The world’s most accurate passive radio-frequency identification device location-tracking
You can book for in-person attendance at Bookwhen, where you will also find the link for Zoom viewing.
The digitisation of retail, healthcare, manufacturing, logistics and supply chain processes means that Internet of Things, smart sensors and other AI driven processes are becoming increasingly prevalent. However, the smallest data inaccuracies can have the most profound operational, safety and security consequences and so existing inventory tracking solutions are typically not fit for purpose because they cannot detect and track tagged items with the required accuracy at speed. Based on his PhD at Cambridge, Sabesan founded PervasID which has developed the world’s most accurate passive Radio-Frequency Identification Device (RFID) location tracking solution powered by Al. Already adopted by major retailers globally, this technology enhances retail inventory management by accurately tracking goods in retail stores using existing tags, replacing time-consuming, labour-intensive manual handheld scanning. Integrated with 3rd party stock management applications giving retailers a graphical map interface showing the location and movement of all tagged goods, the solution improves sales, customer satisfaction, and prevents losses by offering real-time visibility, timely restocking, efficient in-store fulfilment, enhanced click-and-collect and online order services. Furthermore, it can improve full-price sales and decrease end-of-season markdowns.
In the industrial sector, Stanley Black & Decker use the tags to track supplies from their tool cabinets used by aircraft manufacturers. Each cabinet contains over 1,000 tools, and it can result in serious safety incidents if any are left inside an aircraft. It is estimated that Foreign Object Debris (FOD) costs the aviation industry $13 Billion per year in direct and indirect costs, including flight delays, plane changes and fuel inefficiencies. In healthcare, PervasID solutions are being deployed in NHS hospitals for tracking surgical instruments to enhance decontamination and sterilisation processes and for tracking hospital assets to ensure that mission critical medical devices are available at the right place and time, for robust and efficient care. The need for this level of traceability of medical devices has been particularly evident in the COVID-19 pandemic. The solution is predicted to save £billions for NHS hospitals and will save lives.
Dr Sabesan Sithamparanathan OBE, PervasID
Dr Sabesan Sithamparanathan OBE is a multi-award-winning entrepreneur with more than 15 years’ experience in the IoT space. As Founder and President of his Cambridge University spin-out company, PervasID, Sabesan pioneered and developed the world’s most accurate passive RFID technology, creating a world-leading range of products that are transforming entire industries, generating substantial exports and saving lives. Sabesan’s products and endeavours have both national and international benefit and he has become renowned for forging the worlds of academia and business to great effect, for his intellectual and scholarly excellence, and for the entrepreneurial skills that have enabled him to put innovative ideas into practice in a commercially viable way. Sabesan has become an expert in entrepreneurship, strategic business development and innovation leadership. Sabesan was awarded Officer of the Order of the British Empire (OBE) in the King’s 2024 New Year Honours, The Royal Academy of Engineering Silver Medal 2021 for an outstanding and demonstrated personal contribution to UK engineering and Queen’s Award for Enterprise: Innovation 2021. Sabesan is a Fellow of the Royal Academy of Engineering (REng), Intuition of Engineering and Technology (IET) and the ERA Foundation. Sabesan holds a PhD in Engineering from the University of Cambridge and has completed management leadership training at the Harvard Business School.
Attending lectures
The lecture will be preceded by a short presentation from a CSAR PhD Award Winner.
ModelAngelo: Automated atomic model building and protein sequence discovery for cryo-EM maps
Kiarash Jamali, MRC Laboratory of Molecular Biology
Cryo electron microscopy (cryo-EM) produces three-dimensional (3D) maps of the electrostatic potential of biological macromolecules, including proteins. Along with knowledge about the imaged molecules, cryo-EM maps allow de novo atomic modelling, which is typically done through a laborious manual process. Taking inspiration from recent advances in machine learning applications to protein structure prediction, we propose a graph neural network (GNN) approach for automated model building of proteins in cryo-EM maps. The GNN acts on a graph with nodes assigned to individual amino acids and edges representing the protein chain. Combining information from the voxel-based cryo-EM data, the amino acid sequence data, and prior knowledge about protein geometries, the GNN refines the geometry of the protein chain and classifies the amino acids for each of its nodes. Application to test cases shows that our approach outperforms the state-of-the-art and approximates manual building for cryo-EM maps with resolutions better than 3.5 Å. Furthermore, we are able to use a modified form of ModelAngelo to discover protein sequences from the cryo-EM map alone.