Publications by authors named "M Modat"
Background: COVID-19 symptoms may persist beyond acute SARS-CoV-2 infection, as ongoing symptomatic COVID-19 [OSC] (symptom duration 4-12 weeks) and post-COVID syndrome [PCS] (symptom duration ≥12 weeks). Vaccination against SARS-CoV-2 decreases OSC/PCS in individuals subsequently infected with SARS-CoV-2 post-vaccination. Whether vaccination against SARS-CoV-2, or any other vaccinations (such as against influenza) affects symptoms in individuals already experiencing OSC/PCS, more than natural symptom evolution, is unknown.
View Article and Find Full Text PDFPurpose: Subarachnoid haemorrhage is a potentially fatal consequence of intracranial aneurysm rupture, however, it is difficult to predict if aneurysms will rupture. Prophylactic treatment of an intracranial aneurysm also involves risk, hence identifying rupture-prone aneurysms is of substantial clinical importance. This systematic review aims to evaluate the performance of machine learning algorithms for predicting intracranial aneurysm rupture risk.
View Article and Find Full Text PDFArtificial Intelligence Applications for Imaging Metabolic Bone Diseases.
Semin Musculoskelet Radiol
October 2024
Article Synopsis
- AI is revolutionizing medical imaging for metabolic bone diseases (MBDs) like osteoporosis and rare conditions, enhancing diagnosis and management.
- The article analyzes various AI techniques, recent advancements, and their clinical applications, while addressing ethical issues and future developments.
- By combining AI with existing imaging methods, there is potential for improved diagnostic accuracy and patient outcomes in the treatment of MBDs.
Lazy Resampling: Fast and information preserving preprocessing for deep learning.
Comput Methods Programs Biomed
December 2024
Article Synopsis
- Data preprocessing is essential in deep learning, especially in computer vision, but traditional methods can lead to inefficiencies, image quality degradation, and added complexity, particularly in medical imaging.
- Lazy Resampling offers a solution by consolidating multiple spatial preprocessing operations into a single step, minimizing execution time and signal loss, while simplifying the design of preprocessing pipelines.
- Evaluations show that Lazy Resampling reduces information loss and enhances label accuracy, resulting in improved performance for semantic segmentation tasks when training models like UNets.
Background: Predicting diabetic retinopathy (DR) progression could enable individualised screening with prompt referral for high-risk individuals for sight-saving treatment, whilst reducing screening burden for low-risk individuals. We developed and validated deep learning systems (DLS) that predict 1, 2 and 3 year emergent referable DR and maculopathy using risk factor characteristics (tabular DLS), colour fundal photographs (image DLS) or both (multimodal DLS).
Methods: From 162,339 development-set eyes from south-east London (UK) diabetic eye screening programme (DESP), 110,837 had eligible longitudinal data, with the remaining 51,502 used for pretraining.