Zainab Awan | Searcharxiv

arXiv2024

Enhancing Cardiovascular Disease Risk Prediction with Machine Learning Models

Cardiovascular disease remains a leading global cause of mortality, necessitating accurate risk prediction tools. Traditional methods, such as QRISK and the Framingham heart score, exhibit limitations in their ability to incorporate comprehensive patient data, potentially resulting in incomplete risk factor consideration. To address these shortcomings, this study conducts a meticulous review focusing on the application of machine learning models to enhance predictive accuracy. Machine learning models, such as support vector machines, and Random Forest, as well as deep learning techniques like convolutional neural networks and recurrent neural networks, have emerged as promising alternatives. These models offer superior performance, accommodating a broader spectrum of variables and providing precise subgroup-specific predictions. While machine learning integration holds promise for enhancing risk assessment, it presents challenges such as data requirements and computational constraints. Additionally, large language models have revolutionised healthcare applications, augmenting diagnostic precision and patient care. This study examines the core aspects of cardiovascular disease event risk and presents a thorough review of traditional and machine learning models, alongside deep learning techniques, for improved accuracy. It offers a comprehensive survey of relevant datasets, critically compares ML models with conventional approaches, and synthesizes key findings, highlighting their implications for clinical practice. Furthermore, the potential of machine learning and large language models in cardiovascular medicine is undeniable. However, rigorous validation and optimisation are imperative before widespread application in healthcare. This integration promises more accurate and personalised cardiovascular care.

q-bio.GN↗

arXiv2023

Bi-Encoders based Species Normalization -- Pairwise Sentence Learning to Rank

Motivation: Biomedical named-entity normalization involves connecting biomedical entities with distinct database identifiers in order to facilitate data integration across various fields of biology. Existing systems for biomedical named entity normalization heavily rely on dictionaries, manually created rules, and high-quality representative features such as lexical or morphological characteristics. However, recent research has investigated the use of neural network-based models to reduce dependence on dictionaries, manually crafted rules, and features. Despite these advancements, the performance of these models is still limited due to the lack of sufficiently large training datasets. These models have a tendency to overfit small training corpora and exhibit poor generalization when faced with previously unseen entities, necessitating the redesign of rules and features. Contribution: We present a novel deep learning approach for named entity normalization, treating it as a pair-wise learning to rank problem. Our method utilizes the widely-used information retrieval algorithm Best Matching 25 to generate candidate concepts, followed by the application of bi-directional encoder representation from the encoder (BERT) to re-rank the candidate list. Notably, our approach eliminates the need for feature-engineering or rule creation. We conduct experiments on species entity types and evaluate our method against state-of-the-art techniques using LINNAEUS and S800 biomedical corpora. Our proposed approach surpasses existing methods in linking entities to the NCBI taxonomy. To the best of our knowledge, there is no existing neural network-based approach for species normalization in the literature.

cs.CL↗