Alison Q. O'Neil
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Alison Q. O'Neil has a total of 77 co-authors. Here are the top co-authors by number of shared publications:
- Sotirios A. Tsaftaris
The University of Edinburgh
11 shared publications - Pedro Sanchez
School of Engineering, University of Edinburgh, Edinburgh, U.K.
6 shared publications - Xiao Liu
University of Edinburgh
6 shared publications - Jesse Phitidis
Centre for Clinical Brain Sciences University of Edinburgh 49 Little France Crescent EH164SB Edinburgh United Kingdom
4 shared publications - Joanna Marguerite Wardlaw
Centre for Clinical Brain Sciences, UK Dementia Research Institute, University of Edinburgh, Edinburgh, UK
4 shared publications - Maria Del Carmen Valdes Hernandez
Centre for Clinical Brain Sciences, Department of Neuroimaging Sciences, University of Edinburgh
4 shared publications
Latest Publications
- Open Access
- 24 August 2026
- Published by Elsevier in Artificial Intelligence in Medicine
- Vol. 181
Comparative evaluation of training strategies using partially labelled datasets for segmentation of white matter hyperintensities and stroke lesions in FLAIR MRI
- 24 August 2026
- Published by Elsevier in Artificial Intelligence in Medicine
- Vol. 181
Open AccessComparative evaluation of training strategies using partially labelled datasets for segmentation of white matter hyperintensities and stroke lesions in FLAIR MRI
- Open Access
Anatomy-Grounded Weakly Supervised Prompt Tuning for Chest X-ray Latent Diffusion Models
Latent Diffusion Models have shown remarkable results in text-guided image synthesis in recent years. In the domain of natural (RGB) images, recent works have shown that such models can be adapted to various vision-language downstream tasks with little to no supervision involved. On the contrary, text-to-image Latent Diffusion Models remain relatively underexplored in the field of medical imaging, primarily due to limited data availability (e.g., due to privacy concerns). In this work, focusing on the chest X-ray modality, we first demonstrate that a standard text-conditioned Latent Diffusion Model has not learned to align clinically relevant information in free-text radiology reports with the corresponding areas of the given scan. Then, to alleviate this issue, we propose a fine-tuning framework to improve multi-modal alignment in a pre-trained model such that it can be efficiently repurposed for downstream tasks such as phrase grounding. Our method sets a new state-of-the-art on a standard benchmark dataset (MS-CXR), while also exhibiting robust performance on out-of-distribution data (VinDr-CXR). Our code will be made publicly available.
Open AccessAnatomy-Grounded Weakly Supervised Prompt Tuning for Chest X-ray Latent Diffusion Models
Latent Diffusion Models have shown remarkable results in text-guided image synthesis in recent years. In the domain of natural (RGB) images, recent works have shown that such models can be adapted to various vision-language downstream tasks with little to no supervision involved. On the contrary, text-to-image Latent Diffusion Models remain relatively underexplored in the field of medical imaging, primarily due to limited data availability (e.g., due to privacy concerns). In this work, focusing on the chest X-ray modality, we first demonstrate that a standard text-conditioned Latent Diffusion Model has not learned to align clinically relevant information in free-text radiology reports with the corresponding areas of the given scan. Then, to alleviate this issue, we propose a fine-tuning framework to improve multi-modal alignment in a pre-trained model such that it can be efficiently repurposed for downstream tasks such as phrase grounding. Our method sets a new state-of-the-art on a standard benchmark dataset (MS-CXR), while also exhibiting robust performance on out-of-distribution data (VinDr-CXR). Our code will be made publicly available.
- Open Access
Exposure to the organochlorine pesticide cis-chlordane induces ALS-like mitochondrial perturbations in stem cell-derived motor neurons
Amyotrophic Lateral Sclerosis (ALS) is a debilitating and incurable neurodegenerative disease with unsolved etiology. Due to the large proportion of patients lacking direct disease inheritance, understanding the environmental factors that contribute to ALS development is of high priority. Epidemiological studies have implicated pesticides and other environmental exposures as possible contributors to ALS pathogenesis. Recently, our group determined that the organochlorine pesticide cis-chlordane is toxic to human motor neurons in a dose-dependent manner, causing an ALS-like phenotype in culture and animals with a mode of action independent of its known GABA A antagonism. Here, we aimed to characterize downstream motor neuron phenotypes associated with cis -chlordane treatment. We performed bulk RNA sequencing, live imaging, immunofluorescent labeling, and real-time metabolic assays on stem cell-derived motor neurons to assess chlordane-associated phenotypes in vitro. We demonstrate that cis -chlordane treatment causes a highly altered mitochondrial phenotype in motor neurons, including increased production of reactive oxygen species, decreased OCR and ATP production, and loss of mitochondrial membrane potential. We further implicate cis -chlordane as a possible mediator of potent motor neuron damage, with exposure to the pesticide inducing mitochondrial phenotypes akin to those seen in ALS. We suggest that future studies investigating the role of pesticides in ALS development center upon the organochlorine molecules.
Open AccessExposure to the organochlorine pesticide cis-chlordane induces ALS-like mitochondrial perturbations in stem cell-derived motor neurons
Amyotrophic Lateral Sclerosis (ALS) is a debilitating and incurable neurodegenerative disease with unsolved etiology. Due to the large proportion of patients lacking direct disease inheritance, understanding the environmental factors that contribute to ALS development is of high priority. Epidemiological studies have implicated pesticides and other environmental exposures as possible contributors to ALS pathogenesis. Recently, our group determined that the organochlorine pesticide cis-chlordane is toxic to human motor neurons in a dose-dependent manner, causing an ALS-like phenotype in culture and animals with a mode of action independent of its known GABA A antagonism. Here, we aimed to characterize downstream motor neuron phenotypes associated with cis -chlordane treatment. We performed bulk RNA sequencing, live imaging, immunofluorescent labeling, and real-time metabolic assays on stem cell-derived motor neurons to assess chlordane-associated phenotypes in vitro. We demonstrate that cis -chlordane treatment causes a highly altered mitochondrial phenotype in motor neurons, including increased production of reactive oxygen species, decreased OCR and ATP production, and loss of mitochondrial membrane potential. We further implicate cis -chlordane as a possible mediator of potent motor neuron damage, with exposure to the pesticide inducing mitochondrial phenotypes akin to those seen in ALS. We suggest that future studies investigating the role of pesticides in ALS development center upon the organochlorine molecules.
- Open Access
- 15 March 2025
- Published by Elsevier in Computer Methods and Programs in Biomedicine
- Vol. 264, 108715
Automated neuroradiological support systems for multiple cerebrovascular disease markers — A systematic review and meta-analysis
- 15 March 2025
- Published by Elsevier in Computer Methods and Programs in Biomedicine
- Vol. 264, 108715
Open AccessAutomated neuroradiological support systems for multiple cerebrovascular disease markers — A systematic review and meta-analysis
- 8 November 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of Biomedical and Health Informatics
- Vol. 29 (12) , 9051-9059
Zero-Shot Medical Phrase Grounding With Off-the-Shelf Diffusion Models
Localizing the exact pathological regions in a given medical scan is an important imaging problem that traditionally requires a large amount of bounding box ground truth annotations to be accurately solved. However, there exist alternative, potentially weaker, forms of supervision, such as accompanying free-text reports, which are readily available. The task of performing localization with textual guidance is commonly referred to as phrase grounding. In this work, we use a publicly available Foundation Model, namely the Latent Diffusion Model, to perform this challenging task. This choice is supported by the fact that the Latent Diffusion Model, despite being generative in nature, contains cross-attention mechanisms that implicitly align visual and textual features, thus leading to intermediate representations that are suitable for the task at hand. In addition, we aim to perform this task in a zero-shot manner, i.e., without any training on the target task, meaning that the model's weights remain frozen. To this end, we devise strategies to select features and also refine them via post-processing without extra learnable parameters. We compare our proposed method with state-of-the-art approaches which explicitly enforce image-text alignment in a joint embedding space via contrastive learning. Results on a popular chest X-ray benchmark indicate that our method is competitive with SOTA on different types of pathology, and even outperforms them on average in terms of two metrics (mean IoU and AUC-ROC). Source code will be released upon acceptance at https://github.com/vios-s .
- 8 November 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of Biomedical and Health Informatics
- Vol. 29 (12) , 9051-9059
Zero-Shot Medical Phrase Grounding With Off-the-Shelf Diffusion Models
Localizing the exact pathological regions in a given medical scan is an important imaging problem that traditionally requires a large amount of bounding box ground truth annotations to be accurately solved. However, there exist alternative, potentially weaker, forms of supervision, such as accompanying free-text reports, which are readily available. The task of performing localization with textual guidance is commonly referred to as phrase grounding. In this work, we use a publicly available Foundation Model, namely the Latent Diffusion Model, to perform this challenging task. This choice is supported by the fact that the Latent Diffusion Model, despite being generative in nature, contains cross-attention mechanisms that implicitly align visual and textual features, thus leading to intermediate representations that are suitable for the task at hand. In addition, we aim to perform this task in a zero-shot manner, i.e., without any training on the target task, meaning that the model's weights remain frozen. To this end, we devise strategies to select features and also refine them via post-processing without extra learnable parameters. We compare our proposed method with state-of-the-art approaches which explicitly enforce image-text alignment in a joint embedding space via contrastive learning. Results on a popular chest X-ray benchmark indicate that our method is competitive with SOTA on different types of pathology, and even outperforms them on average in terms of two metrics (mean IoU and AUC-ROC). Source code will be released upon acceptance at https://github.com/vios-s .
- 24 July 2024
- Published by Springer Nature
- p. 227-241
Confounder-Aware Image Synthesis for Pathology Segmentation in New Magnetic Resonance Imaging Sequences
- 24 July 2024
- Published by Springer Nature
- p. 227-241
Confounder-Aware Image Synthesis for Pathology Segmentation in New Magnetic Resonance Imaging Sequences
- 27 April 2024
- Published by Springer Nature
- p. 63-72
Clinically Focussed Evaluation of Anomaly Detection and Localisation Methods Using Inpatient CT Head Data
- 27 April 2024
- Published by Springer Nature
- p. 63-72
Clinically Focussed Evaluation of Anomaly Detection and Localisation Methods Using Inpatient CT Head Data
- 26 January 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Medical Imaging
- Vol. 43 (6) , 2169-2179
Compositionally Equivariant Representation Learning
Deep learning models often need sufficient supervision (i.e. labelled data) in order to be trained effectively. By contrast, humans can swiftly learn to identify important anatomy in medical images like MRI and CT scans, with minimal guidance. This recognition capability easily generalises to new images from different medical facilities and to new tasks in different settings. This rapid and generalisable learning ability is largely due to the compositional structure of image patterns in the human brain, which are not well represented in current medical models. In this paper, we study the utilisation of compositionality in learning more interpretable and generalisable representations for medical image segmentation. Overall, we propose that the underlying generative factors that are used to generate the medical images satisfy compositional equivariance property, where each factor is compositional (e.g. corresponds to human anatomy) and also equivariant to the task. Hence, a good representation that approximates well the ground truth factor has to be compositionally equivariant. By modelling the compositional representations with learnable von-Mises-Fisher (vMF) kernels, we explore how different design and learning biases can be used to enforce the representations to be more compositionally equivariant under un-, weakly-, and semi-supervised settings. Extensive results show that our methods achieve the best performance over several strong baselines on the task of semi-supervised domain-generalised medical image segmentation. Code will be made publicly available upon acceptance at https://github.com/vios-s.
- 26 January 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Medical Imaging
- Vol. 43 (6) , 2169-2179
Compositionally Equivariant Representation Learning
Deep learning models often need sufficient supervision (i.e. labelled data) in order to be trained effectively. By contrast, humans can swiftly learn to identify important anatomy in medical images like MRI and CT scans, with minimal guidance. This recognition capability easily generalises to new images from different medical facilities and to new tasks in different settings. This rapid and generalisable learning ability is largely due to the compositional structure of image patterns in the human brain, which are not well represented in current medical models. In this paper, we study the utilisation of compositionality in learning more interpretable and generalisable representations for medical image segmentation. Overall, we propose that the underlying generative factors that are used to generate the medical images satisfy compositional equivariance property, where each factor is compositional (e.g. corresponds to human anatomy) and also equivariant to the task. Hence, a good representation that approximates well the ground truth factor has to be compositionally equivariant. By modelling the compositional representations with learnable von-Mises-Fisher (vMF) kernels, we explore how different design and learning biases can be used to enforce the representations to be more compositionally equivariant under un-, weakly-, and semi-supervised settings. Extensive results show that our methods achieve the best performance over several strong baselines on the task of semi-supervised domain-generalised medical image segmentation. Code will be made publicly available upon acceptance at https://github.com/vios-s.
- 2 December 2023
- Published by Springer Nature
- p. 3-17
Segmentation of White Matter Hyperintensities and Ischaemic Stroke Lesions in Structural MRI
- 2 December 2023
- Published by Springer Nature
- p. 3-17
Segmentation of White Matter Hyperintensities and Ischaemic Stroke Lesions in Structural MRI
- Open Access
Group Distributionally Robust Knowledge Distillation
Knowledge distillation enables fast and effective transfer of features learned from a bigger model to a smaller one. However, distillation objectives are susceptible to sub-population shifts, a common scenario in medical imaging analysis which refers to groups/domains of data that are underrepresented in the training set. For instance, training models on health data acquired from multiple scanners or hospitals can yield subpar performance for minority groups. In this paper, inspired by distributionally robust optimization (DRO) techniques, we address this shortcoming by proposing a group-aware distillation loss. During optimization, a set of weights is updated based on the per-group losses at a given iteration. This way, our method can dynamically focus on groups that have low performance during training. We empirically validate our method, GroupDistil on two benchmark datasets (natural images and cardiac MRIs) and show consistent improvement in terms of worst-group accuracy.
Open AccessGroup Distributionally Robust Knowledge Distillation
Knowledge distillation enables fast and effective transfer of features learned from a bigger model to a smaller one. However, distillation objectives are susceptible to sub-population shifts, a common scenario in medical imaging analysis which refers to groups/domains of data that are underrepresented in the training set. For instance, training models on health data acquired from multiple scanners or hospitals can yield subpar performance for minority groups. In this paper, inspired by distributionally robust optimization (DRO) techniques, we address this shortcoming by proposing a group-aware distillation loss. During optimization, a set of weights is updated based on the per-group losses at a given iteration. This way, our method can dynamically focus on groups that have low performance during training. We empirically validate our method, GroupDistil on two benchmark datasets (natural images and cardiac MRIs) and show consistent improvement in terms of worst-group accuracy.