Available
Project number:
2025_88
Start date:
October 2025
Project themes:
AI, Machine Learning, and Multimodal Data
Main supervisor:
Consultant Clinical Oncologist
Co-supervisor:
Professor Andy King
Additional Information:
Multi-institutional large-scale deep learning-based modelling of mandibular osteoradionecrosis using clinical data and dose distribution volumes: Phase II of the PREDMORN study.
Background Osteoradionecrosis of the jaw (ORNJ) is a severe complication of radiotherapy in patients treated for head and neck cancer (HNC). Conventional normal tissue complication probability (NTCP) models(1) include clinical and demographic factors combined with one-dimensional metrics from the dose-volume histogram (DVH), to find statistical associations with toxicity outcomes. Non-uniform dose distributions, commonly seen in non-target normal structures, are not well represented by the DVH due to lack of spatial information. This is relevant as anatomical and radiobiological heterogeneities will result in sub-regions with different radiation responses(2,3), hence the need to include spatial dose information into NTCP models to improve accuracy. Novelty & Importance Deep learning (DL) can effectively extract dosiomic information that contributes to toxicity prediction (4). For ORN, previous independent single-institution works(5) obtained opposing results which highlighted the need for larger multi-institutional datasets. To address this, the PREDMORN (PREDiction models for mandibular OsteoRadioNecrosis) Consortium was created(6), currently consisting of seven European and North American institutions, to curate the largest and most diverse dataset ever considered worldwide to develop robust and generalisable prediction models for mandibular ORN using large-scale multi-institutional real-world data. The study has two phases; phase I focused on statistical analyses of clinical and DVH data proposing an externally validated logistic regression-based ORN-NTCP model. Phase II is designed to include spatial information from planned radiation dose maps and CT images. Aims & Objectives To complete PREDMORN- phase II, focusing on DL-based modelling using a large-scale multi-institutional dataset including spatial information. Planned Research Method To develop and externally validate a DL-based NTCP model for mandibular ORN using the PREDMORN dataset, incorporate uncertainty quantification into the DL-based NTCP model to assess the confidence of model predictions and develop an online decision support tool to facilitate prospective validation of our model. Quality Control Following the TRIPOD+AI guidelines, our model will be externally and prospectively validated on independent datasets prior to clinical implementation. References 1. van Dijk LV, Abusaif AA, Rigert J, et al. Normal Tissue Complication Probability (NTCP) Prediction Model for Osteoradionecrosis of the Mandible in Patients With Head and Neck Cancer After Radiation Therapy: Large-Scale Observational Cohort. International Journal of Radiation Oncology*Biology*Physics. 2021;111(2):549-558. doi:10.1016/j.ijrobp.2021.04.042 2. Hopewell JW, Trott KR. Volume effects in radiobiology as applied to radiotherapy. Radiotherapy and Oncology. 2000;56(3):283-288. doi:10.1016/S0167-8140(00)00236-X 3. Marks LB, Yorke ED, Jackson A, et al. Use of Normal Tissue Complication Probability Models in the Clinic. International Journal of Radiation Oncology*Biology*Physics. 2010;76(3):S10-S19. doi:10.1016/j.ijrobp.2009.07.1754 4. Ibragimov B, Toesca D, Chang D, Yuan Y, Koong A, Xing L. Development of deep neural network for individualized hepatobiliary toxicity prediction after liver SBRT. Medical Physics. 2018;45(10):4763-4774. doi:10.1002/mp.13122 5. Humbert-Vidan L, Patel V, Andlauer R, King AP, Guerrero Urbano T. Prediction of Mandibular ORN Incidence from 3D Radiation Dose Distribution Maps Using Deep Learning. In: Wu S, Shabestari B, Xing L, eds. Applications of Medical Artificial Intelligence, AMAI 2022. Lecture Notes in Computer Science. Vol 13540. Springer Nature Switzerland AG; 2022:49-58. 6. Humbert-Vidan L, Hansen CR, Fuller CD, et al. Protocol Letter: A multi-institutional retrospective case-control cohort investigating PREDiction models for mandibular OsteoRadioNecrosis in head and neck cancer (PREDMORN). Radiotherapy and Oncology. 2022;176:99-100. doi:10.1016/j.rad
We are now accepting applications for 1 October 2025
How to apply
Candidates should possess or be expected to achieve a 1st or upper 2nd class degree in a relevant subject including the biosciences, computer science, mathematics, statistics, data science, chemistry, physics, and be enthusiastic about combining their expertise with other disciplines in the field of healthcare.
Important information for International Students:
It is the responsibility of the student to apply for their Student Visa. Please note that the EPSRC DRIVE-Health studentship does not cover the visa application fees or the Immigration Health Surcharge (IHS) required for access to the National Health Service. The IHS is mandatory for anyone entering the UK on a Student Visa and is currently £776 per year for each year of study. Further detail can be found under the International Students tab below.
Next Steps
- Applications submitted by the closing date of Thursday 6 February 2025 will be considered by the CDT. We will contact shortlisted applicants with information about this part of the recruitment process.
- Candidates will be invited to attend an interview. Interviews are projected to take place in April 2025.
- Project selection will be through a panel interview chaired by either Professor Richard Dobson and Professor Vasa Curcin (CDT Directors) followed by informal discussion with prospective supervisors.
- If you have any questions related to the specific project you are applying for, please contact the main supervisor of the project directly.
For any other questions about the recruitment process, please email us at
