Papers by Kaveri Kale
KGVL-BART: Knowledge Graph Augmented Visual Language BART for Radiology Report Generation (2023.eacl-main)
Copied to clipboard
| Challenge: | Timely generation of radiology reports and diagnoses is a challenge worldwide due to the enormous number of cases and shortage of radiologists. |
| Approach: | They propose a Knowledge Graph Augmented Vision Language BART model that takes two chest X-ray images and outputs a report with patient-specific findings. |
| Outcome: | The proposed model outperforms state-of-the-art transformer-based models on scoring metrics. |
“Knowledge is Power”: Constructing Knowledge Graph of Abdominal Organs and Using Them for Automatic Radiology Report Generation (2023.acl-industry)
Copied to clipboard
Kaveri Kale, Pushpak Bhattacharyya, Aditya Shetty, Milind Gune, Kush Shrivastava, Rustom Lawyer, Spriha Biswas
| Challenge: | conventional radiology workflows involve dictating diagnosis to transcriptionists, which is prone to delay and error. |
| Approach: | They propose to generate a set of knowledge graphs from a large collection of free-text radiology reports and use them to generate automatic radiology report generation. |
| Outcome: | The proposed model improves the reported BLEU-3, ROUGE-L, METEOR, and CIDEr scores by 2%, 4%, 2% and 2% respectively. |
Replace and Report: NLP Assisted Radiology Report Generation (2023.findings-acl)
Copied to clipboard
| Challenge: | Clinical practice frequently uses medical imaging for diagnosis and treatment. |
| Approach: | They propose a template-based approach to generate radiology reports from radiographs . they use multilabel image classifiers to generate tags, pathological descriptions from tags . |
| Outcome: | The proposed method improves on the most popular radiology report datasets. |