Percorrer por autor "Franco, Tiago"
A mostrar 1 - 10 de 16
Resultados por página
Opções de ordenação
- Anomaly detection using smart shirt and machine learning: a systematic reviewPublication . Nunes, Eduardo; Barbosa, José; Alves, Paulo; Franco, Tiago; Silva, AlfredoIn recent years, the popularity and use of Artificial Intelligence (AI) and significant investments in the Internet of Medical Things (IoMT) will be common to use products such as smart socks, smart pants, and smart shirts. These products are known as Smart Textile or E-textile, which can monitor and collect signals that our body emits. These signals allow it to extract anomalous components using Machine Learning (ML) techniques that play an essential role in this area. This study presents a Systematic Literature Review (SLR) on Anomaly Detection using ML techniques in Smart Shirt. The objectives of the SLR are: (i) to identify what type of anomaly the smart shirt can detect; (ii) what ML techniques are in use; (iii) which datasets are in use; (iv) identify smart shirt or signal acquisition devices worn in the chest region; (v) list the performance metrics used to evaluate the ML model; (vi) the results of the techniques in general; (vii) types of ML algorithms are being applied. The SLR selected eleven primary studies published between January/2017-May/2022. The results showed that six anomalies were identified, with the Fall anomaly being the most cited. The Support Vector Machines (SVM) algorithm is the most used. Most of the primary studies used public or private datasets. The Hexoskin smart shirt was most cited. The most used metric performance was accuracy. Almost all primary studies presented a result above 90%, and all primary studies used the Supervisioned type of ML.
- Approaches to classify knee osteoarthritis using biomechanical dataPublication . Franco, Tiago; Henriques, Pedro Rangel; Alves, Paulo; Pereira, Maria JoãoKnee osteoarthritis (KOA) is a degenerative disease that mainly affects the elderly. The development of this disease is associated with a complex set of factors that cause abnormalities in motor functions. The purpose of this review is to understand the composition of works that combine biomechanical data and machine learning techniques to classify KOA progress. This study was based on research articles found in the search engines Scopus and PubMed between January 2010 and April 2021. The results were divided into data acquisition, feature engineering, and algorithms to synthesize the discovered content. Several approaches have been found for KOA classification with significant accuracy, with an average of 86% overall and three papers reaching 100%; that is, they did not fail once in their tests. The acquisition of data proved to be the divergent task between the works, the most considerable correlation in this stage was the use of the ground reaction force (GRF) sensor. Although three studies reached 100% in the classification, two did not use a gradual evaluation scale, classifying between KOA or healthy individuals. Thus, we can get out of this work that machine learning techniques are promising for identifying KOA using biomechanical data. However, the classification of pathological stages is a complex problem to discuss, mainly due to the difficult access and lack of standardization in data acquisition.
- Automatic Fall Detection with Thermal CameraPublication . Kalbermatter, Rebeca B.; Franco, Tiago; Pereira, Ana I.; Valente, António; Soares, Salviano Pinto; Lima, JoséPeople are living longer, promoting new challenges in healthcare. Many older adults prefer to age in their own homes rather than in healthcare institutions. Portugal has seen a similar trend, and public and private home care solutions have been developed. However, age-related pathologies can affect an elderly person’s ability to perform daily tasks independently. Ambient Assisted Living (AAL) is a domain that uses information and communication technologies to improve the quality of life of older adults. AI-based fall detection systems have been integrated into AAL studies, and posture estimation tools are important for monitoring patients. In this study, the OpenCV and the YOLOv7 machine learning framework are used to develop a fall detection system based on posture analysis. To protect patient privacy, the use of a thermal camera is proposed to prevent facial recognition. The developed system was applied and validated in the real scenario.
- Big data analytics para a classificação do risco de abandono escolar em cursos do ensino superiorPublication . Franco, Tiago; Alves, Paulo; Koscianski, AndréO abandono escolar é uma constante preocupação das instituições de ensino superior, com divergentes e complexos fatores relatados por diversos autores. Conseguimos notar que este problema é bastante abrangente e explorado, já sendo possível encontrar nas instituições setores especializados ou programas de auxílio como psicólogos, auxílio moradia, programa de monitoria, entre outros, buscando minimizar a quantidade de alunos desistentes. Entretanto estas propostas dependem do próprio aluno a buscar ajuda necessária, abrindo uma lacuna para aqueles que não se sentem confortáveis a procurar ou não possuem total conhecimento do próprio caso. Este trabalho propõe um modelo para a identificação prévia dos alunos desistentes, com objetivo de tornar as instituições de ensino aptas a entender melhor os casos de abandono e se possível encaminhá-los a setores especializados. Para tal, utilizamos o Instituto Politécnico de Bragança como estudo de caso que nos forneceu mais 200 milhões de registros relacionados aos alunos matriculados entre 2008 a 2017. Analisamos e processamos a Big Data fornecida com a finalidade de moldá-la como parâmetros de entrada de algoritmos de machine learning. Inicialmente testamos três algoritmos e descobrimos que o random forest demonstra ser o mais eficiente neste contexto. A partir disso, aproveitamos do volume de dados para identificar qual seria melhor ciclo de treino e obtemos que o período de 4 anos consegue atingir melhores resultados. No aprimoramento do modelo adicionamos mais 2 atributos buscando realçar a trajetória escolar do aluno. Para implementação e visualização do modelo, desenvolvemos uma ferramenta de extração de dados e uma aplicação Web, que através de diferentes níveis de acesso, além de conseguir identificar os alunos em risco de abandono, também possibilita aos usuários efetuar análises comparativas entre escolas e cursos por meio de uma página personalizada com estatísticas transformadas em gráficos e tabelas. O estudo se apresenta como uma boa solução para identificação prévia dos alunos em risco de abandono, possibilitando análises e encaminhamentos. O modelo ainda pode ser ampliado a mais parâmetros e tende a obter melhores resultados ao longo dos anos aperfeiçoando através do reforço os atributos criados.
- Biofeedback-based method for real-time fatigue monitoring of kneePublication . Franco, Tiago; Henriques, Pedro; Alves, Paulo; Pereira, Maria João; Leitão, Paulo; Azevedo, NelsonThis paper introduces and implements a method to monitor muscle fatigue in real-time using a wearable biofeedback system to improve muscle rehabilitation treatments. The biofeedback system consists of an electromyography (EMG) sensor to capture muscle activity and two motion sensors to track knee angles. The proposed method for monitoring muscle fatigue involves three steps: (1) recognition of the movement phases during the knee extension exercise; (2) clipping of the EMG signal and calculation of fatigue-related metrics; and (3) normalization of metrics through a calibration process. An experimental session was performed with 10 healthy subjects performing 50 repetitions of the knee extension exercise. Processed data revealed changes in fatigue-related metrics, which align with existing literature. A comparison was also made between real-time and computer processing using raw data. While minor differences were noted between the two processing methods, the mobile app closely mirrored the trajectory of processed data in the cloud, ensuring reliability and consistency. This study advances remote muscle rehabilitation by quantifying muscle fatigue during treatment sessions. Thus, health professionals can tailor treatment plans based on individual patient characteristics, optimizing treatment duration, and reducing injury risk.
- Data acquisition system for a wearable-based fall preventionPublication . Kaizer, Raul; Sestrem, Leonardo; Franco, Tiago; Gonçalves, João; Teixeira, João Paulo; Lima, José; Carvalho, José Augusto; Leitão, PauloReliable ways to treat and monitor patients remotely have been researched and proposed by numerous people. Many of these propositions are under the wearable category due to it usually not requiring deep knowledge to be handled and its durability. Among the many applicable ways, fall monitoring has gained importance as the world population ages and countries aim to increase the quality of life. For it to be possible, there are many ways such as analyzing muscle response, body position, or brain activities, but for most of them, the result ends up being expensive and or inaccurate. With this in mind, this paper brings the development of an acquisition system for electromyography, electrocardiography, body position and temperature. The acquired data is transmitted to the smartphone through Bluetooth Low Energy (BLE) and then sent to a secure cloud to be provided to the physician. In future works, artificial intelligence codes will analyze the data patterns to predict fall occurrences and establish functional electrical stimulation (FES) routines to prevent falls and or treat the patients according to their necessities.
- Data acquisition, conditioning and processing system for a wearable-based biostimulationPublication . Oliveira, Leonardo Sestrem de; Kaizer, Raul; Gonçalves, João; Leitão, Paulo; Teixeira, João Paulo; Lima, José; Franco, Tiago; Carvalho, José AugustoData acquisition by electromyography, as well as the muscle stimulation, has become more accessible with the new developments in the wearable technology and medicine. In fact, for treatments, games or sports, it is possible to find examples of the use of muscle signals to analyse specific aspects related, e.g., to disease, injuries or movement impulses. However, these systems are usually expensive, does not integrate data acquisition with the muscle stimulation and does not exhibit an adaptive control behaviour that consider the pathology and the patient response. This paper presents a wearable system that integrates the signal acquisition and the electrostimulation using dry thin-film titanium-based electrodes. The acquired data is transmitted to a mobile application running on a smartphone by using Bluetooth Low Energy (BLE) technology, where it is analysed by employing artificial intelligence algorithms to provide customised treatments for each patient profile and type of pathology, and taking into consideration the feedback of the acquired electromyography signal. The acquired patient’s data is also stored in a secure cloud database to support the physician to analyse and follow-up the clinical results from the rehabilitation process.
- Decision support system for NMES treatments : a solution designPublication . Franco, TiagoThe preservation of functional capacity in old age is associated with a more active and dignified life. Maintaining this capacity is not a trivial task; several diseases can make it difficult to practice regular physical exercises or make it unfeasible, such as knee osteoarthritis. Neuromuscular electrical stimulation (NMES) is a treatment for muscle rehabilitation positioned as an alternative. However, it is still unclear which electrostimulation configurations can produce the most effective treatment. The literature indicates that the difference in treatment depends on the intrinsic characteristics of the patients. In this scenario, a decision support system is designed to assist physiotherapists in data analyses to create a personalized treatment. The proposed treatment relies on a wearable system with an NMES actuator and biofeedback sensors. Thus, it is expected to adapt the NMES in real-time based on unique patient characteristics, such as muscle fatigue. In addition, the system architecture is designed for the treatment session to be carried out at the patient's home, reducing costs and providing more comfort.
- Decision support systems for lower limb rehabilitation using electrical stimulation—a reviewPublication . Franco, Tiago; Henriques, Pedro; Alves, Paulo; Pereira, Maria JoãoThis paper presents a comprehensive review of Decision Support Systems (DSS) for lower limb rehabilitation using Electrical Stimulation (ES), employing a rigorous two-part methodology. The first part involves a bibliometric analysis of articles from 1980 to 2023, while the second part is a systematic review of studies from 2019 to 2023, addressing six key research questions. The review identifies the main characteristics of DSS, such as data usage, sensitive data protection, reasoning techniques, and validation processes. It highlights the development focus on joint control systems, increasing interest in biofeedback and AI applications, and significant interest in FES-Cycling. Despite advancements, “decision support” remains in the early stages with simple architectures and limited data handling. Conversely, studies show advanced ES control models validated with neurological patients. This article emphasizes the need for sophisticated DSS that integrate data protection, reasoning methods, and patient monitoring to enhance rehabilitation outcomes and identifies significant gaps for future research.
- Implementation of big data analytics tool in a higher education institutionPublication . Franco, Tiago; Alves, Paulo; Pedrosa, Tiago; Pereira, Maria João; Canão, JoséIn search of intelligent solutions that could help improve teaching in higher education, we discovered a set of analyzes that had already been discussed and just needed to be implemented. We believe that this reality can be found in several educational institutions, with paper or mini-projects that deal with educational data and can have positive impacts on teaching. Because of this, we designed an architecture that could extract from multiple sources of educational data and support the implementation of some of these projects found. The results show an important tool that can contribute positively to the teaching institution. Effectively, we can highlight that the implementation of a predictive model of students at risk of dropping out will bring a new analytical vision. Also, the system’s practicality will save managers a lot of time in creating analyzes of the state of the institutions, respecting privacy concerns of the manipulated data, supported by a secure development methodology.