RESUMEN
Our work focuses on tackling the problem of fine-grained recognition with incomplete multi-modal data, which is overlooked by previous work in the literature. It is desirable to not only capture fine-grained patterns of objects but also alleviate the challenges of missing modalities for such a practical problem. In this paper, we propose to leverage a meta-learning strategy to learn model abilities of both fast modal adaptation and more importantly missing modality completion across a variety of incomplete multi-modality learning tasks. Based on that, we develop a meta-completion method, termed as MECOM, to perform multimodal fusion and explicit missing modality completion by our proposals of cross-modal attention and decoupling reconstruction. To further improve fine-grained recognition accuracy, an additional partial stream (as a counterpart of the main stream of MECOM, i.e., holistic) and the part-level features (corresponding to fine-grained objects' parts) selection are designed, which are tailored for fine-grained nature to capture discriminative but subtle part-level patterns. Comprehensive experiments from quantitative and qualitative aspects, as well as various ablation studies, on two fine-grained multimodal datasets and one generic multimodal dataset show our superiority over competing methods. Our code is open-source and available at https://github.com/SEU-VIPGroup/MECOM.
RESUMEN
Objective: To investigate the risk factors of acute radiation-induced lung injury (acute RILI) induced by radiotherapy for esophageal cancer. Methods: A total of 206 patients with esophageal cancer who received radiotherapy in our hospital from January 2017 to March 2020 were selected. The general data such as gender, age, and comorbidities of the patients were collected, as well as the levels of cytokines (TNF-α, TNF-ß, and IL-6) in peripheral blood before radiotherapy; radiotherapy dose-related parameters were recorded during radiotherapy. Follow-up was 12 months after radiotherapy. The patients with induced acute RILI after radiotherapy were set as the observation group (n = 75). Patients without acute RILI after radiotherapy were set as the control group (n = 131). Univariate and multivariate logistic regression analysis was performed on the risk factors of acute RILI induced by radiotherapy for esophageal cancer. Results: Univariate analysis and multivariate logistic regression analysis showed that the combined diabetes, total radiation dose, combined lung disease, physical factors (V30, Dmean), and preradiotherapy cytokine (TNF-α, TNF-ß, and IL-6) elevated level was an independent risk factor for radiotherapy-induced acute RILI in esophageal cancer (P < 0.05). Conclusion: Concomitant diabetes, total radiation dose, lung disease, physical factors (V30, Dmean), and levels of cytokines (TNF-α, TNF-ß, and IL-6) before radiation therapy are risk factors for acute RILI induced by radiation therapy in esophageal cancer. The possibility of acute RILI should be comprehensively assessed according to the patient's condition, and the radiotherapy regimen should be adjusted to reduce and avoid the induction of acute radiation-induced lung injury.
