用三视角皮肤微结构分析，自动判断手机拍的皮肤照能否诊断。
Beyond Blur: A Semantic Tri-view Pipeline for Teledermatology Gradability via Skin Micro-relief
- 通过三视角图像分析皮肤微结构，量化图像质量。
- 在SCIN数据集上达到0.96 AUC和97.7%阳性预测值。
- 适合需要实时筛查不合格照片的远程皮肤科应用。
智能手机皮肤照片在远程皮肤科中不可或缺，但评估其诊断可用性（可分级性）仍是移动医疗流程中的关键瓶颈。皮肤病专家通常需查看多个拍摄视角（局部、斜角、特写）以识别一致的纹理细节，而非依赖单张图像。本文提出语义三视角管道，一种可解释的自动化远程皮肤科可分级性筛查架构，将表皮微结构形式化为可计算的图像质量生物标志物。基于公开的SCIN数据集中的专家标注子集，我们训练了一个轻量级DeepLabV3+模型以分割微结构保真度。这些空间掩码随后通过逻辑回归分类器在最多三个视角间聚合，利用视角冗余提升在非受控手机拍摄条件下的鲁棒性。该方法学习上下文感知、临床可理解的启发式规则，例如在远距离局部视图中惩罚高保真纹理。在预设90%敏感度操作点下，系统明显错误主要反映边界病例中临床主观差异，此时医生依赖非视觉元数据。在SCIN数据集上，性能从多数共识的方差密集型病例的AUC 0.81（PPV 80.6%）提升至光学清晰的一致性病例的AUC 0.96（PPV 97.7%）。总体而言，本工作实现了一种可解释、隐私优先、边缘就绪的系统，可在病例提交时提供实时反馈，提前过滤不可分级的照片集。
原文摘要 · Abstract (English)
Smartphone skin photographs are indispensable to teledermatology, yet assessing the diagnostic suitability of submitted cases (gradability) remains a critical bottleneck in mobile care workflows. Dermatologists routinely review multiple photographic views (regional, angled, and close-up) to identify consistent textural detail rather than relying on a single image. We present the Semantic Tri-view Pipeline, an interpretable architecture for automated teledermatology gradability screening that formalizes epidermal micro-relief as a computable biomarker of image quality. Using an expert-annotated subset of the public SCIN dataset, we train a lightweight DeepLabV3+ model to segment micro-relief fidelity. These spatial masks are then aggregated across up to three case views with a logistic regression classifier, leveraging viewpoint redundancy to support robustness under uncontrolled smartphone acquisition. This approach learns context-aware, clinically intelligible heuristics, such as penalizing high-fidelity texture in regional distance views. Evaluated at a predefined 90% sensitivity operating point, the system's apparent errors largely reflect subjective clinical variance on borderline cases where clinicians rely on non-visual metadata. On SCIN, performance improves from an AUC of 0.81 (80.6% PPV) on variance-heavy majority-consensus cases to 0.96 (97.7% PPV) on optically unambiguous unanimous cases. Overall, this work delivers an interpretable, privacy-by-design, edge-ready system that can provide real-time feedback during case submission to filter ungradable photo sets before review.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。