Vision-knowledge fusion model for multi-domain medical report generation

Article Computer Science, Artificial Intelligence

Knowledge matters: Chest radiology report generation with general and specific knowledge

Shuxin Yang et al.

Summary: Automatic chest radiology report generation is crucial in clinics for reducing the workload of radiologists and preventing misdiagnosis. This paper proposes a knowledge-enhanced approach that utilizes both general and specific medical knowledge to improve the quality of generated reports. Experimental results show that this approach outperforms state-of-the-art methods in terms of performance.

MEDICAL IMAGE ANALYSIS (2022)

Add to Collection

Proceedings Paper Computer Science, Artificial Intelligence

Cross-modal Clinical Graph Transformer for Ophthalmic Report Generation

Mingjie Li et al.

Summary: Automatic generation of ophthalmic reports using data-driven neural networks has the potential to improve clinical practice. However, existing methods neglect the prior clinical knowledge of ophthalmologists. To address this issue, a Cross-modal clinical Graph Transformer (CGT) is proposed for ophthalmic report generation, which incorporates clinical relation triples into visual features to improve the decoding process. By extracting clinical entities and relations from training reports and limiting the impact of knowledge using a visible matrix, the proposed CGT outperforms previous benchmark methods and achieves state-of-the-art performance.

2022 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR 2022) (2022)

Add to Collection

Article Computer Science, Artificial Intelligence

A Survey of the Usages of Deep Learning for Natural Language Processing

Daniel W. Otter et al.

Summary: This article provides a brief introduction to the recent developments in natural language processing driven by deep learning models, discusses their applications and research findings, and offers recommendations for future research directions.

IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS (2021)

Add to Collection

Article Computer Science, Artificial Intelligence

Medical-VLBERT: Medical Visual Language BERT for COVID-19 CT Report Generation With Alternate Learning

Guangyi Liu et al.

Summary: This article proposes using the Medical-VLBERT model for identifying abnormalities in COVID-19 scans and generating medical reports automatically. The model adopts a learning strategy of knowledge pretraining and transferring, achieving state-of-the-art performances on terminology prediction and report generation.

IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS (2021)

Add to Collection

Article Biochemical Research Methods