Multi-Attention and Incorporating Background Information Model for Chest X-Ray Image Report Generation

Huang, Xin; Yan, Fengqi; Xu, Wei; Li, Maozhen

doi:10.1109/access.2019.2947134

Cited by 47 publications

(48 citation statements)

References 31 publications

(34 reference statements)

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Input: data to be read X (1) begin (2) if X is in A2 (3) Move X to the top of A2 (4) else if X is in |A1 out| (5) Delete X from |A1 out| and move it to the top of A2 ( 6) else if X is in |A1 in| (7) Do nothing (8) else//x is not in the cache (9) Request X to disk or network (10) end if (11) end ALGORITHM 2…”

Section: Small File Merge Result Tablementioning

confidence: 99%

Section: Mathematical Problems In Engineeringmentioning

confidence: 99%

“…e electronization, informatization, and intelligentization of hospitals can effectively improve the efficiency of the medical system. Massive electronic medical imaging data provide powerful data support for intelligent auxiliary diagnosis algorithms and also pose new challenges for multisource complex data storage and access [1][2][3][4]. In the new application scenario, doctors use electronic images to replace the traditional film for diagnosis, and patients can view the examination images at any time through various electronic methods (WeChat and AliPay).…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Hadoop-Based Medical Image Storage and Access Method for Examination Series

Huang

Wang³

et al. 2021

Mathematical Problems in Engineering

Self Cite

View full text Add to dashboard Cite

Under the background of electronic medical data, doctors use electronic images to replace the traditional film for diagnosis, and patients can view examination images at any time through various electronic means. The storage and frequent reading of massive data bring new challenges. Given the characteristics of the size and quantity of image files generated by different examination types, different merging strategies are proposed to improve the storage performance of the files; according to the characteristics of medical data with examination as the basic unit, a two-level model combined with medical imaging information is proposed. The indexing mechanism solves the problem that SEQ files cannot be read randomly without an index; given the time characteristics of data access, an improved 2Q algorithm is proposed to cache the prefetched files and the read files in different cache queues, which improves the efficiency of file reading. In the experimental comparison, the proposed algorithm surpasses the baseline method in storage and access performance.

show abstract

Section: Small File Merge Result Tablementioning

confidence: 99%

Section: Mathematical Problems In Engineeringmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Hadoop-Based Medical Image Storage and Access Method for Examination Series

Huang

Wang³

et al. 2021

Mathematical Problems in Engineering

Self Cite

View full text Add to dashboard Cite

show abstract

“…Given the ground truth medical report provided by the radiologists for the input chest X-ray image, existing methods train the encoder-decoder frameworks by minimizing training loss, e.g., crossentropy loss. Due to limited space, please refer to Huang et al (2019); Jing et al (2018) for detailed introduction.…”

Section: Problem Formulationmentioning

confidence: 99%

Contrastive Attention for Automatic Chest X-ray Report Generation

Liu¹,

Yin²,

Wang³

et al. 2021

Findings of the Association for Computational Linguistics: ACL-IJCNLP 2021

View full text Add to dashboard Cite

Recently, chest X-ray report generation, which aims to automatically generate descriptions of given chest X-ray images, has received growing research interests. The key challenge of chest X-ray report generation is to accurately capture and describe the abnormal regions. In most cases, the normal regions dominate the entire chest X-ray image, and the corresponding descriptions of these normal regions dominate the final report. Due to such data bias, learning-based models may fail to attend to abnormal regions. In this work, to effectively capture and describe abnormal regions, we propose the Contrastive Attention (CA) model. Instead of solely focusing on the current input image, the CA model compares the current input image with normal images to distill the contrastive information. The acquired contrastive information can better represent the visual features of abnormal regions. According to the experiments on the public IU-X-ray and MIMIC-CXR datasets, incorporating our CA into several existing models can boost their performance across most metrics. In addition, according to the analysis, the CA model can help existing models better attend to the abnormal regions and provide more accurate descriptions which are crucial for an interpretable diagnosis. Specifically, we achieve the state-ofthe-art results on the two public datasets.

show abstract

“…The model showed promising results according to BLEU, METEOR, ROUGE, and CIDER scores. Other authors in [22] developed a deep learning model for generating chest X-ray imaging reports. The model relied on a multi-attention and bidirectional LSTM to encode images and generate sentences.…”

Section: B Medical Text Generationmentioning

confidence: 99%

A Predictive Text System for Medical Recommendations in Telemedicine: A Deep Learning Approach in the Arabic Context

Habib¹,

Faris²,

Qaddoura

et al. 2021

IEEE Access

View full text Add to dashboard Cite

We are currently witnessing an immense proliferation of natural language processing (NLP) applications. Natural language generation (NLG) has emerged from NLP and is now commonly utilized in various applications, including chatting applications. The objective of this paper is to propose a deep learningbased language generation model that simplifies the process of writing medical recommendations for doctors in an Arabic context, to improve service satisfaction and patient-doctor interactions. The developed language generation model is a predictive text system intended for next word prediction in a telemedicine service. Altibbi a -a digital platform for telemedicine and teleconsultations services in the Middle East and the North Africa (MENA) region-was utilized as a case study for the textual prediction process. The proposed model was trained using data obtained from Altibbi databases related to medical recommendations, particularly gynecology, dermatology, psychiatric diseases, urology, and internist diseases. Variants of deep learning models were implemented and optimized for next word prediction, based on the unidirectional and bidirectional long short-term memory (LSTM and BiLSTM), the one-dimensional convolutional neural network (CONV1D), and a combination of LSTM and CONV1D (LSTM-CONV1D). The algorithms were trained using two versions of the datasets (i.e., 3-gram and 4gram representations) and evaluated in terms of their training accuracy and loss, validation accuracy and loss, and testing accuracy per their matching scores. The proposed models' performances were comparable. CONV1D produced the most promising matching score.

show abstract

Multi-Attention and Incorporating Background Information Model for Chest X-Ray Image Report Generation

Cited by 47 publications

References 31 publications

Hadoop-Based Medical Image Storage and Access Method for Examination Series

Hadoop-Based Medical Image Storage and Access Method for Examination Series

Contrastive Attention for Automatic Chest X-ray Report Generation

A Predictive Text System for Medical Recommendations in Telemedicine: A Deep Learning Approach in the Arabic Context

Contact Info

Product

Resources

About