The study, registered retrospectively on ClinicalTrials.gov as NCT07162168, enrolled 4,152 participants who had undergone noncontrast routine CT scans. The cohort included 1,947 men, with a median age of 59 years spanning a remarkable range from 18 to 99. Rather than relying on slow, labor-intensive manual tracing of the femur, the team trained a deep learning segmentation model based on the nnU-Net framework, a self-configuring architecture widely regarded as one of the most robust tools for biomedical image segmentation. A subset of proximal femurs was first annotated by hand to teach the model exactly where the bone begins and ends in each volumetric scan. Once trained, the automated system could carve out the femoral volume of interest on its own, achieving a Dice similarity coefficient of 0.963 on the test set, a score indicating near-perfect overlap with expert annotations.

With the femur reliably isolated in thousands of scans, the researchers turned to radiomics, the practice of converting medical images into hundreds of quantitative features that describe texture, shape, intensity and spatial patterns within a defined region. Using the open-source PyRadiomics package, they extracted features from each segmented femur, following standardized conventions designed to make radiomic measurements reproducible across scanners and institutions. The idea behind radiomics is deceptively simple: the microscopic architecture of bone, including trabecular thinning, cortical porosity and the redistribution of mineral, leaves fingerprints in the statistical texture of a CT image. Those fingerprints, aggregated across thousands of features, may encode a person’s skeletal age far more richly than a single density measurement.

To test whether these features actually carry information about aging and sex, the team built machine learning models in a development cohort of 2,969 participants and then evaluated them in a completely separate external validation cohort of 1,183 participants, a design that guards against the optimistic results that plague many single-center AI studies. For age, the best performer was a support vector regression model, which estimated chronological age from femoral radiomics with a coefficient of determination of 0.50 in external validation, a mean absolute error of 8.84 years and a root mean squared error of 11.32 years. In plain terms, the model could roughly halve the uncertainty of a blind guess about a person’s age using nothing but the texture of their hip bone.

Sex classification proved even more striking. Two algorithms, support vector classification and the gradient-boosting method XGBoost, both achieved an area under the receiver operating characteristic curve of 0.94 in external validation, with support vector classification showing slightly higher accuracy and F1-score. An AUC of 0.94 means the model distinguished male from female skeletons with high reliability across the entire cohort, capturing structural sexual dimorphism in the femur that goes well beyond simple size differences. Because the features were extracted automatically from routine scans, the result hints that sex-related skeletal architecture is legible in ordinary clinical imaging without any dedicated bone protocol.

Perhaps the most consequential finding concerns the relationship between the radiomics-derived age index and actual bone mineral biology. In the external validation cohort, the researchers compared their radiomic age index against volumetric bone mineral density of the femoral neck measured by quantitative computed tomography, the gold-standard three-dimensional density technique. The radiomics index tracked vBMD more strongly than chronological age did, with a Spearman correlation of minus 0.546 versus minus 0.443 for age alone, a difference the authors confirmed statistically using Steiger’s test (Z = 4.97, P < 0.001). This comparison matters because it suggests the radiomic signature is not merely a proxy for how many birthdays a person has had, but a reflection of genuine structural deterioration in the bone, the kind of change that underlies osteoporosis and hip fracture risk.

The clinical logic here is what researchers call opportunistic imaging. Dual-energy X-ray absorptiometry, the standard test for osteoporosis, is underused: many people at risk never get scanned, and fractures often arrive as the first sign of disease. But a large fraction of older adults undergo abdominal, chest or pelvic CT for unrelated reasons, and those scans routinely capture the proximal femur. If an automated pipeline can convert that incidental imaging into a skeletal age index or a bone quality flag, clinicians could identify patients with accelerated bone aging who would benefit from confirmatory density testing or preventive treatment, all without a single additional scan, extra radiation dose or added visit. The authors caution that their radiomic patterns still need validation against hard clinical outcomes such as incident fractures before such screening becomes practice, but the biological anchoring to vBMD is an encouraging first step.

The technical achievement should not be undersold either. Automated segmentation at a Dice coefficient above 0.96 removes the bottleneck that has historically limited radiomics to small, single-center studies requiring hours of expert contouring. By combining nnU-Net segmentation, standardized PyRadiomics extraction and rigorous external validation across nationwide cohorts, the study offers a template for how imaging biomarkers can be scaled to population level. The work was supported by the Beijing Natural Science Foundation-Fengtai Joint Innovation Project, the National Natural Science Foundation of China Science Center Program and the Shenzhen Medical Research Fund, and it was approved by the ethics committee of Peking University People’s Hospital with informed consent waived for the retrospective use of de-identified images.

There are, of course, caveats to keep in mind. The cohort was drawn from Chinese hospitals, and radiomic features can be sensitive to scanner vendor, reconstruction settings and dose, so multicenter generalization beyond this dataset remains to be demonstrated. The age model’s mean absolute error of nearly nine years is impressive for image-only prediction but far too coarse to serve as a precise biological clock for any individual. And the study measured associations with density, not with fractures, falls or mortality, so the ultimate clinical value hinges on longitudinal follow-up. Still, the direction of travel is clear: the femur that appears incidentally on a routine scan carries a quantifiable structural biography, written in texture patterns that machines can now read at scale.

If subsequent studies confirm that radiomics-derived bone age predicts fractures and responds to osteoporosis treatment, the implications could extend well beyond orthopedics. Bone is a living archive of hormonal change, nutrition, inflammation and mechanical loading, and a cheap, automated readout of its microarchitecture could inform research on aging itself, from sarcopenia to metabolic bone disease. For now, the study stands as a proof of concept that the skeleton’s quiet remodeling, normally invisible in the reading room, can be decoded from data that hospitals already possess, waiting in the archive.

Subject of Research: Automated CT radiomics of the proximal femur for quantifying age- and sex-related bone structural changes

Article Title: Automated proximal femur CT radiomics identifies age- and sex-related radiomic patterns of bone structure: a nationwide multicohort validation study

Article References: Cheng, H., Zhuang, L., Zhang, Y., Ran, Y., Jiang, B., & Kou, Y. (2026). Automated proximal femur CT radiomics identifies age- and sex-related radiomic patterns of bone structure: a nationwide multicohort validation study. BMC Medical Imaging. https://doi.org/10.1186/s12880-026-02836-9

Image Credits: AI Generated

DOI: 10.1186/s12880-026-02836-9

Keywords: radiomics, proximal femur, CT imaging, deep learning, nnU-Net, bone aging, bone mineral density, osteoporosis, machine learning, opportunistic screening, quantitative CT, hip fracture risk

The enzyme in question is protein arginine methyltransferase 6, or Prmt6, the only known writer of a histone mark called H3R2me2a—asymmetric dimethylation of arginine 2 on histone H3. This mark is generally associated with transcriptional repression and is mutually exclusive with the active H3K4 methylation marks, meaning that where H3R2me2a sits, genes tend to stay quiet. In cancer biology, Prmt6 has already earned a reputation as a brake on cellular senescence: by depositing H3R2me2a near genes such as p53 and p21, it suppresses the expression of senescence programmes and keeps cells proliferating. Whether the same anti-ageing mechanism operated in bone, however, had never been tested—until a team working under protocols approved by the Tongji University Committee for the Protection and Use of Laboratory Animals set out to interrogate it in mice.

The researchers built their case on two complementary models of osteoblast ageing. In the first, they exposed cultured newborn mouse skull-derived osteoblasts to 200 micromolar hydrogen peroxide for two hours, inducing oxidative-stress-driven senescence. In the second, they simply passaged the cells repeatedly, letting replicative exhaustion accumulate naturally. Both models worked as intended: western blots and quantitative PCR showed rising levels of the senescence markers P16 and P21, immunofluorescence confirmed their intracellular accumulation, and senescence-associated beta-galactosidase staining revealed a growing fraction of blue-stained, senescent cells. Crucially, as senescence deepened, Prmt6 and its H3R2me2a mark faded from the cells at both the mRNA and protein levels—the first hint that this epigenetic system weakens as osteoblasts age.

To test causality rather than mere correlation, the team isolated osteoblasts from Prmt6 knockout mice. Removing the enzyme crashed H3R2me2a levels, and the senescence programme surged: P16 and P21 climbed, beta-galactosidase-positive cells multiplied, and the expression of osteogenic markers—Alp, Ocn and the master transcription factor Runx2—fell significantly. In practical terms, bone-forming cells without Prmt6 not only looked old; they lost their ability to mature and mineralise matrix. Transcriptome-wide RNA sequencing reinforced the picture. Principal component analysis cleanly separated knockout cells from controls along the first principal component, and differential expression analysis identified 2,478 altered genes, 682 upregulated and 1,796 downregulated. Gene ontology and KEGG pathway enrichment pointed to a triad of disruption: suppressed ossification and collagen matrix assembly, enhanced immune cell migration and TNF-alpha production, and altered PI3K-Akt and Rap1 signalling pathways known to govern osteoblast survival and adhesion.

Among all these changes, one innate immune gene stood out. Cross-referencing the differentially expressed genes against Sting-pathway members yielded two candidates, Sting and Nfkbia, but only Sting was consistently and stably regulated by Prmt6 loss. Sting—the stimulator of interferon genes—is the central adaptor of the cGAS-STING pathway, which normally senses cytoplasmic DNA and launches a type I interferon response. In recent years, the pathway has been implicated in ageing far beyond its anti-infective origins: aberrant Sting activation drives chronic low-grade inflammation, senescence-associated secretory phenotype expression and, in the skeleton, mitochondrial dysfunction through the HK2-VDAC1 axis. In the knockout osteoblasts, Sting mRNA and protein rose sharply, and the same elevation appeared in both chemically induced and replicative senescence models.

The functional test followed: when the researchers knocked out Sting, senescence markers P16 and P21 dropped, the antioxidant enzyme Gpx4 increased, beta-galactosidase-positive cells declined, and the osteogenic markers Alp, Ocn and Runx2 rebounded. In other words, silencing Sting did not merely correlate with healthier osteoblasts—it partially rescued them from senescence and restored differentiation capacity. Sting, it appeared, was not an innocent bystander in the ageing bone but an active driver of the senescent state that Prmt6 loss had unleashed.

Downstream of Sting, the trail led to Ifitm3, an interferon-induced transmembrane protein better known for blocking enveloped viruses such as influenza and dengue by stiffening endosomal and lysosomal membranes. Ifitm3 is a classic interferon-stimulated gene, so when Sting signalling fires, Ifitm3 expression follows. The new data confirmed the dependency: Sting knockdown reduced Ifitm3 at every level measured. And like Sting, Ifitm3 climbed in senescent osteoblasts, in Prmt6-deficient cells, and in the femurs of naturally ageing mice, while Ifitm3 knockout reduced P16 and P21, lowered beta-galactosidase positivity, and boosted osteogenic gene expression. The authors suggest that by altering membrane fluidity, promoting lysosomal activity and disrupting autophagic flux, Ifitm3 may be one of the molecular hands that physically push an osteoblast into senescence—though they caution that the precise effector mechanisms remain to be disentangled.

The animal data tied the cellular story to real bone loss. Using micro-CT scanning at 25-micrometre resolution on femurs from mice aged 2, 12 and 24 months, the team documented the expected march of osteoporosis: trabecular number, bone volume fraction, bone surface density and trabecular thickness all fell with age, while trabecular separation widened. Haematoxylin and eosin staining showed coarsened trabecular edges, enlarged marrow spaces and microfractures. Superimposed on this structural decay was the molecular signature: Prmt6 and H3R2me2a declined progressively in femoral tissue, while Sting and Ifitm3 rose—mirroring exactly the gradients seen in the culture dish.

The authors are candid about the limits of the work. The chromatin accessibility changes at the Sting locus following H3R2me2a loss are currently correlative; direct causality, and the possibility that Prmt6 acts through H3R2me2a-independent routes, await rescue experiments. Likewise, whether Ifitm3 directly executes senescence or merely permits it remains unresolved, with experiments involving cholesterol-binding mutants, lysosomal pH modulators and live autophagic-flux monitoring now under way. Even so, the study assembles the first integrated link between a Prmt6-mediated histone modification, the Sting innate immune pathway and osteoblast senescence in vivo, and it reframes age-related osteoporosis as a disease in which the bone-forming compartment loses its epigenetic defences rather than one driven simply by overactive osteoclasts.

If the axis holds up, the therapeutic implications are considerable. The open chromatin state of the Sting locus could serve as a biomarker of osteoblast ageing, and drugs that restrain Sting signalling or restore H3R2me2a deposition might delay or partially reverse bone loss in the elderly. With population ageing intensifying worldwide and osteoporosis already a leading cause of disability and fracture-related mortality, an epigenetic handle on the disease would be a welcome addition to a therapeutic landscape dominated for decades by antiresorptive agents. The road from mouse femurs to human clinics is long, and the authors themselves flag that long-term efficacy and safety of any such epigenetic strategy must be systematically evaluated—but the mapping of this pathway marks a genuine step forward in understanding why our bones grow old.

Subject of Research: Epigenetic regulation of osteoblast senescence in age-related osteoporosis via the Prmt6/H3R2me2a/Sting/Ifitm3 pathway

Article Title: Prmt6 Deficiency Drives Osteoblast Senescence Promoting Age‐Related Osteoporosis via Epigenetic Remodelling of the H3R2me2a/Sting/Ifitm3 Pathway

Article References: Wang, Y., Ge, X., Chang, S., Wang, D., Xu, K., Xu, H., Liu, X., & Wang, S. (2026). Prmt6 Deficiency Drives Osteoblast Senescence Promoting Age‐Related Osteoporosis via Epigenetic Remodelling of the H3R2me2a /Sting/Ifitm3 Pathway. Journal of Cellular and Molecular Medicine, 30(18), Article e71308. https://doi.org/10.1111/jcmm.71308

Image Credits: AI Generated

DOI: 10.1111/jcmm.71308

Keywords: osteoporosis, osteoblast senescence, Prmt6, H3R2me2a, STING, Ifitm3, epigenetics, cellular ageing, bone metabolism, innate immunity, chromatin remodelling, ageing