Ghazal Rouhafzay

Prof. Dr. Ghazal Rouhafzay

Informatique, Université de Moncton, Moncton, New Brunswick, Canada
orcid

Research Focus & Metrics

Academic Career Length

13Years (2014-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
27 December 2020 in Sensors
Verified
Journal Article
Open access
PDF file
Ghazal RouhafzayGhazal Rouhafzay
Ghazal Rouhafzay
ACAna-Maria Cretu
Ana-Maria Cretu
PPPierre Payeur
Pierre Payeur
Transfer of learning or leveraging a pre-trained network and fine-tuning it to perform new tasks has been successfully applied in a variety of machine intelligence fields, including computer vision, natural language processing and audio/speech recognition. Drawing inspiration from neuroscience research that suggests that both visual and tactile stimuli rouse similar neural networks in the human brain, in this work, we explore the idea of transferring learning from vision to touch in the context of 3D object recognition. In particular, deep convolutional neural networks (CNN) pre-trained on visual images are adapted and evaluated for the classification of tactile data sets. To do so, we ran experiments with five different pre-trained CNN architectures and on five different datasets acquired with different technologies of tactile sensors including BathTip, Gelsight, force-sensing resistor (FSR) array, a high-resolution virtual FSR sensor, and tactile sensors on the Barrett robotic hand. The results obtained confirm the transferability of learning from vision to touch to interpret 3D models. Due to its higher resolution, tactile data from optical tactile sensors was demonstrated to achieve higher classification rates based on visual features compared to other technologies relying on pressure measurements. Further analysis of the weight updates in the convolutional layer is performed to measure the similarity between visual and tactile features for each technology of tactile sensing. Comparing the weight updates in different convolutional layers suggests that by updating a few convolutional layers of a pre-trained CNN on visual data, it can be efficiently used to classify tactile data. Accordingly, we propose a hybrid architecture performing both visual and tactile 3D object recognition with a MobileNetV2 backbone. MobileNetV2 is chosen due to its smaller size and thus its capability to be implemented on mobile devices, such that the network can classify both visual and tactile data. An accuracy of 100% for visual and 77.63% for tactile data are achieved by the proposed architecture.

22 October 2025 in Bioengineering
Verified
Journal Article
Open access
PDF file
HRHajarimino Rakotomanana
Hajarimino Rakotomanana
Ghazal RouhafzayGhazal Rouhafzay
Ghazal Rouhafzay
This scoping review systematically maps the rapidly evolving application of Artificial Intelligence (AI) in Autism Spectrum Disorder (ASD) diagnostics, specifically focusing on computational behavioral phenotyping. Recognizing that observable traits like speech and movement are critical for early, timely intervention, the study synthesizes AI’s use across eight key behavioral modalities. These include voice biomarkers, conversational dynamics, linguistic analysis, movement analysis, activity recognition, facial gestures, visual attention, and multimodal approaches. The review analyzed 158 studies published between 2015 and 2025, revealing that modern Machine Learning and Deep Learning techniques demonstrate highly promising diagnostic performance in controlled environments, with reported accuracies of up to 99%. Despite this significant capability, the review identifies critical challenges that impede clinical implementation and generalizability. These persistent limitations include pervasive issues with dataset heterogeneity, gender bias in samples, and small overall sample sizes. By detailing the current landscape of observable data types, computational methodologies, and available datasets, this work establishes a comprehensive overview of AI’s current strengths and fundamental weaknesses in ASD diagnosis. The article concludes by providing actionable recommendations aimed at guiding future research toward developing diagnostic solutions that are more inclusive, generalizable, and ultimately applicable in clinical settings.

15 April 2019 in IEEE Transactions on Instrumentation and Measurement
Verified
Journal Article
Closed access
Ghazal RouhafzayGhazal Rouhafzay
Ghazal Rouhafzay
ACAna-Maria Cretu
Ana-Maria Cretu
Reproducing the visuo-haptic interaction demonstrated by humans to sense and interpret visual and tactile stimuli is the main motivation of this research on tactile object recognition. An enhanced model of visual attention, consisting of information about color, contrast, curvature, entropy, symmetry, intensity, and orientation is developed to assist in selection of only relevant regions to be probed over the surface of an object. We aim to demonstrate that this visual information can be successfully employed to recognize the object from local tactile data, following the idea of haptic glance exhibited by humans (i.e., object recognition by a limited number of static contacts between the object and a finger). Due to the time-consuming nature of real tactile data acquisition, the process is first simulated and tested over a set of virtual objects prior to testing it on real rigid objects. A series of classifiers is trained and tested for the object recognition task and their performance is compared in terms of accuracy. The k-nearest neighbor classifier is shown to be the most promising algorithm among the tested classifiers for the object recognition task, both for simulated data (85.42% for four objects and 76.37% for six objects) and for real tactile sensor data (66.76% for four objects and 58.97% for six objects) when single imprints are used as a basis for recognition. Employing multiple imprints for object recognition yields a 100% accuracy for support vector machines (SVM) and k-nearest neighbors (kNN) classifiers when at least four imprints are considered.

27 March 2025 in Frontiers in Nutrition
Verified
Journal Article
Open access
PDF file
ARAsal Rouhafzay
Asal Rouhafzay
Ghazal RouhafzayGhazal Rouhafzay
Ghazal Rouhafzay
JJJalila Jbilou
Jalila Jbilou
Accurate dietary intake estimation is crucial for managing weight-related chronic diseases, such as diabetes, where precise measurement of food volume and caloric content is essential. Traditional calorie counting methods are often error-prone and may not meet the specific needs of individuals with diabetes. Recent advancements in computer science offer promising solutions through automated systems that estimate calorie intake from food images using deep learning techniques. These systems provide personalized dietary recommendations, helping individuals with diabetes make informed choices. As smartphones and wearable devices become more accessible, the utilization of electronic apps for dietary monitoring is increasing, highlighting the need for more research into safe, secure, and evidence-based IoT solutions. However, challenges such as standardization, validation across diverse populations, and data privacy concerns need to be addressed. This review focuses on the role of computer science in dietary intake estimation, specifically food segmentation, classification, and volume estimation for calorie calculation. By synthesizing existing literature, this review provides insights into current methods, key challenges, and potential future directions. The review also explores advancements in technology that can improve the accuracy of dietary assessments, contributing to personalized disease management and the prevention of weight-related chronic conditions.

29 March 2019 in Sensors
Verified
Journal Article
Open access
PDF file
Ghazal RouhafzayGhazal Rouhafzay
Ghazal Rouhafzay
ACAna-Maria Cretu
Ana-Maria Cretu
Drawing inspiration from haptic exploration of objects by humans, the current work proposes a novel framework for robotic tactile object recognition, where visual information in the form of a set of visually interesting points is employed to guide the process of tactile data acquisition. Neuroscience research confirms the integration of cutaneous data as a response to surface changes sensed by humans with data from joints, muscles, and bones (kinesthetic cues) for object recognition. On the other hand, psychological studies demonstrate that humans tend to follow object contours to perceive their global shape, which leads to object recognition. In compliance with these findings, a series of contours are determined around a set of 24 virtual objects from which bimodal tactile data (kinesthetic and cutaneous) are obtained sequentially and by adaptively changing the size of the sensor surface according to the object geometry for each object. A virtual Force Sensing Resistor array (FSR) is employed to capture cutaneous cues. Two different methods for sequential data classification are then implemented using Convolutional Neural Networks (CNN) and conventional classifiers, including support vector machines and k-nearest neighbors. In the case of conventional classifiers, we exploit contourlet transformation to extract features from tactile images. In the case of CNN, two networks are trained for cutaneous and kinesthetic data and a novel hybrid decision-making strategy is proposed for object recognition. The proposed framework is tested both for contours determined blindly (randomly determined contours of objects) and contours determined using a model of visual attention. Trained classifiers are tested on 4560 new sequential tactile data and the CNN trained over tactile data from object contours selected by the model of visual attention yields an accuracy of 98.97% which is the highest accuracy among other implemented approaches.

17 October 2016 in 3D Research
Verified
Journal Article
Closed access
PDF file
MCMaude Chagnon-Forget
Maude Chagnon-Forget
Ghazal RouhafzayGhazal Rouhafzay
Ghazal Rouhafzay
ACAna-Maria Cretu
Ana-Maria Cretu
Stéphane Bouchard, Ph.D.Stéphane Bouchard, Ph.D.
Stéphane Bouchard, Ph.D.