https://jeroenvanschependom.be/tag/ms/MSWowchemy (https://wowchemy.com)en-usSat, 01 Jan 2022 00:00:00 +0000https://jeroenvanschependom.be/media/icon_hu898bad6a947579ff95e6bfac4d9757a2_264109_512x512_fill_lanczos_center_3.pnghttps://jeroenvanschependom.be/tag/ms/https://jeroenvanschependom.be/project/artificial-intelligence/Sat, 01 Jan 2022 00:00:00 +0000https://jeroenvanschependom.be/project/artificial-intelligence/<p>We study where artificial intelligence can genuinely add value in clinical neurology, with a focus on cognitive impairment, prognosis, and scalable analysis of multimodal data.</p> <h2 id="why-this-matters" style="font-size: 1.2rem; background: #FFB76B; background: linear-gradient(to right, #FFB76B 0%, #FFA73D 30%, #FF7C00 60%, #FF7F04 100%); -webkit-background-clip: text; -webkit-text-fill-color: transparent;">Why this matters</h2> <p>AI is often presented as a universal solution in healthcare. Our work instead asks where machine learning is robust, clinically meaningful, and realistic in neurological disease.</p> <h2 id="current-directions" style="font-size: 1.2rem; background: #FFB76B; background: linear-gradient(to right, #FFB76B 0%, #FFA73D 30%, #FF7C00 60%, #FF7F04 100%); -webkit-background-clip: text; -webkit-text-fill-color: transparent;">Current directions</h2> <ul> <li>AI for cognitive and clinical stratification</li> <li>Federated learning across distributed clinical datasets</li> <li>Methodological work on generalisability, bias, and clinical utility</li> </ul> <h2 id="earlier-work" style="font-size: 1.2rem; background: #FFB76B; background: linear-gradient(to right, #FFB76B 0%, #FFA73D 30%, #FF7C00 60%, #FF7F04 100%); -webkit-background-clip: text; -webkit-text-fill-color: transparent;">Earlier work</h2> <p>One of my earlier explorations of AI in clinical care evaluated whether machine-learning approaches could detect metabolic syndrome in patients treated with antipsychotic medication. That work highlighted a recurring lesson in medical AI: apparent accuracy can depend strongly on the population under study.</p> <h2 id="on-ai-in-ms-care" style="font-size: 1.2rem; background: #FFB76B; background: linear-gradient(to right, #FFB76B 0%, #FFA73D 30%, #FF7C00 60%, #FF7F04 100%); -webkit-background-clip: text; -webkit-text-fill-color: transparent;">On AI in MS care</h2> <p>We have contributed to the debate on whether AI will transform multiple-sclerosis care in the coming decade. For me, the important question is not whether AI is fashionable, but whether it improves decisions, scales across centres, and remains interpretable enough for clinical use.</p> <h2 id="related-outputs" style="font-size: 1.2rem; background: #FFB76B; background: linear-gradient(to right, #FFB76B 0%, #FFA73D 30%, #FF7C00 60%, #FF7F04 100%); -webkit-background-clip: text; -webkit-text-fill-color: transparent;">Related outputs</h2> <p>See the <a href="https://jeroenvanschependom.be/publication/">Publications</a> page for selected work on AI, federated learning, and digital biomarkers.</p>https://jeroenvanschependom.be/project/cognitiveimpairment/Sat, 01 Jan 2022 00:00:00 +0000https://jeroenvanschependom.be/project/cognitiveimpairment/<p>Cognitive impairment is an important and often under-recognised part of multiple sclerosis and related neurological disorders. A major aim of our work is to understand which cognitive domains are affected, how this evolves over time, and how it can be monitored more objectively.</p> <h2 id="why-this-matters" style="font-size: 1.2rem; background: #FFB76B; background: linear-gradient(to right, #FFB76B 0%, #FFA73D 30%, #FF7C00 60%, #FF7F04 100%); -webkit-background-clip: text; -webkit-text-fill-color: transparent;">Why this matters</h2> <p>Cognitive impairment has a substantial impact on quality of life, work participation, and daily functioning. Yet it remains difficult to capture efficiently in routine care and difficult to treat in a targeted way.</p> <h2 id="characterising-impairment" style="font-size: 1.2rem; background: #FFB76B; background: linear-gradient(to right, #FFB76B 0%, #FFA73D 30%, #FF7C00 60%, #FF7F04 100%); -webkit-background-clip: text; -webkit-text-fill-color: transparent;">Characterising impairment</h2> <p>Our work has shown that information-processing speed occupies a central position in the clinical picture of cognitive dysfunction in multiple sclerosis, and that reduced performance in this domain can influence many broader cognitive outcomes.</p> <h2 id="developing-biomarkers" style="font-size: 1.2rem; background: #FFB76B; background: linear-gradient(to right, #FFB76B 0%, #FFA73D 30%, #FF7C00 60%, #FF7F04 100%); -webkit-background-clip: text; -webkit-text-fill-color: transparent;">Developing biomarkers</h2> <p>We develop multimodal biomarkers of cognitive impairment using:</p> <ul> <li>neurophysiology, including P300-based markers</li> <li>MRI-derived measures of structural damage</li> <li>shape analysis of the corpus callosum</li> <li>functional connectivity and dynamic-network analysis</li> <li>blood and CSF biomarkers</li> </ul> <p>For neurophysiological work specifically, see the <a href="https://jeroenvanschependom.be/project/neurophysiology/">Neurophysiology</a> project page.</p> <h2 id="clinical-monitoring" style="font-size: 1.2rem; background: #FFB76B; background: linear-gradient(to right, #FFB76B 0%, #FFA73D 30%, #FF7C00 60%, #FF7F04 100%); -webkit-background-clip: text; -webkit-text-fill-color: transparent;">Clinical monitoring</h2> <p>A current focus is to make cognitive monitoring more scalable in the clinic, including app-based testing and image-based modelling approaches that can support longitudinal follow-up.</p> <h2 id="treatment-development" style="font-size: 1.2rem; background: #FFB76B; background: linear-gradient(to right, #FFB76B 0%, #FFA73D 30%, #FF7C00 60%, #FF7F04 100%); -webkit-background-clip: text; -webkit-text-fill-color: transparent;">Treatment development</h2> <p>Because cognitive impairment remains difficult to treat, we are exploring new intervention strategies, including combined physical-cognitive training and non-invasive neuromodulation approaches.</p> <h2 id="related-outputs" style="font-size: 1.2rem; background: #FFB76B; background: linear-gradient(to right, #FFB76B 0%, #FFA73D 30%, #FF7C00 60%, #FF7F04 100%); -webkit-background-clip: text; -webkit-text-fill-color: transparent;">Related outputs</h2> <p>See the <a href="https://jeroenvanschependom.be/publication/">Publications</a> page for selected papers on cognitive impairment, cognitive decline, and multimodal biomarkers.</p>https://jeroenvanschependom.be/project/neurophysiology/Sat, 01 Jan 2022 00:00:00 +0000https://jeroenvanschependom.be/project/neurophysiology/<p>Because cognitive impairment is difficult to assess objectively, we study brain activity directly to understand how neurological disease alters large-scale brain function.</p> <h2 id="brain-dynamics" style="font-size: 1.2rem; background: #FFB76B; background: linear-gradient(to right, #FFB76B 0%, #FFA73D 30%, #FF7C00 60%, #FF7F04 100%); -webkit-background-clip: text; -webkit-text-fill-color: transparent;">Brain dynamics</h2> <p>Modern neurophysiology increasingly goes beyond static connectivity analyses. We study transient brain states with distinct spatial, temporal, and spectral signatures, because these short-lived states offer a window on the timescales at which cognition unfolds.</p> <h2 id="spectral-analysis" style="font-size: 1.2rem; background: #FFB76B; background: linear-gradient(to right, #FFB76B 0%, #FFA73D 30%, #FF7C00 60%, #FF7F04 100%); -webkit-background-clip: text; -webkit-text-fill-color: transparent;">Spectral analysis</h2> <p>Oscillatory activity, especially in the alpha range, has repeatedly been linked to cognitive impairment in multiple sclerosis. A central question in our work is how specific these associations are and which aspects of cognition they capture most strongly.</p> <h2 id="excitation-and-inhibition" style="font-size: 1.2rem; background: #FFB76B; background: linear-gradient(to right, #FFB76B 0%, #FFA73D 30%, #FF7C00 60%, #FF7F04 100%); -webkit-background-clip: text; -webkit-text-fill-color: transparent;">Excitation and inhibition</h2> <p>The 1/f spectral slope has been proposed as a marker of excitation-inhibition balance. We study how this measure changes under pharmacological manipulation and whether it can help explain variation in cognitive functioning in neurological disease.</p> <h2 id="why-this-matters-for-biomarkers" style="font-size: 1.2rem; background: #FFB76B; background: linear-gradient(to right, #FFB76B 0%, #FFA73D 30%, #FF7C00 60%, #FF7F04 100%); -webkit-background-clip: text; -webkit-text-fill-color: transparent;">Why this matters for biomarkers</h2> <p>By linking dynamic brain activity to cognition, we aim to build biomarkers that are closer to mechanism than conventional behavioural testing alone.</p> <h2 id="related-outputs" style="font-size: 1.2rem; background: #FFB76B; background: linear-gradient(to right, #FFB76B 0%, #FFA73D 30%, #FF7C00 60%, #FF7F04 100%); -webkit-background-clip: text; -webkit-text-fill-color: transparent;">Related outputs</h2> <p>See the <a href="https://jeroenvanschependom.be/publication/">Publications</a> page for selected work on MEG, transient brain states, oscillatory activity, and spectral slope measures.</p>