The automatic Brain CT reports generation can improve the efficiency and accuracy of diagnosing cranial diseases. However, current methods are limited by 1) coarse-grained supervision: the training data in image-text format lacks detailed supervision for recognizing subtle abnormalities, and 2) coupled cross-modal alignment: visual-textual alignment may be inevitably coupled in a coarse-grained manner, resulting in tangled feature representation for report generation. In this paper, we propose a novel Pathological Graph-driven Cross-modal Alignment (PGCA) model for accurate and robust Brain CT report generation. Our approach effectively decouples the cross-modal alignment by constructing a Pathological Graph to learn fine-grained visual cues and align them with textual words. This graph comprises heterogeneous nodes representing essential pathological attributes (i.e., tissue and lesion) connected by intra- and inter-attribute edges with prior domain knowledge. Through carefully designed graph embedding and updating modules, our model refines the visual features of subtle tissues and lesions and aligns them with textual words using contrastive learning. Extensive experimental results confirm the viability of our method. We believe that our PGCA model holds the potential to greatly enhance the automatic generation of Brain CT reports and ultimately contribute to improved cranial disease diagnosis.

Medical report automatic generation has gained increasing interest recently as a way to help radiologists write reports more efficiently. However, this image-to-text task is rather challenging due to the typical data biases: 1) Normal physiological structures dominate the images, with only tiny abnormalities; 2) Normal descriptions accordingly dominate the reports. Existing methods have attempted to solve these problems, but they neglect to exploit useful information from similar historical cases. In this paper, we propose a novel Cross-modal Contrastive Attention (CMCA) model to capture both visual and semantic information from similar cases, with mainly two modules: a Visual Contrastive Attention Module for refining the unique abnormal regions compared to the retrieved case images; a Cross-modal Attention Module for matching the positive semantic information from the case reports. Extensive experiments on two widely-used benchmarks, IU X-Ray and MIMIC-CXR, demonstrate that the proposed model outperforms the state-of-the-art methods on almost all metrics. Further analyses also validate that our proposed model is able to improve the reports with more accurate abnormal findings and richer descriptions.