A5067679392- Radiomics and Machine Learning in Medical Imaging
- Medical Imaging Techniques and Applications
- Glioma Diagnosis and Treatment
- Advanced X-ray and CT Imaging
- Lung Cancer Diagnosis and Treatment
- Head and Neck Cancer Studies
- Lymphoma Diagnosis and Treatment
- Privacy-Preserving Technologies in Data
- MRI in cancer diagnosis
- Colorectal and Anal Carcinomas
- Sarcoma Diagnosis and Treatment
- Adrenal and Paraganglionic Tumors
- Hepatocellular Carcinoma Treatment and Prognosis
- Aortic aneurysm repair treatments
- Prostate Cancer Diagnosis and Treatment
- Colorectal Cancer Surgical Treatments
- Cancer, Hypoxia, and Metabolism
- Radiation Dose and Imaging
- Brain Tumor Detection and Classification
- Radiopharmaceutical Chemistry and Applications
- Salivary Gland Tumors Diagnosis and Treatment
- Body Composition Measurement Techniques
- Vascular Anomalies and Treatments
- Esophageal Cancer Research and Treatment
- Machine Learning in Healthcare
Leeds Teaching Hospitals NHS Trust
2018-2025
St James's University Hospital
2019-2025
University of Leeds
2011-2025
Leeds General Infirmary
2020-2024
Mid Yorkshire Hospitals NHS Trust
2015
Although machine learning (ML) has shown promise in numerous domains, there are concerns about generalizability to out-of-sample data. This is currently addressed by centrally sharing ample, and importantly diverse, data from multiple sites. However, such centralization challenging scale (or even not feasible) due various limitations. Federated ML (FL) provides an alternative train accurate generalizable models, only numerical model updates. Here we present findings the largest FL study...
The aim was to predict survival of glioblastoma at 8 months after radiotherapy (a period allowing for completing a typical course adjuvant temozolomide), by applying deep learning the first brain MRI completion.
Incidence of anal squamous cell carcinoma (ASCC) is increasing, with curative chemoradiotherapy (CRT) as the primary treatment non-metastatic disease. A significant proportion patients have locoregional failure (LRF), but distant relapse uncommon. Accurate prognostication progression-free survival (PFS) would help personalisation CRT regimens. The study aim was to evaluate novel imaging pre-treatment features, prognosticate for PFS in ASCC. Consecutive ASCC treated intent at a large tertiary...
The aim of this study was to develop and validate an automated pipeline that could assist the diagnosis active aortitis using radiomic imaging biomarkers derived from [18F]-Fluorodeoxyglucose Positron Emission Tomography-Computed Tomography (FDG PET-CT) images. aorta automatically segmented by convolutional neural network (CNN) on FDG PET-CT control patients. dataset split into training (43 aortitis:21 control), test (12 aortitis:5 control) validation (24 aortitis:14 cohorts. Radiomic...
Objective (1): To develop and validate a machine learning (ML) model using radiomic features (RFs) extracted from [18F]FDG PET-CT to predict abdominal aortic aneurysm (AAA) growth rate. Methods (2): This retrospective study included 98 internal 55 external AAA patients undergoing PET-CT. RFs were manual segmentations of AAAs PyRadiomics. Recursive feature elimination (RFE) reduced for optimisation. A multi-layer perceptron (MLP) was developed prediction compared against Random Forest (RF),...
Approximately 30% of patients with diffuse large B-cell lymphoma (DLBCL) will have recurrence. The aim this study was to develop a radiomic based model derived from baseline PET/CT predict 2-year event free survival (2-EFS).Patients DLBCL treated R-CHOP chemotherapy undergoing pre-treatment between January 2008 and 2018 were included. dataset split into training internal unseen test sets (ratio 80:20). A logistic regression using metabolic tumour volume (MTV) six different machine learning...
Abstract Introduction Bronchial artery embolisation (BAE) is an established treatment method for massive haemoptysis. The aim of this study to evaluate the impact BAE on in-hospital outcomes and long-term survival in patients with Methods Retrospective review all cases acute haemoptysis treated by between April 2000 2012 at least a 5 year follow up each patient. Targeted was performed lateralising symptoms, bronchoscopic sites bleeding or angiographic unilateral abnormal vasculature. In...
Relapse occurs in ~20% of patients with classical Hodgkin lymphoma (cHL) despite treatment adaption based on 2-deoxy-2-[18F]fluoro-D-glucose positron emission tomography/computed tomography response. The objective was to evaluate pre-treatment FDG PET/CT-derived machine learning (ML) models for predicting outcome cHL.All cHL undergoing PET/CT at our institution between 2008 and 2018 were retrospectively identified. A 1.5 × mean liver standardised uptake value (SUV) a fixed 4.0 SUV threshold...
The aim of this study was to explore the feasibility assisted diagnosis active (peri-)aortitis using radiomic imaging biomarkers derived from [
Background and purposeMagnetic Resonance Imaging (MRI) exhibits scanner dependent contrast, which limits generalisability of radiomics machine-learning for radiation oncology. Current deep-learning harmonisation requires paired data, retraining new scanners often suffers from geometry-shift alters anatomical information. The aim this study was to investigate style-blind auto-encoders MRI accommodate unpaired training avoid harmonise data previously unseen scanners.Materials methodsA...
To develop a machine learning (ML) model based on radiomic features (RF) extracted from whole prostate gland magnetic resonance imaging (MRI) for prediction of tumour hypoxia pre-radiotherapy.
Incomplete response on FDG PET-CT following (chemo)radiotherapy (CRT) for head and neck squamous cell carcinoma (HNSCC) hinders optimal management. The study assessed the utility of an interval (second look) PET-CT.
Glioblastoma (GBM) has the typical radiological appearance (TRA) of a centrally necrotic, peripherally enhancing tumor with surrounding edema. The objective this study was to determine whether developing GBM displays spectrum imaging changes detectable on routine clinical prior TRA GBM. Patients pre-operative diagnosed (1 January 2014-31 March 2022) were identified from neuroscience center. reviewed by an experienced neuroradiologist. Imaging patterns preceding analyzed. A total 76 out 555...
Fluorine-18 fluorodeoxyglucose (FDG)-positron emission tomography/computed tomography (PET/CT) is widely used for staging high-grade lymphoma, with the time to evaluate such studies varying depending on complexity of case. Integrating artificial intelligence (AI) within reporting workflow has potential improve quality and efficiency. The aims present study were influence an integrated research prototype segmentation tool implemented diagnostic PET/CT reading software speed variable levels...