POND2024
19–20 Sep, 2024
UCL (London, UK)
UCL (London, UK)
| Home | Registration | Program | Directions | Flyer |
| 1230-1250 | Arrivals and Lunch |
| Sandwiches provided. | |
| 1255 | Welcome |
| Neil Oxtoby (UCL) | |
| Welcome to POND 2024! | |
Phenomenological modelling |
|
| Chair: Neil Oxtoby | |
| 1300 | Keynote 1. Disease progression modelling: from descriptive to generative approaches for clinical decision making |
| Marco Lorenzi, Inria Sophia-Antipolis, France | |
| 1345 | A machine learning-based prediction of tau load and distribution in Alzheimer’s disease using plasma, MRI and clinical variables |
| Linda Karlsson, Lund University, Sweden | |
| Tau positron emission tomography (PET) is a reliable neuroimaging technique for assessing regional load of tau pathology in the brain, commonly used in Alzheimer’s disease (AD) research and clinical trials. However, its routine clinical use is limited by cost and accessibility barriers. In this project, we explore using machine learning (ML) models to predict clinically useful tau-PET outcomes from low-cost and non-invasive features, e.g., basic clinical variables, plasma biomarkers, and structural magnetic resonance imaging (MRI). Results demonstrated that models including plasma biomarkers yielded highly accurate predictions of tau-PET burden, with especially high contribution from plasma P-tau217. In contrast, MRI variables stood out as best predictors of asymmetric tau load between the two hemispheres (an example of clinically relevant spatial information). The models showed high generalizability to external test cohorts with data collected at multiple sites. Based on these results, we also propose a proof-of-concept two-step classification workflow, demonstrating how the ML models can be translated to a clinical setting. This study reveals current potential in predicting tau-PET information from scalable cost-effective variables, which could improve diagnosis and prognosis of AD. | |
| 1405 | From disease progression models to improving clinical trials: An application to Alzheimer’s disease |
| Pierre-Emmanuel Poulet, ARAMIS Lab of Inria Paris, France | |
| Disease progression models have shown significant advancements in predicting cognitive decline across various neurodegenerative diseases. However, the challenge of translating these predictive insights into clinical practice remains open. In this talk, we will explore a methodological framework designed to enhance the effectiveness of clinical trials, but also demonstrate how it can inform and refine our modeling choices. We will then delve into the implementation of two strategies of mixture using a spatiotemporal longitudinal mixed-effects model to account for disease heterogeneity. We will showcase the method with an application to an Alzheimer's disease clinical trial. | |
| 1425 | A General and Generative Approach to Spatiotemporal Modeling of Medical Images |
| Lemuel Puglisi, University of Catania, Italy: Homepage | |
| Medical images are the gold standard for quantifying in vivo neurodegeneration caused by brain disorders. Key biomarkers are extracted from medical imaging data to understand the disease dynamics. The latest advancements in generative AI, combined with the growing availability of imaging data, have unlocked the potential to model disease progression directly within high-dimensional medical images. In this presentation, we will explore both the challenges associated with this task and a proposed solution, called Brain Latent Progression (BrLP). | |
| 1445 | BREAK |
| Refreshments provided | |
Recent highlights from UCL (CMIC and DRC) |
|
| Chair: Alex Young | |
| 1515 | Rapid Power Pitches |
| Speakers TBC | |
| 1600 | Estimating the time between Aβ positivity and elevated regional tau in the 1946 British birth cohort |
| Will Coath, UCL, UK | |
| 1615 | Unscrambling disease progression at scale with optimal transport |
| Peter Wijeratne, University of Sussex and UCL, UK: Homepage | |
| 1630 | 15-minute Comfort Break |
Connectivity-based modelling 1 |
|
| Chair: Anna Schroder | |
| 1645 | Demonstration of an open source toolbox for network spreading models: regional amyloid burden promotes tau production in Alzheimer’s disease |
| Ellie Thompson, UCL, UK: Homepage | |
| 1655 | Effects of regional neurotransmitter receptor densities on modelling amyloid and tau accumulation in Alzheimer’s disease with network spreading models |
| Sonja Soskic, UCL, UK | |
| 1705 | A coupled-mechanisms probabilistic inference framework for neurodegenerative disease progression |
| Tiantian He, UCL, UK | |
| 1715 | Day 1 Close |
| 1800 | Dinner |
| 0915 | Welcome |
| TBC (UCL) | |
| Day 1 Recap, Day 2 Overview. | |
Connectivity-based modelling 2 |
|
| Chairs: Neil Oxtoby and Alex Young | |
| 0930 | Keynote 2. Mechanisms of Tau spreading in Alzheimer's disease and primary tauopathies |
| Nicolai Franzmeier, LMU Munich, Germany: Homepage | |
| 1015 | Connectome-based AD Progression Models: From Group-Level to Individualized Predictions |
| Yu Xiao, Lund University, Sweden | |
| The network-based neurodegeneration hypothesis suggests that the onset of Alzheimer's Disease (AD) occurs within key vulnerable regions, whose network connectivity guides the spread of atrophy and disease pathological protein into new areas of the brain. This idea has inspired various disease progression models, particularly focused on structural and functional brain connectivity. In this talk, I will present the work my lab and I have done on AD progression modeling using connectome-based methods. At the group level, I will discuss simulations of tau propagation, highlighting how different brain connectivity and regional biological factors influence its spread throughout the brain. I will then introduce my work on predicting regional atrophy and cognitive decline using individual connectomes, with the goal of providing precise clinical forecasts and facilitating earlier, more targeted interventions for those at risk of AD. | |
| 1030 | BREAK |
| Refreshments provided | |
Beyond the POND: Broader applications |
|
| Chair: Alex Young | |
| 1100 | Title TBC |
| Joe Jacob, UCL, UK | |
| Abstract TBC | |
| 1120 | Title TBC |
| Andre Altmann, UCL, UK | |
| Abstract TBC | |
| 1140 | Neuroanatomical normative modelling in dementia |
| James Cole, UCL, UK | |
| Abstract TBC | |
| 1200 | Departures |