Longer Version for "Deep Context-Encoding Network for Retinal Image Captioning"
This work addresses the problem of reducing workload for ophthalmologists by automating report generation, though it appears incremental as it builds on existing methods with specific enhancements.
The authors tackled the problem of automatically generating medical reports for retinal images by proposing a context-driven encoding network, which achieved state-of-the-art performance with improvements of +16% in BLEU-avg, +10.2% in CIDEr, and +8.6% in ROUGE over baseline models.
Automatically generating medical reports for retinal images is one of the promising ways to help ophthalmologists reduce their workload and improve work efficiency. In this work, we propose a new context-driven encoding network to automatically generate medical reports for retinal images. The proposed model is mainly composed of a multi-modal input encoder and a fused-feature decoder. Our experimental results show that our proposed method is capable of effectively leveraging the interactive information between the input image and context, i.e., keywords in our case. The proposed method creates more accurate and meaningful reports for retinal images than baseline models and achieves state-of-the-art performance. This performance is shown in several commonly used metrics for the medical report generation task: BLEU-avg (+16%), CIDEr (+10.2%), and ROUGE (+8.6%).