论文标题
通过利用温暖的开始来改善胸部X射线报告的生成
Improving Chest X-Ray Report Generation by Leveraging Warm Starting
论文作者
论文摘要
自动从患者的胸部X射线（CXR）中生成报告是减少临床工作量和改善患者护理的有前途解决方案。但是，当前的CXR报告生成器（主要是编码器到二十个模型）缺乏在临床环境中部署的诊断准确性。为了改善CXR报告的生成，我们研究了使用最近开源的计算机视觉和自然语言处理检查点（例如Vision Transformer（VIT）和PubMedbert）的热量开始编码器和解码器。为此，在MIMIC-CXR和IU X射线数据集上评估了每个检查点。我们的实验研究表明，卷积视觉变压器（CVT）Imagenet-21K和蒸馏性生成的预训练的变压器2（DistilGPT2）检查点分别最适合温暖启动编码器和解码器。与最先进的（$ \ MATHCAL {M}^2 $ Transformer Progressive）相比，CVT2DISTILGPT2的CE F-1的提高了8.3 \％，BLEU-4，1.6 \％的Rouge-f in Rouge-f in Rouge-l和1.0 \％\％的提高为1.0 \％。 CVT2DISTILGPT2产生的报告与放射科医生报告的相似性比以前的方法更高。这表明利用温暖的开始可以改善CXR报告的生成。 CVT2DISTILGPT2的代码和检查点可在https://github.com/aehrc/cvt2distilgpt2上找到。
Automatically generating a report from a patient's Chest X-Rays (CXRs) is a promising solution to reducing clinical workload and improving patient care. However, current CXR report generators -- which are predominantly encoder-to-decoder models -- lack the diagnostic accuracy to be deployed in a clinical setting. To improve CXR report generation, we investigate warm starting the encoder and decoder with recent open-source computer vision and natural language processing checkpoints, such as the Vision Transformer (ViT) and PubMedBERT. To this end, each checkpoint is evaluated on the MIMIC-CXR and IU X-Ray datasets. Our experimental investigation demonstrates that the Convolutional vision Transformer (CvT) ImageNet-21K and the Distilled Generative Pre-trained Transformer 2 (DistilGPT2) checkpoints are best for warm starting the encoder and decoder, respectively. Compared to the state-of-the-art ($\mathcal{M}^2$ Transformer Progressive), CvT2DistilGPT2 attained an improvement of 8.3\% for CE F-1, 1.8\% for BLEU-4, 1.6\% for ROUGE-L, and 1.0\% for METEOR. The reports generated by CvT2DistilGPT2 have a higher similarity to radiologist reports than previous approaches. This indicates that leveraging warm starting improves CXR report generation. Code and checkpoints for CvT2DistilGPT2 are available at https://github.com/aehrc/cvt2distilgpt2.
