Phase 2 (January 2023 – December 2023)

Voice Metrics for Discourse Quality Analysis

Authors: Nicolae Jinga, Florica Moldoveanu, Alin Moldoveanu, Anca Morar, Irina Mocanu, Alexandru Butean
Journal: UPB Scientific Bulletin, Vol. 85, Iss. 3, 2023
The paper introduces new discourse quality metrics and an evaluation method, and furthermore, provides a pilot implementation and evaluation in a specific use case – that of public speaking. Voice analysis and discourse quality metrics are important topics as they can solve various problems of modern-day life. In public speaking trainings, it is important to be able to analyze voice and discourse, to offer the speaker valuable feedback and information on how to improve when speaking in public. Several papers have been identified that study this issue, in which rhythm, tone, fluency and clarity, as well as biological signals, are analyzed, to detect emotions or anxiety. A set of voice quality metrics have been defined that can be implemented based on any speech recognition system. The pilot implementation showcases the difference of using various Speech-to-Text (STT) APIs over the proposed metrics and how these affect the discourse evaluation process.
Status: Accepted and Published

On Authentication in Virtual Reality Environments for Rehabilitation and Psychotherapy Systems

Authors: Florina Ungureanu, Bianca Andreea Bordea, Robert Gabriel Lupu, George Vieriu
Conference: IEEE International Conference on e-Health and Bioengineering (EHB 2022), 9-10 Noiembrie 2023
Virtual and Augmented Reality techniques are increasingly used in medical recovery systems and for treating phobias or psychiatric disorders. Considering the specifics of the patients involved in these procedures, authentication in virtual environments raises challenges. This paper deals with some approaches for authentication based mainly on biometrics and coordinated 3D manipulation and pointing. The authentication solution based on ear identification has been integrated into a neuromotor recovery system. This approach was the most comfortable for patients, it was reliable and offered a very good response time.
Status: Accepted and Published

Immersive Phobia Therapy through Adaptive Virtual Reality and Biofeedback

Authors: Alin Moldoveanu, Oana Mitruț, Nicolae Jinga, Cătălin Petrescu, Florica Moldoveanu ,Victor Asavei, Ana Magdalena Anghel, Livia Petrescu
Journal: Appl. Sci. 2023, 13(18), 10365 – spec. iss. Virtual/Augmented Reality and Its Applications, DOI: 10.3390/app131810365
Phobias are a type of anxiety disorder characterized by severe fear of objects and situations. In recent years, virtual reality exposure therapy has emerged as a safer and more convenient method of treating phobias, with the same rate of success as classical therapy, the in vivo exposure to stimuli. Following extensive research concerning the available technologies and after studying the state of the art in the field, we developed the PhoVR (Phobia therapy through Virtual Reality) system during a 2-year long project that involved the participation of a technical university, a biology faculty, and an industry partner specialized in designing 3D applications. Our final prototype incorporates scenarios for acrophobia, claustrophobia, and fear of public speaking therapy, in which the user is required to perform gamified tasks and navigate virtual reality environments with biophysical data acquisition (electrodermal activity and heart rate), automatic anxiety level classification, biofeedback integrated into the scenes’ elements, the dynamic adaptation of the virtual environments, and relaxation techniques made available at any moment. The control panel is an application dedicated to psychotherapists for managing patients’ profiles and therapy sessions. The feedback obtained in a qualitative survey with subjects and psychotherapists who evaluated the prototype validated the system and provided suggestions for refinement.
Status: Accepted and Published

Biophysical Signal Processing for Automatic Anxiety Classification in a Virtual Reality Exposure Therapy System

Authors: Nicolae Jinga, Catalin-Dumitru Petrescu, Oana Mitrut, Alin Moldoveanu, Florica Moldoveanu, Livia Petrescu
Journal: UPB Scientific Bulletin, 2023.
The paper presents a signal processing framework for automatic anxiety level classification in a virtual reality exposure therapy system. Two types of biophysical data (heart rate and electrodermal activity) were recorded, preprocessed and passed through a feature extraction procedure that provided input for the real-time anxiety level classification algorithm. The paper showcases the mathematical and engineering techniques behind the procedures and concludes with the challenges encountered in the research and future development ideas. The proposed method provides a good estimate of the level of anxiety while having at the same time a reduced level of complexity, allowing implementation on equipment with limited computing resources.
Status: Accepted, In process of publication

Gamification of post-stroke neuromotor rehabilitation exercises using hand tracking in VR

Authors: Robert Florin Apavaloaiei, Stefan Daniel Achirei
Journal: Bulletin of the Polytechnic Institute of Iași, 2023.
The main goal of this study is to develop a gaming platform for people who have had a stroke and are going through various stages of neuromotor rehabilitation. The main idea is that regular recovery activities are boring and repetitive, which might make people less interested in being in therapy. By including enjoyable but practical games into workout routines, there is potential to enhance patients’ motivation. To ensure the accuracy and reliability of the system, a validation process was conducted with the help of two stroke patients now undergoing rehabilitation at the Clinical Rehabilitation Hospital in Iasi. The parameters were adjusted based on individual patient characteristics, and a total of 12 tests were conducted, from which various results were obtained. At the end of the clinical trial, feedback was gathered from patients in order to get first-hand perspectives from the intended users of the system and identify areas for potential system enhancement. In the end, this research has demonstrated some of the benefits and potential uses of virtual reality (VR) technology in patient rehabilitation, and the participants in the clinical study have found it to be positive and useful.
Status: Accepted, In process of publication

Gamified Virtual Reality in Post-stroke Neurorehabilitation: A Systematic Review

Authors:Alin Moldoveanu, Iulia-Cristina Stănică, Ana-Magdalena Anghel, Mara-Ilinca-Mădălina Brezeanu
Status: In progress

Phase 1 (June 2022 – December 2022)

On the Design of a Modular Wearable Solution for Assistive Technologies

Authors: Nicolae Alexandru Botezatu, Robert Gabriel Lupu, Simona Caraiman
Conference: IEEE International Conference on e-Health and Bioengineering (EHB 2022), 17-19 November 2022.
The paper introduces the VibroWear architecture, a solution with a modular design at both hardware and software levels, that integrates an energy-aware engine in order to increase operational autonomy. By providing means of defining policies regarding energy consumption at module or I/O function level, the solution aims at adjusting the operational state of the modules (i.e. full, partial or decreased I/O activity, adjustable sample rates).
Status: Accepted and Published

Vocabulary enrichment in mother language after preschoolers’ interaction with a social robot

Authors: Nicoleta Laura Popa, Oana Alexandra Clim, Adina Boaca, Cristian-Tiberius Axinte, Robert-Gabriel Lupu, Georgiana Juravle
Conference: International Conference on Technological Ecosystems for Enhancing Multiculturality (TEEM 2022), 19-21 October 2022.
In the last decade, a number of studies conducted with preschoolers interacting with social robots suggest positive effects on foreign language learning, attitudes towards technology, and the recognition of geometrical shapes and colours. However, the effects of similar interactions for the improvement of mother tongue acquisition among children are rather understudied. The present study explores the effect of the child-robot interaction on the enrichment of mother tongue vocabulary, based on a quasi-experimental methodology. Two groups of Romanian preschoolers (N= 40) interacted with the social robot NAO either individually or as members of a group, based on a custom-developed script. As Romanian language is not covered by the robot’s software, NAO was programmed for this purpose. Preschoolers were individually interviewed pre and post-interaction, and asked to create ‘short stories about a robot’. The qualitative data were analysed with QDA Miner Lite, based on the total number of words in the story, words in the field of technology, and human relationships. Within-participants results were quantitatively explored. The findings suggest promising effects regarding the richness of vocabulary after the interaction with the social robot. Findings are discussed in the context of nowadays usage of technology in early childhood education.
Status: Accepted, In process of publication

Evaluation of non-textual metrics for public speaking analysis and training

Authors: Nicolae Jinga, Alin Moldoveanu, Florica Moldoveanu, Anca Morar, Alexandru Butean
Journal: UPB Scientific Bulletin, December 2022.
Status: Submitted, Under evaluation

A systematic survey of gamification strategies in VR-based neuromotor rehabilitation

Authors: Mara Ilinca Mădălina Chirașcu, Iulia Stănică, Alin Moldoveanu, Florica Moldoveanu
Journal: Frontiers in Psychology: Gamification as Intervention Strategy for Neuropsychological Rehabilitation, December 2022.
Status: In progress