1. Background:
Lung and oesophageal cancers remain leading causes of cancer mortality, with limited long-term survival despite multimodal treatments. Advances in genomics and medical imaging generate vast and diverse datasets that, if effectively integrated, could revolutionise prognosis and treatment personalisation. However, the heterogeneity of these data demands novel computational frameworks capable of unifying biological, clinical, and imaging information.
2. Novelty & Importance:
This project pioneers an integrative AI/ML framework for these cancer, extending state-of-the-art methods in deep learning and mathematical optimisation. Building on recent innovations in supervised concrete autoencoders and pathway-based modelling, it will deliver interpretable, multimodal analytics that reveal molecular subtypes, predict outcomes, and support individualised therapeutic decisions
3. Aims & Objectives:
The research will develop a scalable platform for multimodal data analysis encompassing imaging, genomic, and clinical sources. Objectives include constructing a comprehensive knowledge graph of thoracic cancer data, discovering novel biomarkers through multi-omic integration, and developing predictive models using Bayesian and ensemble ML methods. The final deliverable will be a validated, open-source informatics pipeline for clinical decision support in precision oncology.
Through cross-disciplinary collaboration between informatics and cancer imaging, this PhD project will advance both methodological innovation in AI-driven bioinformatics and tangible clinical impact for patients with thoracic malignancies.
References:
1. Wong CW, Chaudhry A. Radiogenomics of lung cancer. J Thorac Dis. 2020 Sep;12(9):5104-5109. doi: 10.21037/jtd-2019-pitd-10.
2. Shui L, Ren H, Yang X, Li J, Chen Z, Yi C, Zhu H, Shui P. The Era of Radiogenomics in Precision Medicine: An Emerging Approach to Support Diagnosis, Treatment Decisions, and Prognostication in Oncology. Front Oncol. 2021 Jan 26;10:570465. doi: 10.3389/fonc.2020.570465.
3. Zanfardino, M., Franzese, M., Pane, K. et al. Bringing radiomics into a multi-omics framework for a comprehensive genotype–phenotype characterization of oncological diseases. J Transl Med 17, 337 (2019). https://doi.org/10.1186/s12967-019-2073-2
4. Sieswerda MS, Bermejo I, Geleijnse G, Aarts MJ, Lemmens VEPP, De Ruysscher D, Dekker ALAJ, Verbeek XAAM. Predicting Lung Cancer Survival Using Probabilistic Reclassification of TNM Editions With a Bayesian Network. JCO Clin Cancer Inform. 2020 May;4:436-443. doi: 10.1200/CCI.19.00136..