Keynote Speaker

Title: 

         AI for Solving Real-World Challenges in Brain Imaging

Abstract:  

MRI is an important and versatile medical imaging modality. Brain MRI is used to diagnose and prognose a range of different disorders ranging from developmental disorders, degenerative disorders, stroke and cancers. In this talk I will focus on four domains where recent developments in machine learning and artificial intelligence have made important contributions: Image Quality Transfer allows to improve the image quality of archived historic scans but also the images generated by older, less powerful MRI scanners; Segmentation is routinely used on brain MRIs to identify specific brain regions and their size and novel methods show broad generalizability of segmentation algorithms to various MRI contrasts; Harmonization ensures that data obtained at different sites and with different scanners can be analysed jointly; Normative Modelling aims to learn how healthy brains are "put together" and change as they mature and age in order to detect abnormal processes. 

Short Bio:  

Andre Altmann is Professor of Biomedical Data Science and AI at UCL. He trained in Computer Science at RWTH Aachen before completing his PhD in Computational Biology at the Max Planck Institute for Informatics. Prior to joining UCL, he held research positions at the Max Planck Institute of Psychiatry and Stanford University. At UCL, he founded the COMBINE (COMputational Biology in Imaging and geNEtics) Lab within the UCL Hawkes Institute.

His research focuses on understanding how molecular and genetic mechanisms shape brain structure and function in both healthy and diseased populations. In particular, his work investigates the biological basis of neuropsychiatric and neurodegenerative disorders, including depression and Alzheimer’s disease, through the integration of neuroimaging, molecular genetics, and machine learning. He is also interested in developing AI-driven biomarkers for early disease detection, patient stratification, and personalized treatment, with the long-term goal of improving precision medicine approaches in brain health.

Short Bio:  

Alex Zhou is the Founder and CEO of Linkerbot, a serial entrepreneur with a distinctive multi-disciplinary background. A graduate of the elite Gifted Youth Class at Huazhong University of Science and Technology, his expertise spans mathematics, computer science, and engineering. With over 15 years of experience bridging consumer internet products and robotics R&D, he has a proven track record of leading a product to a user base of 300 million.

Today, he leads Linkerbot to become a global leader in dexterous hands and dexterous creation foundation models. The company fully covers the three major technical routes of tendon-driven, linkage, and direct-drive systems, achieving a monthly production peak of over 4,000 units and capturing more than 80% of the global high-DoF dexterous hand market. It has completed multi-billion RMB cumulative financing. Driven by the mission of “creating everything” and the vision of realizing “Doraemon’s magical pocket,” the company pursues full-stack, in-depth R&D from underlying materials and actuators to physical creation foundation models. Zhou has led the development of the Linker Hand series, the Open TeleDex teleoperation system, the LinkerSkillNet skill library, and the Linker Genesis foundation model, setting global benchmarks in dexterous manipulation and intelligent creation.

Title: 

           Linkerbot: From Hand to Data to Action -- Opening the Data Flywheel for General Dexterous Manipulation

Abstract:

This talk will introduce Linkerbot, the global leader in high-DoF dexterous hands with 80% market share and a 500+ R&D team. We present our latest direct-drive innovations, O20 and O30, alongside two new releases for CVPR 2026: UMI-Dex, an open-source, isomorphic portable data collection solution, and DexScale, a large-scale multimodal dataset for universal dexterous manipulation. Both are designed to bypass retargeting and fuel vision-tactile policy learning. We invite the research community to leverage our open-source data and join our Scholar & Genius Programs, offering hands-on access to industry-leading hardware and technical support. Together, we advance embodied intelligence from hand to action.

Short Bio:  

Mandy Ma is the Head of Research Outreach at Linkerbot, who brings over eight years of experience in building an industry‑academia innovation ecosystem. She leads Linkerbot’s flagship “Scholar Program” and “Genius Researcher Program,” connecting top‑tier researchers worldwide to co‑develop cutting‑edge exploratory projects. Mandy also orchestrates Linkerbot’s strategic presence at premier academic conferences (e.g., IROS, CVPR), oversees the company’s participation in academic competitions, and collaborates with universities on joint curriculum design. Holding an MPhil from the University of Cambridge’s Institute for Manufacturing and a B.Eng. from Beijing Institute of Technology, Mandy bridges robotics and AI research with industry needs. Over eight years, she has turned joint R&D into commercial products and built sustainable talent pipelines, including postdoctoral programs.