# [Presentation]HMILab Presentation **1. RadGraph: Extracting Clinical Entities and Relations from Radiology Reports** ![image](https://hackmd.io/_uploads/ryyjvtmB6.png) :::info **Author** : Saahil Jain, Ashwin Agrawal, Adriel Saporta, Pierre Chambon, Yuhao Zhang, Matthew P. Lungren, Andrew Y. Ng, Curtis P. Langlotz(Standford University) Steven QH Truong, Du Nguyen Duong, Tan Bui(VinBrain, VinUniversity) Pranav Rajpurkar(Harvard University) **Paper Link** : https://arxiv.org/abs/2106.14463 **reference Link** : https://physionet.org/content/radgraph/1.0.0/ ::: --- **2. Longitudinal Data and a Semantic Silimarity Reward for Chest X-ray Report Generation** ![image](https://hackmd.io/_uploads/B1jFtHWJ0.png) :::info **Author** : Aaron Nicolson, Jason Dowling, Bevan Koopman The Australian e-Health Research Centre, CSIRO Health and Biosecurity, Brisbane, Australia **Paper Link** : https://www.sciencedirect.com/science/article/pii/S0933365723001471 ::: - we propose a CXR report generator that integrates aspects of the radiolgoist workflow and is trained with our proposed reward for reinforcement learning - conditioning on multiple CXRs from a patient's study, and differentiating between report sections with section embeddings and separator tokens. --- **3. Evaluating progress in automatic chest X-ray radiology report generation** ![image](https://hackmd.io/_uploads/SkK2GuVr6.png) :::info **Author** : Feiyang Yu,Mark Endo,Rayan Krishnan,Ian Pan, Andy Tsai, Eduardo Pontes Reis, Eduardo Kaiser Ururahy Nunes Fonseca,4 Henrique Min Ho Lee,4 Zahra Shakeri Hossein Abad,5 Andrew Y. Ng,1 Curtis P. Langlotz,6 Vasantha Kumar Venugopal,7 and Pranav Rajpurkar8,10, Department of Computer Science, Stanford University, Stanford, CA 94305, USA Department of Radiology, Brigham and Women’s Hospital, Boston, MA 02115, USA Department of Radiology, Boston Children’s Hospital, Harvard Medical School, Boston, MA 02115, USA Cardiothoracic Radiology Group, Hospital Israelita Albert Einstein, Sa˜ o Paulo, Sa˜ o Paulo 05652, Brazil 5Dalla Lana School of Public Health, University of Toronto, Toronto, ON M5T 3M7, Canada AIMI Center, Stanford University, Stanford, CA 94304, USA CARPL.ai, New Delhi, Delhi 110016, India Department of Biomedical Informatics, Harvard Medical School, Boston, MA 02115, USA These authors contributed equally Lead contact **Paper Link** : https://www.cell.com/patterns/pdf/S2666-3899(23)00157-5.pdf ::: we analyze the failure modes of the metrics to understand their limitations and provide guidance for metric selection and interpretation. This study establishes RadGraph F1 and RadCliQ as meaningful metrics for guiding future research in radiology report generation. --- **4. UniXGen: A Unified Vision-Language Model for Multi-View Chest X-ray Generation and Report Generation** :::info **Published on 23 Feb 2023** Hyungyung Lee, Da Young Lee, Wonjae Kim, Jin-Hwa Kim, Tackeun Kim, Jihang Kim, Leonard Sunwoo, Edward Choi, KAIST, Deep-in-SIght Co., Naver AI Lab, Seoul National University Bundang Hospital **Paper Link** : https://arxiv.org/pdf/2302.12172.pdf ::: - we design a unified model for bidirectional chest X-ray and reports generation by adopting a vector quantization method to discretize chest X-rays into discrete visual tokens and formulating both tasks as sequence generation tasks. - we introduce several special tokens to generate chest X-rays with specific views that can be useful when the desired views are unavailable. Furthermore, UniXGen can flexibly take various inputs from single to multiple views to take advantage of the additional findings available in other X-ray views. **5. LLM-CXR : Instruction-Finetuned LLM for CXR Image Understanding and Generation** ![image](https://hackmd.io/_uploads/rJoT1UZyR.png) :::info **Published on 16 Jan 2024** Suhyeon Lee∗ , Won Jun Kim∗ , Jinho Chang & Jong Chul Ye Korea Advanced Institute of Science & Technology(KAIST) **Paper Link** : https://arxiv.org/pdf/2305.11490.pdf :::