Feature engineering vs. deep learning for paper section identification: Toward applications in Chinese medical literature

Zhou, Sijia; Li, Xin

doi:10.1016/j.ipm.2020.102206

Cited by 19 publications

(13 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Several studies have been conducted for classifying radiologic reports as positive or negative for a given disease [1,10,31,32] or for classifying various diagnoses from medical records written in Chinese [13]. Most of the studies used a CNN-based model and showed a better performance than did our model [1,10,31,32].…”

Section: Discussionmentioning

confidence: 91%

“…Especially in deep learning, convolutional neural network (CNN)-based models have shown significant accuracy in extracting pulmonary embolism [10] and pulmonary infection from medical reports [1]. The model can be used to classify diagnosis from whole medical records even when they are written in the Chinese language [13], and a recurrent neural network-based model has been used for classifying stroke and identifying its location [14]. However, the use of bilingual clinical reports is common for EHRs in non-English-speaking countries.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

An Attention Model With Transfer Embeddings to Classify Pneumonia-Related Bilingual Imaging Reports: Algorithm Development and Validation

et al. 2021

View full text Add to dashboard Cite

Background In the analysis of electronic health records, proper labeling of outcomes is mandatory. To obtain proper information from radiologic reports, several studies were conducted to classify radiologic reports using deep learning. However, the classification of pneumonia in bilingual radiologic reports has not been conducted previously. Objective The aim of this research was to classify radiologic reports into pneumonia or no pneumonia using a deep learning method. Methods A data set of radiology reports for chest computed tomography and chest x-rays of surgical patients from January 2008 to January 2018 in the Asan Medical Center in Korea was retrospectively analyzed. The classification performance of our long short-term memory (LSTM)–Attention model was compared with various deep learning and machine learning methods. The area under the receiver operating characteristic curve (AUROC), area under the precision-recall curve, sensitivity, specificity, accuracy, and F1 score for the models were compared. Results A total of 5450 radiologic reports were included that contained at least one pneumonia-related word. In the test set (n=1090), our proposed model showed 91.01% (992/1090) accuracy (AUROCs for negative, positive, and obscure were 0.98, 0.97, and 0.90, respectively). The top 3 performances of the models were based on FastText or LSTM. The convolutional neural network–based model showed a lower accuracy 73.03% (796/1090) than the other 2 algorithms. The classification of negative results had an F1 score of 0.96, whereas the classification of positive and uncertain results showed a lower performance (positive F1 score 0.83; uncertain F1 score 0.62). In the extra-validation set, our model showed 80.0% (642/803) accuracy (AUROCs for negative, positive, and obscure were 0.92, 0.96, and 0.84, respectively). Conclusions Our method showed excellent performance in classifying pneumonia in bilingual radiologic reports. The method could enrich the research on pneumonia by obtaining exact outcomes from electronic health data.

show abstract

Section: Discussionmentioning

confidence: 91%

Section: Introductionmentioning

confidence: 99%

An Attention Model With Transfer Embeddings to Classify Pneumonia-Related Bilingual Imaging Reports: Algorithm Development and Validation

et al. 2021

View full text Add to dashboard Cite

show abstract

“…However, given the diversity of publication venues, submission guidelines, and even different authors' writing preferences, it is estimated that only around 25% of published biomedical articles are structured and the used structure scheme varies [11,12,22]. As a result, several approaches have been proposed throughout the years to segment article abstracts into coherent sections [1,22,24,36]. In this paper, we adopt a standard classification approach [24] for identifying article sectionswhenever section information is missing -and instead focus on how to leverage this structure for effective retrieval.…”

Section: Structured Fine-tuningmentioning

confidence: 99%

Structured Fine-Tuning of Contextual Embeddings for Effective Biomedical Retrieval

Ueda

Santos

Macdonald

et al. 2021

Proceedings of the 44th International ACM SIGIR Conference on Research and Development in Information Retrieval

View full text Add to dashboard Cite

Biomedical literature retrieval has greatly benefited from recent advances in neural language modeling. In particular, fine-tuning pretrained contextual language models has shown impressive results in recent biomedical retrieval evaluation campaigns. Nevertheless, current approaches neglect the inherent structure available from biomedical abstracts, which are (often explicitly) organised into semantically coherent sections such as background, methods, results, and conclusions. In this paper, we investigate the suitability of leveraging biomedical abstract sections for fine-tuning pretrained contextual language models at a finer granularity. Our results on two TREC biomedical test collections demonstrate the effectiveness of the proposed structured fine-tuning regime in contrast to a standard fine-tuning that does not leverage structure. Through an ablation study, we show that models fine-tuned on individual sections are able to capture potentially useful word contexts that may be otherwise ignored by structure-agnostic models. CCS CONCEPTS• Applied computing → Bioinformatics; • Information systems → Retrieval models and ranking.

show abstract

Section: Introductionmentioning

confidence: 99%

“…When CAD is used to assist diagnosis, effective feature engineering can be realized with the help of doctors, which makes it possible for some classical machine learning methods with better understanding to achieve better performance than deep learning models [ 10 ]. Appropriate features can be obtained through feature selection algorithms [ 11 ], selection methods based on physician experience [ 10 ], or other methods. On the other hand, many models based on deep neural networks may hinder the efficiency of the interaction between doctors and the system due to the incomprehensible nature of its decision-making process [ 12 , 13 ], while highly complex models are also not conducive to the physician’s adjustment to reduce diagnostic bias [ 14 , 15 ].…”

Section: Introductionmentioning

confidence: 99%

An Adaptive Deep Ensemble Learning Method for Dynamic Evolving Diagnostic Task Scenarios

et al. 2021

Diagnostics

View full text Add to dashboard Cite

Increasingly, machine learning methods have been applied to aid in diagnosis with good results. However, some complex models can confuse physicians because they are difficult to understand, while data differences across diagnostic tasks and institutions can cause model performance fluctuations. To address this challenge, we combined the Deep Ensemble Model (DEM) and tree-structured Parzen Estimator (TPE) and proposed an adaptive deep ensemble learning method (TPE-DEM) for dynamic evolving diagnostic task scenarios. Different from previous research that focuses on achieving better performance with a fixed structure model, our proposed model uses TPE to efficiently aggregate simple models more easily understood by physicians and require less training data. In addition, our proposed model can choose the optimal number of layers for the model and the type and number of basic learners to achieve the best performance in different diagnostic task scenarios based on the data distribution and characteristics of the current diagnostic task. We tested our model on one dataset constructed with a partner hospital and five UCI public datasets with different characteristics and volumes based on various diagnostic tasks. Our performance evaluation results show that our proposed model outperforms other baseline models on different datasets. Our study provides a novel approach for simple and understandable machine learning models in tasks with variable datasets and feature sets, and the findings have important implications for the application of machine learning models in computer-aided diagnosis.

show abstract

Feature engineering vs. deep learning for paper section identification: Toward applications in Chinese medical literature

Cited by 19 publications

References 12 publications

An Attention Model With Transfer Embeddings to Classify Pneumonia-Related Bilingual Imaging Reports: Algorithm Development and Validation

An Attention Model With Transfer Embeddings to Classify Pneumonia-Related Bilingual Imaging Reports: Algorithm Development and Validation

Structured Fine-Tuning of Contextual Embeddings for Effective Biomedical Retrieval

An Adaptive Deep Ensemble Learning Method for Dynamic Evolving Diagnostic Task Scenarios

Contact Info

Product

Resources

About