CMVFT:一种多尺度注意力引导框架用于增强多视角角膜地形图中的疑似圆锥角膜分类
本研究为回顾性横断面研究,共纳入573张角膜地形图,涵盖正常、疑似圆锥角膜和圆锥角膜眼。研究设计角膜多视角融合Transformer(CMVFT),整合七张标准角膜地形图特征,经预训练ResNet-50提取单视角表征后由多尺度注意力模块(MSAM)精炼,再经融合Transformer建模跨视角长程依赖。实验评估显示CMVFT能在属性重叠的特征空间中有效区分疑似病例,消融研究证实MSAM和融合Transformer对稳健多视角特征整合均不可或缺。
要点
- CMVFT整合七张标准角膜地形图,经MSAM和融合Transformer实现多视角特征融合。
- 实验评估显示该框架能有效区分疑似圆锥角膜病例。
- 消融研究证实MSAM与融合Transformer对多视角特征整合均不可或缺。
核心结论
- 本研究首次将Transformer驱动的多视角融合方法应用于角膜地形图分析,CMVFT在弥补小样本场景固有表征差距方面显示出识别疑似圆锥角膜病例并支持早期干预策略的潜力,并对早期临床干预具有前瞻性意义。
研究设计与样本
- 回顾性横断面研究,共纳入573张角膜地形图,代表正常、疑似或圆锥角膜眼。
- 设计角膜多视角融合Transformer(CMVFT),整合七张标准角膜地形图特征;预训练ResNet-50提取单视角表征,经自定义多尺度注意力模块(MSAM)精炼,再经融合Transformer建模跨视角长程依赖。
主要发现
- 实验评估表明CMVFT能在属性重叠的特征空间中有效区分疑似病例。
- 消融研究证实MSAM和融合Transformer对稳健多视角特征整合均不可或缺,成功弥补小数据集中的潜在表征不足。
临床怎么用
- 原文未据此给出推荐。
局限性
- 原文未报告局限性。
以上解读依据论文摘要(全文未开放获取或许可不允许翻译转载)。
原文
OBJECTIVE: To develop a multiview fusion framework that effectively identifies suspect keratoconus cases and facilitates the possibility of early clinical intervention. DESIGN: Retrospective cross-sectional study. SUBJECTS: A total of 573 corneal topography maps representing eyes classified as normal, suspect, or keratoconus. METHODS: We designed the corneal multiview fusion transformer (CMVFT), which integrates features from seven standard corneal topography maps. A pretrained ResNet-50 extracts single-view representations that are further refined by a custom-designed multiscale attention module (MSAM). This integrated design specifically compensates for the representation gap commonly encountered when applying Transformers to small-sample corneal topography datasets by dynamically bridging local convolution-based feature extraction with global self-attention mechanisms. A subsequent fusion Transformer then models long-range dependencies across views for comprehensive multiview feature integration. MAIN OUTCOME MEASURES: The primary measure was the framework's ability to differentiate suspect cases from normal and keratoconus cases, thereby creating a pathway for early clinical intervention. RESULTS: Experimental evaluation demonstrated that CMVFT effectively distinguishes suspect cases within a feature space characterized by overlapping attributes. Ablation studies confirmed that both the MSAM and the fusion Transformer are essential for robust multiview feature integration, successfully compensating for potential representation shortcomings in small datasets. CONCLUSIONS: This study is the first to apply a Transformer-driven multiview fusion approach in corneal topography analysis. By compensating for the representation gap inherent in small-sample settings, CMVFT shows promise in enabling the identification of suspect keratoconus cases and supporting early intervention strategies, with prospective implications for early clinical intervention.
引用
- Lu Y, Li B, Zhang Y, et al. CMVFT: A Multiscale Attention-Guided Framework for Enhanced Keratoconus Suspect Classification in Multiview Corneal Topography. Am J Ophthalmol. 2025. PMID: 40803643. doi:10.1016/j.ajo.2025.08.008
供执业医师工作参考,不替代临床判断;数值与分级以指南原文为准。