Dr Joshua Astley
Clinical Medicine, School of Medicine and Population Health
Postdoctoral Research Associate
Full contact details
Clinical Medicine, School of Medicine and Population Health
Polaris
18 Claremont Crescent
Sheffield
S10 2TA
- Profile
-
I obtained a degree in Mechanical Engineering from the University of Sheffield in 2019 before securing a fully-funded Faculty of Medicine Post Graduate Research Committee Scholarship to complete a PhD within the POLARIS group at the University of Sheffield. I completed my PhD in 2023 and in recognition of my PhD thesis, I received the prestigious Institute of Physics Best PhD Thesis in Medical Physics award. I am currently working as a Postdoctoral Research Associate in Pulmonary MR Image Computing Science at the University of Sheffield.
- Research interests
-
My research interests focus on lung image analysis methods, primarily in MR imaging, alongside biomarker prediction in patients with lung disease. I have published several peer reviewed research articles and conference abstracts on the use of neural networks in medical imaging applications, such as image segmentation, synthesis and survival analysis, with the aim of improving patient care. I have a keen interest in the scientific method and producing research that is rigorously validated and statistically sound. I am excited by the developing fields of explainable AI, multi-modal fusion models, uncertainty awareness and AI ethics.
- Publications
-
Show: Featured publications All publications
Featured publications
Journal articles
- Explainable deep learning-based survival prediction for non-small cell lung cancer patients undergoing radical radiotherapy. Radiotherapy and Oncology, 193, 110084-110084. View this article in WRRO
- PhysVENeT: a physiologically-informed deep learning-based framework for the synthesis of 3D hyperpolarized gas MRI ventilation. Scientific Reports, 13. View this article in WRRO
- Implementable deep learning for multi-sequence proton MRI lung segmentation: a multi-center, multi-vendor, and multi-disease study. Journal of Magnetic Resonance Imaging. View this article in WRRO
- A dual-channel deep learning approach for lung cavity estimation from hyperpolarized gas and proton MRI. Journal of Magnetic Resonance Imaging. View this article in WRRO
- Large-scale investigation of deep learning approaches for ventilated lung segmentation using multi-nuclear hyperpolarized gas MRI. Scientific Reports, 12(1). View this article in WRRO
- Deep learning in structural and functional lung image analysis.. British Journal of Radiology, 95(1132). View this article in WRRO
- A hybrid model‐ and deep learning‐based framework for functional lung image synthesis from multi‐inflation CT and hyperpolarized gas MRI. Medical Physics. View this article in WRRO
Conference proceedings papers
- 3D deep convolutional neural network-based ventilated lung segmentation using multi-nuclear hyperpolarized gas MRI. Thoracic Image Analysis, Vol. 12502 (pp 24-35). Lima, Peru, 8 October 2020 - 8 October 2020. View this article in WRRO
All publications
Journal articles
- Explainable deep learning-based survival prediction for non-small cell lung cancer patients undergoing radical radiotherapy. Radiotherapy and Oncology, 193, 110084-110084. View this article in WRRO
- PhysVENeT: a physiologically-informed deep learning-based framework for the synthesis of 3D hyperpolarized gas MRI ventilation. Scientific Reports, 13. View this article in WRRO
- Implementable deep learning for multi-sequence proton MRI lung segmentation: a multi-center, multi-vendor, and multi-disease study. Journal of Magnetic Resonance Imaging. View this article in WRRO
- A dual-channel deep learning approach for lung cavity estimation from hyperpolarized gas and proton MRI. Journal of Magnetic Resonance Imaging. View this article in WRRO
- Large-scale investigation of deep learning approaches for ventilated lung segmentation using multi-nuclear hyperpolarized gas MRI. Scientific Reports, 12(1). View this article in WRRO
- Deep learning in structural and functional lung image analysis.. British Journal of Radiology, 95(1132). View this article in WRRO
- A hybrid model‐ and deep learning‐based framework for functional lung image synthesis from multi‐inflation CT and hyperpolarized gas MRI. Medical Physics. View this article in WRRO
Conference proceedings papers
- P177 129Xe MRI phenotyping and longitudinal change in patients with asthma and/or COPD and normal pulmonary function tests. ‘You ain’t seen nothing yet’ – Imaging across COPD, nodules and lung cancer screening
- P176 Bronchodilator response discordance in patients with asthma and/or COPD assessed by 129Xe-MRI and spirometry. ‘You ain’t seen nothing yet’ – Imaging across COPD, nodules and lung cancer screening
- Xe-MRI bronchodilator response to assess disease severity in patients with asthma and/or COPD. Respiratory function technologists/scientists
- Longitudinal change in lung physiology and 129Xe MRI over 1 year in patients with asthma and/or COPD. Clinical respiratory physiology, exercise and functional imaging
- Lung physiology and 129Xe MRI data driven clustering of patients with asthma and/or COPD. Clinical respiratory physiology, exercise and functional imaging
- Bronchodilator response discordance in patients with asthma and/or COPD using Xe-MRI and spirometry. Respiratory function technologists/scientists
- A comparison of 129Xe MRI and advanced lung function testing in patients with asthma and /or COPD: The NOVELTY ADPro substudy. Imaging
- 3D deep convolutional neural network-based ventilated lung segmentation using multi-nuclear hyperpolarized gas MRI. Thoracic Image Analysis, Vol. 12502 (pp 24-35). Lima, Peru, 8 October 2020 - 8 October 2020. View this article in WRRO
- Multi-modal assessment of dose-related changes in regional lung function in non-small cell lung cancer patients receiving radiotherapy. ISMRM Annual Meeting, 7 May 2022 - 12 May 2022.
- Longitudinal comparison of quantitative UTE lung MRI and CT biomarkers in interstitial lung disease. ISMRM Annual Meeting, 7 May 2022 - 12 May 2022.
- 129Xe MRI patterns of lung function in patients with asthma and/or COPD in the NOVELTY study. ISMRM Annual Meeting, 7 May 2022 - 12 May 2022.
- Deep learning-based synthesis of hyperpolarized gas MRI ventilation from 3D multi-inflation proton MRI. ISMRM Annual Meeting, 7 May 2022 - 12 May 2022.
- A multi-channel deep learning approach for lung cavity estimation from hyperpolarized gas and proton MRI. ISMRM Annual Meeting, 7 May 2022 - 12 May 2022.
- Current projects
- Longitudinal prediction of lung biomarkers via a diverse range of clinical, demographic and imaging data derived from a large cohort of patients with asthma and/or COPD acquired as part of a sub-study of the NOVELTY study https://noveltystudy.com/
- A CNN-based end-to-end approach for image-based biomarker prediction from 129Xe-MRI and 1H-MRI with the aim of improving functional lung clinical workflows.
- Use of explainable AI techniques, such as grad-CAM and SHAP, to provide improved trust in algorithms amongst clinicians and patients.
- Investigating the use of fusion models for multi-modal prediction and survival analysis in lung cancer radiotherapy.