Carlos Cunha

PIDI/CISeD/2023/011 • Modelos de Inteligência Artificial para Deteção de Stress Crónico e Padrões de Movimento em Ambientes de Atividades da Vida Diária

Leave a Comment / [CDS] Connected Digital Society, concluído, interno, nacional, não competitivo, não financiado, viseu /

Principal Researcher:
Rui Pedro Duarte
Duration: 2022 – 2024

Cised team members
Carlos Augusto da Silva Cunha

For a long time, stress has been considered an important health factor that affects the quality of life. Several studies indicate that excessive and continuous stress can trigger or worsen several diseases, such as cancer and cardiovascular. Stress is a response to a stressor, an external stimulus or condition that causes an emotional, physical, or psychological change in the individual. However, even though momentary stress can be easily identified, it is hard for people to recognize that they are living a stressful life. Multiple sclerosis (MS) is a chronic disease that affects individuals. It has symptoms on diverse levels, such as vision, fatigue, numbness and tingling, muscle spasms, stiffness and weakness, mobility problems, problems with thinking, learning, planning, and even depression and anxiety. With this, a few critical questions need to be raised: Is there any relation between stress and MS exacerbations? What are the relevant biomarkers to identify continuous stress? Are they reliable? Are their measurements little intrusive?
Thus, it is vital to measure stress and identify its types, in order to monitor and help people to deal with the manifestations of stress. It becomes crucial to identify continuous stress, as there are studies that indicate that it has a negative impact, while a systemic stress situation may even be rewarded. The monitoring process of this continuous stress needs to be controlled and with low intrusion in the user, so that it does not have negative effects on their lifestyle. For that, there are several biomarkers to identify stress in people, but not all are recommended. The most indicated biomarkers are heart rate (HR) and electrical conductivity, which can be monitored by devices such as smart watches. With this, it is possible to train artificial intelligence models to create a solution to detect long-term stress, which does not harm people's quality of life.

, ,

Machine Learning and Food Security: Insights for Agricultural Spatial Planning in the Context of Agriculture 4.0

Leave a Comment / 2022, Papers in Indexed Journals | Scopus and/or WoS, viseu /

Martinho, V. J. P. D., Cunha, C. A. d. S., Pato, M. L., Costa, P. J. L., Sánchez-Carreira, M. C., Georgantzís, N., Rodrigues, R. N., Coronado, F. (2022).
Machine Learning and Food Security: Insights for Agricultural Spatial Planning in the Context of Agriculture 4.0.
Applied Sciences, 12, 11828.
https://doi.org/10.3390/app122211828

Thought on Food: A Systematic Review of Current Approaches and Challenges for Food Intake Detection

Leave a Comment / 2022, Papers in Indexed Journals | Scopus and/or WoS, viseu /

Neves, P. A., Simões, J., Costa, R., Pimenta, L., Gonçalves, N.J., Albuquerque, C., Cunha, C., Zdravevski, E., Lameski, P., Garcia, N. M., et al. (2022).
Thought on Food: A Systematic Review of Current Approaches and Challenges for Food Intake Detection.
Sensors, 22: 6443.
https://doi.org/10.3390/s22176443

Intelligent System to Monitoring Diet Plans in Patients with Dementia such as Alzheimer's

Leave a Comment / 2021, Livros e Capítulos de Livro de Edição Nacional, viseu /

Cunha, C. (2021).
Sistema Inteligente de Controlo de Planos Alimentares para Doentes de Alzheimer e Outras Demências.
In M. J. Amante, S. Barros Fonseca, L. Araújo, P. Xavier, J. Carreira (Orgs.), Demência e COVID-19: Contributos do IV Seminário Internacional Alzheimer e outras Demências (pp. 57-66).
Viseu, Portugal: Escola Superior de Educação do Instituto Politécnico de Viseu.

PROJ/IPV/ID&I/002 • Sistema Inteligente de Controlo de Planos Alimentares para Doentes de Alzheimer e Outras Demências

[CDS] Connected Digital Society, concluído, interno, não competitivo, não financiado, viseu /

Principal Researcher: Carlos Augusto Cunha
Duration: 2020 – 2022

Cised team members
Valter Alves
Rui Pedro Duarte

Funding
CGD; PV

Alzheimer's disease is a progressive loss of mental function, characterized by degeneration of brain tissue, including loss of nerve cells, accumulation of an abnormal protein, and development of neurofibrillary braids. Alzheimer's patients become dependent on other people, even for the most basic tasks. Controlling feeding and hydrating an Alzheimer's patient is thus a crucial task performed by the person who supports their daily routine, called Informal Caregiver (IC).

Conditions of malnutrition, super nutrition, and dehydration are common in people with diseases causing dementia since the loss of their autonomy also manifests itself in the level of their inability to demonstrate food needs. The support of a nutritionist in the preparation and follow-up of a food plan aligned with the needs of the patient is, therefore, fundamental. The monitoring of the food plan is undoubtedly a process that demands from Cl a lot of discipline and the ability to deal with possible circumstantial adaptations, such as replacing foods prescribed in the food plan with other equivalents or changing the quantity of water consumed as a function of ambient temperature.

This project addresses the problem of the creation and monitoring of diet plans in patients with dementia such as Alzheimer's, through the creation of a computer solution. It allows the creation of nutritional plans by the nutritionists using a Web application and the follow-up of these plans by the ICs through an app to significantly increase the quality of life of the patient. The mobile app will be able to send to the CI notifications regarding proper nutrition and hydration in due moments, accompanied by the control of hydration using an intelligent water bottle, will allow ensuring the compliance of the indications of the nutritionist. Besides, the application will suggest alternatives to plan foods if they are unavailable. Another central feature of the solution is the dynamic adaptation of water administration to the patient as a function of the environmental conditions automatically collected by temperature and humidity sensors.

In scientific terms, the project includes the creation of innovative models for adapting food plans using machine learning algorithms and integration approaches, preprocessing and evaluation of data quality collected by external sensors.

, , ,

PROJ/IPV/ID&I/007 • Sistema Inteligente de Controlo de Planos Alimentares para o Desporto

[CDS] Connected Digital Society, concluído, interno, nacional, não competitivo, não financiado, viseu /

Principal Researcher:
Carlos Cunha
Duration: 2019 – 2022

Cised team members
Rui Pedro Duarte
Valter Alves

Funding
CGD; PV

The recent growth in the pursuit of sporting activities, motivated by a widespread increase in the perception of the importance of maintaining physical fitness, campaigns specifically aimed at combating physical inactivity, and opportunities created by the revelation of lesser known modalities, brought forward fundamental questions such as the correct nutrition of the practitioners. Numerous institutions involved in the practice of physical activity, which have since flourished, have been integrating these concerns into their scope, including through nutritionists.

For a nutritionist, the elaboration and monitoring of a food plan in line with the needs of the individual present two critical problems: 1) obtaining biometric data, eating habits and energy consumption to create the food plan, and 2) monitoring and the dynamic adaptation of the food plan.

Sports nutrition is one of the most complex areas of nutrition, since it requires observation of a rather comprehensive set of metrics, encompassing the athlete's physical aspects, physical activity, and eating habits. The use of measuring devices for specific parameters of physical activity represents a common practice among athletes. The integration of data automatically collected by these devices with other data not directly observable, such as dietary habits and subjective metrics, is part of the complexity of creating a global register that can be used by the nutritionist during the development of the plan to feed. Also, at a stage after the preparation of the food plan, the need to adapt it may arise. For example, variations in temperature or physical intensity may involve quick changes in the energy or hydration needs of an individual. In such situations, the data collected by the devices could be used to dynamically adjust the plan and send alerts, informing the sportsman of the need to eat food or water at the right time.

The objectives of this project include the creation of (a) a Web application where the nutritionist can record and follow up on dietary plans and (b) a mobile app for sportspersons, which can collect data provided by smart devices or manually entered, and where they can plan and receive notifications. In scientific terms, the project includes the creation of innovative models for adapting food plans using machine learning algorithms and integration approaches, preprocessing and evaluation of data quality collected by external devices.

, , ,
Scroll to Top