Optimising patient specific treatment plans for ultrasound ablative therapies in the abdomen (OptimUS)
Lead Research Organisation:
University College London
Department Name: Mechanical Engineering
Abstract
Surgical and ablative technologies are the most effective local therapies for solid malignancies. The significant side effects associated with surgical interventions have led to an ongoing quest for safer, more efficient, and better tolerated alternatives. In recent years, there has been a notable shift away from open surgery towards less invasive procedures such as laparoscopic and robotic surgery, and from there to other methods for in situ tumour destruction, often involving energy based destruction. These include embolisation, radiofrequency, microwave and laser ablation, cryoablation and HIFU. HIFU is a procedure which uses high-amplitude ultrasound to thermally ablate a localised region of tissue. For abdominal applications, the ultrasound is typically generated by a focused transducer located extracorporealy. As the ultrasound propagates through tissue and at high acoustic intensities, absorption of the energy can induce targeted tissue necrosis within a well-defined volume without damaging the overlying tissue.
The aim of OptimUS is to develop a novel mathematical and treatment planning framework for tumours of the abdomen which are to be ablated using HIFU. This new framework will be purpose built for a CE marked commercially available MR-guided & monitored HIFU system, which includes a multi-element transducer. It will make use of high-performance computing capabilities which will deliver optimal treatment plans both accurately and fast. This framework will greatly extend and reinforce the role of HIFU as a completely non-invasive ablative modality, and radically improve the treatment of kidney, pancreas and liver tumours, of uterine fibroids and renal sympathetic denervation (RSD).
World-leading investigators and partners have been assembled to collaborate on this project, from University College London, University of Cambridge, the Institute of Cancer Research UK and Oxford HIFU Unit. All bring unique yet complementary expertise in the fields of mathematics, engineering, physics, oncology and computing to solve a series of complex challenges, which will lead to a significant increase in the quality of life and life expectancy of patients. In addition, the development of HIFU treatment planning methods will lead to more cost-effective cancer treatment protocols, saving the NHS time, resources and money.
The aim of OptimUS is to develop a novel mathematical and treatment planning framework for tumours of the abdomen which are to be ablated using HIFU. This new framework will be purpose built for a CE marked commercially available MR-guided & monitored HIFU system, which includes a multi-element transducer. It will make use of high-performance computing capabilities which will deliver optimal treatment plans both accurately and fast. This framework will greatly extend and reinforce the role of HIFU as a completely non-invasive ablative modality, and radically improve the treatment of kidney, pancreas and liver tumours, of uterine fibroids and renal sympathetic denervation (RSD).
World-leading investigators and partners have been assembled to collaborate on this project, from University College London, University of Cambridge, the Institute of Cancer Research UK and Oxford HIFU Unit. All bring unique yet complementary expertise in the fields of mathematics, engineering, physics, oncology and computing to solve a series of complex challenges, which will lead to a significant increase in the quality of life and life expectancy of patients. In addition, the development of HIFU treatment planning methods will lead to more cost-effective cancer treatment protocols, saving the NHS time, resources and money.
Planned Impact
Secondary liver cancer is estimated to affect more than 30,000 patients in the UK each year alone. Hepatocellular carcinoma, the most common form of primary liver cancer, is the fastest growing cancer in the UK. Liver cancer accounts for 3% of UK cancer deaths. Most cases of liver cancer are inoperable and the sufferer has an average life expectancy of less than one year. Cancers of the pancreas are often so far advanced at the time of diagnosis that patients are usually considered to be unsuitable for a surgical procedure. As for liver cancer, surgery is a lengthy procedure and its risks render it unsuitable for the majority of patients. For the treatment of liver and pancreatic cancer, targeted non-invasive surgical methods such as high-intensity focused ultrasound (HIFU) will lead to a significant increase in the quality of life and life expectancy of patients.
Kidney cancer is the fifth most common cancer in the UK, and its incidence is rising. 50% of people affected by renal cancer in England and Wales do not survive beyond 10 or more years and almost 6 in 10 kidney cancer patients receive major surgical resection as part of their cancer treatment. Kidney cancer surgery can be risky and is associated with complications in more than 20% of cases. Through optimal treatment planning, challenges of treating tumours of the kidney will be overcome.
Fibroids are common, with around 1 in 3 women developing them at some point in their life in the UK. They most often occur in women aged 30-50. Improvements in HIFU treatment planning methods for uterine fibroids through advanced treatment planning using numerical simulations will extend the patient base which could benefit from this modality and thus add more leverage towards NICE approval and clinical translation.
It is estimated that 1 billion individuals worldwide suffer from hypertension, and hypertension-related complications are recognised as the major cause of morbidity and mortality. Renal sympathetic denervation (RSD) is a procedure which has been shown to be of benefit for diseases associated with sympathetic overactivity, such as insulin resistance, arrhythmia, and heart failure. It is traditionally a catheter based intervention and can lead to complications. The prospect of carrying out this procedure non-invasively through a rigorous HIFU treatment planning protocol, would represent significant progress.
The development and experimental validation of a full wave treatment planning software for completely non-invasive ablative therapies of the abdomen using HIFU, and which can integrate seamlessly into a clinical system, will thus bear considerable impact. Mathematical and physical sciences research required to develop precise, non-invasive physical interventions to remove diseased tissue, via interventions such as HIFU will lead to more cost-effective cancer treatment protocols, saving the NHS time, resources and money.
Through our recently funded EPSRC-funded ThUNDDAR Network, we will engage with clinicians from different fields of oncology, radiologists, medical physicists and engineers to disseminate and discuss our findings. We will do this through organisation of specialist meetings, seed funding for pilot studies that cross disciplinary and institutional boundaries, and the production of web based information about therapy ultrasound for the benefit of industry, clinical users, health commissioners and patient groups.
Throughout the duration of the project, opportunities for patenting relevant innovations will arise which will benefit academia and UK plc. This will include (1) novel designs of HIFU array transducers, (2) novel treatment planning protocols and (3) specific tissue mimicking phantoms. Furthermore, the development of improved numerical schemes will enhance the open-source BEM++ and FEniCS software.
Kidney cancer is the fifth most common cancer in the UK, and its incidence is rising. 50% of people affected by renal cancer in England and Wales do not survive beyond 10 or more years and almost 6 in 10 kidney cancer patients receive major surgical resection as part of their cancer treatment. Kidney cancer surgery can be risky and is associated with complications in more than 20% of cases. Through optimal treatment planning, challenges of treating tumours of the kidney will be overcome.
Fibroids are common, with around 1 in 3 women developing them at some point in their life in the UK. They most often occur in women aged 30-50. Improvements in HIFU treatment planning methods for uterine fibroids through advanced treatment planning using numerical simulations will extend the patient base which could benefit from this modality and thus add more leverage towards NICE approval and clinical translation.
It is estimated that 1 billion individuals worldwide suffer from hypertension, and hypertension-related complications are recognised as the major cause of morbidity and mortality. Renal sympathetic denervation (RSD) is a procedure which has been shown to be of benefit for diseases associated with sympathetic overactivity, such as insulin resistance, arrhythmia, and heart failure. It is traditionally a catheter based intervention and can lead to complications. The prospect of carrying out this procedure non-invasively through a rigorous HIFU treatment planning protocol, would represent significant progress.
The development and experimental validation of a full wave treatment planning software for completely non-invasive ablative therapies of the abdomen using HIFU, and which can integrate seamlessly into a clinical system, will thus bear considerable impact. Mathematical and physical sciences research required to develop precise, non-invasive physical interventions to remove diseased tissue, via interventions such as HIFU will lead to more cost-effective cancer treatment protocols, saving the NHS time, resources and money.
Through our recently funded EPSRC-funded ThUNDDAR Network, we will engage with clinicians from different fields of oncology, radiologists, medical physicists and engineers to disseminate and discuss our findings. We will do this through organisation of specialist meetings, seed funding for pilot studies that cross disciplinary and institutional boundaries, and the production of web based information about therapy ultrasound for the benefit of industry, clinical users, health commissioners and patient groups.
Throughout the duration of the project, opportunities for patenting relevant innovations will arise which will benefit academia and UK plc. This will include (1) novel designs of HIFU array transducers, (2) novel treatment planning protocols and (3) specific tissue mimicking phantoms. Furthermore, the development of improved numerical schemes will enhance the open-source BEM++ and FEniCS software.
Organisations
- University College London (Lead Research Organisation)
- ETH Zurich (Collaboration)
- DURHAM UNIVERSITY (Collaboration)
- Pontifical Catholic University of Chile (Collaboration)
- Institute of Cancer Research UK (Collaboration)
- Stanford University (Collaboration)
- University of Utah (Collaboration)
- University of North Carolina at Chapel Hill (Collaboration)
- UNIVERSITY OF CAMBRIDGE (Collaboration)
- University of Eastern Finland (Collaboration)
- Pierre and Marie Curie University - Paris 6 (Collaboration)
- Oxford University Hospitals NHS Trust (Project Partner)
Publications
Aubry JF
(2022)
Benchmark problems for transcranial ultrasound simulation: Intercomparison of compressional wave models.
in The Journal of the Acoustical Society of America
Betcke T
(2018)
Boundary element methods with weakly imposed boundary conditions
Betcke T
(2022)
Hybrid coupling of finite element and boundary element methods using Nitsche's method and the Calderon projection
in Numerical Algorithms
Betcke T
(2022)
Boundary Element Methods for Helmholtz Problems With Weakly Imposed Boundary Conditions
in SIAM Journal on Scientific Computing
Betcke T
(2019)
Boundary Element Methods with Weakly Imposed Boundary Conditions
in SIAM Journal on Scientific Computing
Burman E
(2019)
A fully discrete numerical control method for the wave equation
Burman E
(2019)
A stabilized finite element method for inverse problems subject to the convection-diffusion equation. I: diffusion-dominated regime
in Numerische Mathematik
Burman E
(2018)
Primal-Dual Mixed Finite Element Methods for the Elliptic Cauchy Problem
in SIAM Journal on Numerical Analysis
Burman E
(2020)
Stability estimate for scalar image velocimetry
Burman E
(2023)
Stability estimate for scalar image velocimetry
in Journal of Inverse and Ill-posed Problems
Burman E
(2020)
A finite element data assimilation method for the wave equation
in Mathematics of Computation
Burman E
(2022)
Unfitted hybrid high-order methods for the wave equation
in Computer Methods in Applied Mechanics and Engineering
Burman E
(2021)
A Hybridized High-Order Method for Unique Continuation Subject to the Helmholtz Equation
in SIAM Journal on Numerical Analysis
Burman E
(2018)
A finite element data assimilation method for the wave equation
Burman E
(2019)
Unique continuation for the Helmholtz equation using stabilized finite element methods
in Journal de Mathématiques Pures et Appliquées
Burman E
(2021)
Convergence Analysis of Hybrid High-Order Methods for the Wave Equation
in Journal of Scientific Computing
Burman E
(2019)
Cut finite elements for convection in fractured domains
in Computers & Fluids
Burman E
(2019)
Cut topology optimization for linear elasticity with coupling to parametric nondesign domain regions
in Computer Methods in Applied Mechanics and Engineering
Burman E
(2019)
Hybridized CutFEM for Elliptic Interface Problems
in SIAM Journal on Scientific Computing
Burman E
(2020)
A Fully Discrete Numerical Control Method for the Wave Equation
in SIAM Journal on Control and Optimization
Burman E
(2021)
Hybrid High-Order Methods for the Acoustic Wave Equation in the Time Domain
in Communications on Applied Mathematics and Computation
Description | We have created OptimUS, which is an open-source treatment simulation platform which models the propagation of ultrasonic waves in heterogeneous media. It features treatment planning capabilities based on constrained optimisation techniques. A multi-domain boundary element formulation with hybrid CPU/GPU architecture has been implemented and tested on clinically relevant scenarios, which can run in minutes on a standalone workstation. Nonlinearities are treated in the focal region using an efficient method based on a boundary integral formulation of the Westervelt equation. |
Exploitation Route | We have created an open source repository for OptimUS (at https://github.com/optimuslib/optimus). This will allow academic and clinical users full access to the treatment planning and simulation platform that we have put together with this funding. We will continue to further refine and develop this platform. |
Sectors | Healthcare |
Description | OptimUS has been used to gain insight into MR thermometry and thermocouple measurements on a bone phantom exposed to the field of a Philips Sonalleve HIFU treatment platform. This work was carried out in conjunction with Dr Fiammetta Fedele's team at Guys and St Thomas' Hospital and its aim was to quantify the mechanisms by which HIFU can help with palliative care of bone metastases and to better inform the treatment plan. OptimUS has also been part of a multinational benchmarking exercise for identifying the best software for transcranial ultrasound therapy treatment planning. OptimUS has also been used by several different research groups around the world, such as INSERM Lab-tau in Lyon France and The institute of Cancer Research, UK for their design and interpretation of experimental work. |
Sector | Healthcare |
Impact Types | Societal |
Description | Therapy Ultrasound Network for Drug Delivery & Ablation Research (ThUNDDAR) |
Amount | £676,000 (GBP) |
Funding ID | EP/N026942/1 |
Organisation | Engineering and Physical Sciences Research Council (EPSRC) |
Sector | Public |
Country | United Kingdom |
Start | 08/2016 |
End | 08/2020 |
Title | OptimUS |
Description | Library of routines for the propagation of ultrasonic waves in heterogeneous media with a Python interface. Accurate simulation of ultrasound interaction with human anatomical features and pathologies will reduce the reliance on the use of animal models in the long run. |
Type Of Material | Computer model/algorithm |
Year Produced | 2021 |
Provided To Others? | Yes |
Impact | The computer models in the repository will soon be released to the public. The date for going public has been pushed back due to the Covid pandemic, but it is hoped it will happen in 2022. |
URL | https://github.com/UCL/OptimUS_Bempp |
Description | Benchmark problems for transcranial ultrasound simulation |
Organisation | ETH Zurich |
Country | Switzerland |
Sector | Academic/University |
PI Contribution | The team is participating in a multi-centred, international benchmarking exercise for simulating transcranial transmission of ultrasound. Our team is using the platform developed on this grant (OptimUS) to compare against 10 other modelling tools on a range of benchmarks of increasing geometric complexity. |
Collaborator Contribution | The partners each contribute their own suite of routines and modelling tools for the inter-comparison study. |
Impact | A journal paper has been submitted to the Journal of The Acoustical Society of America. It is in the reviewing process. |
Start Year | 2020 |
Description | Benchmark problems for transcranial ultrasound simulation |
Organisation | Pierre and Marie Curie University - Paris 6 |
Country | France |
Sector | Academic/University |
PI Contribution | The team is participating in a multi-centred, international benchmarking exercise for simulating transcranial transmission of ultrasound. Our team is using the platform developed on this grant (OptimUS) to compare against 10 other modelling tools on a range of benchmarks of increasing geometric complexity. |
Collaborator Contribution | The partners each contribute their own suite of routines and modelling tools for the inter-comparison study. |
Impact | A journal paper has been submitted to the Journal of The Acoustical Society of America. It is in the reviewing process. |
Start Year | 2020 |
Description | Benchmark problems for transcranial ultrasound simulation |
Organisation | Stanford University |
Country | United States |
Sector | Academic/University |
PI Contribution | The team is participating in a multi-centred, international benchmarking exercise for simulating transcranial transmission of ultrasound. Our team is using the platform developed on this grant (OptimUS) to compare against 10 other modelling tools on a range of benchmarks of increasing geometric complexity. |
Collaborator Contribution | The partners each contribute their own suite of routines and modelling tools for the inter-comparison study. |
Impact | A journal paper has been submitted to the Journal of The Acoustical Society of America. It is in the reviewing process. |
Start Year | 2020 |
Description | Benchmark problems for transcranial ultrasound simulation |
Organisation | University of Eastern Finland |
Country | Finland |
Sector | Academic/University |
PI Contribution | The team is participating in a multi-centred, international benchmarking exercise for simulating transcranial transmission of ultrasound. Our team is using the platform developed on this grant (OptimUS) to compare against 10 other modelling tools on a range of benchmarks of increasing geometric complexity. |
Collaborator Contribution | The partners each contribute their own suite of routines and modelling tools for the inter-comparison study. |
Impact | A journal paper has been submitted to the Journal of The Acoustical Society of America. It is in the reviewing process. |
Start Year | 2020 |
Description | Benchmark problems for transcranial ultrasound simulation |
Organisation | University of North Carolina at Chapel Hill |
Department | Pediatric Allergy and Immunology North Carolina |
Country | United States |
Sector | Academic/University |
PI Contribution | The team is participating in a multi-centred, international benchmarking exercise for simulating transcranial transmission of ultrasound. Our team is using the platform developed on this grant (OptimUS) to compare against 10 other modelling tools on a range of benchmarks of increasing geometric complexity. |
Collaborator Contribution | The partners each contribute their own suite of routines and modelling tools for the inter-comparison study. |
Impact | A journal paper has been submitted to the Journal of The Acoustical Society of America. It is in the reviewing process. |
Start Year | 2020 |
Description | Benchmark problems for transcranial ultrasound simulation |
Organisation | University of Utah |
Country | United States |
Sector | Academic/University |
PI Contribution | The team is participating in a multi-centred, international benchmarking exercise for simulating transcranial transmission of ultrasound. Our team is using the platform developed on this grant (OptimUS) to compare against 10 other modelling tools on a range of benchmarks of increasing geometric complexity. |
Collaborator Contribution | The partners each contribute their own suite of routines and modelling tools for the inter-comparison study. |
Impact | A journal paper has been submitted to the Journal of The Acoustical Society of America. It is in the reviewing process. |
Start Year | 2020 |
Description | Development of frequency-robust preconditioning of boundary integral equations for ultrasound propagation |
Organisation | Pontifical Catholic University of Chile |
Country | Chile |
Sector | Academic/University |
PI Contribution | This work has been a true collaboration, supported by two visits by Dr. Elwin van 't Wout to UCL and a visit by the PDRA on this project to Chile. The following were mainly contributed by our team: Unified design of different boundary integral formulations and Consistent application of Calderón and OSRC preconditioning to all boundary integral formulations. |
Collaborator Contribution | a. A suitable and novel preconditioner for the so-called PMCHWT formulation b. Computational validation of conditioning at high frequencies |
Impact | So far three papers have been published and one more has been submitted and is being reviewed. |
Start Year | 2018 |
Description | Discussion of a joint proposal on treatment planning for trans-cranial HIFU |
Organisation | Durham University |
Department | School of Engineering and Computing Sciences |
Country | United Kingdom |
Sector | Academic/University |
PI Contribution | The team at UCL submitted a joint proposal to the EPSRC for a large grant in collaboration with Durham University. This application was not successful but we are encouraged by the positive feedback by the majority of referees who were supportive of the scientific aspects of the proposal. We plan to re-visit this proposal with improvements based on the feedback from the referees. We are now seeking to expand this collaboration by including clinical collaborators from France, the USA and Italy. We aim to re-direct the effort towards modelling of the Blood-Brain Barrier opening for drug delivery to the brain, rather than high intensity ablation of the tumour. |
Collaborator Contribution | The team at UCL produced a joint proposal to the EPSRC for a large grant in collaboration with Durham University. |
Impact | A grant proposal to the EPSRC. |
Start Year | 2018 |
Description | Formation of a consortium for Ultrasound ablation of bone tumours |
Organisation | Institute of Cancer Research UK |
Country | United Kingdom |
Sector | Academic/University |
PI Contribution | Using the findings of this project, we have formed a large consortium with clinical partners from University of Oxford, Imperial College, Royal Marsden Hospital, Stanford University, and University of Rome Sapienza, as well as academics from Durham University and the Institute of Cancer Research. The research aims of this consortium will be to discover why certain types of bone tumours (such as osteoid osteoma) respond well to ultrasound thermal ablation whereas others, mainly malignant forms of bone tumours (osteosarcomas) cannot be treated. This will be mainly aimed at paediatric tumours. The final objective will be to define the parameter space in which all forms of bone tumour could be treatable using high intensity focused ultrasound. |
Collaborator Contribution | The partners bring expert medical and clinical knowledge, as well as guidance on the likelihood of clinical success for the proposed approach. |
Impact | No outputs yet. |
Start Year | 2021 |
Description | Optimus |
Organisation | University of Cambridge |
Department | Department of Engineering |
Country | United Kingdom |
Sector | Academic/University |
PI Contribution | UCL brings to this partnership the following: Therapeutic ultrasound expertise, experience in treatment planning development, a large-scale open-source boundary element library and expertise in FEM-BEM coupling. |
Collaborator Contribution | Garth Wells (university of Cambridge) is one of the main developers of the FEniCS project (http://fenicsproject.org/). The FEniCS project is an open source software library for the automated solution of partial differential equations using the finite element method. FEniCS has an excellent user interface and employs latest software technologies including code generation to provide the user with a flexible and efficient way of solving partial differential equations. |
Impact | In addition to several papers in top scientific journals and numerous presentations at international conferences, this collaboration has produced an open source suite of programmes for treatment planning in therapeutic ultrasound. The collaboration is multi-disciplinary, bringing together expertise in engineering, medical physics and mathematics. |
Start Year | 2017 |
Title | OptimUS |
Description | OptimUS is an open- source treatment simulation platform which models the nonlinear propagation of ultrasonic waves in heterogeneous media. It features treatment planning capabilities based on constrained optimisation techniques. |
Type Of Technology | Software |
Year Produced | 2021 |
Open Source License? | Yes |
Impact | OptimUS represents a disruptive innovation in fast and accurate patient specific therapeutic ultrasound treatment planning. |