I-Corps: Developing A 3D Total Body Imaging and Analysis System for Early Detection of Skin Cancer

I-Corps：开发用于早期检测皮肤癌的 3D 全身成像和分析系统

• 批准号: 2115095
• 负责人: Xianfeng Gu
• 金额: $ 5万
• 依托单位: SUNY at Stony Brook
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2021
• 资助国家: 美国
• 起止时间: 2021-03-15 至 2022-08-31
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=2115095&HistoricalAwards=false
• 关键词: Corps; Developing; 3D; Total; Body

The broader impact/commercial potential of this I-Corps project is the development of a high-speed and high-precision 3D imaging and skin analysis system. Skin cancer is the most common cancer in the US, with 5.5 million new cases diagnosed in 2019. Research shows that skin cancers are highly treatable if detected early. Early detection of melanoma saves lives and improves the treatment outcomes by reducing the risk of spreading cancer to other parts of the body. Early detection of nonmelanoma skin cancers minimizes disfigurement and enhances the quality of life and productivity for many patients. The proposed technology may be a new tool for dermatologists and make skin cancer screening more affordable by optimizing the skin evaluation process, supporting teledermatology services, and lowering the overall costs of care. In addition, there may be other applications of this 3D imaging technology in skin evaluation for cosmetic product development, planning and evaluation of orthodontic and cosmetic procedures, real-time monitoring of radiotherapy patients, facial expression and virtual character creation for movie and virtual reality game productions, and dynamic facial recognition for identity and property protection. This I-Corps project is based on the development of a high-speed, high-accuracy 3D scanner using multi-wavelength, phase-shifting structured light and an automated skin analysis software using computational conformal geometry. The proposed 3D scanner is designed to capture the human body’s geometry at high speed (up to 180 frames/second) and with high precision (depth resolution of 0.2mm). Visual identification of skin changes in 2D is impractical. When the viewing angle is altered, the individual’s posture varies, or the surface is deformed, image registration becomes challenging and unstable. The proposed solution is based on extensive research in computational conformal geometry. Due to the shape-preserving property of conformal mapping, 3D geometric analysis may be accomplished by processing 2D conformal images, which is easier and faster. Independent of their geometric or topological complexities, shapes in real physical worlds may be conformally transformed into one of three canonical shapes (i.e., unit sphere, Euclidean plane, or planar disk). Because of the uniformity, the design of all geometric algorithms can be simplified, resulting in dramatically improved efficiency, accuracy, and robustness. By combining high-performance 3D scanning and advanced conformal mapping techniques, this foundational platform technology may enable many applications.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

这个I-Corps项目更广泛的影响/商业潜力是开发高速和高精度的3D成像和皮肤分析系统。皮肤癌是美国最常见的癌症，2019年诊断出550万新发病例。研究表明，如果早期发现，皮肤癌是高度可治疗的。早期发现黑色素瘤可以挽救生命，并通过降低癌症扩散到身体其他部位的风险来改善治疗效果。非黑色素瘤皮肤癌的早期检测可以最大限度地减少毁容，提高许多患者的生活质量和生产力。这项技术可能是皮肤科医生的一种新工具，通过优化皮肤评估过程，支持远程皮肤病学服务，降低整体护理成本，使皮肤癌筛查更加经济实惠。此外，这种3D成像技术还可以在美容产品开发的皮肤评估、正畸和美容程序的规划和评估、放射治疗患者的实时监控、电影和虚拟现实游戏制作的面部表情和虚拟角色创建以及身份和财产保护的动态面部识别中应用。 这个I-Corps项目基于使用多波长相移结构光的高速、高精度3D扫描仪和使用计算保形几何的自动皮肤分析软件的开发。拟议的3D扫描仪旨在以高速（高达180帧/秒）和高精度（深度分辨率为0.2 mm）捕获人体几何形状。在2D中视觉识别皮肤变化是不切实际的。当视角改变、个人姿势改变或表面变形时，图像配准变得具有挑战性且不稳定。所提出的解决方案是基于广泛的研究计算保形几何。由于保形映射的保形特性，可以通过处理2D保形图像来完成3D几何分析，这更容易和更快。独立于它们的几何或拓扑复杂性，真实的物理世界中的形状可以共形地变换成三种规范形状之一（即，单位球面、欧几里得平面或平面圆盘）。由于一致性，所有几何算法的设计都可以简化，从而显着提高效率、准确性和鲁棒性。通过结合高性能3D扫描和先进的保角映射技术，这一基础平台技术可以实现许多应用。该奖项反映了NSF的法定使命，并通过使用基金会的知识价值和更广泛的影响审查标准进行评估，被认为值得支持。

项目成果

A geometric variational framework for computing optimal transportation maps, I

用于计算最佳交通地图的几何变分框架，I

• DOI: 10.4310/mcgd.2021.v1.n2.a3
• 发表时间: 2021
• 期刊: Computation and Geometry of Data
• 作者: An, Dongsheng;Lei, Na;Cui, Li;Su, Kehua;Xu, Xiaoyin;Luo, Feng;Gu, Xianfeng;Yau, Shing-Tung
• 通讯作者: Yau, Shing-Tung

An End-to-End Conditional Generative Adversarial Network Based on Depth Map for 3D Craniofacial Reconstruction

• DOI: 10.1145/3503161.3548254
• 发表时间: 2022-10
• 期刊: Proceedings of the 30th ACM International Conference on Multimedia
• 作者: Niankai Zhang;Junli Zhao;Fuqing Duan;Zhenkuan Pan;Zhongke Wu;Mingquan Zhou;Xianfeng Gu

End-to-End Evidential-Efficient Net for Radiomics Analysis of Brain MRI to Predict Oncogene Expression and Overall Survival

用于脑 MRI 放射组学分析的端到端循证网络，用于预测癌基因表达和总体生存率

• 发表时间: 2022
• 期刊: Medical Image Computing and Computer Assisted Intervention -- MICCAI 2022
• 作者: Feng, Yingjie;Wang, Jun;An, Dongsheng;Gu, Xianfeng;Xu, Xiaoyin;Zhang, Min
• 通讯作者: Zhang, Min

A novel GCN-based point cloud classification model robust to pose variances

• DOI: 10.1016/j.patcog.2021.108251
• 发表时间: 2022
• 期刊: Pattern Recognit.
• 作者: Huafeng Wang;Yaming Zhang;Wanquan Liu;X. Gu;Xin Jing;Zicheng Liu

Conformal topology optimization of multi-material ferromagnetic soft active structures using an extended level set method

• DOI: 10.1016/j.cma.2021.114394
• 发表时间: 2022-02
• 期刊: Computer Methods in Applied Mechanics and Engineering
• 影响因子: 7.2
• 作者: Jiawei Tian;Manqi Li;Zhong Han;Yong Chen;X. Gu;Q. Ge;Shikui Chen