量子卷积框架SQCDL-WIoT筛查乳腺癌准确率98.14%：《科学报告》在BreaKHis数据集40倍放大下曲线下面积0.9891，11月13日刊出
《科学报告》2026年11月13日刊出一项研究，提出一个名为SQCDL-WIoT的可扩展框架，把可穿戴物联网与量子卷积深度学习结合起来作多类别乳腺癌筛查。框架包含两个各有5个量子比特的变分量子电路，可嵌入ResNet-50骨干网络。作者为Ramudu、Saturi与Ramadevi等人，论文数字对象标识为10.1038/s41598-026-73447-1，2026年5月4日收稿、9月23日接受、11月13日刊出。
研究在BreaKHis数据集上作分层五折交叉验证，并在CBIS-DDSM数据集上作零样本迁移学习。期刊摘要报告，在BreaKHis数据集40倍放大倍率下，SQCDL-WIoT在与全部基线方法的比较中取得最高准确率98.14%，宏平均敏感度为97.63%，宏平均特异度为98.87%，曲线下面积为0.9891。
作者说明，这项研究只使用了公开的去标识化数据集，即BreaKHis与CBIS-DDSM，以及完全合成的模拟可穿戴物联网与临床元数据，没有招募人类受试者，也没有收集或分析可识别的患者信息。
（来源：《科学报告》）
The SQCDL-WIoT Quantum Convolutional Framework Screens Breast Cancer at 98.14 Percent Accuracy: An AUC of 0.9891 at 40x Magnification on BreaKHis in Scientific Reports, Published November 13
A study published in Scientific Reports on November 13, 2026, proposes SQCDL-WIoT, a scalable, wearable-based internet of things enabled quantum convolutional deep learning framework capable of multi-class breast cancer screening. The proposed method includes two variational quantum circuits with five qubits each that can be embedded into a ResNet-50 backbone. The authors are Ramudu, Saturi, Ramadevi and colleagues, and the paper carries digital object identifier 10.1038/s41598-026-73447-1, was received on May 4, 2026, accepted on September 23, 2026, and published on November 13, 2026. Using stratified five-fold cross-validation on the BreaKHis dataset and zero-shot transfer learning on the CBIS-DDSM dataset, SQCDL-WIoT demonstrated the highest accuracy of 98.14 percent, with macro-sensitivity of 97.63 percent, macro-specificity of 98.87 percent, and an area under the curve of 0.9891, when compared to all other baseline methods at 40x magnification on BreaKHis, according to the journal abstract. The authors stated that the study exclusively used publicly available de-identified datasets, namely BreaKHis and CBIS-DDSM, together with fully synthetic simulated wearable IoT and clinical metadata, that no human participants were recruited, and that no identifiable patient information was collected or analyzed.
(Source: Scientific)
来源：《科学报告》
发布：GSM·美国（发布链接）
图片来源：图：一名女性正在接受乳腺X光检查、技师在旁操作设备。National Cancer Institute，公共领域，经 Wikimedia Commons。 / Photo: National Cancer Institute, public domain, via Wikimedia Commons.
GSM Global · 本篇英文版 English edition of this story
