Katy Szczepura
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Katy Szczepura has a total of 136 co-authors. Here are the top co-authors by number of shared publications:
- Peter Hogg
University of Salford, United Kingdom
20 shared publications - Edwin R. Chilvers
National Heart and Lung Institute, Imperial College London, London, UK
6 shared publications Prina Ruparelia
Department of Respiratory Medicine, University of Cambridge School of Clinical Medicine, Cambridge, UK
6 shared publications- John D. Thompson
School of Health Sciences, University of Salford, Salford, M6 6PU, United Kingdom
5 shared publications - Adrien Michael Peters
Department of Nuclear Medicine, King’s College Hospital, London, UK
4 shared publications Heok K. Cheow
Department of Radiology, University of Cambridge School of Clinical Medicine, Addenbrooke's Hospital, Cambridge, UK
4 shared publications
Latest Publications
- 15 January 2026
- Published by Springer Nature
- p. 85-118
AI in Projectional Radiography
- Open Access
- 1 September 2025
- Published by Korean Society of Ultrasound in Medicine in Ultrasonography
- Vol. 44 (5) , 400-407
Measuring diaphragmatic excursion using 4-dimensional ultrasound: a feasibility study
Four-dimensional ultrasound (4DUS) could provide more accurate characterisation of diaphragm function than existing M-mode ultrasound approaches. Therefore, the aim of this study was to investigate the feasibility of a novel method for tracking diaphragm excursion from 4DUS data. 4DUS was acquired from 12 participants who exhibited a range of breathing patterns. A custom algorithm tracked, reconstructed, and parameterized diaphragm movement using a sphere of time-varying radius. To validate the algorithm, a randomly selected slice of the sphere was sampled and compared to visual analysis. Agreement between the visual analysis and the algorithm was characterised using a Bland-Altman analysis. A root mean squared error (RMSE) metric was also calculated to quantify the fit between the ultrasound data and the spherical parametrisation. There was good agreement between the automated algorithm and visual analysis (bias, 0.09 cm; 95% limits of agreement, -0.44 to 0.25 cm). The RMSE metric was low (0.9-1.5 mm) across the 12 participants, demonstrating that the sphere was a good fit to the measured 4DUS data. This study demonstrates the feasibility of automated tracking of diaphragmatic excursion from 4DUS data using a sphere of time-varying radius. This technique may prove useful for diagnosing and monitoring breathing dysfunction.
- 1 September 2025
- Published by Korean Society of Ultrasound in Medicine in Ultrasonography
- Vol. 44 (5) , 400-407
Open AccessMeasuring diaphragmatic excursion using 4-dimensional ultrasound: a feasibility study
Four-dimensional ultrasound (4DUS) could provide more accurate characterisation of diaphragm function than existing M-mode ultrasound approaches. Therefore, the aim of this study was to investigate the feasibility of a novel method for tracking diaphragm excursion from 4DUS data. 4DUS was acquired from 12 participants who exhibited a range of breathing patterns. A custom algorithm tracked, reconstructed, and parameterized diaphragm movement using a sphere of time-varying radius. To validate the algorithm, a randomly selected slice of the sphere was sampled and compared to visual analysis. Agreement between the visual analysis and the algorithm was characterised using a Bland-Altman analysis. A root mean squared error (RMSE) metric was also calculated to quantify the fit between the ultrasound data and the spherical parametrisation. There was good agreement between the automated algorithm and visual analysis (bias, 0.09 cm; 95% limits of agreement, -0.44 to 0.25 cm). The RMSE metric was low (0.9-1.5 mm) across the 12 participants, demonstrating that the sphere was a good fit to the measured 4DUS data. This study demonstrates the feasibility of automated tracking of diaphragmatic excursion from 4DUS data using a sphere of time-varying radius. This technique may prove useful for diagnosing and monitoring breathing dysfunction.
- Open Access
- 12 May 2025
- Published by Wiley in Expert Systems
- Vol. 42 (6)
Approaches to Automatic Classification, Detection and Segmentation of Breast Arterial Calcification Using Deep Learning
Objective: Cardiovascular disease (CVD) is the leading cause of premature death in the United Kingdom with one type, coronary artery disease, killing more than two times as many women as breast cancer. Recently, researchers have noted that breast arterial calcification (BAC), which is regularly observed as an incidental finding on mammograms, could be used to risk‐stratify women for CVD. However, identifying BAC is known to be a tedious, expensive and time‐consuming process. Thus, this paper investigates deep learning models for BAC classification, object detection and segmentation. Methodology: A data set, annotated under the guidance of two consultant radiologists, was created using data augmentation. This was used to evaluate several alternative deep learning models. Results: A modified ResNet22 classification network achieved a test accuracy of 80%, indicating that this method could be used as a flag for the presence or absence of BAC. We also used this network for feature extraction in a YOLOv4 BAC object detection network. Despite improving on a recent similar study, this latter network performed poorly with very low average precision scores at several thresholds. More promising was our DeepLabv3+‐based BAC segmentation network, which reached similar high global accuracy scores to three recent studies and a BFScore of over 70% specifically for BAC. It also performed satisfactorily on an unseen data set. Conclusions: These results show the potential for using classification and segmentation models as part of a pipeline for detecting BAC.
- 12 May 2025
- Published by Wiley in Expert Systems
- Vol. 42 (6)
Open AccessApproaches to Automatic Classification, Detection and Segmentation of Breast Arterial Calcification Using Deep Learning
Objective: Cardiovascular disease (CVD) is the leading cause of premature death in the United Kingdom with one type, coronary artery disease, killing more than two times as many women as breast cancer. Recently, researchers have noted that breast arterial calcification (BAC), which is regularly observed as an incidental finding on mammograms, could be used to risk‐stratify women for CVD. However, identifying BAC is known to be a tedious, expensive and time‐consuming process. Thus, this paper investigates deep learning models for BAC classification, object detection and segmentation. Methodology: A data set, annotated under the guidance of two consultant radiologists, was created using data augmentation. This was used to evaluate several alternative deep learning models. Results: A modified ResNet22 classification network achieved a test accuracy of 80%, indicating that this method could be used as a flag for the presence or absence of BAC. We also used this network for feature extraction in a YOLOv4 BAC object detection network. Despite improving on a recent similar study, this latter network performed poorly with very low average precision scores at several thresholds. More promising was our DeepLabv3+‐based BAC segmentation network, which reached similar high global accuracy scores to three recent studies and a BFScore of over 70% specifically for BAC. It also performed satisfactorily on an unseen data set. Conclusions: These results show the potential for using classification and segmentation models as part of a pipeline for detecting BAC.
- Open Access
- 15 November 2024
- Published by Elsevier in Radiography
- Vol. 31 (1) , 131-136
The impact of technical and environmental conditions on the quality assessment in mammography
- 15 November 2024
- Published by Elsevier in Radiography
- Vol. 31 (1) , 131-136
Open AccessThe impact of technical and environmental conditions on the quality assessment in mammography
- Open Access
- 24 July 2024
- Published by Elsevier in Radiography
- Vol. 30 (5) , 1308-1316
Exploring end of life care provision during medical imaging in hospitals: Analysis of survey data from the UK radiography workforce
- 24 July 2024
- Published by Elsevier in Radiography
- Vol. 30 (5) , 1308-1316
Open AccessExploring end of life care provision during medical imaging in hospitals: Analysis of survey data from the UK radiography workforce
- Open Access
- 14 February 2024
- Published by Elsevier in Radiography
- Vol. 30 (2) , 666-672
An exploration of factors involved in the roll out of a digital application in breast services: A case study approach
- 14 February 2024
- Published by Elsevier in Radiography
- Vol. 30 (2) , 666-672
Open AccessAn exploration of factors involved in the roll out of a digital application in breast services: A case study approach
- Open Access
- 3 November 2023
- Published by Elsevier in Radiography
- Vol. 30 (1) , 132-140
Understanding experiences of the radiography workforce delivering medical imaging as part of patients’ end of life care: An exploratory qualitative interview study
- 3 November 2023
- Published by Elsevier in Radiography
- Vol. 30 (1) , 132-140
Open AccessUnderstanding experiences of the radiography workforce delivering medical imaging as part of patients’ end of life care: An exploratory qualitative interview study
- 29 September 2023
- Published by Elsevier in Journal of Tissue Viability
- Vol. 32 (4) , 577-584
Skin tears in mammography: A narrative review
- 29 September 2023
- Published by Elsevier in Journal of Tissue Viability
- Vol. 32 (4) , 577-584
Skin tears in mammography: A narrative review
- 2 November 2022
- Published by Springer Nature
- p. 233-245
Image Quality, System Optimisation and Quality Control
- 2 November 2022
- Published by Springer Nature
- p. 233-245
Image Quality, System Optimisation and Quality Control
- 2 November 2022
- Published by Springer Nature
- p. 295-307
Practical Mammography