Dynamic optimisation of tumor's dense-dose scheduling via closed-loop control and beyond
Lead Research Organisation:
University of Southampton
Department Name: Sch of Electronics and Computer Sci
Abstract
Closed-loop control (CLC) theory is widely used in different industries (robotics, automotive, aerospace, etc.) and to a less extent in clinical settings, despite its success story in the automation of glucose control for diabetic patients through an artificial pancreas. This proposal looks to jointly define a future research agenda driven by inputs from oncological experts (directly and through observation) for the deployment of CLC in decision making for tumour growth control in a heterogeneous environment. This will be presented to the medical and academic communities via multiple channels at the end of this collaborative research. The research proposed herein covers a 12-month preliminary feasibility study enabled by two 2-month periods of collaborative research with the Memorial Sloan Kettering Cancer Centre (MSKCC) in New York, funded by the OTG.
Publications
Belkhatir Z
(2023)
Nonlinear Impulsive Control Design for Biologically Grounded NSM Tumor Growth Model Using Exact Linearized Mapping
in IFAC-PapersOnLine
| Description | This EPSRC overseas travel grant (OTG) award funded a visit to Memorial Sloan Kettering Cancer Center (MSKCC) in New York. The visit was very fruitful as it enabled the PI to discuss and collaborate with clinicians, biologists and radiologists on the topic of adaptive cancer therapy and beyond. The visit was a big enabler to co-design the current challenges hindering to move closed-loop control theory a step forward towards clinics. One of the biggest achievement of this visit was to formulate a clinically sound learning-based closed-loop control problem that will support the systematic adaptation decision-making of cancer chemotherapy. The visit has also opened further discussion and collaborative work on cancer-related research questions. A bigger follow-up project would help to address the questions formulated during the funded travel visit. An interdisciplinary collaborative efforts led by the PI, who was funded by this OTG award, are being undertaken. |
| Exploitation Route | The key outcome of this research visit was to initiate collaborations with clinical experts on a topic of interest to them using technological tools from AI and closed-loop control theory. Efforts are being made to address the co-formulated research questions that will enable a more systematic personalized adaptation of cancer chemotherapy scheduling that is envisioned to improve cancer patients care and their response to therapy in the longer term. |
| Sectors | Healthcare Other |
| Description | Quantitative cancer-MSKCC |
| Organisation | Memorial Sloan Kettering Cancer Center |
| Country | United States |
| Sector | Academic/University |
| PI Contribution | This EPSRC travel grant funding (visit conducted during 2023) enabled me to visit Memorial Sloan Kettering Cancer Center (MSKCC) and re-initiate discussions and joint collaborations with existing and new collaborators (including clinicians, biologists and radiologists in addition to mathematicians/computer scientists) on the topic of adaptive cancer therapy and robust radiological cancer imaging analysis. I have contributed during the funded visit to the formulation of the research gaps and challenges related to combining learning and control theory to solve the problem of systematically adapting concentration and injection time for chemotherapy scheduling. This has resulted a follow-up standard funding application being submitted to address the identified technical gaps during the conducted visit with planned first of its kinds in-vitro experiments to validate the learning-based closed-loop technology to be developed. I have also contributed to the development of a computational pipeline that aims to characterize the impact of different variability parameters (e.g., kernel reconstruction differences and integral tube current settings), considered major obstacles to the clinical use of radiomic features, on the radiomic features extracted from computed tomography (CT) scans. |
| Collaborator Contribution | Collaborators from MSKCC have contributed by providing knowledge-based expertise (clinical, radiological and medical physics) to the control and learning-based problems that were formulated. They greatly helped in having a clinically sound problem worth being investigated to help having a more personalised chemotherapy to take forward to the next stage of trial. They have also provided CT scans data of a 3D-printed phantom that have been used to study the robustness of radiomic features with respect to variability existing, e.g., between vendors, scanners, protocols, and even reconstruction software versions. |
| Impact | * Funding application has been submitted in which MSKCC is the partner who agreed to provide in-kind contributions of ~£48K to develop and work on the formulated research questions during the travel visit. * A paper titled, "Cancer radiomic feature variations due to reconstruction kernel choice and integral tube current" by Elfried Salanon, Anqi Fu, Aditya P Apte, Usman Mahmoud, Zehor Belkhatir, Amita Shukla-Dave, Joseph O Deasy. A preprint is available at: https://www.biorxiv.org/content/10.1101/2024.06.04.596806v1. * Another paper titled, "Nonlinear impulsive control design for biologically grounded NSM tumor growth model using exact linearized mapping" by Zehor Belkhair, Soulaimane Berkane, Larry Norton, Allen Tannenbaum has been published in IFAC PapersOnLine. |
| Start Year | 2023 |
| Description | Engage with clinicians and biologists at Weill Cornell Medicine and MSKCC |
| Form Of Engagement Activity | A talk or presentation |
| Part Of Official Scheme? | No |
| Geographic Reach | International |
| Primary Audience | Other audiences |
| Results and Impact | I have been invited during the awarded travel grant visit to MSKCC to deliver a talk and engage with medical/biological experts within the Meyer Cancer Center at Weill Cornell Medicine, New York. This has sparked interesting questions and discussion that have helped in shaping a clinically/biologically sound research questions. |
| Year(s) Of Engagement Activity | 2023,2024 |
