient who is in a waiting room and who, for example, is mainly framed in a ‘profile’ of a person who tends to be calm, at the time in which he / she is identified as “guite agitated”, with data 141’ significantly higher (where with “significantly higher” it is intended for example values greater than 150%) than the upper limit of the typical range (i.e. given by the mean of the historical measurements for said patient) of data 141’ for his / her ‘profile’, could be agitated for fear of the possible outcome of the visit that is about to take place, for not having following therapy in a correct manner or for external causes. In this case, the Al 120 will compare the data 141’ with the data 104’ acguired during the visit. For example, in the event in which the body language (analyzed by the algorithm 160) of the patient should reveal fear, the Al 120 might understand that the state of agitation detected with the data 141’ is connected to the fear of the outcome of the visit and therefore could leave the frame of the patient unaltered in the ‘profiles’ of the same already associated therewith and, conseguently leave unaltered the evaluation of his / her ‘adherence’. In the event in which the data 104’ instead detect a body language (analyzed by the algorithm 160) indicative of a person who hides something and / or LEIBO. / 67e2025 who lies, the Al 120 could comprise that the state of agitation detected with the data 141’ is connected to not having attentively followed the therapy and could, therefore, update the evaluation of the ‘adherence’ of the patient, reducing the value thereof. In each case, the Al will evaluate the manner in which the state of agitation will be reflected on the future ‘adherence’.

[0237] Specifically, in FIG. 1, a condition is shown in which a patient 2 is situated in a waiting room (upper left frame in FIG. 1) and compiles a ‘guestionnaire’ 106 by means of said front-end application 103, which in FIG. 1 is accessible by means of a tablet owned by the patient 2. The compilation of said ‘guestionnaire’ 106 generates data 106’. In FIG. 1 said patient also wears said portable device 140 which is a smartwatch, with said accelerometer 141 mounted at its interior, which generates data 141’.

[0238] In several preferred embodiments of the present invention, said video acguisition unit 104 is employed for the video monitoring of a patient hospitalized at a ‘healthcare facility’. Said monitoring adapted to provide an objective information piece on the ‘adherence’ of the patient to a prescribed therapy and to provide an observational sample that constitutes a base for carrying out forecasts of future ‘adherence’ in the scope of the same hospitalization setting and, as proxy, in other settings, in particular the home.

[0239] In several embodiments of the present invention, the video acguisition unit 104 employed for the monitoring of a patient hospitalized at a ‘healthcare facility’ is used together with said accelerometer 141, which is worn by said patient in the form of a bracelet and / or of a ring, and the data 141’ are adapted to provide an indication of movements relative to the preparation and / or self-administration of drugs. Said data 104’ and said data 141’ are integrated with each other and compared by said analysis algorithm 160 and / or said movement algorithms and / or said Al 120 in order to obtain objective data relative to the intake of drugs by the patient. The movement of bringing drugs to and from the mouth, by way of a non-limiting or non-binding example, can be a predefined movement whose information (positioning of the LEIBO. / 67e2025 accelerometer 141, execution speed) are saved in a database 110. The analysis algorithm 160 and / or the movement algorithms and / or the Al 120 are involved with determining the correspondence, with tolerance ranges, between the data 141’ acguired in real time on the hospitalized patient and the predefined data saved in said database 110. In these embodiments of the present invention, the data 141’ and the data 104’ allow carrying out a double verification with regard to the intake of drugs, by way of a non-limiting or non-binding example, in pills, by a hospitalized patient. The double verification of this data allows having objective data (relative to whether the drugs were taken or not, and to the moment they were taken) that can be used for constructing ‘guestionnaires’ 106 adapted to reguest information from the patient on said objective data, and therefore have, through the comparison with the data 141’ and the data 104’, a measurement of the level of objectivity of the selfreported data and of the awareness and perception of the patient with respect to different dimensions (context, guantity, times) of the intake of the drugs.

[0240] Advantageously, according to the present invention, the sending of said ‘guestionnaire’ 106 to the patient 2 who is situated in the waiting room allows distracting him from the portable device 140 by making him concentrate on the ‘guestionnaire’ 106 and thus facilitating a more natural registration (i.e. less conditioned by the patient’s concentrating on the portable device 140 to which he / she might not be accustomed to using) of the data 141’, ensured by the unconscious movements which the patient 2 makes during the compilation of said ‘guestionnaire’ 106. The compilation of the ‘guestionnaire’ 106 can provide for the simultaneous activation of the audio unit 105 (microphone) and / or of the video unit 104 of the portable device 140 for a multidimensional evaluation of the same compilation.

[0241] As shown in FIG. 1, said portable device 140 interacts with the fixed device 150 in order to generate data 145’. In the second frame of FIG. 1 (bottom left), said patient 2 is shown within the the visit room of the ‘healthcare worker’ 1. Also in this case, the LEIBO. / 67e2025 devices 140 and 150 interact in order to generate data 145’. In addition, in this case, the presence of said units 104 and 105 allows the acquisition respectively of data 104’ and 105’ which are subsequently processed by said algorithm 160 for the analysis of verbal and extra-verbal language. As shown in FIG. 1, the data 106’ and 141’ directly affect the values of said ‘predictors’ 50, while the data 145’ affect said agendas 130. The results of the processing of the algorithm 160 and the updated values of the ‘predictors’ 50 are constantly acquired by said Artificial Intelligence 120 for the prediction of the ‘adherence’ of the patient 2 to a ‘procedure’ and of the ‘appropriateness’ of the same ‘procedure’, by way of a non-limiting or non-binding example, of the ‘adherence’ to a therapy and of the ‘appropriateness’ of the therapy itself.

[0242] A vast range of ‘predictors’ 50 is employed, also by means of automatic learning techniques, in order to measure, predict and update the ‘adherence’ in its different dimensions and, consequently, also in the dimension of the possible missed appointment and cancellation of an appointment, by a patient, with respect to the programmed ‘procedure’. The ‘predictors’ 50 can represent a great quantity of data relative to the patients and / or ‘caregivers’ and / or ‘healthcare workers’ and / or ‘healthcare facilities’ as well as to external data sources, not directly correlated with the patient and / or the ‘caregiver’ and / or the ‘healthcare worker’ and / or the ‘healthcare facility’, such as, by way of a non-limiting or non-binding example, weather conditions, road traffic conditions, etc.

[0243] Preferably, the ‘predictors’ 50 are variables indicative of demographic data, social data, psychological data, health data, location and planning. The demographic ‘predictors’ 50 comprise, for example, age, sex, civil state and size of the family and can be deduced, as a non-limiting manner by said data 104’, 105’ and / or 106’. The social ‘predictors’ 50 comprise, for example, insurance coverage, direct costs of the patient, financial constraints, ‘adherence’ to the previous visits and these can be deduced, as a non-limiting example, from the interactions registered between said LEIBO. / 67e2025 devices 140 and 150 and from said data 106’. The psychological ‘predictors’ 50 comprise, for example, attitude to bad news, attitude to treatments, relationship with the doctor and they can be deduced from said objective data (like the data 104’, 105’ and / or 141’) and said self-reported data 106’. The health ‘predictors’ 50 comprise, for example, the specialization of the ‘healthcare facility’ and / or of the ‘healthcare worker’, the type of visit, the change in the health state and the subjective health state and can be deduced from said objective data and said self-reported data. The location ‘predictors’ 50 comprise, for example, the location and the traffic conditions, the transport means, the weather conditions, the travel wait time. The planning ‘predictors’ 50 comprise, for example, date and hour of the appointment, tardiness in the scheduled appointment, time remaining before the appointment, planned wait time, several of which can be deduced from the interactions registered between the devices 140 and 150. Each ‘predictor’ 50 is associated, by said Al 120, with a relative probability and a relative weight for the patient.

[0244] The tracking of patients, ‘caregivers’ and / or ‘healthcare workers’ through said portable devices 140 and fixed devices 150, as well as the determination of the steps actually followed by a ‘healthcare process’, are operated according to the following steps:

[0245] - association of the identity of the patient, ‘caregiver’ and / or ‘healthcare worker’ with said portable electronic recognition device 140, provided, by way of a nonlimiting or non-binding example, with Bluetooth Low Energy technology;

[0246] - identification of said portable electronic recognition device 140 within a ‘healthcare facility’ by means of said plurality of fixed electronic recognition devices 150 provided with Bluetooth Low Energy technology. Each of said fixed electronic recognition devices 150 defines a distinct position in said ‘healthcare facility’;

[0247] - identification of the path of said patient, ‘caregiver’ and / or ‘healthcare worker’ within the said ‘healthcare facility’ by incorporating the detections of the portable LEIBO. / 67e2025 device 140 by each fixed device 150;

[0248] - identification of the timing (travel time) relative to said identified path;

[0249] - determination of the actual steps of the ‘healthcare process’ based on the detections of said fixed devices 150 and of the identification of the timing. In this step, a fixed device 150 which detects said portable device 140 at a first division of the ‘healthcare facility’ for time similar to only the passage, by way of a nonlimiting or non-binding example, of the patient, is not taken into account in determining an actual step of the ‘healthcare process’, but only in determining times. For example, the time employed between two successive detections of the portable device 140 of the patient, by the fixed devices 150, will be employed for evaluating the timing of the passage from a zone “overseen” by a fixed device 150 to another and an average of the timing values will be taken in order to carry out estimates that can also affect said appointment time slots. On the contrary, instead, a fixed device 150 which detects said portable device 140 at a second division of the ‘healthcare facility’ for an extended time comparable therefore to the stay time, still as an example, of the patient in said second division, is taken under consideration and scheduled as a step actually performed of the ‘healthcare process’. By way of a non-limiting or non-binding example, if said second division should be the radiology division, in the ‘healthcare process’ the actual step will appear for the visit at the radiology division, for example for carrying out an X-ray examination;

[0250] - comparison between said pre-established steps and said actual steps of the ‘healthcare process’, by way of a non-limiting or non-binding example, relative to said reserved ‘healthcare process’.

[0251] With reference to FIG. 2, another schematic representation of a system 100 according to the present invention is shown. The system 100, shown in FIG. 2, comprises said computerized apparatus 101 to which said back-end structure 102 and said front-end application 103 refer, and which comprises said database 110, said LEIBO. / 67e2025

[0252] Artificial Intelligence (Al) 120 and said agendas 130 of the ‘procedures’. FIG. 2 shows how a patient 2 with a portable device 140 is detected by the fixed devices 150 placed within a ‘healthcare facility’, and how these communicate position and time of said detection to said computerized apparatus 101.

[0253] In an example of operation of the system 100 according to the present invention, the interaction between portable device 140 and fixed device 150 allows a ‘healthcare facility’ and / or said Al 120 to know the state of crowding of a waiting room in which said fixed device 150 is placed. In addition, in this operating example, the interaction between portable device 140, fixed device 150 and agendas 130 allows a ‘healthcare facility’ and / or said Al 120 to know if a patient has arrived on time for an appointment and how many minutes before or after the time of the defined appointment has the patient been waiting in the waiting room. Advantageously, knowing if the patient is early or late for his appointment allows updating the framing thereof in his / her ‘characterization’ (see Definitions). For example, for a patient who is sometimes late, the Al 120 will update his / her frame in the behavioral ‘profile’, attributing a greater value to the percentage of correspondence to a behavioral ‘profile’ of a person who is generally disorganized and reducing the percentage value of correspondence to a behavioral ‘profile’ of a person who is generally punctual. In addition, in the event in which the patient is always late, notwithstanding an initial framing in a behavioral ‘profile’ of a punctual person, the Al 120 provides to reduce the percentage of correspondence of the patient to the ‘profile’ of a punctual person to a value egual to or close to 0; in the latter case, the Al 120 could generate a prediction of the ‘adherence’ of the patient to the ‘healthcare process’ that is lower than that initial. Advantageously, the ‘healthcare facility’ and / or the Al 120 can evaluate the time that a patient, just arrived in waiting room, must wait before his visit. Said evaluation is carried out based on the estimated average time for a visit and calculated based on the historic record of the visits carried out by the patients analyzing the average stay time of the portable devices 140 in the area of interest of the fixed device 150 placed LEIBO. / 67e2025 within the examination hall and clinical space. Advantageously, the Al 120 can advise a ‘healthcare facility’ to administer with a ‘healthcare worker’ or otherwise, or automatically (e.g. by digitally sending him or carrying out a call with synthesized voice), to a patient who is in the waiting room, and for which a wait time for example egual to 15 minutes has been estimated, a ‘guestionnaire’ 106 with a compilation time of 10 minutes. Preferably, said ‘guestionnaire’ 106 comprises guestions relative to the objective situation in which the patient is situated, as a non-limiting example: “Is the waiting room crowded?” and / or “Have you been waiting a long time for your turn?”. Said guestions, relative to the objective situation, are adapted to provide an indication of the attention and / or memory and / or perception of the patient and / or of the ‘caregiver’ and / or of the ‘healthcare worker’ (referred to the subjective perception) and contribute to refining the ‘characterization’ (see Definitions) thereof. The guestions, given the specific situation, as a non-limiting example, are selected by the Al 120 based on the variables deemed most suitable for optimizing said ‘characterization’ thereof, the evaluation of the expectations, of the health state and of the treatments underway and, finally, of the intentions.

[0254] In another operating example of the system 100, the exit of the patient from the clinic in which he / she has been subjected to a clinical test is registered by the interaction between said portable device 140 and various fixed devices 150. Advantageously, the Al 120 can advise a ‘healthcare facility’ to administer, with a ‘healthcare worker’ or otherwise, or automatically, a ‘guestionnaire’ 106 to said patient who has just exited the clinic, such ‘guestionnaire’ 106 containing guestions relative to said test just completed. In this example, having just exited from the clinic, the memories of the patient will tend to be more vivid and the responses more coherent with that which actually occurred. Also in this example, the ‘guestionnaire’ 106 can contain guestions with regard to objective facts, for which by way of example the data 104’ is available, in order to understand the attention and / or memory and / or the perception of the patient and / or of the ‘caregiver’ and / or of the ‘healthcare worker’ and refine the LEIBO. / 67e2025

[0255] ‘characterization’ thereof (see Definitions).

[0256] Said system 100 allows predicting and continuously and automatically updating the ‘adherence’ of a patient to a ‘healthcare path’, a ‘healthcare process’, a ‘healthcare procedures’ on the basis of objective data (derived from the analysis: by the first algorithm, the geolocation data, interaction, 145’ and data extracted from the ‘log files’; by the algorithm 160, of the data 104’ and 105’; by the movement algorithms of the data 141’) and of subjective / self-referred data (such as the data 106’). As a function of the evaluation of the ‘adherence’ of the patient and of the ‘characterization’ of the patient, ‘caregiver’ and / or ‘healthcare worker’, said Al 120 varies the types and / or the values of the attributes of the ‘contact framework’ with the patient, ‘caregiver’, ‘healthcare worker’ and / or ‘healthcare facility’, in order to increase the value of said ‘adherence’ and, conseguently, in order to improve the outcome of a ‘healthcare path’, a ‘healthcare process’, a ‘healthcare procedures’ of the specific patient.

[0257] The use of the devices 140 and 150, by way of a non-limiting or non-binding example, with Bluetooth Low Energy technology, allows the collection of data in real time, exceeding the limits of the manual collection of the data reguired by other solutions. On such matter, the ‘healthcare process’ management system 100 is attained, by way of a non-limiting or non-binding example, by means of Bluetooth Low Energy technology, i.e. by means of a wireless technology personal area network with the object of providing an energy consumption and a cost that is considerably reduced, maintaining a communication interval similar to the conventional Bluetooth technology.

[0258] The ‘healthcare process’ management system 100 is therefore adapted to identify the path and the timing relative to the path of the patient within the ‘healthcare facility’ based on the portable electronic recognition device 140, and to determine the actual steps of the ‘healthcare process’ based on the path and on the timing identified, comparing the pre-established steps with the actual steps of the LEIBO. / 67e2025

[0259] ‘healthcare process’.

[0260] According to one embodiment of the present invention, said fixed electronic recognition devices 150 are sensors of Beacon type.

[0261] According to one embodiment of the present invention, like that shown in FIG. 2, said portable electronic recognition device 140 is a sensor of Beacon type. When a patient carries out the check-in to a ‘healthcare facility’, he / she is given a Beacon portable device 140, to bring with him up to check-out. While the Beacon portable device 140 is with the patient, information is collected, in real time, by way of a non-limiting or non-binding example, such as: participation in the reserved ‘healthcare procedure’, check-in, interested time slot, patient punctuality, time spent in the waiting room, punctuality of the doctor, duration of the treatment, date and hour of the check-out. Due to the fixed electronic recognition devices 150 (or gateways) positioned along the path in the ‘healthcare facility’, in fact, the patients’ steps can all be traced.

[0262] The data collected by means of devices 140 and 150 is analyzed with the objective of bringing to light possible inefficiencies in the mode in which the ‘healthcare path’ of the patient is planned and managed. In such a manner, by way of a non-limiting or non-binding example, by means of the Bluetooth Low Energy technology or UWB and the association of the identity of the patient with a portable electronic recognition device 140, the movements of the patient inside the ‘healthcare facility’ can be automatically tracked and, conseguently, it is possible to extract, always automatically, the steps of the ‘healthcare process’ attained by the patient himself / herself and verify the ‘adherence’ with the steps of the reguested ‘healthcare process’.

[0263] With reference to FIG. 5, a patient 2 is shown who compiles a ‘guestionnaire’ 106 and is monitored by means of the system 100. As mentioned, the system 100 also allows monitoring the ‘patient self-governing procedures’; in FIG. 5, specifically, a patient 2 is shown who, on his own, responds to a ‘guestionnaire’ 106 through a front-end application 103 that is available on a person tablet with a video camera (video LEIBO. / 67e2025 acquisition unit 104); said patient 2, by compiling said ‘questionnaire’ 106 and allowing the acquisition of the data 104’ by means of said video acquisition unit 104, contributes to his / her ‘characterization’ (see Definitions) by the computerized apparatus 101.

[0264] In several preferred embodiments of the present invention, said video 104 and / or audio 105 acquisition units acquire data 104’ and / or 105’ relative to said patient 2 and to a ‘caregiver’ thereof.

[0265] In several preferred embodiments, said ‘questionnaires’ 106 are simultaneously administered to said patient 2 and to said ‘caregiver’ in order to acquire data 106’ directly from said patient 2 and said ‘caregiver’. By way of a non-limiting or nonbinding example, the questions of the ‘questionnaires’ 106 can be different between patient 2 and ‘caregiver’, in particular if one intends to verify, on one hand, the judgement of the ‘caregiver’ on the state, also psychophysical, of the patient 2 and, on the other hand, the support that the same ‘caregiver’ can offer to said patient 2 and the relative effect on the ‘adherence’. Said data 106’, acquired by the ‘caregiver’, adapted to provide an indication of the congruity / coherence of the data 106’ acquired by the patient 2 with respect to other data acquired by said patient 2 or, in any case, an indication of the possible alteration of the perception of the patient 2 due to its psychophysical state at the time of the compilation of the ‘questionnaire’ 106.

[0266] With reference to FIG. 6, a scheme is shown representative of the process for generating a ‘questionnaire’ 106 in an automated manner.

[0267] In several embodiments of the present invention, like those shown in FIGS. 6-7, said ‘questionnaires’ 106 are generated through artificial intelligence and / or chatbot services 601 which use algorithms of Artificial lntelligence / ‘machine learning’ / deep learning like the chatbot GPT services by Open Al®.

[0268] Said chatbot 601, with an Artificial Intelligence, automatically generate said ‘questionnaires’ 106, on the basis, by way of a non-limiting or non-binding example, LEIBO. / 67e2025 of data 141’ and / or 145’, and more generally on the basis of objective and selfreferred data, on the basis of data 602 of the patient contained in said database 110 and of the ‘adherence’ 30 calculated for the patient. Specifically, said chatbot 601 are adapted to provide a ‘questionnaire’ 106 on the basis of a textual input 610 which, in various embodiments of the present invention, can be generated by a ‘healthcare worker’ and / or, automatically, by the Al 120 of the computerized apparatus 101, including, by way of a non-limiting or non-binding example, data relative to the clinical condition and to the ‘characterization’ (see Definitions) of the patient, the desired compilation times and / or the desired number of questions of the ‘questionnaire’ 106 and / or introductory questions and / or questions for which there are already responses provided by the objective data such as, by way of a nonlimiting or non-binding example, the data 104’, 105’ and / or 141’. Said input 610, automatically provided by said Al 120 to said chatbot 601 (such as for example said chatbot GPT by Open Al®), is generated and / or updated instantaneously by requiring the writing of a ‘questionnaire’ 106 with a compilation time lower than a specific time and / or a desired number of questions and / or specific questions.

[0269] In several embodiments of the present invention, the chatbots 601 further generate text outputs and / or audio outputs (synthesized) for “exchanges of questions and responses” (see ‘Communication’ in ‘Contact framework’, Definitions section), for example “information questions” (i.e. “Am I disturbing you? Shall we be in touch at another time?”), in order to exchange “factual information” (see ‘Communication’ in ‘Contact framework’, Definitions section) (i.e. “You should approach the waiting room since it will shortly be your turn”), “opinions and view points” (see ‘Communication’ in ‘Contact framework’, Definitions section) (i.e. “Seems this is the first time you have had this symptom, perhaps you should speak about this directly with the doctor”), “personal emotions and experiences” (see ‘Communication’ in ‘Contact framework’,

[0270] Definitions section) (i.e. “I am quite relieved to hear that you no longer have any pain”), “social and courtesy exchanges” (see ‘Communication’ in ‘Contact LEIBO. / 67e2025 framework’, Definitions section) (i.e. “I thank you very much for this conversation”). The chatbot 601 then associates, as a function of the ‘characterization’ (see Definitions), the generation of the ‘communication’ of the ‘contact’ with each of the ‘participants’ and continuously updates it as a function of the ‘adherence’ of the patient.

[0271] In several embodiments of the present invention, said ‘questionnaire’ 106 is administered in the form of a “motivational interview” i.e. a motivational interview adapted, by way of a non-limiting or non-binding example, to investigate the perception of the patient with regard to an ambivalent position. By way of a nonlimiting or non-binding example, a patient with an ambivalent position, afraid to start a therapy but who also wishes to start it in order to feel better, receives a motivational interview in the form of a ‘questionnaire’ 106 in which the possible side effects are better illustrated of the therapy to be started together with the probability that they will be verified, facilitating the awareness of the motivations underlying said ambivalent position and the selection of a position.

[0272] FIG. 6 shows how from the data 145’, extrapolated from said database 110, an average wait time 645 is acquired, for example in a waiting room. In addition, said average wait time 645 is also influenced by the extracted from the agendas 130, evaluating if a patient 2 is early, on time or late. Where said data 145’, whose acquisition is shown with a dashed arrow in FIG. 6, is relative to the measurements, preferably specific for said waiting room, of the interactions between the devices 140 and 150 of all the patients 2 who have used and use the system 100. The Al 120 acquires said average wait time 645 and requests a ‘questionnaire’ 106 from said chatbot 601, with a compilation time 646 lower than said average wait time 645. At this point, from said database 110, said Al 120 acquire said value of ‘adherence’ 30 for said patient 2. When said ‘adherence’ 30 is below a fixed (fixed / defined by an administrator of the system 100) minimum threshold value 630 (e.g. equal to 70%), said Al 120 requests said chatbot 601 to insert in said ‘questionnaire’ 106 questions 650 regarding the LEIBO. / 67e2025 measurements pursuant to data 145’ and / or 141’, such as for example asking said patient if he / she has been in the waiting room for a long time or if the waiting room is crowded (information that can be confirmed from the data 145’) and / or if he / she is very agitated (information which can be confirmed by the data 141’). From said database 110, said Al 120 extracts also data 602 relative to the patient 2 such to be able to include in said input 610 data relative to the age of the patient and his / her pathology, to his / her clinical situation and to the type of visit to be carried out. Said input 610 provided to said chatbot 601 allows generating said ‘guestionnaire’ 106.

[0273] With reference to FIG. 7, a scheme is shown that is representative of a process for generating an input 610 in an automated manner.

[0274] Said chatbots 601 are adapted to provide a ‘guestionnaire’ 106 on the basis of said text input 610, in which the compilation time 646 is influenced by data 105’ acguired from the audio acguisition devices.

[0275] In several embodiments of the invention like that shown in FIG. 7, the patient 2 receives an audio call by a ‘healthcare facility’ (e.g. secretary operators of the organization), by a ‘healthcare worker’ 1 or by a programmable call apparatus 701 adapted to carry out calls automatically when it receives an input from said Artificial Intelligence 120. Said programmable call apparatus 701 is a software or the set of reguested software / hardware for carrying out “robocalls”, by way of a non-limiting or non-binding example, the CallMaker service by Call Maker Dialer®. Said Artificial Intelligence 120 can send an input to said programmable call apparatus 701 in order to carry out a call to said patient 2. Said input is sent upon analysis of the data 602 of the patient 2 present in said database 110 and reporting, among the other data, the hours when the patient 2 is not available (e.g. working hours of the patient). The first guestions carried out by said programmable call apparatus 701 are adapted to verify the availability of said patient 2 to respond to a ‘guestionnaire’ 106 and to understand the setting where he / she is situated. Said algorithm 160, by analyzing the environmental noises 702 of the call, can in fact understand if an interlocutor LEIBO. / 67e2025

[0276] (such as for example the patient 2) is situated outside or inside a closed place and if he / she is in the presence or not in the presence of other people. Said analysis of the environmental noises 702 and / or the analysis of objective data made by the first algorithm, by way of a non-limiting or non-binding example, of the interaction data between portable devices 140 adapted to provide data of face-to-face ‘contacts’ inside and outside a ‘healthcare facility’, being employed to define the availability of a ‘patient’ and / or ‘caregiver’ and / or ‘healthcare worker’ to proceed with a ‘contact’, i.e. the possibility to proceed with the ‘contact’ or to resend said ‘contact’. Said analysis of said algorithm 160 of said environmental noises 702 and / or said analysis of said first algorithm of said objective data defining parts of said text input 610 relative to the compilation time and / or to the number of guestions and / or to the guestions of said guestionnaire 106. Said parts of said text input 610 being updated in real time based on the modification of the environment context where said interlocutor, e.g. the patient 2, is situated, by way of a non-limiting or non-binding example, on the basis of new data of face-to-face ‘contacts’ provided from the interaction data between portable devices 140. When the patient 2 is situated outside in a crowded place 710 or inside a crowded place 720, the ‘guestionnaire’ 106 is “resent” in order to preserve the privacy of the patient 2 himself / herself, communicating to the patient 2 that he / she will be recontacted, acguiring preferably available time slots and updating said data in said database 110 and in the memory of said programmable call apparatus 701. Said algorithm 160 also processes the responses 703 of said patient 2, which can allow (Y) or not allow (N) responding to the ‘guestionnaire’ 106 at the time of the call. When the patient 2 does not allow (N) responding to the ‘guestionnaire’ 106, this is “resent”, asking the patient 2, by way of a non-limiting or non-binding example, two times and days when he / she is available to be recontacted and updating said data in said database 110. When the patient 2 allows (Y) responding to the ‘guestionnaire’ 106, said algorithm 160, from the analysis of the environmental noise 702 determines the ‘environmental contexts’ where the LEIBO. / 67e2025 patient 2 is situated and, consequently, regulates the time of compilation 646 of the ‘questionnaire’ 106. When the patient 2 is situated in an external place that is not crowded 711, said compilation time 646 is brought to below the thresholds that are manually programmable and / or definable by said Al 120, in these cases the compilation time 646 will preferably be comprised between 30s and 180s, in order to allow asking, as a non-limiting and non-binding example, one or two questions to the patient 2. When the patient 2 is situated in a closed uncrowded place 721, said compilation time 646 is regulated based on the wait time 645 and / or based on the ‘characterization’ (see Definitions) of the patient 2 and the input 610 is provided (in the previously described modes) such that the chatbot 601 attains the ‘questionnaire’ 106. In the case shown in FIG. 7, said input 610 is textual and software Text To Speech is employed for converting the questions processed by the chatbot 601 for the ‘questionnaire’ 106 and for its administration by means of voice call by the programmable call apparatus 701.

[0277] In several embodiments, the ‘healthcare facility’ comprises a contact center with human operators. Said contact center is adapted to register the call by using an audio acquisition unit 105 and it transfers the data 105’ to the computerized apparatus 101. On the basis of said data 105’, the algorithm 160 carries out an analysis of the ‘contact’ between operator of contact center and patient. In the cases in which said algorithm 160 and said Al 120 determine, from the analysis of said data 105’, a poor affinity between patient and said operator of contact center, the computerized apparatus 101 sends notifications to the ‘healthcare facility’ (in the example, to the contact center) in order to prevent the same operator in the future from being interfaced with the patient with whom there was little affinity. The analysis operated by said algorithm 160 and said Al 120 can be expressed in various graphical and / or textual representation forms (also by means of “summary”) and can be made available, by way of a non-limiting or non-binding example, automatically and in real time on said computerized apparatus 101 and sent to said ‘healthcare facility’ (in the LEIBO. / 67e2025 example, to the contact center of the ‘healthcare facility’) and / or to said contact center operator, so as to underline the critical nature of the ‘contact’, facilitating, through operating suggestions and indications, the attainment of the ‘objectives’ of the ‘contact’.

[0278] With reference to FIG. 3, a block diagram is shown that is representative of a method 200 for managing the ‘healthcare processes’ according to the present invention.

[0279] The method 200 for managing the ‘healthcare processes’ shown in FIG. 3 comprises the steps of:

[0280] - one or more steps of acguisition and analysis 201 of objective data and / or selfreported data; said objective data and / or self-reported data being analyzed by said first analysis algorithm, by said algorithm 160, by said movement algorithms and / or by said Al 120 in order to automatically attribute to a patient, one or more ‘caregivers’ and / or a ‘healthcare worker’, a ‘contact framework’ with the relative attributes and to automatically attribute ‘personality states’ and / or ‘profiles’ and / or ‘states of change’ for said patient, said ‘caregiver’ and / or said ‘healthcare worker’; said attributes of the ‘contact framework’ can also be paired with the ‘healthcare facility’;

[0281] - one or more steps of selection 202 of a ‘contact framework’, by means of ‘machine learning’, in which types and / or values of the attributes of a ‘contact framework’ are selected that are suitable for the patient, ‘caregiver’, ‘healthcare worker’ and / or ‘healthcare facility’, by said database 110, based on the objective and / or self-referred data acguired and continuously updated for said patient, ‘caregiver’, ‘healthcare worker’ and / or ‘healthcare facility’ in the ‘contacts’ completed with said patient, ‘caregiver’, ‘healthcare worker’ and / or ‘healthcare facility’ and / or based on the values of the ‘predictors’ and / or based on the ‘characterization’ (see Definitions) of the patient, of the ‘caregiver’ and / or of the ‘healthcare worker’.

[0282] Said selection step 202 provides for a first identification (in the first “iteration” of the cycle shown in FIG. 3) of the ‘contact framework’ by means of statistical data LEIBO. / 67e2025 and then provides for said identification on the basis of the objective and / or selfreferred data. The selection step 202 indeed provides for, following ‘contacts’ with the patient, ‘caregiver’, ‘healthcare worker’ and / or ‘healthcare facility’, an updating of said ‘contact framework’ on the basis of collected objective and selfreferred data and on the basis of relative processing operated by said first algorithm, by said algorithm 160 and by said movement algorithms during said ‘contacts’;

[0283] - one or more adeguacy prediction steps 203 of the ‘contact framework’ selected in said steps of selection and updating 202, in which a probability of adeguacy is calculated by employing said objective self-referred data and / or said ‘predictors’ 50; said adeguacy prediction steps 203 of the ‘contact framework’ being attained by employing said Artificial Intelligence (Al) engine 120; said probability of adeguacy being assigned to each attribute of the ‘contact framework’;

[0284] - one or more steps of adeguacy verification 204 of the ‘contact framework’, by means of ‘machine learning’; said steps of adeguacy verification 204 provide for confirming said probability of adeguacy and / or the updating thereof for each attribute of the ‘contact framework’;

[0285] - one or more steps of verification of the outcomes 205 of the ‘contact’ and of the actions following said outcomes of the ‘contact’; said steps of verification of the outcomes 205 of the ‘contact’ and of the conseguent actions are attained on the basis of the objective and self-referred data and by employing said Artificial Intelligence (Al) engine 120 and / or said first algorithm and / or said algorithm 160 and / or said movement algorithms;

[0286] - one or more steps of outcome forecast 206 of said identified ‘procedure’, in which a probability of positive or negative outcome of the identified ‘procedure’ is calculated by employing said ‘predictors’ 50 associated with the said patient and / or said ‘caregiver’ and / or said ‘healthcare worker’ and / or said ‘healthcare facility’. By “outcome” of the ‘procedure’, it is intended herein the result, whether LEIBO. / 67e2025 positive or negative, in terms of health of the patient which, following the ‘procedure’, can be improved, unchanged or worsened. Said “outcome” is a function of the ‘appropriateness’ of the ‘procedure’ (and of the ‘guidelines’ employed for said ‘procedure’) and of the ‘adherence’ of the patient to the ‘procedure’ itself. The probability (whether positive or negative) of “outcome” of the identified ‘procedure’ is calculated by employing said ‘predictors’ 50 associated with said patient and / or said ‘caregiver’ and / or said ‘healthcare worker’ and / or said ‘healthcare facility’ in which each ‘predictor’ 50 has a relative probability (intended as the probability that said ‘predictor’ 50 has a certain relevance for the patient) and a weight (intended as incidence or value that the single ‘predictor’ 50 has with with respect to the other ‘predictors’ 50).

[0287] By way of a non-limiting or non-binding example, a relative probability percentage is calculated by using technigues, like the “importance feature” or the correlation coefficient, to determine how much each ‘predictor’ 50 contributes to the forecast of the outcome. A higher percentage of probability indicates that the ‘predictor’ 50 is more influential in the analysis.

[0288] By way of a non-limiting or non-binding example, the weight or incidence is calculated by using technigues such as the regression coefficients or the weights assigned by the ‘machine learning’ models. A higher weight also indicates that the ‘predictor’ 50 has a greater incidence on the outcome with respect to the other ‘predictors’ 50 in the model.

[0289] Preferably, the forecast step 206 is conducted by employing said Artificial Intelligence (Al) engine 120 adapted to identify the probability of ‘adherence’ and modify the relative probabilities or relative weights for each ‘predictor’ 50 for each patient and to add or eliminate one or more ‘predictors’ 50 in said database 110. Preferably, the Artificial Intelligence (Al) engine 120 employs automatic learning algorithms and algorithms for recognition and / or processing of the natural language (algorithm 160) and extra-verbal language in order to modify the relative LEIBO. / 67e2025 probabilities or the relative weights for each ‘predictor’ 50 for each patient and to add or eliminate one or more ‘predictors’ 50.

[0290] Said forecast step 206 is preferably automated by extrapolating data from the ‘contacts’, by way of a non-limiting example, with the patient, independent of the type of ‘communication’ or ‘channel’ employed.

[0291] As already specified above, the statistical representation of the probabilities of each ‘predictor’ 50, i.e. the relative weight, is to be intened as exemplifying but not limiting of the types of evaluation of the probabilities, which can also be non- statistical, such as for examle q ualitative evaluations;

[0292] - one or more steps of verification 207 of “outcome” of the identified ‘procedure’, carried out and / or being completed, by means of ‘machine learning’. As a function of the detection, by way of a non-limiting example, of the conditions of the patient, the method 200 provides for the assignment of scores to urgency indicators and ‘appropriateness’ of the ‘procedure’ and their updating as well the assignment of a score capable for the ‘adherence’ of the patient to the ‘procedure’ and relative sub-steps on the basis of the ‘predictors’ 50 and of the acguired objective and selfreferred data. In such a manner, it is possible to measure, verify and correlate the outcome of the ‘procedure’ with the actual level of ‘adherence’ of the patient. Said steps of verification 207 comprising the following substeps: o registration and monitoring of the data of prescription, of the signs, of the symptons and clinical data of said patient; o assignment of a cumulative score or distinct scores to urgency indicators and ‘appropriateness’ of said identified ‘procedure’ for said patient; o assignment of a value of ‘adherence’ of said patient to the identified ‘procedure’, as evaluated based on the data collected in the preceding steps.

[0293] - one or more steps of updating 208 said database 110 with types of ‘characterizations’ (see Definitions) of the patients, ‘caregivers’ and / or ‘healthcare workers’ and / or types and values of attributes of the ‘contact LEIBO. / 67e2025 framework’ for said patients, ‘caregivers’, ‘healthcare worker’ and / or ‘healthcare facilities’ as a function of the ‘predictors’ 50 and objective and self-referred data (such as by way of example the data 104’, 105’, 145’ and / or 141’) acquired, in which a probability is assigned for suitability to said patients, said ‘caregivers’ and / or said ‘healthcare workers’ relative to said ‘characterizations’ (see Definitions) and said attributes of the ‘contact framework’; types and values of the ‘contact framework’ can also be paired with the ‘healthcare facility’; said steps of updating being attained by employing said Artificial Intelligence (Al) engine 120.

[0294] The database 110 of the ‘time contexts’ (time slots) of the ‘healthcare procedures’ is typically defined as a function of the agendas 130 and hence of the organization and of the ‘location contexts’ of said ‘healthcare procedures’, of the planning of the activities as well as of the availability of the ‘healthcare workers’ and of the reservations already in place, being changeable as a function of the availability variables of the ‘healthcare workers’ and of the actual completion of the remote ‘healthcare procedures’.

[0295] A reservation step of an appointment according to the present invention can be attained, for example, by means of a ‘healthcare process’ management system in which a computerized system comprises the aforesaid database 110 of ‘location contexts’ and of ‘time contexts’ (time slots) of the ‘healthcare procedures’ and a back-end structure 102 is provided for the ‘healthcare workers’, for example in the form of computerized electronic devices capable ofbeing connected to a computerized apparatus 101, for the modification of the ‘location contexts’ and of the ‘time contexts’ (time slots) available to ‘healthcare procedures’ that can be completed. Furthermore, a front-end application 103 can be provided to patients, for example in app form for computerized mobile devices, for the reservation and management of the ‘healthcare procedures’ as well as for completing further steps of the method 200, object of the present invention, according to that described in more detail hereinbelow. LEIBO. / 67e2025

[0296] Different architectures can also be provided as a function of the technical needs for implementation of the process, object of the method 200 according to the present invention.

[0297] As described above, the aforesaid database 110 of ‘characterizations’ (see Definitions) and of types and values of the attributes of the ‘contact framework’ can be opertively connected with a computerized apparatus 101 of a ‘healthcare process’ management system 100, possibly joined in a single database 110. The front-end application 103 of the patient can, in several embodiments of the invention, allow only partial access to one of the databases 110 in order to implement the preference defined by the patient himself / herself.

[0298] Considering that there are different causes for low ‘adherence’ correlated with the planning process (from here, the usual activities and / or operating reminder modes for minimizing no-show causes, overbooking for minimizing no-show effects, open access for avoiding no-shows, etc.) but that there can be many other causes and correlations of said low ‘adherence’, the present invention also considers and manages the steps of pre-programming as well as those of post-programming and post-dispensing, as detailed hereinbelow.

[0299] The process for selecting the ‘healthcare facility’ and / or of the ‘healthcare worker’ (pre-programming) allows minimizing the causes of no-show relative to emotional factors, guality of the service and suitability of the therapies, which can therefore be effectively managed by selecting the most suitable ‘healthcare facility’ and / or the ‘healthcare worker’. In such a manner, in addition to the suitable ‘communication modes’, it is possible to identify the ‘healthcare worker’ most suitable for increasing the ‘adherence’ of the patient to the ‘healthcare procedures’.

[0300] The described probabilities of suitability, statistically evaluated, are to be intended as representative and exemplifying of a form of evaluation of the same, without being limiting for the invention, which can provide for different forms of evaluation, even of non-statistical type, such as for example qualitative evaluations. LEIBO. / 67e2025

[0301] A combination of statistical and automatic learning techniques is used for identifying the aforesaid elements as well as, as described in more detail hereinbelow, for identifying the patterns, intentional or otherwise, which lead to a lack of ‘adherence’ to the ‘healthcare procedures’.

[0302] In several preferred embodiments of the present invention, said objective and / or self-referred data, by way of a non-limiting or non-binding example, said data 104’, 105’ and 106’ are used for assigning, to a ‘healthcare worker’ 1, one or more ‘characterizations’ (see Definitions). In these embodiments of the present invention, ‘questionnaires’ 106 are also proposed to said ‘healthcare worker’ 1 so as to obtain data 106’ for a ‘characterization’ (see Definitions).

[0303] The ‘characterization’ of the ‘healthcare worker’ 1 is defined by a certain percentage of suitability with the ‘characterization’ of the patient 2. When said suitability percentage exceeds a level definable by an administrator of the servize and / or by said Al 120, said ‘healthcare worker’ 1 results associable with the patient 2 and can be selected during said step of selecting 202 the contact framework.

[0304] Preferably, the steps of updating 208 are attained by employing said Artificial Intelligence (Al) engine 120 adapted to identify or modify said probabilities of suitability and to add or eliminate, in said database 110, ‘characterizations’ (see Definitions), ‘personality states’, ‘profiles’, ‘states of change’ and / or types and values of the attributes of the ‘contact framework’. The Artificial Intelligence (Al) engine 120 is adapted to identify said probability of said patient belonging to a ‘characterization’ (see Definitions) and the probability of ‘adherence’ to the identified ‘procedure’ and to modify said relative probabilities or said relative weights for each ‘predictor’ 50 for each patient and to add or eliminate, in said database 110, one or more ‘predictors’ 50. The Artificial Intelligence (Al) engine 120 identifies a level of satisfaction of said patient through the data 106’ and the processing conducted by the first algorithm on the geolocation data, interaction, ‘log files’ and on the data 145’, by the algorithm 160 on the data 104’ and 105’ and by the movement algorithms on the data 141’. LEIBO. / 67e2025

[0305] Preferably, said Artificial Intelligence (Al) engine 120 employs automatic learning algorithms, for recognition and / or processing of the natural language and extraverbal language (such as for example said algorithm 160) in order to extract and process the information adapted to identify or modify said probabilities of suitability and / or of ‘adherence’, to add or eliminate, in the database 110, one or more ‘characterizations’, types of ‘personality states’ and / or ‘profiles’ and / or ‘states of change’ and / or types and values of attributes of the ‘contact framework’ and / or to modify said relative probabilities or said relative weights for each ‘predictor’ 50 for each patient and to add or eliminate, in the database 110, one or more ‘predictors’ 50. When a patient reguests an appointment and / or when the patient is prescribed ‘healthcare procedures’, the ‘healthcare process’ management system 100 implements the method 200 described above, defining a list of ‘healthcare workers’ who can better meet the needs and preferences of the patient, needs and preferences which can be explicitly expressed and / or dynamically deduced based on the objective and self-referred analyzed data, such as for example data 104’, 105’, 106’ and 141’, as described above.

[0306] Post-programming and post-dispensing evaluations of the ‘healthcare procedures’ or downstream of the relative canceling or no-show, are necessary for evaluating the quality with respect to the steps for scheduling and completing of said ‘healthcare procedures’ and the effectiveness of the further prescribed and followed ‘procedures’, as is perceived by the patient.

[0307] In several embodiments of the present invention like that shown in FIG.4, the method 200 for managing the ‘healthcare processes’ comprises one or more steps of analysis 209 of the ‘procedure’ reguested, in which a level of satisfaction of said patient is identified through said data 106’ of ‘guestionnaires’ 106, comprising ad hoc guestions regarding the level of satisfaction, and the processing of the first algorithm, of the algorithm 160 and of the movement algorithms on the objective and self-referred data determined during the ‘contacts’ with said patient. LEIBO. / 67e2025

[0308] During the method 200 for managing the ‘healthcare processes’, there is a continuous updating of agendas 130 of the ‘procedure’ by means of ‘machine learning’ technigues. By agendas 130 of the ‘procedures’, it is intended sets of data regarding ‘location contexts’ and ‘time contexts’ (time slots) compatible with said ‘healthcare worker’. Said agendas 130 of the ‘procedures’ comprising locations and time slots compatible with said ‘location context’, with said ‘time context’ (appointment time slot) and with said ‘healthcare worker’, identified for said patient. Said agenda 130 adapted to show said ‘healthcare worker’ all the appointments with all the patients and all the relative automatic updating. Said agenda 130 showing said patient only the slot relative to its appointment and hiding the names of other patients.

[0309] Said agendas 130 are automatically updated by said computerized apparatus 101 based on the interaction data, the data 145’ and said data 104’ and / or 105’. Specifically, said computerized apparatus 101 updates, in said agenda 130, an appointment relative to a ‘procedure’ as “concluded” or as no-show when a patient results or does not result - from the interaction data and / or from the data 145’ and / or data 104’ and / or 105’ - to have entered and exited from the visit room of a hospital, division, clinic or other. In this manner, when a first patient exits from a visit room before the end of his / her slot, said agenda 130 automatically updates the state of the visit as “concluded”. When a second patient, who has an appointment in the slot following that of the first patient, results, from the interaction data and / or from the data 145’ and / or from the data 104’ and / or 105’, already in the waiting room, the same is notified on said front-end application 103 and / or on the portable device 140 of the possibility of entering directly in the visit room; when said second patient, results, from the interaction data, from the data 145’ and / or from the data 104’ and / or 105’, far from the waiting room or from the ‘healthcare facility’, a third or subseguent patient, who has an appointment in the subseguent slot following that of the second or another patient and which results from said interaction data and / or LEIBO. / 67e2025 said data 145’ and / or said data 104’ and / or 105’ in the waiting room, is notified on said front-end application 103 and / or on said portable device 140 and invited to directly access said visit room. Alternatively, said second or subsequent patient receives a notification on said front-end application 103 and / or on said portable device 140 in which he / she requests the availability to arrive before his / her appointment, based on the early exit of said first or subsequent patient. Said notification in which the availability is requested to arrive earlier at the appointment being sent on said front-end application 103 and / or on said portable device 140 also on the basis of the forecast of said no-show and on the time of said early exit of said first or subsequent patient, carried out by said algorithm 160 and / or said Al 120 on the data 104’ and / or 105’ and / or on the interaction data and / or 145’ of said first or subsequent appointment. When said second or subsequent patient directly accesses the visit room or provides his / her availability to arrive early, said agenda 130 is updated with the new start and end hours of the appointment for said second or subsequent patient and, in time order, all the patients with time slots following that of said second or subsequent patient receive an analogous notification on said frontend application 103 and / or on said portable device 140. When said second or subsequent patient does not directly access the visit room or denies his / her availability to arrive early, said agenda 130 remains unchanged. Analogously, when said first patient exits from the visit room after the termination of the time slot previously assigned thereto, said second patient and all the subsequent patients receive, on said front-end application 103 and / or on said portable device 140, a notification relative to the tardiness. Said late notification being sent on said frontend application 103 and / or on said portable device 140 also on the basis of the forecast of said tardiness carried out by said algorithm 160 and / or said Al 120 on the data 104’ and / or 105’ and / or on the interaction data and / or 145’ of said first or subsequent appointment.

[0310] The structure and the updating of the agendas 130 is articulated in the following LEIBO. / 67e2025 steps:

[0311] - forecast and updating of the demand of the ‘procedure’ as a function of the historical data also including data that is continuously acguired / updated (ongoing data) in the management of said ‘procedures’ reguested and / or completed;

[0312] - identification, planning and updating of the offer of the ‘procedure’ suitable for meeting the forecast of the demand;

[0313] - identification, classification and updating of the different ‘characterizations’ (see Definitions) of the patients as a function of the ‘adherence’ data acguired in the management, historical and continuously updated, of said ‘procedures’;

[0314] - identification, classification and updating of the set of ‘location contexts’ and of ‘time contexts’ (types and time slots) suitable for completing said offered ‘procedures’, for the ‘characterizations’ (see Definitions) of the patients and for the planning of the offer as a function of the data, historical and continuously updated, also of ‘adherence’, acguired in the management of said ‘procedures’ and / or preferences, explicit or implicit, of the ‘healthcare workers’ and / or of the patients;

[0315] - planning and updating of the set of the agendas 130, i.e. of the ‘location contexts’ and of the ‘time contexts’ (types and time slots) of the ‘procedures’, identified as a function of the data acguired in the management, historical and continuously updated, of said ‘procedures’, of the set of ‘location contexts’ and of ‘time contexts’ (types and time slots) as well as of the availability of the ‘healthcare workers’ and of the actual completing of said ‘procedures’ and of different probabilities of ‘adherence’ of the patients to the ‘location contexts’ and to the ‘time contexts’ (types and time slots) assigned.

[0316] Finally, it is clear that modifications, additions or variations that are obvious for a man skilled in the art can be made to the invention described up to now, without departing from the protective scope that is provided by the enclosed claims.

Claims

LEIBO. / 67e2025Claims1. Healthcare process management system (100), with at least a healthcare worker and at least a patient; said system (100) forecasting, promoting, measuring, evaluating, optimizing, and continuously updating said patient’s adherence to one or more healthcare processes; said system (100) comprising:- at least a computerized apparatus (101) suitable of hosting one or more databases (110), one or more Al (120) and one or more agendas (130) of procedures; said computerized apparatus (101) being managed and accessible with a back-end structure (102) and a front-end application (103); said Al (120) being trained on the basis of historical data present in the databases (110) and / or on the basis of data present in scientific literature; said system (100) characterized in that it comprises:- at least a portable device (140) associated with said patient and / or one or more caregivers and / or said healthcare worker to electronically track their position, path and contacts inside and outside a healthcare facility and the times thereof; said contacts being interactions between at least two participants; said interactions being face-to-face and / or through paper, electronic or digital means; said participants comprising said patient, caregiver, healthcare worker and / or automated computer systems; said portable device (140) comprising a geolocation unit suitable of providing geolocation data and / or a wireless communication unit suitable of providing interaction data between multiple portable devices (140); the mutual interaction between said portable devices (140) with wireless technology or the interaction between said portable devices (140) and said geolocation unit providing face-to-face contact data inside and / or outside a healthcare facility;- log files relating to contacts of said patient and / or one or more caregivers and / or a healthcare worker and / or a healthcare facility; said log filesLEIBO. / 67e2025 comprising one or more types and / or values of one or more attributes of a contact framework associated with said patient and / or caregiver and / or healthcare worker and / or healthcare facility; said contact framework being a partial or total set of attributes characterizing one or more contacts; said attributes being types and actions to manage said adherence of said patient, one or more participants, one or more objectives of said contact, one or more rules of a contact, types and contents of said contact, communication modes during a contact, types and means of communication of said contact, types and levels of relationship between said participants, relationship dynamics between said participants, types and levels of involvement of said participants, one or more durations of said contact, location context of said contact, time context of said contact;- at least a first algorithm; said first algorithm is used to analyze geolocation data, interaction data and data extracted from said log files;- predictors (50) contained in said databases (110); said predictors (50) being computer variables relating to data of a patient and / or his / her caregiver and / or a healthcare worker and / or a healthcare facility; said predictors (50) depending on objective data; said geolocation data, said interaction data, said data extracted from said log files and the processing of said first algorithm being objective data; said predictors (50) and said objective data providing data through which said Al (120) associates to said patient, caregiver and / or healthcare worker one or more personality states and / or profiles and / or one or more states of change; said predictors (50), said objective data and said first algorithm providing data through which said Al (120) manage said adherence of said patient; said adherence management comprising continuous operations of forecasting, promoting, measuring, evaluating, optimizing and updating the adherence of said patient.

2. Healthcare process management system (100) according to the preceding claimLEIBO. / 67e20251, characterized in that it further comprises a plurality of fixed devices (150) positioned in said healthcare facility; said fixed devices (150) detecting the passage and stay time, within one or more areas, of portable devices (140) with wireless technology, and communicating said passage and stay time to said computerized apparatus (101); the interaction between said portable devices (140) and said fixed devices (150) providing data (145’) indicating the position, path and stay times of said portable devices (140) in the areas monitored by the fixed devices (150); the mutual interaction between said portable devices (140) and said fixed devices (150) providing mutual interaction data relating to face- to-face contacts within a healthcare facility; said first algorithm further analyzing said data (145’) and said mutual interaction data between said portable devices (140) and said fixed devices (150); said log files comprising one or more types and / or values of one or more attributes of said contact framework associated with said patient and / or caregiver and / or healthcare worker and / or said healthcare facility; said data (145’) and said mutual interaction data updating said log files; said objective data further comprising said data (145’) and said mutual interaction data; said objective data comprising said data (145’) and said mutual interaction data providing further data through which said Al (120) manages said adherence of said patient.

3. Healthcare process management system (100) according to any of the preceding claims, characterized in that said Al (120) associates and continuously updates one or more attributes of each contact framework to increase the value of said adherence for said patient.

4. Healthcare process management system (100) according to any of the preceding claims, characterized in that it further comprises guestionnaires (106) administered to the patient and / or caregiver and / or healthcare worker; said guestionnaires (106) providing self-reported data (106’); said predictors (50) further depending on said self-reported data (106’) collected by saidLEIBO. / 67e2025 questionnaires (106); said self-reported data (106’) providing further data through which said Al (120) associates to said patient, caregiver and / or healthcare worker one or more personality states and / or profiles and / or one or more states of change; said questionnaires (106) comprising one or more questions on said objective data to obtain data relating to the attention and / or memory and / or perception of the patient and / or caregiver and / or healthcare worker; said data (106’) updating said log files; said Al (120) operating a comparison between said objective data and said self-reported data (106’); said comparison, and said self-reported data (106’) collected from said questionnaires (106) providing further data through which said Al (120) manages said adherence of said patient.

5. Healthcare process management system (100) according to any of the preceding claims, characterized in that it further comprises:- one or more audio acquisition units (105) suitable for acquiring the audio of a contact between a patient and / or one or more of his / her caregivers and / or a healthcare worker and / or a healthcare facility; said audio acquisition unit (105) suitable for acquiring audio data (105’) and sending them to said computerized apparatus (101);- at least an algorithm (160), comprised in said computerized apparatus (101), for the analysis of verbal and extra-verbal language, suitable of analyzing said data (105’); said algorithm (160) providing further data through which said Al (120) manages said adherence of said patient; said data (105’) being objective data; said data (105’) further updating said log files; said objective data comprising said data (105’) providing further data through which said Al (120) manages said adherence of said patient.

6. Healthcare process management system (100) according to the preceding claim 5, characterized in that it further comprises one or more video acquisition units (104) suitable of acquiring images and audio of a patient and / or one or moreLEIBO. / 67e2025 caregivers and / or a healthcare worker and / or healthcare facility; said video acquisition unit (104) suitable of acquiring audio and video data (104’) and sending them to said computerized apparatus (101); said objective data further comprising said data (104’); said algorithm (160) further analyzing said data (104’); said data (104’) further updating said log files; said objective data comprising said data (104’) providing further data through which said Al (120) manages said adherence of said patient.

7. Healthcare process management system (100) according to the preceding claim 6, characterized in that it further comprises wearable smart glasses comprising a video acquisition unit (104); said wearable smart glasses being worn by said patient and / or caregiver and / or healthcare worker; said video acquisition unit (104) suitable of capturing a viewing area of said patient and / or caregiver and / or healthcare worker; said Al (120) analyzing the data (104’) of said video acquisition unit (104) and determining when an object / document / screen / point of a healthcare facility remains in said viewing area for a defined time.

8. Healthcare process management system (100) according to any of the preceding claims, characterized in that said portable device (140) comprises an accelerometer (141); said accelerometer (141) performing movement measurements and transmitting detected data (141’) to said computerized apparatus (101); said computerized apparatus (101) comprising movement algorithms; said movement algorithms providing further data through which said Al (120) manages said adherence of said patient; said movement algorithms processing and analyzing data (141’) acquired by said accelerometer (141); said movement algorithms determining the state of agitation of patients and / or caregivers and / or healthcare workers; said objective data comprising said data (141’); said data (141’) further updating said log files; said objective data comprising said data (141’) providing further data through which said Al (120) manages said adherence of said patient.LEIBO. / 67e20259. Healthcare process management system (100) according to any of the preceding claims, comprising said guestionnaires (106), according to the preceding claim 4; said system (100) being characterized in that it further comprises a chatbot (601) with an artificial intelligence suitable of generating said guestionnaires (106) on the basis of objective and / or self-reported data, data (602) of the patient and / or the caregiver and / or the healthcare worker and / or the healthcare facility contained in said databases (110), on the basis of one or more adherence values (30) and a textual input (610); said textual input (610) being provided via said computerized apparatus (101) and / or automatically generated by said Al (120) for said chatbot (601) comprising data relating to one or more personality states and / or profiles and / or states of change of the patient and / or caregiver and / or healthcare worker, comprising desired times for completing a guestionnaire (106) and / or comprising a desired number of guestions in the guestionnaire (106), comprising a clinical condition of the patient, comprising introductory guestions and / or comprising guestions for which objective data are available.

10. Healthcare process management system (100) according to previous claims 5 and 9, characterized in that said algorithm (160) analyses environmental noises (702) from the data (105’) of a call and said first algorithm analyses objective data, both said algorithms understanding whether an interlocutor is outside or in a closed place and / or whether he / she is in the presence of other people; said Al (120) carrying out a comparison between the processing of said first algorithm carried out on objective data and the analysis of said algorithm (160) of said environmental noises (702); said analysis of the environmental noises (702) and / or of the objective data and / or said comparison between said environmental noises (702) and said objective data defining the availability of a patient and / or caregiver and / or healthcare worker to proceed with a contact; said analysis of said algorithm (160) of said environmental noises (702) and / orLEIBO. / 67e2025 said analysis of said first algorithm of said objective data and / or said comparison operated by said Al (120) defining parts of said textual input (610) relating to the compilation time and / or the number of guestions and / or to said guestions of said guestionnaire (106).

11. Healthcare process management system (100) according to the previous claims 6 and 8, characterized in that said video acguisition unit (104) monitors on video a patient hospitalized in a healthcare facility; said accelerometer (141) being worn by said patient in the form of a bracelet and / or a ring; said data (141’) providing an indication of movements related to the preparation and / or selfadministration of drugs; said objective data comprising a comparison, carried out by said Al (120) and / or by said analysis algorithm (160) and / or by said movement algorithms, between said data (104’) and said data (141’) to ascertain the intake of drugs by the patient.

12. Healthcare process management system (100) according to any of the preceding claims, characterized in that said Al (120) on the basis of objective and / or self-reported data and / or predictors (50) and / or on the basis of personality states and / or profiles and / or states of change associated with said patient and / or caregiver and / or healthcare worker, proceeds automatically with a contact with a patient, a caregiver and / or a healthcare worker and / or a healthcare facility or proceeds semi-automatically by suggesting to a patient, a caregiver and / or a healthcare worker and / or another healthcare facility worker to proceed with said contact; said contact comprising the administration of a guestionnaire (106) according to the preceding claim 4.

13. Method (200) for managing healthcare processes, suitable of exploiting a system (100) according to any of the preceding claims, characterized in that it comprises: one or more acguisition and analysis steps (201) of objective data and / or self-reported data; said objective data and / or self-reported data beingLEIBO. / 67e2025 analyzed by said first algorithm and / or by said algorithm (160) according to previous claim 5 and / or by said movement algorithms according to previous claim 8 and / or by said Al (120) to automatically assign to a patient and / or one or more caregivers and / or a healthcare worker and / or a healthcare facility a contact framework with the related attributes and to define personality states and / or profiles and / or states of change for said patient and / or said caregiver and / or said healthcare worker; said contact framework and said personality states, profiles and states of change being updated on the basis of the acguisition of subseguent objective and / or self-reported data;- one or more selection steps (202) of a contact framework, by means of machine learning, in which types and / or values of the attributes of said contact framework are chosen from said databases (110) for the patient and / or the caregiver and / or the healthcare worker and / or the healthcare facility, on the basis of objective and / or self-reported data, continuously updated, acguired or available for said patient and / or caregiver and / or healthcare worker and / or healthcare facility, and / or on the basis of the values of the predictors (50), and / or on the basis of the types of personality states and / or profiles and / or states of change, attributed to said patient and / or caregiver and / or healthcare worker on the basis of said data present in said databases (110); said selection steps (202) aimed at providing for an initial identification of the contact framework by means of statistical data;- one or more adeguacy prediction steps (203) of the selected contact framework in said selection steps (202), wherein a probability of adeguacy is calculated using said objective, self-reported data and / or said predictors (50) and / or said personality states, profiles and states of change; said adeguacy prediction steps (203) of the contact framework being realized using said Artificial Intelligence (Al) engine (120); said probability ofLEIBO. / 67e2025 adequacy being assigned to each attribute of the contact framework;- one or more adequacy verification steps (204) of the contact framework by means of machine learning; said adequacy verification steps (204) comprise the confirmation of said adequacy probabilities and / or their updating for each attribute of the contact framework;- one or more verification steps of the outcomes (205) of the contact and of the actions resulting from said outcomes; said verification steps of the outcomes (205) of the contact and of the actions resulting from said outcomes are carried out on the basis of objective and self-reported data and using said Artificial Intelligence (Al) engine (120) and / or said first algorithm and / or said algorithm (160) and / or said movement algorithms;- one or more outcome prediction steps (206) of an identified procedure, wherein a probability of a positive or negative outcome of the identified procedure is calculated using said predictors (50); said probability of a positive or negative outcome is a function of the patient’s adherence to the procedure and the appropriateness of the procedure; said outcome prediction steps (206) being realized using said Artificial Intelligence (Al) engine (120);- one or more verification steps (207) of the outcome of the procedure identified, carried out and / or being carried out, by means of machine learning; said verification steps (207) involve the assignment of a cumulative score or distinct scores to indicators of urgency and appropriateness of said procedure identified for said patient; said verification steps (207) involve the assignment of one or more values of patient adherence to the procedure on the basis of the predictors (50), objective data and self-reported data and personality states and / or profiles and / or states of change;- one or more updating steps (208) of said database (110) of types of personality states and / or profiles and / or states of change and / or typesLEIBO. / 67e2025 and / or values of attributes of the contact framework that can be matched to said patient and / or caregiver and / or healthcare worker based on the predictors (50) and the acquired data, where each type of personality states and / or profiles and / or states of change and / or type and / or value of said identified attributes is assigned a probability of suitability to said patient and / or caregiver and / or healthcare worker; said updating steps (208) being implemented using said Artificial Intelligence (Al) engine (120).

14. Method (200) for managing healthcare processes according to the preceding claim 13 characterized in that it further comprises one or more analysis steps (209) of the requested procedure, in which a level of satisfaction of said patient and / or caregiver and / or healthcare worker is identified through said data (106’) of questionnaires (106), according to the previous claim 4, ad hoc and the processing of the first algorithm, of the algorithm (160) and of the movement algorithms on the objective and self-reported data; said analysis steps (209) being carried out using said Artificial Intelligence (Al) engine (120).

Images