Engineering the Digital Universe

Speaker: Dr Sepideh Chakaveh
Time: 13:00-14:00
Seminar Room 2, Wolfson College & Online (Teams link).

Biography:

Dr Sepideh (Sepi) Chakaveh is the CEO/Founder of Pixsellar & the inventor of PixselChat, as well as an academic at the University of Oxford, teaching Data Science, AI & Software Engineering. Sepi is Research Common Room Member of Wolfson College Oxford, member of the advisory board of the Royal Holloway College Cyber Security Doctoral Training Centre, the winner of Everywoman Innovator Award 2020, and Technology Gamechanger 2021, Winner of The Innovation Excellence Streaming Platform of the Year 2022 and the Winner of Video AI Technology Specialists of the Year 2022 and 5V DeepTech Ambassodor 2023. Sepi has a degree in Electronics Engineering & a PhD in Experimental Astrophysics & Space Sciences. She has worked both in the UK & Germany as senior scientist/academic including, Imperial College, Max Planck Institute in Heidelberg, Fraunhofer Society & Southampton University where she was the co-founder and the former Director of Southampton Data Science Academy, the first UK on-line Data Science academy. Sepi is the inventor of PixselChat a multiple award bearing next generation of data communication platforms with embedded translation capabilities allowing everyone to speak with others without knowing their language. PixselChat is endorsed by the innovate UK as a global successful product created by a UK start-up company Pixsellar.

Abstract:

Engineering the Digital Universe invites us to reimagine our world as a layered, evolving construct built not from bricks and mortar but from data, information and knowledge. This talk explores the architecture of the digital universe—its foundational levels, from physical infrastructures and network protocols to platforms, social machines, and emergent digital societies.

We examine how data flows across these layers, becoming the lifeblood of digital experience, enabling everything from logistics to language. Central to this universe are social machines—complex assemblies where people and technology co-create meaning, knowledge, and value.

While AI occupies a vital position within this ecosystem, driving new capabilities in perception, prediction, and automation, it does not define the digital universe. Instead, it is one of many forces shaping its ongoing evolution.

This talk unpacks the engineering mindset required to responsibly and purposefully construct and steward the digital universe, and argues for a vision that balances computation with compassion, scale with ethics, and automation with accountability.

The session concludes with a live demonstration of real-world systems developed by the author's company, showcasing how these principles are being applied in practice to build meaningful, scalable digital products which solve real-life problems.

From Thermograms to Concepts: Using Concept Bottleneck Models to Balance Accuracy and Interpretability in Clinical AI Thermography

Speaker: Mustafa Alghali
Time: 13:00-14:00
Seminar Room 3, Wolfson College & Online (Teams link).

Biography:

Mustafa Alghali is the Co-founder and Chief Technology Officer of Amplifai Health, a Saudi-based healthcare technology company established in 2023. With a background in computer science and a focus on applied artificial intelligence, Mustafa has contributed to cutting-edge AI research at Mila – Quebec AI Institute and developed large-scale AI systems at Unity Technologies. As a technical visionary, he now leads the R&D team at Amplifai Health in building TFScan an AI-powered thermal imaging system designed for the early detection of diabetic foot ulcers. Amplifai Health is a healthtech company pioneering AI-driven thermal imaging to enable early, non-invasive detection of conditions like diabetic foot ulcers (DFUs). Their flagship solution, TFScan, is SFDA-approved and ISO-certified, combining portable thermography with AI-powered analysis to deliver instant risk assessments. This empowers clinicians with timely insights and improves patient outcomes. With a mission to make proactive health diagnosis globally accessible, Amplifai Health partners with healthcare systems—beginning in Saudi Arabia—to reduce preventable complications through scalable, data-informed tools.

Abstract:

This presentation, "From Thermograms to Concepts: Using Post hoc Concept Bottleneck Models (PCBMs) to Balance Accuracy and Interpretability in Clinical AI Thermography," will explore the critical role of interpretable AI in clinical settings, particularly within the evolving field of thermography.

The discussion will introduce the core idea of concept-based models (CBMs), highlighting their inherent interpretability while acknowledging the practical limitations they face in real-world deployments. Building on this, we present Post hoc Concept Bottleneck Models (PCBMs), demonstrating how they overcome these challenges. We'll discuss PCBMs' two primary approaches to reveal decision pathways and biases.

The discussion then centers on the specific complexities of clinical thermography. Given its evolving potential and the multifactorial risk factors that necessitate longitudinal clinical trials, we will illustrate how PCBMs can provide a crucial tool for understanding which predefined, guideline-driven concepts genuinely inform stratified risk assessments. Finally, Practical implementation is addressed by highlighting how PCBMs can be applied to thermographic data across longitudinal outcome studies. And how this approach not only supports outcome validation but also strengthens clinician engagement. By enabling healthcare professionals to interrogate, validate, and even influence the model’s conceptual focus.