Track papers across labs, topics, and trends.
Search papers, labs, and topics across Lattice.
To resolve the severe morphological ambiguity in radiograph-based bone tumor diagnosis, this paper establishes a dual-input multi-task architecture that bridges YOLO-derived lesion crops with full-field radiographs via bidirectional cross-modal attention. Jointly optimizing for lesion segmentation and subtype classification allows fine-grained margin boundaries to be dynamically contextualized by surrounding skeletal anatomy. Evaluated on the 3,746-patient BTXRD benchmark, the approach attains a 0.896 Dice similarity coefficient, a 0.928 macro-averaged F1-score, and a 0.999 AUC for malignant osteosarcomas.
Bone tumor classification reaches a 0.999 osteosarcoma AUC by simply forcing local lesion crops and full radiographs to bidirectionally attend to one another during joint multi-task learning.
Primary bone tumors are rare but clinically aggressive neoplasms whose diagnosis from radiographs is challenged by heterogeneous morphology, subtle lesion margins, and overlapping bone structures. To address the limitations of existing single-view models, we present a dual-input, multi-task learning framework that, to our knowledge, is the first to apply bidirectional cross-modal attention between a lesion crop and the full radiograph for joint segmentation and subtype classification. Using the multi-institutional Bone Tumor X-ray Radiograph Dataset (BTXRD, n=3,746), we employ a YOLO-based detector to generate regions of interest, which are paired with full images as inputs to a dual-stream DenseNet121 architecture. Features are integrated via a novel cross-modal attention fusion strategy, refined by Hierarchical Multi-scale Feature Fusion, effectively balancing fine-grained lesion detail with global anatomical context. Evaluated on a held-out patient-level test split, the model demonstrates superior performance over single-input baselines, achieving an overall Dice Similarity Coefficient of 0.896 and a macro-averaged classification F1-score of 0.928. Notably, the system exhibits exceptional sensitivity for malignant osteosarcoma (AUC 0.999), validating the potential of dual-stream context modeling to support radiologists in accurate, early decision-making.