James H. F. Rudd

James H. F. Rudd

Auto-generated profile, based on publicly available information

James H. F. Rudd

Department of Medicine, University of Cambridge, Cambridge, UK

Research Focus & Metrics

Academic Career Length

26Years (2001-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
15 March 2021 in Nature Machine Intelligence
Verified
Journal Article
Open access
PDF file
MRMichael Roberts
Michael Roberts
DDDerek Driggs
Derek Driggs
MTMatthew Thorpe
Matthew Thorpe
JGJulian Gilbey
Julian Gilbey
MYMichael Yeung
Michael Yeung
CSCarola-Bibiane Schönlieb
Carola-Bibiane Schönlieb
Machine learning methods offer great promise for fast and accurate detection and prognostication of coronavirus disease 2019 (COVID-19) from standard-of-care chest radiographs (CXR) and chest computed tomography (CT) images. Many articles have been published in 2020 describing new machine learning-based models for both of these tasks, but it is unclear which are of potential clinical utility. In this systematic review, we consider all published papers and preprints, for the period from 1 January 2020 to 3 October 2020, which describe new machine learning models for the diagnosis or prognosis of COVID-19 from CXR or CT images. All manuscripts uploaded to bioRxiv, medRxiv and arXiv along with all entries in EMBASE and MEDLINE in this timeframe are considered. Our search identified 2,212 studies, of which 415 were included after initial screening and, after quality screening, 62 studies were included in this systematic review. Our review finds that none of the models identified are of potential clinical use due to methodological flaws and/or underlying biases. This is a major weakness, given the urgency with which validated COVID-19 models are needed. To address this, we give many recommendations which, if followed, will solve these issues and lead to higher-quality model development and well-documented manuscripts.

11 June 2002 in Circulation
Verified
Journal Article
Closed access
PDF file
JRJ.H.F. Rudd
J.H.F. Rudd
EWE.A. Warburton
E.A. Warburton
TFT.D. Fryer
T.D. Fryer
HJH.A. Jones
H.A. Jones
JCJ.C. Clark
J.C. Clark
PWP.L. Weissberg
P.L. Weissberg
Background — Atherosclerotic plaque rupture is usually a consequence of inflammatory cell activity within the plaque. Current imaging techniques provide anatomic data but no indication of plaque inflammation. The glucose analogue [ 18 F]-fluorodeoxyglucose ( 18 FDG) can be used to image inflammatory cell activity non-invasively by PET. In this study we tested whether 18 FDG-PET imaging can identify inflammation within carotid artery atherosclerotic plaques. Methods and Results — Eight patients with symptomatic carotid atherosclerosis were imaged using 18 FDG-PET and co-registered CT. Symptomatic carotid plaques were visible in 18 FDG-PET images acquired 3 hours post- 18 FDG injection. The estimated net 18 FDG accumulation rate (plaque/integral plasma) in symptomatic lesions was 27% higher than in contralateral asymptomatic lesions. There was no measurable 18 FDG uptake into normal carotid arteries. Autoradiography of excised plaques confirmed accumulation of deoxyglucose in macrophage-rich areas of the plaque. Conclusions — This study demonstrates that atherosclerotic plaque inflammation can be imaged with 18 FDG-PET, and that symptomatic, unstable plaques accumulate more 18 FDG than asymptomatic lesions.

1 February 2014 in The Lancet
Verified
Journal Article
Open access
NJNikhil V Joshi
Nikhil V Joshi
AVAlex T Vesey
Alex T Vesey
MWMichelle C Williams
Michelle C Williams
ASAnoop S V Shah
Anoop S V Shah
PCPatrick A Calvert
Patrick A Calvert
DNDavid E Newby
David E Newby
The use of non-invasive imaging to identify ruptured or high-risk coronary atherosclerotic plaques would represent a major clinical advance for prevention and treatment of coronary artery disease. We used combined PET and CT to identify ruptured and high-risk atherosclerotic plaques using the radioactive tracers (18)F-sodium fluoride ((18)F-NaF) and (18)F-fluorodeoxyglucose ((18)F-FDG). In this prospective clinical trial, patients with myocardial infarction (n=40) and stable angina (n=40) underwent (18)F-NaF and (18)F-FDG PET-CT, and invasive coronary angiography. (18)F-NaF uptake was compared with histology in carotid endarterectomy specimens from patients with symptomatic carotid disease, and with intravascular ultrasound in patients with stable angina. The primary endpoint was the comparison of (18)F-fluoride tissue-to-background ratios of culprit and non-culprit coronary plaques of patients with acute myocardial infarction. In 37 (93%) patients with myocardial infarction, the highest coronary (18)F-NaF uptake was seen in the culprit plaque (median maximum tissue-to-background ratio: culprit 1·66 [IQR 1·40-2·25] vs highest non-culprit 1·24 [1·06-1·38], p<0·0001). By contrast, coronary (18)F-FDG uptake was commonly obscured by myocardial uptake and where discernible, there were no differences between culprit and non-culprit plaques (1·71 [1·40-2·13] vs 1·58 [1·28-2·01], p=0·34). Marked (18)F-NaF uptake occurred at the site of all carotid plaque ruptures and was associated with histological evidence of active calcification, macrophage infiltration, apoptosis, and necrosis. 18 (45%) patients with stable angina had plaques with focal (18)F-NaF uptake (maximum tissue-to-background ratio 1·90 [IQR 1·61-2·17]) that were associated with more high-risk features on intravascular ultrasound than those without uptake: positive remodelling (remodelling index 1·12 [1·09-1·19] vs 1·01 [0·94-1·06]; p=0·0004), microcalcification (73% vs 21%, p=0·002), and necrotic core (25% [21-29] vs 18% [14-22], p=0·001). (18)F-NaF PET-CT is the first non-invasive imaging method to identify and localise ruptured and high-risk coronary plaque. Future studies are needed to establish whether this method can improve the management and treatment of patients with coronary artery disease. Chief Scientist Office Scotland and British Heart Foundation.

15 May 2019 in PLOS ONE
Verified
Journal Article
Open access
PDF file
AAAhmed M. Alaa
Ahmed M. Alaa
TBThomas Bolton
Thomas Bolton
EAEmanuele Di Angelantonio
Emanuele Di Angelantonio
JRJames H. F. Rudd
James H. F. Rudd
MSMihaela van der Schaar
Mihaela van der Schaar
Identifying people at risk of cardiovascular diseases (CVD) is a cornerstone of preventative cardiology. Risk prediction models currently recommended by clinical guidelines are typically based on a limited number of predictors with sub-optimal performance across all patient groups. Data-driven techniques based on machine learning (ML) might improve the performance of risk predictions by agnostically discovering novel risk predictors and learning the complex interactions between them. We tested (1) whether ML techniques based on a state-of-the-art automated ML framework (AutoPrognosis) could improve CVD risk prediction compared to traditional approaches, and (2) whether considering non-traditional variables could increase the accuracy of CVD risk predictions. Using data on 423,604 participants without CVD at baseline in UK Biobank, we developed a ML-based model for predicting CVD risk based on 473 available variables. Our ML-based model was derived using AutoPrognosis, an algorithmic tool that automatically selects and tunes ensembles of ML modeling pipelines (comprising data imputation, feature processing, classification and calibration algorithms). We compared our model with a well-established risk prediction algorithm based on conventional CVD risk factors (Framingham score), a Cox proportional hazards (PH) model based on familiar risk factors (i.e, age, gender, smoking status, systolic blood pressure, history of diabetes, reception of treatments for hypertension and body mass index), and a Cox PH model based on all of the 473 available variables. Predictive performances were assessed using area under the receiver operating characteristic curve (AUC-ROC). Overall, our AutoPrognosis model improved risk prediction (AUC-ROC: 0.774, 95% CI: 0.768-0.780) compared to Framingham score (AUC-ROC: 0.724, 95% CI: 0.720-0.728, p < 0.001), Cox PH model with conventional risk factors (AUC-ROC: 0.734, 95% CI: 0.729-0.739, p < 0.001), and Cox PH model with all UK Biobank variables (AUC-ROC: 0.758, 95% CI: 0.753-0.763, p < 0.001). Out of 4,801 CVD cases recorded within 5 years of baseline, AutoPrognosis was able to correctly predict 368 more cases compared to the Framingham score. Our AutoPrognosis model included predictors that are not usually considered in existing risk prediction models, such as the individuals’ usual walking pace and their self-reported overall health rating. Furthermore, our model improved risk prediction in potentially relevant sub-populations, such as in individuals with history of diabetes. We also highlight the relative benefits accrued from including more information into a predictive model (information gain) as compared to the benefits of using more complex models (modeling gain). Our AutoPrognosis model improves the accuracy of CVD risk prediction in the UK Biobank population. This approach performs well in traditionally poorly served patient subgroups. Additionally, AutoPrognosis uncovered novel predictors for CVD disease that may now be tested in prospective studies. We found that the “information gain” achieved by considering more risk factors in the predictive model was significantly higher than the “modeling gain” achieved by adopting complex predictive models.

1 October 2011 in The Lancet
Verified
Journal Article
Closed access
ZFZahi A Fayad
Zahi A Fayad
VMVenkatesh Mani
Venkatesh Mani
MWMark Woodward
Mark Woodward
DKDavid Kallend
David Kallend
MAMarkus Abt
Markus Abt
ATAhmed Tawakol
Ahmed Tawakol

1 April 2012 in JACC
Verified
Journal Article
Closed access
MDMarc R. Dweck
Marc R. Dweck
MCMarcus W.L. Chow
Marcus W.L. Chow
NJNikhil V. Joshi
Nikhil V. Joshi
MWMichelle C. Williams
Michelle C. Williams
CJCharlotte Jones
Charlotte Jones
DNDavid E. Newby
David E. Newby