Upcoming Events

Currently there are no events.

Past Events

Mar

2025

Advancing AI Innovation and Governance (2:30 pm - 4:00 pm)

Title: Advancing AI Innovation and Governance

Speaker: Haruna Isah, Associate Director, Centre for Applied AI (CAAI) Sheridan College, Oakville, ON

Time: Wednesday. March 12th, 2:30pm

Talk Team's link: Click Here

Abstract: The Sheridan’s Centre for Applied AI (CAAI) is dedicated to pioneering the development and application of Artificial Intelligence (AI) across diverse sectors, including health care, telecommunications, finance, education, entertainment, and retail. CAAI sparks impact and advances real-world solutions by bringing together industry and community partners, researchers, students and change-makers to harness the transformative power of AI through collaborative research and industry partnerships. This talk will delve into CAAI's multifaceted research initiatives, highlighting key projects in the broader field of AI such as machine learning, computer vision, and natural language processing to specific areas such as generative and agentic AI. To further spark interest in AI research and innovation, we will also examine the critical need for AI governance frameworks and security tools. As AI-powered solutions are increasingly deployed in real-world applications, they introduce new security vulnerabilities, ethical dilemmas, regulatory challenges, and concerns about public trust. This discussion will explore how robust AI governance, encompassing AI security, ethics, responsible AI, and trustworthy AI, can ensure that AI systems are not only powerful and efficient but also safe, fair, and aligned with societal values.

Short Bio: Dr. Haruna Isah is the Associate Director of Sheridan’s Centre for Applied AI (CAAI). He is an experienced leader in AI and cybersecurity with over a decade of expertise in applied research, strategic planning, and fostering innovation. His work bridges academia and industry, delivering transformative AI solutions while creating value for students, empowering faculty, and enhancing community outcomes. Before his time at CAAI, Haruna was Research Associate and Talent Development Manager at the Canadian Institute for Cybersecurity. He was also an IBM-SOSCIP Postdoctoral Fellow at Queen's University. His commitment to advancing artificial intelligence and cybersecurity is evident in his numerous publications and presentations at international conferences. He has also served as a judge for AI and cybersecurity competitions, contributing his expertise to evaluating innovative solutions in these fields. His education includes a PhD in Computing and an MSc in Software Engineering, both at the University of Bradford in the UK.

View poster here

Mar

2025

Creating Immersive Smart Spaces that Care: Computer Science and Creativity for a Science Fiction World (2:30 pm - 3:30 pm)

Title: Creating Immersive Smart Spaces that Care: Computer Science and Creativity for a Science Fiction World.

Speaker: Alexis Morris, Acadia Alumni, Associate Professor, Faculty of Arts & Science (Digital Futures), OCAD University, Toronto

Time: Wednesday. March 5th, 2:30pm

Talk Team's link: Click Here

Abstract: Ours is a fantastic time, filled with advancements driven by computer science, enabling artificial intelligence, world sensing and control, and deep immersive connections to become possible. We are on the cusp of converging revolutions that transform our relationship to the world around us in exciting ways. New possibilities open for computer scientists to embrace design-science and new forms of creation with technology. This talk is about where these themes meet for our everyday environments, with a focus on designing mixed reality smart-spaces that will eventually come alive. Join me in this conversation about a future where science fiction becomes reality.

Short Bio: Dr. Morris is an Associate Professor in the Digital Futures program at OCAD University, director of the Adaptive Context Environments (ACE) Lab, and the Canada Research Chair in the Internet of Things (Tier II) 2018-2023. Morris’ research is in the design and development of immersive smart spaces, using hybrid mixed reality and Internet of Things architectures. His present research explores the applied uses of these hybrid environments, across multiple domains, with a new focus toward spaces that care for their inhabitants. Dr. Morris’ expertise bridges a cross-section of approaches in artificial intelligence, virtual and augmented reality, context awareness, adaptive risk management, multi-agent systems modelling, human-agent interaction, and pervasive technologies.

Access full pdf here.

Jan

2025

The Promise of Dream Technology (2:30 pm - 3:30 pm)

Talk by Ken Raj Leslie

Assistant Professor, Department of Psychology, Acadia University, Wolfville, NS

Date: January 22, 2025 Time: 2:30pm – 3:30pm Location: Room 113, Carnegie Hall Acadia University

kenneth.leslie@acadiau.ca

https://psychology.acadiau.ca/ken neth-leslie.html

Title: The Promise of Dream Technology

By Dr. Ken Raj Leslie (Acadia University)

Abstract: The discovery of REM sleep in 1953 revolutionized our understanding of dreaming. Subsequent work in the 1980s showed that dreamers can become aware they are dreaming and influence the dream, i.e., lucid dreaming. Early attempts at lucid dream technology failed commercially. The advent of wearable technology has created new opportunities for dream tech. Dr. Leslie founded MotionBed Inc. in 2021 as a sleep tech startup. He is developing an Apple Watch and iPhone system called DreamDirector, that helps users remember and influence their dreams. Dr. Leslie has been working with the Acadia Institute for Data Analytics (AIDA) since 2022 to develop real- time REM sleep detection technology for the purpose of dream engineering.

Brief Bio: Dr. Leslie is interested in applying discoveries and insights from psychology and neuroscience to create value in the real world. He has worked as a consumer neuroscientist in the private sector and is an expert in using brain and body measures to understand human responses. Dr. Leslie has also used fMRI to study the human mirror neuron system and its role in unconscious mimicry and empathy. He also has experience recording from individual neurons and helped identify a novel form of synaptic plasticity called synaptic scaling (Turrigiano et al., 1998). Dr. Leslie has an abiding interest in sleep and dreaming, and has studied dream incorporation, lucid dreaming, and the role of the vestibular system in sleepiness. In 2021 Dr. Leslie founded a start-up company called MotionBed Inc. in Kentville that is developing a consumer dream directing system, called DreamDirector.AI, that will allow users to influence their own dreams.

Access poster here

Nov

2024

Enhancing Interpretability in DL Models for Breast Cancer Detection (11:00 am)

Title: Enhancing Interpretability in DL Models for Breast Cancer Detection

Abstract: Breast cancer is one of the most common cancers found primarily in women, affecting 1 in 8 women in Canada over their lifetime, according to the Public Health Agency of Canada. The early diagnosis of breast cancer can reduce the spread of cancer cells and the risk of death. Machine learning (ML) systems can help healthcare professionals diagnose breast cancer with better accuracy by seeing the underlying patterns in the images that the radiologist might find difficult to detect early on. While deep learning (DL) models can obtain high accuracies, their black-box nature inhibits their ability to explain how they came to their conclusion and their regions of interest (ROI), creating distrust between DL systems and the radiologist. This research employs two CNN model architectures - AlexNet and Houby & Yassin's CNN – to classify recently obtained mammographic images from the KAU-BCMD dataset into six BI-RAD categories (0 - 6). To resolve the problem, SHAP, an explainability approach based on Shapley values, has been used to assess the role that each feature has towards the prediction process. There were two SHAP variants that were applied: PartitionSHAP and DeepSHAP. The method of employing the dataset through the models is by 1) preprocessing the dataset, 2) classifying the dataset in the BI-RADs 0-6, and 3) obtaining the SHAP plots identifying the model's ROIs. The proposed methods resulted in high-performance metrics for AlexNet and Houby & Yassin's CNN with an AUC and F1-score of 0.983 and 0.888, and 0.982 and 0.891, respectively.

Nov

2024

A Comparison of Autoencoders and Variational Autoencoders for Anomaly Detection in Dermoscopic Images (11:00 am)

Title: A Comparison of Autoencoders and Variational Autoencoders for Anomaly Detection in Dermoscopic Images

Abstract:

The early detection and diagnoses of skin abnormalities are crucial for effective treatment and management of skin diseases. This paper explores the application of deep learning techniques for skin tissue analysis, focusing on the detection of abnormalities from dermoscopic images. Unsupervised learning methods like Autoencoders (AE) and Variational Autoencoders (VAE) save resources by eliminating the need for labeled data, making them more efficient and scalable than supervised learning. We compare the performance of AE and VAE architectures in developing a robust model capable of distinguishing between benign and malignant skin lesions.

This study uses the HAM10000 dataset of 10,015 dermoscopic images, divided into seven classes representing both benign and malignant diagnostic categories. The dataset was split into benign (normal) and malignant (anomalous) cases. The models were trained to learn features of the normal data and generate reconstructions of these images. The reconstruction error measures how accurately the model interprets the image features. An optimal decision boundary is chosen to classify images as benign or malignant based on their reconstruction error.

The AE and VAE models were evaluated using accuracy, F1-score, and False Negative Rate (FNR). Minimizing FNR is crucial in healthcare as it indicates missed malignant cases. Experimental results, averaged over 30 training runs, demonstrate that the VAE outperforms the AE in accuracy (69.31% vs. 68.80%), while the AE surpasses the VAE with a higher F1-score (70.30% vs. 69.45) and a lower FNR (26.15% vs. 30.19%). These findings suggest the AE architecture is promising for automated skin cancer detection, potentially leading to more accurate and timely diagnoses and improved patient outcomes.

Nov

2024

Advancing Fruit Breeding using Computational Tools (2:30 pm - 3:30 pm)