Search

International Classification of Diseases (ICD) serves as the foundation for generating comparable global disease statistics across regions and over time. The process of ICD coding involves assigning codes to diseases based on clinical notes, which can describe a patient's condition in a standard way. However, this process is complicated by the vast number of codes and the intricate taxonomy of ICD codes, which are hierarchically organized into various levels, including chapter, category, subcategory, and its subdivisions. Many existing studies focus solely on predicting subcategory codes, ignoring the hierarchical relationships among codes. To address this limitation, we propose a multitask learning model that trains multiple classifiers for different code levels, while also capturing the relations between coarser and finer-grained labels through a reinforcement mechanism. Our approach is evaluated on both English and Chinese benchmark dataset, and we demonstrate that our method achieves competitive performance with baseline models, particularly in terms of macro-F1 results. These findings suggest that our approach effectively leverages the hierarchical structure of ICD codes to improve disease code prediction accuracy. Analysis of attention mechanism shows that multigranularity attention of our model captures crucial feature of input text on different granularity levels, which can provide reasonable explanations for the prediction results. [ABSTRACT FROM AUTHOR]

Motivation There is mounting evidence that the subcellular localization of lncRNAs can provide valuable insights into their biological functions. In the real world of transcriptomes, lncRNAs are usually localized in multiple subcellular localizations. Furthermore, lncRNAs have specific localization patterns for different subcellular localizations. Although several computational methods have been developed to predict the subcellular localization of lncRNAs, few of them are designed for lncRNAs that have multiple subcellular localizations, and none of them take motif specificity into consideration. Results In this study, we proposed a novel deep learning model, called LncLocFormer, which uses only lncRNA sequences to predict multi-label lncRNA subcellular localization. LncLocFormer utilizes eight Transformer blocks to model long-range dependencies within the lncRNA sequence and shares information across the lncRNA sequence. To exploit the relationship between different subcellular localizations and find distinct localization patterns for different subcellular localizations, LncLocFormer employs a localization-specific attention mechanism. The results demonstrate that LncLocFormer outperforms existing state-of-the-art predictors on the hold-out test set. Furthermore, we conducted a motif analysis and found LncLocFormer can capture known motifs. Ablation studies confirmed the contribution of the localization-specific attention mechanism in improving the prediction performance. Availability and implementation The LncLocFormer web server is available at http://csuligroup.com:9000/LncLocFormer. The source code can be obtained from https://github.com/CSUBioGroup/LncLocFormer. [ABSTRACT FROM AUTHOR]

Determining drug–drug interactions (DDIs) is an important part of pharmacovigilance and has a vital impact on public health. Compared with drug trials, obtaining DDI information from scientific articles is a faster and lower cost but still a highly credible approach. However, current DDI text extraction methods consider the instances generated from articles to be independent and ignore the potential connections between different instances in the same article or sentence. Effective use of external text data could improve prediction accuracy, but existing methods cannot extract key information from external data accurately and reasonably, resulting in low utilization of external data. In this study, we propose a DDI extraction framework, instance position embedding and key external text for DDI (IK-DDI), which adopts instance position embedding and key external text to extract DDI information. The proposed framework integrates the article-level and sentence-level position information of the instances into the model to strengthen the connections between instances generated from the same article or sentence. Moreover, we introduce a comprehensive similarity-matching method that uses string and word sense similarity to improve the matching accuracy between the target drug and external text. Furthermore, the key sentence search method is used to obtain key information from external data. Therefore, IK-DDI can make full use of the connection between instances and the information contained in external text data to improve the efficiency of DDI extraction. Experimental results show that IK-DDI outperforms existing methods on both macro-averaged and micro-averaged metrics, which suggests our method provides complete framework that can be used to extract relationships between biomedical entities and process external text data. [ABSTRACT FROM AUTHOR]

Bainite transformation start temperature (Bs) is an important index to measure the properties of bainitic steel. Based on the experimental results of bainite transformation behavior, the atomic scale characteristics are introduced, and the influence and prediction of different component content on Bs are analyzed by machine learning algorithm. The results show that Bs decreased significantly with the increase in C content (0−0.6wt.%) and Si content (0−0.2 wt.%), while the tendence remains almost unchanged when the Si content is greater than 0.2 wt.%. Furthermore, according to the analysis of atomic scale features, Bs has the strongest dependence on the number of valence electrons and the radius change rate relative to iron. The combination with the above two atomic scale features show the best model performance. The relationship between these two features and Bs is positively proportional, and Bs rises with the increase in their values. Extracting the valuable information about the relationship between Bs and element characteristics from the collected experimental data is of great significance to provide theoretical foundation of possible direction for the advances of designing the excellent properties in steels. [ABSTRACT FROM AUTHOR]

Address matching, which aims to match unstructured addresses with standard addresses in an address database, is a key part of geocoding. The core problem of address matching corresponds to text matching in natural language processing. Existing rule-based methods require human-designed templates and thus, have limited applicability. Machine learning and deep learning-based methods ignore the hierarchical relations between address elements, which easily misclassify semantically similar but geographically different locations. We note that the hierarchy of address elements can fill the semantic gap in address matching. Inspired by how humans discriminate addresses, we propose a multi-task learning approach. The approach jointly recognises the address elements and matches the addresses to incorporate the hierarchical relations between the address elements into the neural network. Simultaneously, we introduce a priori information on the hierarchical relationship of address elements through the conditional random field model. Experimental results on the benchmark datasets Shenzhen Address Database and Jiangsu-Hunan Address Dataset demonstrate the effectiveness of our approach. We achieved state-of-the-art F1 scores (i.e. the harmonic mean of precision and recall) of 99.0 and 94.2 on the two datasets, respectively. [ABSTRACT FROM AUTHOR]