Samuel Withey

The Royal Marsden NHS Foundation Trust · Department of Diagnostic Radiology

Machine learning approaches hold great potential for the automated detection of lung nodules in chest radiographs, but training the algorithms requires vary large amounts of manually annotated images, which are difficult to obtain. Weak labels indicating whether a radiograph is likely to contain pulmonary nodules are typically easier to obtain at s...

Purpose: To develop and test an artificial intelligence (AI) system, based on deep convolutional neural networks (CNNs), for automated real-time triaging of adult chest radiographs on the basis of the urgency of imaging appearances. Materials and Methods: An AI system was developed by using 470 388 fully anonymized institutional adult chest radiog...

Pheochromocytoma (PC) and paraganglioma (PGL) are rare neu- roendocrine tumors that occur throughout the body from the base of the skull to the pelvis. Sympathetic catecholamine-secreting tu- mors may be associated with hyperadrenergic symptoms and long- term morbidity if they are untreated.Typically biochemically silent, head and neck PGLs may res...

Renal cysts are a common imaging finding, often incidental. Ultrasound, CT and MRI are the main modalities responsible for renal cyst detection and characterization. These modalities often play a complementary role in modern radiological practice, each of them with strengths and limitations. In view of a recently proposed ‘multimodality’ update to...

Indeterminate nonpalpable focal testicular lesions have emerged as a clinical problem with the increasing use of scrotal ultrasound, particularly in the context of infertility. Conventional morphological ultrasound and colour Doppler have been unreliable at differentiating benign from malignant lesions. Multiparametric ultrasound (mpUS) comprises r...

Background Directly-injected therapies (DIT) include a broad range of agents within a developing research field in cancer immunotherapy, with encouraging clinical trial results in various tumour subtypes. Currently, the majority of such therapies are only available within clinical trials; however, more recently, talimogene laherparepvec (T-VEC, Iml...

10500 Background: Incidence of prostate cancer (PCa) is increasing, but there is no internationally agreed population screening program. Studies using an age-based PSA approach show a high rate of false-positive results as well as over-diagnosis of indolent PCa. Genome wide association studies identify common germline variants to calculate a polyge...

5070 Background: Radiographic Progression-free survival (rPFS) derived from Computer Tomography (CT) and Bone Scan (BS) using Prostate Cancer Working Group (PCWG) criteria is utilized as an intermediate clinical endpoint for overall survival (OS) benefit in mCRPC. Whole Body MRI (WBMRI) is emerging as a potentially superior response biomarker, allo...

Background: Artificial intelligence (AI) systems can potentially aid the diagnostic pathway of prostate cancer by alleviating the increasing workload, preventing overdiagnosis, and reducing the dependence on experienced radiologists. We aimed to investigate the performance of AI systems at detecting clinically significant prostate cancer on MRI in...

As the management of gastrointestinal malignancy has evolved, tumor response assessment has expanded from size-based assessments to those that include tumor enhancement, in addition to functional data such as those derived from PET and diffusion-weighted imaging. Accurate interpretation of tumor response therefore requires knowledge of imaging moda...

Background Personalising management of primary oesophageal adenocarcinoma requires better risk stratification. Lack of independent validation of proposed imaging biomarkers has hampered clinical translation. We aimed to prospectively validate previously identified prognostic grey-level co-occurrence matrix (GLCM) CT features for 3-year overall surv...

Purpose The urethra is a critical structure in prostate radiotherapy planning; however, it is impossible to visualise on CT. We developed a surrogate urethra model (SUM) for CT-only planning workflow and tested its geometric and dosimetric performance against the MRI-delineated urethra (MDU). Methods The SUM was compared against 34 different MDUs...

Simple Summary The correct delineation of disease is a critical step in radiotherapy planning and delivery. In the past, the main focus has been on the primary tumour and any involved lymph nodes. Our study assessed whether the accuracy of gross tumour volume (GTV) contouring for patients with rectal cancer can be improved using an MRI reporting sy...

Background T2* mapping can characterize tumor hypoxia, which may be associated with resistance to therapy. Acquiring T2* maps during MR‐guided radiotherapy could inform treatment adaptation by, for example, escalating the dose to resistant sub‐volumes. Purpose The purpose of this work is to demonstrate the feasibility of the accelerated T2* mappin...

Introduction: Better predictive markers are needed to deliver individualized care for patients with primary esophagogastric cancer. This exploratory study aimed to assess whether pre-treatment imaging parameters from dynamic contrast-enhanced MRI and 18F-fluorodeoxyglucose (18F-FDG) PET/CT are associated with response to neoadjuvant therapy or out...

Objectives Radiomic models present an avenue to improve oesophageal adenocarcinoma assessment through quantitative medical image analysis. However, model selection is complicated by the abundance of available predictors and the uncertainty of their relevance and reproducibility. This analysis reviews recent research to facilitate precedent-based mo...

Background: Trimodality therapy including HDR-Brachytherapy (TMT-HDR) is a management option used in select European Centres for muscle invasive bladder cancer (MIBC). Suitable patients have good bladder function, solitary tumours ≤ 5cm, no CIS, no pelvic lymphadenopathy, and are located away from the bladder trigone. Methods: We conducted a retros...

Radiological investigations are essential in the management of oesophageal and gastro-oesophageal junction cancers. The current multi modal combination of computed tomography (CT), 18F-fluorodeoxyglucose positron emission tomography combined with CT (PET/CT) and endoscopic ultrasound (EUS) has limitations, which hinders the prognostic and predictiv...

Bladder paragangliomas (bPGL) are rare neuroendocrine tumors arising from the sympathetic paraganglia present in the bladder wall. Bladder PGLs are typically submucosal or intramural but when subserosal may not be readily visible at cystoscopy. The average size at presentation is 3.9 cm (range 1.0–9.1 cm). When small, bPGL are usually spherical, we...

Background: 18F-fluorodeoxyglucose positron emission tomography/magnetic resonance imaging (18F-FDG PET/MRI) may improve cancer staging by combining sensitive cancer detection with high-contrast resolution and detail. We compared the diagnostic performance of 18F-FDG PET/MRI to 18F-fluorodeoxyglucose positron emission tomography/computed tomograph...

Escherichia coli (E. coli)-related urosepsis associated with a ureteric stone has been shown to cause a systemic bacteraemia that can spread to other parts of the body. Hematogenous spread of infection is the most common cause of pyogenic spondylodiscitis. A 74-year-old female presented with acute left-sided flank pain and was found to have an obst...

Background ¹⁸F-Fluorodesoxyglucose Positron-emission tomography magnetic resonance imaging (18F-FDG PET/MRI) may improve cancer staging by combining sensitive cancer detection with high contrast resolution and detail. We compared the diagnostic performance of 18F-FDG PET/MRI to ¹⁸F-Fluorodesoxyglucose Positron-emission tomography computed tomograph...

[This corrects the article DOI: 10.1016/j.ejro.2018.07.002.].

Machine learning approaches hold great potential for the automated detection of lung nodules on chest radiographs, but training algorithms requires very large amounts of manually annotated radiographs, which are difficult to obtain. The increasing availability of PACS (Picture Archiving and Communication System), is laying the technological foundat...

Purpose: Locoregional stage remains a relatively poor outcome discriminator in primary oesophageal cancer undergoing curative treatment. We aimed to assess the relationship between tumour perfusion, metabolism & volume with nodal stage. Methods and Materials: Following ethical approval and informed consent, DCE-MRI and 18F-FDG PET/CT were performe...

Purpose: Oesophageal cancer has poor survival rates despite curative treatment. We hypothesized that a low perfusion-high metabolism phenotype may be associated with poorer response to neoadjuvant therapy. We aimed to assess the relationship between tumour perfusion and metabolism with therapy response. Methods and Materials: Following ethical app...

Convolutional neural networks (CNNs) have been successfully employed in recent years for the detection of radiological abnormalities in medical images such as plain x-rays. To date, most studies use CNNs on individual examinations in isolation and discard previously available clinical information. In this study we set out to explore whether Long-Sh...

Aims: Ablation therapies are an innovative nephron-sparing alternative to radical nephrectomy for early stage renal cancers, although determination of treatment success is challenging. We aimed to undertake a systematic review of the literature to determine whether assessment of tumour perfusion may improve response assessment or alter clinical man...

Convolutional neural networks (CNNs) have been successfully employed in recent years for the detection of radiological abnormalities in medical images such as plain x-rays. To date, most studies use CNNs on individual examinations in isolation and discard previously available clinical information. In this study we set out to explore whether Long-Sh...

Purpose The chest x-ray is the most commonly performed radiology study, yet interpretation can be challenging. Deep learning systems for computer-aided detection (CAD) hold great potential but have been limited by the need for large numbers of radiologist-annotated radiographs. Other authors have described some success with report classification or...

Motivated by the need to automate medical information extraction from free-text radiological reports, we present a bi-directional long short-term memory (BiLSTM) neural network architecture for modelling radiological language. The model has been used to address two NLP tasks: medical named-entity recognition (NER) and negation detection. We investi...

Lesser sac herniation is a rare phenomenon, where the bowel protrudes through the epiploic foramen into the lesser sac. We describe the case of a 55-year-old male who presented with acute abdominal pain and in whose case the subtle findings of lesser sac herniation were missed during CT scan reporting. Re-review of the images after the patient’s co...

A patient with chronic obstructive pulmonary disease (COPD) and pulmonary fibrosis presented dyspnoeic with productive cough and large-volume haemoptysis, 1 month after coronary stenting and commencement of clopidogrel. A chest radiograph showed a wellcircumscribed opacity in the left lower zone with surrounding consolidation, where previously an e...

Motivated by the need to automate medical information extraction from free-text radiological reports, we present a bi-directional long short-term memory (BiLSTM) neural network architecture for modelling radiological language. The model has been used to address two NLP tasks: medical named-entity recognition (NER) and negation detection. We investi...