A5084685551- Context-Aware Activity Recognition Systems
- Non-Invasive Vital Sign Monitoring
- Topic Modeling
- Natural Language Processing Techniques
- Biomedical Text Mining and Ontologies
- IoT-based Smart Home Systems
- IoT and Edge/Fog Computing
- ECG Monitoring and Analysis
- Bluetooth and Wireless Communication Technologies
- Mobile Health and mHealth Applications
- Wireless Body Area Networks
- Indoor and Outdoor Localization Technologies
- Physical Activity and Health
- User Authentication and Security Systems
- Target Tracking and Data Fusion in Sensor Networks
- Neurology and Historical Studies
- Energy Efficient Wireless Sensor Networks
- Emotion and Mood Recognition
- Advanced Text Analysis Techniques
- Biomedical and Engineering Education
- Vehicle emissions and performance
- Digital Mental Health Interventions
- Stroke Rehabilitation and Recovery
- Optical Imaging and Spectroscopy Techniques
- Text Readability and Simplification
Centro San Giovanni di Dio Fatebenefratelli
2023-2024
Istituti di Ricovero e Cura a Carattere Scientifico
2023-2024
University of Brescia
2015-2019
Institute of Electrical and Electronics Engineers
2018
University of Gävle
2018
Brescia University
2016
Causal relationship between physical activity and prevention of several diseases has been known for some time. Recently, attempts to quantify dose-response health show that automatic tracking quantification the exercise efforts not only help in motivating people but improve conditions as well. However, no commercial devices are available weight training calisthenics. This work tries overcome this limit, exploiting machine learning technique (particularly Linear Discriminant Analysis, LDA)...
In the era of digital healthcare, huge volumes textual information generated every day in hospitals constitute an essential but underused asset that could be exploited with task-specific, fine-tuned biomedical language representation models, improving patient care and management. For such specialized domains, previous research has shown fine-tuning models stemming from broad-coverage checkpoints can largely benefit additional training rounds over large-scale in-domain resources. However,...
In the last few years, several wearables appeared in market, for fitness and healthcare applications. Such smart devices have been proposed as a possible solution lowering costs of healthcare, leading to mHealth revolution. typical scenario, each wearable, embedding sensors, processing units communication modules, adopts smartphone data collection, displaying, remote communication. this paper, authors modify paradigm simplifying (e.g., relying only on simple analog front ends interfaces)...
The wider and availability of powerful, low-cost mobile devices (e.g., smartphones or tablets) is deeply changing healthcare, so that the mHealth term has been coined. announcement healthcare projects by market big players as Apple Samsung confirms this trend. In particular, opportunity to collect reliable patient data automatically allows enhance patient/user self-management helps in better delivering therapies. paper, authors propose an innovative architecture for a smart pill-dispenser...
A possible solution to reduce healthcare costs in a world that is aging the adoption of telehealth technologies. Several solutions appeared, past suggesting use smart devices (mobile phones or tablets) for supporting telehealth. Unfortunately, elderly people may have difficulties such even if user interfaces are becoming friendly and intuitive. This work introduces concept “Smartstone”, (low-cost) smartphone as simple, effective, portable gateway/edge server mobile towards cloud Internet...
One of Smart City's goals is to reduce healthcare spending, also by prolonging the active and independent lives elderly. Often elderly live alone; their lifestyle, including proper adherence therapies, often controlled relatives or friends caregivers daily weekly telephone calls. In addition investigating energy environment infrastructure services, project Brescia Living wants experience new remote monitoring services. The constantly monitored a system wearable home sensors; smartphone,...
The introduction of computerized medical records in hospitals has reduced burdensome activities like manual writing and information fetching. However, the data contained are still far underutilized, primarily because extracting from unstructured textual takes time effort. Information Extraction, a subfield Natural Language Processing, can help clinical practitioners overcome this limitation by using automated text-mining pipelines. In work, we created first Italian neuropsychiatric Named...
Location based services (LBS) have emerged as an active area of research aimed to filter and organize the information present on internet. This work deals with LBS applied industrial Internet Things (IoT) applications, ranging from distributed measurement systems (DMS) human-machine interface (HMI). The goal is investigate available techniques determine location assets (e.g., sensors actuators) then characterize some application scenarios. In detail, use cases been implemented exploiting...
The monitoring of vital parameters for athletes all disciplines is becoming more and essential, to improve performances ensure a constant control athletes' physical conditions, thus preventing health risks injuries. However, especially in amatorial competitions, the low available budget limits leads use inexpensive ineffective instrumentation. In this paper, low-cost system, based on commercial smartdevice, purposes context car races proposed. A single-lead electrocardiographic trace, body...
Detection of R-peaks in an electrocardiogram (ECG) acquisition is the primary goal any algorithm for automatic processing ECG signals. Several methods have been proposed past to accomplish this task, but usually they are designed use clinical situations, where strict real-time requirements not always needed and availability devices with high computational resources a problem. The recent broad success personal portable health monitoring has opened new scenario, which algorithms need be...
Traditional healthcare solutions oblige patients to stay in hospital; bio-signal monitoring and treatment dosing require the presence of specialized personnel. In turn, this scenario implies high-cost, especially for chronic diseases or fragile people surveillance. Wearable sensor-based may be an effective approach lowering costs, since they offer good performance can used directly at home, e.g., permanent self-monitoring. addition, their communication capabilities make easier exchange...
Alzheimer Disease (AD) represents one of the main causes neurodegenerative dementia in population aged 60+. This fact, together with growing number elderly forecasted for next decades, will increase its relevance medium-term future. In this view, numerous disease-modifying treatments are being developed, which have been shown to be effective only if administered early disease course. Thus, identification typology afflicting a patient is primary importance. Despite several techniques proposed...
Ambient assisted-living (AAL) systems, including wearables and home automation, could improve the quality of life fragile people. As analyzed in this paper, typical architecture AAL systems is based on wireless sensors networks proprietary gateways it can be complex, expensive, hardly includes interoperable multivendor devices. Emerging Android-based help reducing costs, but flexibility interoperability issues will still remain. On basis fact, development an system by a framework-oriented...
Nowadays system architectures based on Internet of Things (IoT) are becoming a pervasive topic in several applications. Smartphones integrate growing number high quality sensors and often they used as gateways for external sensors, wearable or ambient assisted living. Several recent projects about smart cities living experiencing sensory data management by means smartphones, resulting large flow battery power concerns. In these architectures, using embedded within the smartphone is trivial,...
Thanks to the diffusion of Radio Frequency Identification (RFID) and Quick Response (QR) technologies, distributed selective Human Machine Interfaces (HMI) are appearing in industrial environments as well some museums other application scenarios. By approaching a smartphone suitable tag or symbol, page website is shown user can access useful information without any action. Recently, indoor localization tools have been proposed, based on Wi-Fi Bluetooth (BT), locating user. Even if...
The introduction of computerized medical records in hospitals has reduced burdensome activities like manual writing and information fetching. However, the data contained are still far underutilized, primarily because extracting from unstructured textual takes time effort. Information Extraction, a subfield Natural Language Processing, can help clinical practitioners overcome this limitation by using automated text-mining pipelines. In work, we created first Italian neuropsychiatric Named...
Nowadays smartphones and tablets are pervasive in our lives, used to fulfil several of daily tasks. One approach is their usage as sensory systems; by exploiting embedded sensors, it possible get data about users. Typically, such sensors high end, relatively easy use them means Applications. Data access handled Android, which decreases sensor performance, sampling rate, but greatly eases fruition information. On the other hand, many useful available, today, only externally from smartphone;...
In the era of digital healthcare, huge volumes textual information generated every day in hospitals constitute an essential but underused asset that could be exploited with task-specific, fine-tuned biomedical language representation models, improving patient care and management. For such specialized domains, previous research has shown fine-tuning models stemming from broad-coverage checkpoints can largely benefit additional training rounds over large-scale in-domain resources. However,...
Abstract Borderline Personality Disorder (BPD) is a complex mental condition. Individuals with BPD have an average of three lifetime suicide attempts, and 10% them die by suicide. Understanding risk factors linked to suicidal behaviors crucial for effective intervention strategies. In recent years, machine learning (ML) approaches predicting in persons disorders been developed, but reliable, BPD-specific tool lacking. this work, we developed DRAMA-BPD (Detecting Risk Attempts Machine...
The advent of computerized medical recording systems in healthcare facilities has made data retrieval tasks easier, compared to manual recording. Nevertheless, the potential information contained within records remains largely untapped, mostly due time and effort required extract from unstructured documents. Natural Language Processing (NLP) represents a promising solution this challenge, as it enables use automated text-mining tools for clinical practitioners. In work, we present...