BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Brain age prediction and early neurodegeneration detection using c
 ontrastive learning on brain biomechanics - Jakob Träuble\, Department of
  Chemical Engineering and Biotechnology\, University of Cambridge
DTSTART:20260302T123000Z
DTEND:20260302T133000Z
UID:TALK243712@talks.cam.ac.uk
CONTACT:Dace Apšvalka
DESCRIPTION:*Speaker:* Jakob Träuble\, Department of Chemical Engineering
  and Biotechnology\, University of Cambridge\, UK.\n\n*Title:* Brain age p
 rediction and early neurodegeneration detection using contrastive learning
  on brain biomechanics\n\n*Abstract:* One of the main reasons why drugs fo
 r neurodegenerative diseases often fail is that treatment typically begins
  only after symptoms have appeared—by which point significant\, and poss
 ibly irreversible\, damage may have already occurred. Non-invasive imaging
  techniques\, such as Magnetic Resonance Imaging (MRI)\, have previously b
 een explored for presymptomatic diagnosis\, but with limited success. More
  recently\, Magnetic Resonance Elastography (MRE)—a technique capable of
  mapping the brain's biomechanical properties\, including stiffness and da
 mping ratio—has shown promise in detecting early changes. However\, curr
 ent studies have been limited by small sample sizes\, and a lack of robust
  algorithms capable of accurately interpreting data under such constraints
 .\n\nWe developed a self-supervised contrastive regression framework train
 ed on 3D MRE-derived stiffness and damping ratio maps from 311 healthy ind
 ividuals (aged 14–90) and evaluated its performance against structural 3
 D T1-weighted MRI. Brain age predictions were used to compute brain age ga
 ps (BAGs)\, quantifying deviations from normative ageing trajectories. We 
 applied the models to Alzheimer's disease (AD\, n = 11) and mild cognitive
  impairment (MCI\, n = 20) cohorts\, and analysed whole-brain and region-s
 pecific predictions using occlusion-based saliency maps and subcortical se
 gmentation.\n\nIn our controlled experimental setting\, MRE combined with 
 contrastive learning provides a sensitive\, non-invasive biomarker of brai
 n ageing and neurodegeneration\, outperforming MRI and differentiating dis
 ease stage–specific biomechanical signatures. Regional brain age gap pro
 filing may have the potential to identify at-risk\, cognitively normal ind
 ividuals\, which could facilitate timely intervention trials in the future
 \, pending longitudinal validation. \n\n*Bio:* Jakob Träuble is a PhD stu
 dent in Biotechnology at the University of Cambridge\, working in the Mole
 cular Neuroscience Group led by Prof. Gabriele Kaminski Schierle. His rese
 arch uses machine learning to study neuronal activity and brain elasticity
 \, with applications to neurodegenerative diseases. He holds a BSc in Phys
 ics from the University of Munich and an MPhil in Biotechnology from Cambr
 idge.\n\n*Venue:* MRC CBU West Wing Seminar Room and Zoom https://us02web.
 zoom.us/j/82385113580?pwd=RmxIUmphQW9Ud1JBby9nTDQzR0NRdz09 (Meeting ID: 82
 3 8511 3580\; Passcode: 299077)\n
LOCATION: MRC-CBU\, 15 Chaucer Road\, Cambridge
END:VEVENT
END:VCALENDAR