Multimodal VAEs have recently received significant attention as generative models for weakly-supervised learning with multiple heterogeneous modalities. In parallel, VAE-based methods have been explored as probabilistic approaches for clustering tasks. Our work lies at the intersection of these two research directions. We propose a novel multimodal VAE model, in which the latent space is extended to learn data clusters, leveraging shared information across modalities. Our experiments show that our proposed model improves generative performance over existing multimodal VAEs, particularly for unconditional generation. Furthermore, our method favourably compares to alternative clustering approaches, in weakly-supervised settings. Notably, we propose a post-hoc procedure that avoids the need for our method to have a priori knowledge of the true number of clusters, mitigating a critical limitation of previous clustering frameworks.

Emanuele Palumbo, Sonia Laguna, Daphné Chopard, Julia E Vogt

ICML 2023 Workshop on Structured Probabilistic Inference/Generative Modeling

23.06.2023

Link

Multimodal VAEs have recently received significant attention as generative models for weaklysupervised learning with multiple heterogeneous modalities. In parallel, VAE-based methods have been explored as probabilistic approaches for clustering tasks. Our work lies at the intersection of these two research directions. We propose a novel multimodal VAE model, in which the latent space is extended to learn data clusters, leveraging shared information across modalities. Our experiments show that our proposed model improves generative performance over existing multimodal VAEs, particularly for unconditional generation. Furthermore, our method favorably compares to alternative clustering approaches, in weakly-supervised settings. Notably, we propose a post-hoc procedure that avoids the need for to have a priori knowledge of the true number of clusters, mitigating a critical limitation previous clustering frameworks.

Emanuele Palumbo, Sonia Laguna, Daphné Chopard, Julia E Vogt

ICML 2023 Workshop DeployableGenerativeAI

23.06.2023

Link

Electronic health records contain a wealth of valuable information for improving healthcare. There are, however, challenges associated with clinical text that prevent computers from maximising the utility of such information. While deep learning (DL) has emerged as a practical paradigm for dealing with the complexities of natural language, applying this class of machine learning algorithms to clinical text raises several research questions. First, we tackled the problem of data sparsity by looking into the task of adverse event detection. As these events are rare, examples thereof are lacking. To compensate for data scarcity, we leveraged large pre-trained language models (LMs) in combination with formally represented medical knowledge. We demonstrated that such a combination exhibits remarkable generalisation abilities despite the low availability of data. Second, we focused on the omnipresence of short forms in clinical texts. This typically leads to out-of-vocabulary problems, which motivates unlocking the underlying words. The novelty of our approach lies in its capacity to learn how to automatically expand short forms without resorting to external resources. Third, we investigated data augmentation to address the issue of data scarcity at its core. To the best of our knowledge, we were one of the firsts to investigate population-based augmentation for scheduling text data augmentation. Interestingly, little improvement was seen in fine-tuning large pre-trained LMs with the augmented data. We suggest that, as LMs proved able to cope well with small datasets, the need for data augmentation was made redundant. We conclude that DL approaches to clinical text mining should be developed by fine-tuning large LMs. One area where such models may struggle is the use of clinical short forms. Our method to automating their expansion fixes this issue. Together, these two approaches provide a blueprint for successfully developing DL approaches to clinical text mining in low-data regimes.

Daphné Chopard

PhD Thesis

15.03.2023

Link

Motivation: Global acronyms are used in written text without their formal definitions. This makes it difficult to automatically interpret their sense as acronyms tend to be ambiguous. Supervised machine learning approaches to sense disambiguation require large training datasets. In clinical applications, large datasets are difficult to obtain due to patient privacy. Manual data annotation creates an additional bottleneck. Results: We proposed an approach to automatically modifying scientific abstracts to (i) simulate global acronym usage and (ii) annotate their senses without the need for external sources or manual intervention. We implemented it as a web-based application, which can create large datasets that in turn can be used to train supervised approaches to word sense disambiguation of biomedical acronyms. Availability and implementation: The datasets will be generated on demand based on a user query and will be downloadable from https://datainnovation.cardiff.ac.uk/acronyms/.

Maxim Filimonov, Daphné Chopard, Irena Spasić

Bioinformatics

26.05.2022

LinkPubmedDOI

Background: Pharmacovigilance and safety reporting, which involve processes for monitoring the use of medicines in clinical trials, play a critical role in the identification of previously unrecognized adverse events or changes in the patterns of adverse events. Objective: This study aims to demonstrate the feasibility of automating the coding of adverse events described in the narrative section of the serious adverse event report forms to enable statistical analysis of the aforementioned patterns. Methods: We used the Unified Medical Language System (UMLS) as the coding scheme, which integrates 217 source vocabularies, thus enabling coding against other relevant terminologies such as the International Classification of Diseases–10th Revision, Medical Dictionary for Regulatory Activities, and Systematized Nomenclature of Medicine). We used MetaMap, a highly configurable dictionary lookup software, to identify the mentions of the UMLS concepts. We trained a binary classifier using Bidirectional Encoder Representations from Transformers (BERT), a transformer-based language model that captures contextual relationships, to differentiate between mentions of the UMLS concepts that represented adverse events and those that did not. Results: The model achieved a high F1 score of 0.8080, despite the class imbalance. This is 10.15 percent points lower than human-like performance but also 17.45 percent points higher than that of the baseline approach. Conclusions: These results confirmed that automated coding of adverse events described in the narrative section of serious adverse event reports is feasible. Once coded, adverse events can be statistically analyzed so that any correlations with the trialed medicines can be estimated in a timely fashion.

Daphné Chopard, Matthias S Treder, Padraig Corcoran, Nagheen Ahmed, Claire Johnson, Monica Busse, Irena Spasić, others

JMIR Medical Informatics

24.12.2021

LinkPubmedDOI

Despite its proven efficiency in other fields, data augmentation is less popular in the context of natural language processing (NLP) due to its complexity and limited results. A recent study (Longpre et al., 2020) showed for example that task-agnostic data augmentations fail to consistently boost the performance of pretrained transformers even in low data regimes. In this paper, we investigate whether data-driven augmentation scheduling and the integration of a wider set of transformations can lead to improved performance where fixed and limited policies were unsuccessful. Our results suggest that, while this approach can help the training process in some settings, the improvements are unsubstantial. This negative result is meant to help researchers better understand the limitations of data augmentation for NLP.

Daphné Chopard, Matthias S Treder, Irena Spasić

Proceedings of the Second Workshop on Insights from Negative Results in NLP

01.11.2021

LinkDOICode

Abbreviations and acronyms are shortened forms of words or phrases that are commonly used in technical writing. In this study we focus specifically on abbreviations and introduce a corpus-based method for their expansion. The method divides the processing into three key stages: abbreviation identification, full form candidate extraction, and abbreviation disambiguation. First, potential abbreviations are identified by combining pattern matching and named entity recognition. Both acronyms and abbreviations exhibit similar orthographic properties, thus additional processing is required to distinguish between them. To this end, we implement a character-based recurrent neural network (RNN) that analyses the morphology of a given token in order to classify it as an acronym or an abbreviation. A siamese RNN that learns the morphological process of word abbreviation is then used to select a set of full form candidates. Having considerably constrained the search space, we take advantage of the Word Mover’s Distance (WMD) to assess semantic compatibility between an abbreviation and each full form candidate based on their contextual similarity. This step does not require any corpus-based training, thus making the approach highly adaptable to different domains. Unlike the vast majority of existing approaches, our method does not rely on external lexical resources for disambiguation, but with a macro F-measure of 96.27% is comparable to the state-of-the art.

Daphné Chopard, Irena Spasić

Statistical Language and Speech Processing: 7th International Conference, SLSP 2019, Ljubljana, Slovenia, October 14--16, 2019, Proceedings 7

14.10.2019

LinkDOI