
8/ Together, these studies form a unified step forward: large‑scale, automated musculoskeletal phenotyping that connects spinal curvature, muscle composition, metabolic health, and disease risk. 🧵 End.

Professor of Metabolic Imaging, applying precision phenotyping to disease processes. @RCOptimalHealth, University of Westminster. All views my own.

8/ Together, these studies form a unified step forward: large‑scale, automated musculoskeletal phenotyping that connects spinal curvature, muscle composition, metabolic health, and disease risk. 🧵 End.

7/ Read the full paper here: doi.org/10.1038/s41598-026-6… Huge congratulations to 1. Marili Niglas, 2. Marjola Thanaj and 3. Brandon Whitcher for leading these research projects.

6/ Paper 3 introduces an automated 3D deep‑learning framework for precise segmentation of 20 hip and thigh muscles across >13,000 MRI scans. It reveals sex‑specific differences, heterogeneity in fat infiltration, distinct T2D signatures, and longitudinal changes in muscle health.

5/ Type 2 diabetes showed sex‑specific deformation patterns: inward in men, outward in women. Some spatial features even predicted incident T2D, pointing to new imaging biomarkers for risk stratification. doi.org/10.1002/rco2.70046

4/ Paper 2 focuses on the gluteus maximus, combining conventional MRI with 3D regional morphology. Distinct shape changes were associated with physical activity, frailty, osteoporosis, alcohol intake, grip strength; Not apparent with just volume or fat fraction alone.

3/ These highlight the strong connections between muscle quality, metabolic health, and spinal alignment doi.org/10.1007/s00586-026-1…

2/ Scoliosis was dramatically under‑recognised: only 0.5% self‑reported scoliosis, yet automated MRI detected curvature >10° in 28% of participants. Curvature was linked to age, paraspinal muscle fat infiltration, visceral adiposity, and chronic back pain.

THREAD: We have 3 new papers, one shared goal; mapping musculoskeletal health at scale. 1/ Our new study in the European Spine Journal of fully automated Cobb angle estimation across 33,889 @uk_biobank MRI scans, enabling population‑level assessment of spinal curvature.

MRI shows a 4–5% muscle/lean mass decline over ~2.5 years — missed by DXA. Bottom line: DXA is practical at scale, but MRI gives a more precise picture of lean mass change, especially in ageing and longitudinal studies.

New paper out! nature.com/articles/s43856-0… Using data from 32,961 @uk_biobank participants, we compared DXA vs MRI for body composition. Key takeaways: DXA works well for fat measures. It overestimates lean mass vs MRI.

Using deep learning on 44k @uk_biobank abdominal MRIs, we mapped heart and vascular structures, linked these traits to CVD risk, and uncovered 15 new genetic associations. Abdominal MRI is a powerful, untapped tool for understanding cardiovascular health rdcu.be/e5f3C

I was ridiculously excited to meet @AuntMinnie today! You never know what’s going to happen next at @RSN. There is just so much to explore. A fantastic first day #RSNA25

We’re proud to share the publication of a new paper in the Journal of Abdominal Radiology on automated MRI quantification of body fat and skeletal muscle. Study highlights: ✅ For clinicians: A rapid, single-slice assessment of body fat and muscle, strongly correlated with gold-standard volumetric measures, to support routine monitoring, triage, and decision-making. 💊 For pharma: Trial-ready volumetric MRI biomarkers for body composition, validated against expert manual segmentations and interchangeable across methods and vendors, enabling objective, scalable endpoints in obesity, cardiometabolic and musculoskeletal trials 👏 Congratulations to Magdalena Nowak, Luis Núñez, Charles E. Hill, Tim Pagliaro, John McGonigle, Marili Niglas, Camilo Bell-Bradford, Matthew Robson, Helena Thomaides Brears, E. Louise Thomas, and Jimmy D. Bell on this important contribution. 🔗 link.springer.com/article/10… 👉 BBC bbc.co.uk/news/articles/c8d6… #Perspectum #BodyComposition #MedicalImaging #ClinicalResearch #Pharma #ClinicalTrials #UKBiobank #Collaboration

All speaker presentations from last month's 100K Imaging Project celebration at the Royal Society are now available to view ✅

Wonderful to see out graduating students walk across the stage at the Royal Festival Hall this afternoon. What a joyous celebration of their achievement, congratulations to all of our graduates and their families 🎓@UniWestminster

Please to share our latest @RCOptimalHealth paper led by Elena Sorokin and @YaghootkarH now published in @FrontGenetics @LifeSciWestmin frontiersin.org/journals/gen…