Developing a humanoid bioreactor for tendon tissue engineering

Lead Research Organisation: University of Oxford
Department Name: Botnar Research Centre


Tendon diseases, like other musculoskeletal disorders, represent a growing social and economic burden as our population is aging. They often result in tears, causing pain and disability. Surgical repairs are performed at an increasing rate but patient's outcomes are not promising, with 40% of repairs failing at the shoulder joint due to poor tissue healing. Patients with major tissue loss have particularly little chances of recovering. A promising repair strategy is the use of engineered tendon grafts. Tissue engineering involves the development of bioreactors that generate tendon tissue in vitro using the patient's cells, scaffolds and mechanical stimulation. However, more advanced bioreactors are needed to provide functional tendon grafts. Current bioreactors mostly provide uniaxial cyclic loadings, while evidence suggests that they should provide multiaxial stresses, similar to those found physiologically. In this context, musculoskeletal humanoid robots have the potential to apply realistic stresses. These robots replicate the inner structures of the human body such as muscles, tendons and bones. They have seen major developments in recent years, making it now possible to consider their use for unexplored applications in medicine. The aim of this research project is to investigate the potential of using musculoskeletal humanoid robots as a platform for musculoskeletal tissue engineering and in particular for tendon engineering. The overall hypothesis is that humanoid robots will enable the provision of physiological mechanical stimulation and that, as a result, they will lead to engineered tendons that are more functional than those produced with current stretch bioreactors. To demonstrate this, we are proposing to: (1) design a flexible bioreactor chamber compatible with musculoskeletal humanoids, (2) adapt an existing humanoid shoulder for our tendon engineering applications, (3) define the loading regimes of the humanoid bioreactor based on shoulder rehabilitation exercises, (4) produce tendon constructs with the novel system and demonstrate improvements compared to current bioreactors. Computational modelling and motion analysis will be used to support our work. This pioneering project is a step towards functional and personalised grafts to improve patient outcomes and reduce costs to the society.

Planned Impact

Shoulder pain caused by rotator cuff tendon tear is responsible for prolonged periods of disability, absence from work and inability to carry out even basic household activities. Rotator cuff tendon tears are found in around 15% of 60 year olds, 25% of 70 year olds and 30% of 80 year olds. With the aging population and increased participation of the elderly in the labour force, their burden has become even more important. A recent study on health economics of rotator cuff estimated that successful repair would result in lifetime societal saving of $3.44 billion in the U.S. alone.

Approximately 10,000 of these rotator cuff tendon operations are carried out in the UK every year at a cost to the NHS of around £6800 per operation. However, conventional surgical repair (typically using sutures and anchors) still results in around 40% of failure. For patient with major tissue loss or experiencing a re-tear, conventional repair failure rates are even higher and, instead, tendon autograft transplantation may be performed. However, despite encouraging results, this approach remains rare mainly due to concern related to donor-site morbidity.

A regenerative medicine solution that provides engineered tendon grafts fabricated in vitro with the patient's cells would eliminate this main concern and would provide a reliable source of autografts for patients suffering either from a partial tear or from a large tears involving major tissue loss. The potential benefit to patients may include reduction of current failure rate of tissue repair, faster recovery and better tissue healing.

While current bioreactors used for fabricating engineered tendon have shown great potential, so far they have failed to produce grafts that are clinically relevant. This is in part because of the inadequate mechanical stimulation that they provide and the unsuitable dimensions of the constructs. In this context, humanoid musculoskeletal robots become relevant as they have the potential to closely mimic both the structure and the movements of the patient's body and thus to provide physiological mechanical stimulation to the tissue constructs. If the approach is proven to be promising, the long-term results of this research might lead to engineered grafts with improved functionality that are clinically relevant.

This may provide a long-term solution for the repair of the rotator cuff, which would translate into safer and more cost-effective patient care. But it could also have widespread applications for the repair of other musculoskeletal tissues including other tendons, ligaments, meniscus, and even cartilage and bone. A reliable and unlimited source of autografts is sought in a vast array of clinical areas and therefore the approach proposed here could potentially offer great benefits to patients through improved treatment for many conditions.


10 25 50
Description 1) We have developed a novel bioreactor chamber and have filed a UK patent application in September 2019, achieving deliverable D1. The chamber design is unique compared to existing bioreactor chambers as it is able to undergo multi-axial mechanical stimulation.
In WP1, we have been completed tasks T1.1 and T1.2 (Design a flexible bioreactor chamber with CAD software, Prototyping and manufacturing of the chamber with 3D printers, respectively). Task T1.3 (Implementation of mathematical models to optimise the chamber) has been initiated.

2) We have modified the humanoid shoulder provided by our partner in Munich to improve its mimicry of the human shoulder, achieving deliverable D2. For this we merged the CAD drawings of the original arm to scans of a scapula and humerus to produce a hybrid shoulder that meets our needs. A real shoulder implant has also been added to the shoulder to enable smooth motions between components.
In WP2, we have been completed tasks T2.1 and T2.3 (Adapt the humanoid shoulder and Defining progressive loading regimes based on physiotherapy rehabilitation exercises, respectively).

3) By combining our chamber and robotic arm, and by adding additional bioreactor components (incubator, pump, tubing, and connectors), we have produced an early prototype of the humanoid bioreactor system. This means milestone M1 has been achieved. The scaffold material used in the chamber is made of electrospun fibres, which have the ability to mimic the extracellular matrix of tendon tissue. We are currently preparing a research manuscript that introduces this humanoid bioreactor system. We believe that communications linked to the project (talks, media coverage, networking, etc.) will be facilitated by this first publication as until now dissemination of our current findings has been limited due to IP related considerations.
Exploitation Route Academic routes:
We expect that this project will create numerous opportunities for science, technology and medicine, such as in tissue engineering and regenerative medicine, basic cell/material/drug research (more physiologically relevant in vitro model), rehabilitation (through exploring loading regimes), mathematics and computer engineering (new models and validation opportunities), biomechanics (improved understanding of the shoulder joint) and robotics (development of improved musculoskeletal humanoids).

Non-academic route:
In the long term, grafts engineered with this approach could support soft tissue repair (such as tendon repair), with the potential to offer personalised and anatomically specific grafts. If applicable to other soft tissues, this strategy could further stimulate the UK economy, by taking part to a global market for soft tissue repair estimated to reach £16 billion in 2023.
Sectors Healthcare,Pharmaceuticals and Medical Biotechnology

Description Robotics group (Munich) 
Organisation Technical University of Munich
Country Germany 
Sector Academic/University 
PI Contribution By proposing to use musculoskeletal humanoids for medical applications, we have created a huge opportunity for the Munich team's robotic systems to be useful (so far these had no concrete applications). The TUM are now interested in developing more clinically relevant robotic systems, such as developing robotic arms that mimic closer the morphology and motion of human shoulders. Many of the latest developments that have taken place at the TUM have been influenced by our vision and requirements.
Collaborator Contribution The TUM has provided continuous support to ensure the proper functioning of the robotic arm during our experiments. This was done through providing both hardware and software support (about 1h/week in average). They have also provided the CAD drawings of their robotic arm necessary for us to bring our adaptations to the system.
Impact - Deliverable D2 achieved: adaptation of the MSK humanoid shoulder for building the humanoid bioreactor prototype.
Start Year 2019
Description BRC open day 
Form Of Engagement Activity Participation in an open day or visit at my research institution
Part Of Official Scheme? No
Geographic Reach Local
Primary Audience Patients, carers and/or patient groups
Results and Impact Our group held a stand on biomaterials and tissue engineering at the Oxford Biomedical Research Centre open day at the Nuffield Orthopedic Centre, exposing our research to about 30 patients and public members. This sparked many questions and discussion about our technologies in development.
Year(s) Of Engagement Activity 2019
Description Departmental news article 
Form Of Engagement Activity Engagement focused website, blog or social media channel
Part Of Official Scheme? No
Geographic Reach Local
Primary Audience Other audiences
Results and Impact Short news article produced by department to announce the project launch following the EPSRC award.
Year(s) Of Engagement Activity 2019
Description IF Oxford Science and Ideas Festival 
Form Of Engagement Activity Participation in an activity, workshop or similar
Part Of Official Scheme? No
Geographic Reach Local
Primary Audience Public/other audiences
Results and Impact Our group held a stand on biomaterials and tissue engineering at the IF Oxford Science and Ideas Festival in October 2019, exposing our research to 50+ public members. This sparked many questions and discussion about our technologies in development.
Year(s) Of Engagement Activity 2019
Description School class visit (Oxford Science) 
Form Of Engagement Activity Participation in an open day or visit at my research institution
Part Of Official Scheme? No
Geographic Reach Local
Primary Audience Schools
Results and Impact Our group hosted fifteen school girls (~yr10) on behalf on of Science Oxford (and through Engineering Science), with the idea is to inspire them into an engineering career path.
Year(s) Of Engagement Activity 2019