A comprehensive in silico approach to measure spatio-temporal changes of bone tissue in mouse models of osteoporosis and osteoarthritis

Project background

Osteoporosis and osteoarthritis are the most common chronic diseases of the musculoskeletal system and impair the quality of life of millions of people in the UK. Osteoporosis results in severe bone loss, reduced bone strength and an increased risk of fracture. In osteoarthritis, the cartilage of the joints thin and degrade resulting in stiffness and pain, which can also result in bone loss. Treatments for osteoporosis are successful in only a proportion of patients and there is no cure for osteoarthritis. Preclinical studies for new interventions are often performed in mice, where aspects of osteoporosis and osteoarthritis are induced by invasive surgery. New treatments ideally need to increase bone strength but measuring bone strength is invasive and requires large numbers of animals so preclinical studies usually rely on simplified measures. With previous NC3Rs funding, a protocol using longitudinal micro-computed tomography (microCT) imaging was developed allowing the entire mouse tibia to be imaged non-invasively, which avoids the need for sacrificing mice at every experimental time point.

Why we funded it

This Project Grant aims to develop computational models for use with longitudinal microCT imaging protocols to predict mouse tibia mechanical properties (e.g. stiffness and strength) in preclinical assessments of interventions for osteoporosis and osteoarthritis. These models will be disseminated internationally using a semi-automated service on a web-interface.

The previously developed microCT imaging protocol resulted in a 63% reduction in the number of mice required for an experiment measuring changes of bone mineral content. Extending these to experiments detecting early changes in the subchondral bone of osteoarthritic joints will save approximately 190 mice per year in the Skelet.AL facility at the University of Sheffield. To induce joint phenotypes similar to those observed in osteoarthritis patients, one knee of a mouse is destabilised surgically, which is classified as a moderate procedure under the Animals (Scientific Procedures) Act 1986. Ovariectomies are also a common procedure to mimic the effects of loss of oestrogen in menopausal women with osteoporosis. Extending the protocol to include experiments such as those in this Project Grant through the development of the new computational models, automated procedures and dissemination through the web-interface would allow a comparable level of reduction both in Dr Dall’Ara’s laboratory and other groups performing similar studies.  

Research methods

In clinical research, bone strength can be measured using finite element computational models derived from data obtained in longitudinal CT scans. In this study Dr Dall’Ara’s group will create a similar pipeline for preclinical applications starting from microCT images of the mouse tibia. Three different models will be evaluated and validated against ex vivo data to determine which model is the best predictor for bone strength. The model will account for bone geometry, density distribution and microarchitecture and be used under different loading scenarios to predict bone strength and fracture risk. Imaging processing protocols will then be extended to determine whether it is possible to detect small changes in the subchondral bone of the knee joint preceding cartilage degeneration in OA. These will be measured by combining in vivo microCT images, image registration and automatic image processing comparing healthy mice to mice that have undergo destabilisation of the medial meniscus.

This project aims to improve the current preclinical assessment of interventions for bone diseases (i.e. osteoporosis, OP, and osteoarthritis, OA), by developing more accurate longitudinal measurements in mice and evaluating endpoints, which are more relevant to predict the outcomes of the treatment in patients, and therefore facilitate the clinical translation.

We will develop and validate in vivo microCT based computational models for prediction of the mouse tibia strength, which is more related to risk of fracture than standard bone density measurements.

We will also develop a semi-automated web-service for allowing other researchers worldwide to perform non-invasive estimations of bone strength and spatio-temporal measurements of bone properties, which are more accurate than those performed with standard cross-sectional experimental designs, and lead therefore to a 63% reduction of the number of mice to be used. In this project we will convert the manual procedure we can use only in our laboratory into a semi-automatic approach, share the developed scripts with expert bone modellers, and offer a web-service to all researchers worldwide, who are willing to use this approach but do not have the expertise.

Finally we will extend the previously developed protocols for accurate measurements of early changes of knee subchondral bone in OA in vivo. We will evaluate the longitudinal changes induced by destabilization of the medial meniscus, the most common OA mouse model, and we will validate the outcomes of the longitudinal approach against ex vivo measurements of cartilage and bone from a standard cross-sectional experiment. We are confident that this will extend the reduction of required number of mice for projects, which aim to understand OA pathogenesis, to create new less invasive OA animal models and to develop new interventions for this disease.

Cheong VS et al. (2021). Positive interactions of mechanical loading and PTH treatments on spatio-temporal bone remodelling. Acta Biomaterialia, in press. doi: 10.1016/j.actbio.2021.09.035

Cheong VS, Kadirkamanathan V and Dall'Ara E (2021). The Role of the Loading Condition in Predictions of Bone Adaptation in a Mouse Tibial Loading Model. Frontiers in Bioengineering and Biotechnology, in press. doi: 10.3389/fbioe.2021.676867

Oliviero S et al. (2020). Optimization of the failure criterion in micro-Finite Element models of the mouse tibia for the non-invasive prediction of its failure load in preclinical applications. Journal of the Mechanical Behavior of Biomedical Materials, in press. doi: 10.1016/j.jmbbm.2020.104190

Rakowski AG et al. (2020). ChronoMID—Cross-modal neural networks for 3-D temporal medical imaging data. PLoS One 15(2):e0228962. doi: 10.1371/journal.pone.0228962

Roberts BC et al. (2020). PTH(1–34) treatment and/or mechanical loading have different osteogenic effects on the trabecular and cortical bone in the ovariectomized C57BL/6 mouse. Scientific Reports 10: e8889. doi: 10.1038/s41598-020-65921-1

Cheong VS et al. (2019). A novel algorithm to predict bone changes in the mouse tibia properties under physiological conditions. Biomechanics and Modeling in Mechanobiology. doi: 10.1007/s10237-019-01266-7 

Oliviero S et al. (2019). Effect of repeated in vivo microCT imaging on the properties of the mouse tibia. PLoS One 14(11):e0225127. doi: 10.1371/journal.pone.0225127

Roberts BC et al. (2019). The longitudinal effects of ovariectomy on the morphometric, densitometric and mechanical properties in the murine tibia: A comparison between two mouse strains. Bone 127:260-270. doi: 10.1016/j.bone.2019.06.024

Viceconti M and Dall’Ara E (2019). From bed to bench: How in silico medicine can help ageing research.  Mech Ageing Dev 177:103-108. doi: 10.1016/j.mad.2018.07.001

Zhang Y et al. (2019). A new method to monitor bone geometry changes at different spatial scales in the longitudinal in vivo μCT studies of mice bones. PLoS One 14(7):e0219404. doi: 10.1371/journal.pone.0219404

Oliviero S I (2018). Validation of finite element models of the mouse tibia using digital volume correlation. J Mech Behav Biomed Mater. 81:172-184.  doi: 10.1016/j.jmbbm.2018.06.022

Giorgi M and Dall'Ara E (2018). Variability in strain distribution in the mice tibia loading model: A preliminary study using digital volume correlation. Med Eng Phys. 62:7-16. doi: 10.1016/j.medengphy.2018.09.001


Back to top
Project grant




University of Sheffield

Grant reference number


Award date

Dec 2017 - Apr 2021

Grant amount