Community knowledge graph abstraction for enhanced link prediction: A study on PubMed knowledge graph.

OBJECTIVE: As new knowledge is produced at a rapid pace in the biomedical field, existing biomedical Knowledge Graphs (KGs) cannot be manually updated in a timely manner. Previous work in Natural Language Processing (NLP) has leveraged link prediction to infer the missing knowledge in general-purpose KGs. Inspired by this, we propose to apply link prediction to existing biomedical KGs to infer missing knowledge. Although Knowledge Graph Embedding (KGE) methods are effective in link prediction tasks, they are less capable of capturing relations between communities of entities with specific attributes (Fanourakis et al., 2023). METHODS: To address this challenge, we proposed an entity distance-based method for abstracting a Community Knowledge Graph (CKG) from a simplified version of the pre-existing PubMed Knowledge Graph (PKG) (Xu et al., 2020). For link prediction on the abstracted CKG, we proposed an extension approach for the existing KGE models by linking the information in the PKG to the abstracted CKG. The applicability of this extension was proved by employing six well-known KGE models: TransE, TransH, DistMult, ComplEx, SimplE, and RotatE. Evaluation metrics including Mean Rank (MR), Mean Reciprocal Rank (MRR), and Hits@k were used to assess the link prediction performance. In addition, we presented a backtracking process that traces the results of CKG link prediction back to the PKG scale for further comparison. RESULTS: Six different CKGs were abstracted from the PKG by using embeddings of the six KGE methods. The results of link prediction in these abstracted CKGs indicate that our proposed extension can improve the existing KGE methods, achieving a top-10 accuracy of 0.69 compared to 0.5 for TransE, 0.7 compared to 0.54 for TransH, 0.67 compared to 0.6 for DistMult, 0.73 compared to 0.57 for ComplEx, 0.73 compared to 0.63 for SimplE, and 0.85 compared to 0.76 for RotatE on their CKGs, respectively. These improved performances also highlight the wide applicability of the extension approach. CONCLUSION: This study proposed novel insights into abstracting CKGs from the PKG. The extension approach indicated enhanced performance of the existing KGE methods and has applicability. As an interesting future extension, we plan to conduct link prediction for entities that are newly introduced to the PKG.

Improvement of intervention information detection for automated clinical literature screening during systematic review.

Systematic literature review (SLR) is a crucial method for clinicians and policymakers to make their decisions in a flood of new clinical studies. Because manual literature screening in SLR is a highly laborious task, its automation by natural language processing (NLP) has been welcomed. Although intervention is a key information for literature screening, NLP models for its detection in previous works have not shown adequate performance. In this work, we first design an algorithm for automated construction of high-quality intervention labels by utilizing information retrieved from a clinical trial database. We then design another algorithm for improving model's recall and F1 score by imposing adaptive weights on training instances in the loss function. The intervention detection model trained on the weighted datasets is tested with the Evidence-Based Medicine NLP (EBM-NLP) corpus, and shows 9.7% and 4.0% improvements respectively in recall and F1 score compared to the previous state-of-the-art model on the corpus. The proposed algorithms can boost automation of literature screening during SLR in the clinical domain.

Social media and pharmacovigilance: A review of the opportunities and challenges.

Adverse drug reactions come at a considerable cost on society. Social media are a potentially invaluable reservoir of information for pharmacovigilance, yet their true value remains to be fully understood. In order to realize the benefits social media holds, a number of technical, regulatory and ethical challenges remain to be addressed. We outline these key challenges identifying relevant current research and present possible solutions.

A qualitative exploration of barriers to efficient and effective structured medication reviews in primary care: Findings from the DynAIRx study.

INTRODUCTION: Structured medication reviews (SMRs), introduced in the United Kingdom (UK) in 2020, aim to enhance shared decision-making in medication optimisation, particularly for patients with multimorbidity and polypharmacy. Despite its potential, there is limited empirical evidence on the implementation of SMRs, and the challenges faced in the process. This study is part of a larger DynAIRx (Artificial Intelligence for dynamic prescribing optimisation and care integration in multimorbidity) project which aims to introduce Artificial Intelligence (AI) to SMRs and develop machine learning models and visualisation tools for patients with multimorbidity. Here, we explore how SMRs are currently undertaken and what barriers are experienced by those involved in them. METHODS: Qualitative focus groups and semi-structured interviews took place between 2022-2023. Six focus groups were conducted with doctors, pharmacists and clinical pharmacologists (n = 21), and three patient focus groups with patients with multimorbidity (n = 13). Five semi-structured interviews were held with 2 pharmacists, 1 trainee doctor, 1 policy-maker and 1 psychiatrist. Transcripts were analysed using thematic analysis. RESULTS: Two key themes limiting the effectiveness of SMRs in clinical practice were identified: 'Medication Reviews in Practice' and 'Medication-related Challenges'. Participants noted limitations to the efficient and effectiveness of SMRs in practice including the scarcity of digital tools for identifying and prioritising patients for SMRs; organisational and patient-related challenges in inviting patients for SMRs and ensuring they attend; the time-intensive nature of SMRs, the need for multiple appointments and shared decision-making; the impact of the healthcare context on SMR delivery; poor communication and data sharing issues between primary and secondary care; difficulties in managing mental health medications and specific challenges associated with anticholinergic medication. CONCLUSION: SMRs are complex, time consuming and medication optimisation may require multiple follow-up appointments to enable a comprehensive review. There is a need for a prescribing support system to identify, prioritise and reduce the time needed to understand the patient journey when dealing with large volumes of disparate clinical information in electronic health records. However, monitoring the effects of medication optimisation changes with a feedback loop can be challenging to establish and maintain using current electronic health record systems.

The DynAIRx Project Protocol: Artificial Intelligence for dynamic prescribing optimisation and care integration in multimorbidity.

Background: Structured Medication Reviews (SMRs) are intended to help deliver the NHS Long Term Plan for medicines optimisation in people living with multiple long-term conditions and polypharmacy. It is challenging to gather the information needed for these reviews due to poor integration of health records across providers and there is little guidance on how to identify those patients most urgently requiring review. Objective: To extract information from scattered clinical records on how health and medications change over time, apply interpretable artificial intelligence (AI) approaches to predict risks of poor outcomes and overlay this information on care records to inform SMRs. We will pilot this approach in primary care prescribing audit and feedback systems, and co-design future medicines optimisation decision support systems. Design: DynAIRx will target potentially problematic polypharmacy in three key multimorbidity groups, namely, people with (a) mental and physical health problems, (b) four or more long-term conditions taking ten or more drugs and (c) older age and frailty. Structured clinical data will be drawn from integrated care records (general practice, hospital, and social care) covering an ∼11m population supplemented with Natural Language Processing (NLP) of unstructured clinical text. AI systems will be trained to identify patterns of conditions, medications, tests, and clinical contacts preceding adverse events in order to identify individuals who might benefit most from an SMR. Discussion: By implementing and evaluating an AI-augmented visualisation of care records in an existing prescribing audit and feedback system we will create a learning system for medicines optimisation, co-designed throughout with end-users and patients.

Fine-Tuning Word Embeddings for Hierarchical Representation of Data Using a Corpus and a Knowledge Base for Various Machine Learning Applications.

Word embedding models have recently shown some capability to encode hierarchical information that exists in textual data. However, such models do not explicitly encode the hierarchical structure that exists among words. In this work, we propose a method to learn hierarchical word embeddings (HWEs) in a specific order to encode the hierarchical information of a knowledge base (KB) in a vector space. To learn the word embeddings, our proposed method considers not only the hypernym relations that exist between words in a KB but also contextual information in a text corpus. The experimental results on various applications, such as supervised and unsupervised hypernymy detection, graded lexical entailment prediction, hierarchical path prediction, and word reconstruction tasks, show the ability of the proposed method to encode the hierarchy. Moreover, the proposed method outperforms previously proposed methods for learning nonspecialised, hypernym-specific, and hierarchical word embeddings on multiple benchmarks.

DeepGraphMolGen, a multi-objective, computational strategy for generating molecules with desirable properties: a graph convolution and reinforcement learning approach.

We address the problem of generating novel molecules with desired interaction properties as a multi-objective optimization problem. Interaction binding models are learned from binding data using graph convolution networks (GCNs). Since the experimentally obtained property scores are recognised as having potentially gross errors, we adopted a robust loss for the model. Combinations of these terms, including drug likeness and synthetic accessibility, are then optimized using reinforcement learning based on a graph convolution policy approach. Some of the molecules generated, while legitimate chemically, can have excellent drug-likeness scores but appear unusual. We provide an example based on the binding potency of small molecules to dopamine transporters. We extend our method successfully to use a multi-objective reward function, in this case for generating novel molecules that bind with dopamine transporters but not with those for norepinephrine. Our method should be generally applicable to the generation in silico of molecules with desirable properties.

Jointly learning word embeddings using a corpus and a knowledge base.

Methods for representing the meaning of words in vector spaces purely using the information distributed in text corpora have proved to be very valuable in various text mining and natural language processing (NLP) tasks. However, these methods still disregard the valuable semantic relational structure between words in co-occurring contexts. These beneficial semantic relational structures are contained in manually-created knowledge bases (KBs) such as ontologies and semantic lexicons, where the meanings of words are represented by defining the various relationships that exist among those words. We combine the knowledge in both a corpus and a KB to learn better word embeddings. Specifically, we propose a joint word representation learning method that uses the knowledge in the KBs, and simultaneously predicts the co-occurrences of two words in a corpus context. In particular, we use the corpus to define our objective function subject to the relational constrains derived from the KB. We further utilise the corpus co-occurrence statistics to propose two novel approaches, Nearest Neighbour Expansion (NNE) and Hedged Nearest Neighbour Expansion (HNE), that dynamically expand the KB and therefore derive more constraints that guide the optimisation process. Our experimental results over a wide-range of benchmark tasks demonstrate that the proposed method statistically significantly improves the accuracy of the word embeddings learnt. It outperforms a corpus-only baseline and reports an improvement of a number of previously proposed methods that incorporate corpora and KBs in both semantic similarity prediction and word analogy detection tasks.

Causality Patterns for Detecting Adverse Drug Reactions From Social Media: Text Mining Approach.

BACKGROUND: Detecting adverse drug reactions (ADRs) is an important task that has direct implications for the use of that drug. If we can detect previously unknown ADRs as quickly as possible, then this information can be provided to the regulators, pharmaceutical companies, and health care organizations, thereby potentially reducing drug-related morbidity and saving lives of many patients. A promising approach for detecting ADRs is to use social media platforms such as Twitter and Facebook. A high level of correlation between a drug name and an event may be an indication of a potential adverse reaction associated with that drug. Although numerous association measures have been proposed by the signal detection community for identifying ADRs, these measures are limited in that they detect correlations but often ignore causality. OBJECTIVE: This study aimed to propose a causality measure that can detect an adverse reaction that is caused by a drug rather than merely being a correlated signal. METHODS: To the best of our knowledge, this was the first causality-sensitive approach for detecting ADRs from social media. Specifically, the relationship between a drug and an event was represented using a set of automatically extracted lexical patterns. We then learned the weights for the extracted lexical patterns that indicate their reliability for expressing an adverse reaction of a given drug. RESULTS: Our proposed method obtains an ADR detection accuracy of 74% on a large-scale manually annotated dataset of tweets, covering a standard set of drugs and adverse reactions. CONCLUSIONS: By using lexical patterns, we can accurately detect the causality between drugs and adverse reaction-related events.

Learning linear transformations between counting-based and prediction-based word embeddings.

Despite the growing interest in prediction-based word embedding learning methods, it remains unclear as to how the vector spaces learnt by the prediction-based methods differ from that of the counting-based methods, or whether one can be transformed into the other. To study the relationship between counting-based and prediction-based embeddings, we propose a method for learning a linear transformation between two given sets of word embeddings. Our proposal contributes to the word embedding learning research in three ways: (a) we propose an efficient method to learn a linear transformation between two sets of word embeddings, (b) using the transformation learnt in (a), we empirically show that it is possible to predict distributed word embeddings for novel unseen words, and

An iterative approach for the global estimation of sentence similarity.

Measuring the similarity between two sentences is often difficult due to their small lexical overlap. Instead of focusing on the sets of features in two given sentences between which we must measure similarity, we propose a sentence similarity method that considers two types of constraints that must be satisfied by all pairs of sentences in a given corpus. Namely, (a) if two sentences share many features in common, then it is likely that the remaining features in each sentence are also related, and (b) if two sentences contain many related features, then those two sentences are themselves similar. The two constraints are utilized in an iterative bootstrapping procedure that simultaneously updates both word and sentence similarity scores. Experimental results on SemEval 2015 Task 2 dataset show that the proposed iterative approach for measuring sentence semantic similarity is significantly better than the non-iterative counterparts.

A cross-lingual similarity measure for detecting biomedical term translations.

Bilingual dictionaries for technical terms such as biomedical terms are an important resource for machine translation systems as well as for humans who would like to understand a concept described in a foreign language. Often a biomedical term is first proposed in English and later it is manually translated to other languages. Despite the fact that there are large monolingual lexicons of biomedical terms, only a fraction of those term lexicons are translated to other languages. Manually compiling large-scale bilingual dictionaries for technical domains is a challenging task because it is difficult to find a sufficiently large number of bilingual experts. We propose a cross-lingual similarity measure for detecting most similar translation candidates for a biomedical term specified in one language (source) from another language (target). Specifically, a biomedical term in a language is represented using two types of features: (a) intrinsic features that consist of character n-grams extracted from the term under consideration, and (b) extrinsic features that consist of unigrams and bigrams extracted from the contextual windows surrounding the term under consideration. We propose a cross-lingual similarity measure using each of those feature types. First, to reduce the dimensionality of the feature space in each language, we propose prototype vector projection (PVP)--a non-negative lower-dimensional vector projection method. Second, we propose a method to learn a mapping between the feature spaces in the source and target language using partial least squares regression (PLSR). The proposed method requires only a small number of training instances to learn a cross-lingual similarity measure. The proposed PVP method outperforms popular dimensionality reduction methods such as the singular value decomposition (SVD) and non-negative matrix factorization (NMF) in a nearest neighbor prediction task. Moreover, our experimental results covering several language pairs such as English-French, English-Spanish, English-Greek, and English-Japanese show that the proposed method outperforms several other feature projection methods in biomedical term translation prediction tasks.

Metaphor interpretation using paraphrases extracted from the web.

Interpreting metaphor is a hard but important problem in natural language processing that has numerous applications. One way to address this task is by finding a paraphrase that can replace the metaphorically used word in a given context. This approach has been previously implemented only within supervised frameworks, relying on manually constructed lexical resources, such as WordNet. In contrast, we present a fully unsupervised metaphor interpretation method that extracts literal paraphrases for metaphorical expressions from the Web. It achieves a precision of [Formula: see text], which is high for an unsupervised paraphrasing approach. Moreover, the method significantly outperforms both the baseline and the selectional preference-based method of Shutova employed in an unsupervised setting.