University Professor, PhD, Engineer, Project Manager
(1993 – Present)
A patent is not an award. It is a document describing a technical solution precisely enough that someone else could reproduce it, and new enough that it does not already exist. That definition sets the standard: an invention has to work outside the room where it was conceived.
The inventions registered here share one concern. They take a technology that performs well in a laboratory and make it usable by someone who is not an engineer - a patient at home, a care worker, a person living with a disability. That translation is the difficult part, and it is where most promising prototypes stop.
The three areas are artificial intelligence applied to health, medical Internet of Things and remote rehabilitation, and large-scale analysis of clinical data. They are not separate: a sensor that measures a movement produces data, the data feed a predictive model, and the model is only useful if it returns something a clinician can act on.
The patent work is always accompanied by peer-reviewed publications, and the reason is practical rather than academic. Without experimental evidence a health-related invention remains a hypothesis, and no regulator, hospital or investor will treat it as anything else.
Metrics updated in real time on Google Scholar.
The problem. Assistive technology is normally designed around one specific impairment. It stops being useful the moment the condition changes, or combines with another.
The approach. A platform that reads the context and adapts to the person, instead of asking the person to adapt to the tool. Inclusivity is the design criterion, not a feature added at the end.
The problem. Rehabilitation works when it is continuous, but it requires physical presence. Patients who live far away, who cannot travel, or who need months of treatment simply stop.
The approach. Hardware and software that bring the session home: sensors that measure how a movement is actually performed rather than merely recording attendance, and a channel that returns to the therapist data precise enough to adjust the programme.
The problem. Clinical data already exist in enormous quantity, but they live apart: the medical record, the wearable device, the laboratory, the questionnaire. Kept apart, they predict nothing.
The approach. An architecture that brings them together and makes them analysable, with attention to the factor that matters more than the algorithm in daily practice: the quality and traceability of the data going in.
A multimedia, multichannel information system, recognised by the Italian Presidency of the Council of Ministers among the best applications for citizen services.
Blockchain certification for the EQAVET and EDCI frameworks: traceability and validation of professional competences across European systems.
Certified project manager in over 200 complex projects
national and international
Deep Learning, NLP, Computer Vision, AutoML
Spark, Hadoop, ETL, Data Warehouse
Biomedical sensors, edge computing, MQTT
PMBOK, Agile, Waterfall, CPM/PERT
Wearable, FHIR/HL7, telerehabilitation
SAP, CRM, BI, cloud architectures
Data security, GDPR, infrastructure protection
Scientific method, human-AI relationship
Critical thinking, innovation governance
Multidisciplinary coordination, EU/IT funds
Technology transfer, results valorisation
Lecturing, e-learning, scientific materials
LinkedIn
X / Twitter