A5050834387- Radiomics and Machine Learning in Medical Imaging
- AI in cancer detection
- Ovarian cancer diagnosis and treatment
- Health Systems, Economic Evaluations, Quality of Life
- COVID-19 diagnosis using AI
- Brain Metastases and Treatment
- Economic and Financial Impacts of Cancer
- Glioma Diagnosis and Treatment
- Frailty in Older Adults
- Semantic Web and Ontologies
- Business Process Modeling and Analysis
- Cell Image Analysis Techniques
- Cancer survivorship and care
- Management of metastatic bone disease
- Machine Learning in Healthcare
- Cervical Cancer and HPV Research
- Colorectal Cancer Screening and Detection
- Advanced X-ray and CT Imaging
- Artificial Intelligence in Healthcare and Education
- Image Retrieval and Classification Techniques
- Advances in Oncology and Radiotherapy
- Advanced Radiotherapy Techniques
- Generative Adversarial Networks and Image Synthesis
- Lung Cancer Diagnosis and Treatment
- Digital Imaging for Blood Diseases
St James's University Hospital
2020-2025
Leeds Teaching Hospitals NHS Trust
2013-2024
University of Leeds
2019-2024
Artificial intelligence has the potential to transform radiotherapy workflow, resulting in improved quality, safety, accuracy and timeliness of delivery. Several commercially available artificial intelligence-based auto-contouring tools have emerged recent years. Their clinical deployment raises important considerations for oncologists, including quality assurance validation, education, training job planning. Despite this, there is little literature capturing views oncologists with respect...
Artificial intelligence (AI) in healthcare describes algorithm-based computational techniques which manage and analyse large datasets to make inferences predictions. There are many potential applications of AI the care older people, from clinical decision support systems that can identification delirium records wearable devices predict risk a fall. We held four meetings clinicians researchers. Three priority areas were identified for application people. These included: monitoring early...
The Cancer and Aging Research Group (CARG) score was developed to predict severe chemotherapy-induced toxicity risk in older adults; validation study results have varied. Tolerance of Anti-cancer Systemic Therapy the Elderly sought evaluate CARG prospectively a chemotherapy-naïve UK population. This multicentre, prospective, observational recruited patients aged ≥65 years commencing first-line chemotherapy for any solid organ malignancy or setting. Baseline demographics established frailty...
Objectives: To characterise treatment patterns and survival outcomes for patients with locally advanced or metastatic malignancy of the urothelial tract during a period immediately preceding widespread use immune checkpoint inhibitors in UK Patients Methods: We retrospectively examined electronic case notes attending Leeds Cancer Centre, carcinoma, receiving chemotherapy between January 2003 March 2017. Patient characteristics, patterns, were collected. Summary descriptive statistics...
Governments and medical associations across the world, including US Food Drug Administration, UK Medicines Healthcare products Regulatory Agency, Royal College of Radiologists, European Society Radiology, believe advent health technologies associated with artificial intelligence (AI) will be most radical change in how care is delivered our lifetime.1Royal RadiologistsRCR position statement on intelligence.http://www.rcr.ac.uk/posts/rcr-position-statement-artificial-intelligenceDate: 2018Date...
Weakly-supervised classification of histopathology slides is a computationally intensive task, with typical whole slide image (WSI) containing billions pixels to process. We propose Discriminative Region Active Sampling for Multiple Instance Learning (DRAS-MIL), efficient method using attention scores focus sampling on highly discriminative regions. apply this the diagnosis ovarian cancer histological subtypes, which an essential part patient care pathway as different subtypes have genetic...
The area of process change over time is a particular concern in healthcare, where patterns care emerge and evolve response to individual patient needs. We propose structured approach analyse that suitable for the complex domain healthcare. Our applies qualitative comparison at three levels abstraction: holistic perspective (process model), middle-level (trace), fine-grained detail (activity). aim was detect points, localise characterise change, unravel/understand evolution. illustrate using...
Large pretrained transformers are increasingly being developed as generalised foundation models which can underpin powerful task-specific artificial intelligence models. Histopathology show promise across many tasks, but analyses have been limited by arbitrary hyperparameters that were not tuned to the specific task/dataset. We report most rigorous single-task validation conducted date of a histopathology model, and first performed in ovarian cancer subtyping. Attention-based multiple...
1521 Background: Older adults have a higher risk of developing chemotherapy (CTx) related toxicity. The Cancer Aging Research Group (CARG) score was developed and validated in the USA to predict severe CTx induced toxicity older adults; subsequent validation studies had varying results. TOASTIE study sought evaluate CARG prospectively United Kingdom (UK) population. Methods: This multicentre, prospective, observational recruited patients aged ≥65 years commencing first-line neo-adjuvant,...
Computer vision models are increasingly capable of classifying ovarian epithelial cancer subtypes, but they differ from pathologists by processing small tissue patches at a single resolution. Multi-resolution graph leverage the spatial relationships multiple magnifications, learning context for each patch. In this study, we conduct most thorough validation model subtyping to date. Seven were tuned and trained using five-fold cross-validation on set 1864 whole slide images (WSIs) 434 patients...
The use of linked healthcare data in research has the potential to make major contributions knowledge generation and service improvement. However, using for secondary purposes raises legal ethical concerns relating confidentiality, privacy protection rights. Using a linkage anonymisation approach that processes lawfully line with best practice create an anonymous (non-personal) dataset can address these concerns, yet there is no set defining all steps involved such flow end-to-end. We aimed...
Since the emergence of COVID-19, deep learning models have been developed to identify COVID-19 from chest X-rays. With little no direct access hospital data, AI community relies heavily on public data comprising numerous sources. Model performance results exceptional when training and testing open-source surpassing reported capabilities in pneumonia-detection prior outbreak. In this study impactful are trained a widely used tested an external test set dataset, for task classifying X-rays...
Introduction More people are living with and beyond a cancer diagnosis. There is limited understanding of the long-term effects treatment on quality life personal household finances when compared to without cancer. In separate protocol we have proposed link de-identified data from electronic primary care hospital records for large population survivors matched controls. this current protocol, propose linkage Patient Reported Outcomes Measures above subset population. The aim study investigate...
Introduction This study is a retrospective evaluation of the performance deep learning models that were developed for detection COVID-19 from chest x-rays, undertaken with goal assessing suitability such systems as clinical decision support tools. Methods Models trained on National Chest Imaging Database (NCCID), UK-wide multi-centre dataset 26 different NHS hospitals and evaluated independent multi-national datasets. The considers technical contributors to model error potential bias. Model...