Below you will find a list of scientific and other publications that are direct results of the research done in the ALAMEDA project.

All publications are freely downloadable and disseminated in open access modalities


Published: 29 September 2023

Monitoring and Predicting Health Status in Neurological Patients: The ALAMEDA Data Collection Protocol

Authors: A. Sorici, L. Bajenaru, I. Georgiana Mocanu, A. M. Florea, P. Tsakanikas, A. C. Ribigan, L. Pedullà and A. Bougea

(1) Objective: We explore the predictive power of a novel stream of patient data, combining wearable devices and patient reported outcomes (PROs), using an AI-first approach to classify the health status of Parkinson’s disease (PD), multiple sclerosis (MS) and stroke patients (collectively named PMSS).

(2) Background: Recent studies acknowledge the burden of neurological disorders on patients and on the healthcare systems managing them. To address this, effort is invested in the digital transformation of health provisioning for PMSS patients.

(3) Methods: We introduce the data collection journey within the ALAMEDA project, which continuously collects PRO data for a year through mobile applications and supplements them with data from minimally intrusive wearable devices (accelerometer bracelet, IMU sensor belt, ground force measuring insoles, and sleep mattress) worn for 1–2 weeks at each milestone.

Download pdf: full document

Published: 28 September 2023

FoGGAN: Generating Realistic Parkinson’s Disease Freezing of Gait Data Using GANs


N. Peppes, P. Tsakanikas, E. Daskalakis, T. Alexakis, E. Adamopoulou and K. Demestichas 

Data scarcity in the healthcare domain is a major drawback for most state-of-the-art technologies engaging artificial intelligence. The unavailability of quality data due to both the difficulty to gather and label them as well as due to their sensitive nature create a breeding ground for data augmentation solutions. Parkinson’s Disease (PD) which can have a wide range of symptoms including motor impairments consists of a very challenging case for quality data acquisition.

Download pdf: full document

Published: 14 September 2023

Towards Detecting Freezing of Gait Events Using Wearable Sensors and Genetic Programming

Authors: A. N. Tarekegn, F. Alaya Cheikh , M. Sajjad and M. Ullah

Freezing of gait (FOG) is one of the most common manifestations of advanced Parkinson’s disease. It represents a sudden interruption of walking forward associated with an increased risk of falling and poor quality of life. Evolutionary algorithms, such as genetic programming (GP), have been effectively applied in modelling many real-world application domains and diseases occurrence. In this paper, we explore the application of GP for the early detection of FOG episodes in patients with Parkinson’s disease. The study involves the analysis of FOG by exploiting the statistical and time-domain features from wearable sensors, followed by automatic feature selection and model construction using GP.


Download pdf: full document

Published: 2 August 2023

Enhancing Human Activity Recognition Through Sensor Fusion and Hybrid Deep Learning Model

Authors: A. N. Tarekegn, M. Ullah F. Alaya Cheikh and M. Sajjad 


Wearable-based human activity recognition (HAR) is essential for several applications, such as health monitoring, physical training, and rehabilitation. However, most HAR systems presently depend on a single sensor, typically a smartphone, due to its widespread use. To improve performance and adapt to various scenarios, this study focuses on a smart belt equipped with acceleration and gyroscope sensors for detecting activities of daily living (ADLs). The collected data was pre-processed, fused and used to train a hybrid deep learning model incorporating a CNN and BiLSTM network. We evaluated the effect of window length on recognition accuracy and conducted a performance analysis of the proposed model.


Download pdf: full document

Published: 19 June 2023

Shared Decision-Making to Improve Health-Related Outcomes for Adults with Stroke Disease

Authors: L. Bajenaru, A. Sorici, I. Georgiana Mocanu, A. Magda Florea, F. Anca Antochi, A. Cristina Ribigan

Stroke is one of the leading causes of disability and death worldwide, a severe medical condition for which new solutions for prevention, monitoring, and adequate treatment are needed. This paper proposes a SDM framework for the development of innovative and effective solutions based on artificial intelligence in the rehabilitation of stroke patients by empowering patients to make decisions about the use of devices and applications developed in the European project ALAMEDA.

Download pdf: full document

Published: 8 June 2023

Ince-PD Model for Parkinson’s Disease Prediction Using MDS-UPDRS I & II and PDQ-8 Score

Authors: N.Tsolakis, C. Maga-Nteve, G. Meditskos, S. Vrochidis and I. Kompatsiaris

Parkinson’s Disease (PD) is one of the most prevalent and complex neurodegenerative disorders. Timely and accurate diagnosis is essential for the effectiveness of the initial treatment and improvement of the patients’ quality of life. Since PD is an incurable disease, the early intervention is important to delay the progression of symptoms and severity of the disease. This paper aims to present Ince-PD, a new, highly accurate model for PD prediction based on Inception architectures for time-series classification, using wearable data derived from IoT sensor-based recordings and surveys from the mPower dataset. The fea- ture selection process was based on the clinical knowledge shared by the medical experts through the course of the EU funded project ALAMEDA. The algorithm predicted total MDS-UPDRS I & II scores with a mean absolute error of 1.97 for time window and 2.27 for patient, as well as PDQ-8 scores with a mean absolute error of 2.17 for time window and 2.96 for patient. Our model demonstrates a more effective and accurate method to predict Parkinson Disease, when compared to some of the most significant deep learning algorithms in the literature.

Download pdf: full document

Published: 3 May 2023

InSEption: A Robust Mechanism for Predicting FoG Episodes in PD Patients

Authors: D. Dimoudis, N. Tsolakis, C. Magga-Nteve, G. Meditskos, S. Vrochidis and I. Kompatsiaris

The integration of IoT and deep learning provides the opportunity for continuous monitoring and evaluation of patients’ health status, leading to more personalized treatment and improved quality of life. This study explores the potential of deep learning to predict episodes of freezing of gait (FoG) in Parkinson’s disease (PD) patients. Initially, a literature review was conducted to determine the state of the art; then, two inception-based models, namely LN-Inception and InSEption, were introduced and tested using the Daphnet dataset and an additional novel medium-sized dataset collected from an IMU (inertial measuring unit) sensor. The results show that both models performed very well, outperforming or achieving performance comparable to the state-of-the-art. In particular, the InSEption network showed exceptional performance, achieving a 6% increase in macro F1 score compared to the inception-only-based counterpart on the Daphnet dataset. In a newly introduced IMU dataset, InSEption scored 97.2% and 98.6% in terms of F1 and AUC, respectively. This can be attributed to the added squeeze and excitation blocks and the domain-specific oversampling methods used for training. The benefits of using the Inception mechanism for signal data and its potential for integration into wearable IoT are validated.

Download pdf: full document

Published: 30 March 2023

An Intelligent Ecosystem for Improving Brain Disease Monitoring of Patients Using Wearable Devices and Artificial Intelligence

Authors: A. Sorici, L. Bajenaru, I. Mocanu, A. Florea

Nowadays, neurological diseases represent a medical emergency for which new prevention, monitoring and adequate treatment solutions are needed. ALAMEDA project proposes a monitoring solution for patients with Parkinson’s disease, Multiple Sclerosis, and Stroke, using multiple sensors and specific applications to collect information on the condition of health and aspects of lifestyle, activity level, sociability and mood. The paper describes the ALAMEDA infrastructure architecture and  all processes to support the dedicated AI toolkit, along with all relevant concepts, components and interactions. Data collection methods are central to the objectives of the ALAMEDA project, as they are responsible for the acquisition of all data necessary for patient monitoring and evaluation during the pilot period. In this sense, the data flow for the data collection service for smart bracelets and smart insoles is presented, as well as how users interact with these devices.

Download pdf: full document

Published: 20 Febrary 2023

A Comprehensive Survey on Deep Facial Expression Recognition: Challenges, Applications, and Future Guidelines

Authors: M. Sajjad, Faouzi Alaya Cheikh, M. Ullah, G. Christodoulou, M. Hijji, K. Muhammad, J. J.P.C. Rodrigues

Facial expression recognition (FER) is an emerging and multifaceted research topic. Applications of FER in healthcare, security, safe driving, and so forth have contributed to the credibility of these methods and their adoption in human-computer interaction for intelligent outcomes.
Computational FER mimics human facial expression coding skills and conveys important cues that complement speech to assist listeners. Similarly, FER methods based on deep learning and artificial intelligence (AI) techniques have been developed with edge modules to ensure efficiency and real-time processing. To this end, numerous studies have explored different aspects of FER.

Download pdf: full document

Published: 18 Febrary 2023

AI and IoT Enabled Sleep Stage Classification

Authors: D. Zografakis, P. Tsakanikas, I. Roussaki, K.-M. Giannakopoulou

Sleep is a key aspect affecting health, cognitive functionality, and human psychology on all occasions. There- fore, on the one hand, sleep greatly impacts the quality of life, while on the other hand poor health and/or psychology often deteriorate the quality of sleep. Moving beyond the golden standard for sleep studies, i.e. polysomnography, and building on the current state of the art in wearables, this paper aims to propose a deep learning approach that focuses on sleep stage classification, introducing the timeseries related information input to the classification. In this respect, smartwatch sensor measurements are used and a series of meth- ods have been tested. The proposed approach constitutes a preliminary work on sleep stage classification introducing a novel approach of feature engineering incorporating the time-related information concerning the transition of the sleep stages via a Long Short-Term Memory (LSTM) encoding of the accelerometer data from smartwaches.

Download pdf: full document

Published: 5 Febrary 2023

Ambiental Factors in Parkinson’s Disease Progression: A Systematic Review

Authors: A. Bougea, N. Papagiannakis, A. Simitsi, E. Panagiotounakou, C. Chrysovitsanou, E. Angelopoulou, C. Koros, L. Stefanis

Parkinson’s disease (PD) is a progressive neurodegenerative disease with motor and non-motor symptoms characterized by complex interactions between genetic and environmental factors. These factors influence the onset and varying evolution rate of PD, however the exact mechanisms remain unclear. Given the ineffective symptomatic treatments of PD, the research priority is focused on prevention and the recognition of modifiable environmental risk factors including pesticides, heavy metals, viruses, and air pollution.

Download pdf: full document

Published: 30 January 2023

Deep learning based speech emotion recognition for Parkinson patient

Authors: H. Khan, M. Ullah, Fadi Al-Machot, Faouzi Alaya Cheikh, M. Sajjad

This research aims to develop a system that can accurately identify common SEs which are important for PD patients, such as anger, happiness, normal, and sadness. We proposed a novel lightweight deep model to predict common SEs.

The adaptive wavelet thresholding method is employed for pre-processing the audio data. Fur- thermore, we generated spectrograms from the speech data instead of directly processing voice data to extract more dis- criminative features. The proposed method is trained on gen- erated spectrograms of the IEMOCAP dataset.

Download pdf: full document

Published: 22 November 2022

An Affective Multi-Modal Conversational Agent for Non Intrusive Data Collection from Patients with Brain Diseases

Authors: C. Chira, E. Mathioudis, C. Michailidou, P. Agathangelou, G. Christodoulou, I. Katakis, E. Kontopoulos, and K. Avgerinakis

This paper presents Zenon, an affective, multi-modal conversational agent (chatbot) specifically designed for treatment of brain diseases like multiple sclerosis and stroke.

Zenon collects information from patients in a non-intrusive way and records user sentiment using two different modalities: text and video. A user-friendly interface is designed to meet users’ needs and achieve an efficient conversation flow. What makes Zenon unique is the support of multiple languages, the combination of two information sources for tracking sentiment, and the deployment of a semantic knowledge graph that ensures machine-interpretable information exchange.

Download pdf: full document

Published: 20 June 2022

Monitoring Neurological Disorder Patients via Deep Learning Based Facial Expressions Analysis

Authors: M. Munsif, M. Ullah, Bilal Ahmad, M. Sajjad, Faouzi Alaya Cheikh

Facial expression (FE) is the most natural and convincing source to communicate human emotions, providing valuable insides to the observer while assessing the emotional incongruities. In health care, the FE of the patient (specifically of neurological disorders (NDs) such as Parkinson’s, Stroke, and Alzheimer’s) can assist the medical doctor in evaluating the physical condition of a patient, such as fatigue, pain, and sadness.

Download pdf: full article (pp. 412-423)

Published: 29 June 2022

Balancing between holistic and cumulative sentiment classification

Authors: P. Agathangelou, I. Katakis

Our method, by combining convolutional, recurrent and attention neural networks can extract rich linguistic patterns that reveal the user’s sentiment towards the entity under review. We evaluate our method in nine datasets that represent both binary and multi-class classification tasks.

Download pdf: full document


Published: 13 May 2022

Usability of the Virtual Supermarket Test for Older Adults with and without Cognitive Impairment

Authors: S. Zygourisa, S. Segkoulia, A. Triantafyllidisa, D. Giakoumisa

This study conducted a preliminary usability assessment of the Virtual Supermarket Test (VST), a serious gamebased self-administered cognitive screening test for mild cognitive impairment (MCI).

Download pdf: full document


Published: 1 April 2022

A Semantic Technologies Toolkit for Bridging Early Diagnosis and Treatment in Brain Diseases: Report from the Ongoing EU-Funded Research Project ALAMEDA

Authors: C. Maga-Nteve, E. Kontopoulos, N.TsolakisI. KatakisE. MathioudisP. MitziasK. AvgerinakisG. MeditskosA. KarakostasS. Vrochidis, I. Kompatsiaris 

Semantic Web technologies are increasingly being deployed in various e-health scenarios, prominently due to their inherent capacity to harmonize heterogeneous information from diverse sources and devices, as well as their capability to provide meaningful interpretations and higher-level insights. This paper reports on ongoing work in the recently started EU-funded project ALAMEDA towards a semantic toolkit for bridging the gap between early diagnosis and treatment in a variety of brain diseases.

Download pdf: full document


Published: 24 February 2022

Internet of Things Technologies and Machine Learning Methods for Parkinson’s Disease Diagnosis, Monitoring and Management: A Systematic Review

Authors: K-M. Giannakopoulou, I. Roussaki, K. Demestichas

This paper presents a state-of-the-art systematic review of the recent advances in the Internet of Things and Artificial Intelligence applications on Parkinson’s Disease, including the subdomains of machine learning and deep learning.

(This article belongs to the Special Issue AI and IoT Enabled Solutions for Healthcare)


Download pdf: full article


Published: 22 December 2022

Ontology based semantic model for health data interpretation

Authors: C. Maga-Nteve, N. Tsolakis, G. Meditskos, A. Karakostas, S. Vrochidis, I. Kompatsiaris

In this paper we propose a new semantic data model in which health information derived from Parkinson’s, Multiple Sclerosis, and Stroke (PMSS) patients is systematically analyzed to generate and improve knowledge that will be transferred to patient care in order to design and develop innovative health risk prediction and intervention tools


Download pdf: full article