Dataset Information

Impact of Automatic Query Generation and Quality Recognition Using Deep Learning to Curate Evidence From Biomedical Literature: Empirical Study.

ABSTRACT: BACKGROUND:The quality of health care is continuously improving and is expected to improve further because of the advancement of machine learning and knowledge-based techniques along with innovation and availability of wearable sensors. With these advancements, health care professionals are now becoming more interested and involved in seeking scientific research evidence from external sources for decision making relevant to medical diagnosis, treatments, and prognosis. Not much work has been done to develop methods for unobtrusive and seamless curation of data from the biomedical literature. OBJECTIVE:This study aimed to design a framework that can enable bringing quality publications intelligently to the users' desk to assist medical practitioners in answering clinical questions and fulfilling their informational needs. METHODS:The proposed framework consists of methods for efficient biomedical literature curation, including the automatic construction of a well-built question, the recognition of evidence quality by proposing extended quality recognition model (E-QRM), and the ranking and summarization of the extracted evidence. RESULTS:Unlike previous works, the proposed framework systematically integrates the echelons of biomedical literature curation by including methods for searching queries, content quality assessments, and ranking and summarization. Using an ensemble approach, our high-impact classifier E-QRM obtained significantly improved accuracy than the existing quality recognition model (1723/1894, 90.97% vs 1462/1894, 77.21%). CONCLUSIONS:Our proposed methods and evaluation demonstrate the validity and rigorousness of the results, which can be used in different applications, including evidence-based medicine, precision medicine, and medical education.

SUBMITTER: Afzal M

PROVIDER: S-EPMC6928703 | biostudies-literature | 2019 Dec

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Impact of Automatic Query Generation and Quality Recognition Using Deep Learning to Curate Evidence From Biomedical Literature: Empirical Study.

Afzal Muhammad M Hussain Maqbool M Malik Khalid Mahmood KM Lee Sungyoung S

JMIR medical informatics 20191209 4

<h4>Background</h4>The quality of health care is continuously improving and is expected to improve further because of the advancement of machine learning and knowledge-based techniques along with innovation and availability of wearable sensors. With these advancements, health care professionals are now becoming more interested and involved in seeking scientific research evidence from external sources for decision making relevant to medical diagnosis, treatments, and prognosis. Not much work has ...[more]

PMID: 31815673

Similar Datasets

Project description:BackgroundTraining deep learning (DL) models to automatically recognize diseases in nasopharyngeal MRI is a challenging task, and optimizing the performance of DL models is difficult.PurposeTo develop a method of training anatomical partition-based DL model which integrates knowledge of clinical anatomical regions in otorhinolaryngology to automatically recognize diseases in nasopharyngeal MRI.Study typeSingle-center retrospective study.PopulationA total of 2485 patients with nasopharyngeal diseases (age range 14-82 years, female, 779[31.3%]) and 600 people with normal nasopharynx (age range 18-78 years, female, 281[46.8%]) were included.Sequence3.0 T; T2WI fast spin-echo sequence.AssessmentFull images (512 × 512) of 3085 patients constituted 100% of the dataset, 50% and 25% of which were randomly retained as two new datasets. Two new series of images (seg112 image [112 × 112] and seg224 image [224 × 224]) were automatically generated by a segmentation model. Four pretrained neural networks for nasopharyngeal diseases classification were trained under the nine datasets (full image, seg112 image, and seg224 image, each with 100% dataset, 50% dataset, and 25% dataset).Statistical testsThe receiver operating characteristic curve was used to evaluate the performance of the models. Analysis of variance was used to compare the performance of the models built with different datasets. Statistical significance was set at P < 0.05.ResultsWhen the 100% dataset was used for training, the performances of the models trained with the seg112 images (average area under the curve [aAUC] 0.949 ± 0.052), seg224 images (aAUC 0.948 ± 0.053), and full images (aAUC 0.935 ± 0.053) were similar (P = 0.611). When the 25% dataset was used for training, the mean aAUC of the models that were trained with seg112 images (0.823 ± 0.116) and seg224 images (0.765 ± 0.155) was significantly higher than the models that were trained with full images (0.640 ± 0.154).Data conclusionThe proposed method can potentially improve the performance of the DL model for automatic recognition of diseases in nasopharyngeal MRI.Level of evidence4 TECHNICAL EFFICACY STAGE: 1.

Project description:BackgroundA variety of external factors might seriously degrade PET image quality and lead to inconsistent results. The aim of this study is to explore a potential PET image quality assessment (QA) method with deep learning (DL).MethodsA total of 89 PET images were acquired from Peking Union Medical College Hospital (PUMCH) in China in this study. Ground-truth quality for images was assessed by two senior radiologists and classified into five grades (grade 1, grade 2, grade 3, grade 4, and grade 5). Grade 5 is the best image quality. After preprocessing, the Dense Convolutional Network (DenseNet) was trained to automatically recognize optimal- and poor-quality PET images. Accuracy (ACC), sensitivity, specificity, receiver operating characteristic curve (ROC), and area under the ROC Curve (AUC) were used to evaluate the diagnostic properties of all models. All indicators of models were assessed using fivefold cross-validation. An image quality QA tool was developed based on our deep learning model. A PET QA report can be automatically obtained after inputting PET images.ResultsFour tasks were generated. Task2 showed worst performance in AUC,ACC, specificity and sensitivity among 4 tasks, and task1 showed unstable performance between training and testing and task3 showed low specificity in both training and testing. Task 4 showed the best diagnostic properties and discriminative performance between poor image quality (grade 1, grade 2) and good quality (grade 3, grade 4, grade 5) images. The automated quality assessment of task 4 showed ACC = 0.77, specificity = 0.71, and sensitivity = 0.83, in the train set; ACC = 0.85, specificity = 0.79, and sensitivity = 0.91, in the test set, respectively. The ROC measuring performance of task 4 had an AUC of 0.86 in the train set and 0.91 in the test set. The image QA tool could output basic information of images, scan and reconstruction parameters, typical instances of PET images, and deep learning score.ConclusionsThis study highlights the feasibility of the assessment of image quality in PET images using a deep learning model, which may assist with accelerating clinical research by reliably assessing image quality.

Project description:BACKGROUND:Biomedical named-entity recognition (BioNER) is widely modeled with conditional random fields (CRF) by regarding it as a sequence labeling problem. The CRF-based methods yield structured outputs of labels by imposing connectivity between the labels. Recent studies for BioNER have reported state-of-the-art performance by combining deep learning-based models (e.g., bidirectional Long Short-Term Memory) and CRF. The deep learning-based models in the CRF-based methods are dedicated to estimating individual labels, whereas the relationships between connected labels are described as static numbers; thereby, it is not allowed to timely reflect the context in generating the most plausible label-label transitions for a given input sentence. Regardless, correctly segmenting entity mentions in biomedical texts is challenging because the biomedical terms are often descriptive and long compared with general terms. Therefore, limiting the label-label transitions as static numbers is a bottleneck in the performance improvement of BioNER. RESULTS:We introduce DTranNER, a novel CRF-based framework incorporating a deep learning-based label-label transition model into BioNER. DTranNER uses two separate deep learning-based networks: Unary-Network and Pairwise-Network. The former is to model the input for determining individual labels, and the latter is to explore the context of the input for describing the label-label transitions. We performed experiments on five benchmark BioNER corpora. Compared with current state-of-the-art methods, DTranNER achieves the best F1-score of 84.56% beyond 84.40% on the BioCreative II gene mention (BC2GM) corpus, the best F1-score of 91.99% beyond 91.41% on the BioCreative IV chemical and drug (BC4CHEMD) corpus, the best F1-score of 94.16% beyond 93.44% on the chemical NER, the best F1-score of 87.22% beyond 86.56% on the disease NER of the BioCreative V chemical disease relation (BC5CDR) corpus, and a near-best F1-score of 88.62% on the NCBI-Disease corpus. CONCLUSIONS:Our results indicate that the incorporation of the deep learning-based label-label transition model provides distinctive contextual clues to enhance BioNER over the static transition model. We demonstrate that the proposed framework enables the dynamic transition model to adaptively explore the contextual relations between adjacent labels in a fine-grained way. We expect that our study can be a stepping stone for further prosperity of biomedical literature mining.

Project description:Information regarding the physical interactions among proteins is crucial, since protein-protein interactions (PPIs) are central for many biological processes. The experimental techniques used to verify PPIs are vital for characterizing and assessing the reliability of the identified PPIs. A lot of information about PPIs and the experimental methods are only available in the text of the scientific publications that report them. In this study, we approach the problem of identifying passages with experimental methods for physical interactions between proteins as an information retrieval search task. The baseline system is based on query matching, where the queries are generated by utilizing the names (including synonyms) of the experimental methods in the Proteomics Standard Initiative-Molecular Interactions (PSI-MI) ontology. We propose two methods, where the baseline queries are expanded by including additional relevant terms. The first method is a supervised approach, where the most salient terms for each experimental method are obtained by using the term frequency-relevance frequency (tf.rf) metric over 13 articles from our manually annotated data set of 30 full text articles, which is made publicly available. On the other hand, the second method is an unsupervised approach, where the queries for each experimental method are expanded by using the word embeddings of the names of the experimental methods in the PSI-MI ontology. The word embeddings are obtained by utilizing a large unlabeled full text corpus. The proposed methods are evaluated on the test set consisting of 17 articles. Both methods obtain higher recall scores compared with the baseline, with a loss in precision. Besides higher recall, the word embeddings based approach achieves higher F-measure than the baseline and the tf.rf based methods. We also show that incorporating gene name and interaction keyword identification leads to improved precision and F-measure scores for all three evaluated methods. The tf.rf based approach was developed as part of our participation in the Collaborative Biocurator Assistant Task of the BioCreative V challenge assessment, whereas the word embeddings based approach is a novel contribution of this article.Database URL: https://github.com/ferhtaydn/biocemid/.

Project description:BACKGROUND:Automatic identification of term variants or acceptable alternative free-text terms for gene and protein names from the millions of biomedical publications is a challenging task. Ontologies, such as the Cardiovascular Disease Ontology (CVDO), capture domain knowledge in a computational form and can provide context for gene/protein names as written in the literature. This study investigates: 1) if word embeddings from Deep Learning algorithms can provide a list of term variants for a given gene/protein of interest; and 2) if biological knowledge from the CVDO can improve such a list without modifying the word embeddings created. METHODS:We have manually annotated 105 gene/protein names from 25 PubMed titles/abstracts and mapped them to 79 unique UniProtKB entries corresponding to gene and protein classes from the CVDO. Using more than 14 M PubMed articles (titles and available abstracts), word embeddings were generated with CBOW and Skip-gram. We setup two experiments for a synonym detection task, each with four raters, and 3672 pairs of terms (target term and candidate term) from the word embeddings created. For Experiment I, the target terms for 64 UniProtKB entries were those that appear in the titles/abstracts; Experiment II involves 63 UniProtKB entries and the target terms are a combination of terms from PubMed titles/abstracts with terms (i.e. increased context) from the CVDO protein class expressions and labels. RESULTS:In Experiment I, Skip-gram finds term variants (full and/or partial) for 89% of the 64 UniProtKB entries, while CBOW finds term variants for 67%. In Experiment II (with the aid of the CVDO), Skip-gram finds term variants for 95% of the 63 UniProtKB entries, while CBOW finds term variants for 78%. Combining the results of both experiments, Skip-gram finds term variants for 97% of the 79 UniProtKB entries, while CBOW finds term variants for 81%. CONCLUSIONS:This study shows performance improvements for both CBOW and Skip-gram on a gene/protein synonym detection task by adding knowledge formalised in the CVDO and without modifying the word embeddings created. Hence, the CVDO supplies context that is effective in inducing term variability for both CBOW and Skip-gram while reducing ambiguity. Skip-gram outperforms CBOW and finds more pertinent term variants for gene/protein names annotated from the scientific literature.

Dataset Information

Impact of Automatic Query Generation and Quality Recognition Using Deep Learning to Curate Evidence From Biomedical Literature: Empirical Study.

Publications

Impact of Automatic Query Generation and Quality Recognition Using Deep Learning to Curate Evidence From Biomedical Literature: Empirical Study.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets