2D/3D HYBRID TECHNOLOGY AND SOFTWARE SYSTEM FOR DETECTION OF ABNORMALITIES FROM C

描述（由申请人提供）：该项目的目标是开发一种新的2D/3D混合软件系统和技术，用于检测胶囊内窥镜图像的异常。将3D信息添加到2D胃肠道图像将帮助临床医生感知病变区域的大小和形状，从而帮助更准确的诊断。胶囊内窥镜是最近的趋势，并引起越来越多的制造商和医疗从业者的兴趣。然而，目前还没有开发出3D或2D/3D组合图像分析工具来进一步发挥该技术的优势。所提出的新型2D/3D混合技术的优点包括“相互增强的2D/3D算法能够更快，更可靠地分析胶囊图像；”使用改进的阴影方法和使用3D注册的3D飞越建模，从2D胶囊图像快速3D重建；自动分类和检测算法，将视频分割成不同的解剖部分，并在更短的时间内检测溃疡、出血和其他异常；”交互式图形用户界面和2D/3D可视化功能；模块化和开放的软件设计架构。据给定成像有限公司称，全球已有超过82万名患者使用他们的PillCam胶囊进行了胶囊内窥镜检查。胶囊内窥镜是一个快速增长的百万美元市场。在SBIR项目下开发的2D/3D混合软件系统将是第一个利用2D和3D信息快速准确诊断胶囊内窥镜图像异常的软件工具。拟议的系统和技术的商业化潜力是巨大的。公共卫生相关性：该SBIR项目的研究将产生一种令人兴奋的新型2D/3D混合技术和软件系统，可显着减少基于视频胶囊的内窥镜的评估时间并提高检测精度。提出的方法将有助于医生感知解剖标志，以及患者胃肠道病变、息肉等异常的颜色、形状和大小，更准确地诊断疾病的严重程度。作为第一家同时提供胶囊内窥镜成像二维和三维解决方案的公司，我们相信我们产品的市场潜力是巨大的。

DESCRIPTION (provided by applicant): The goal of this project is to develop a novel 2D/3D hybrid software system and technology for detection of abnormalities from capsule endoscopic images. Adding 3D information to 2D gastrointestinal images will help clinicians to perceive the size and shape of diseased areas aiding more accurate diagnosis. Capsule endoscopy is a recent trend and is generating growing interest among manufacturers and medical practitioners. However, no 3D or 2D/3D combined image analysis tools have been developed to further the advantages of this technique. The advantages of the proposed novel 2D/3D hybrid technology include " Mutually augmented 2D/3D algorithms enabling faster and more reliable analysis of capsule images; " Rapid 3D reconstruction from 2D capsule images using improved shape from shading approach and 3D fly-through modeling using 3D registration; " Automated classification and detection algorithms to segment the video into different anatomic sections and detect ulcers, bleeding, and other abnormalities in less time; " Interactive graphical user interface and 2D/3D visualization capability; " Modular and open software design architecture. According to Given Imaging Ltd., more than 820,000 patients worldwide have gone through capsule endoscopy procedure using their PillCam capsules. Capsule endoscopy is a rapidly-growing million-dollar market. The 2D/3D hybrid software system developed under this SBIR project will be the first software tool that utilizes both 2D and 3D information to rapidly and accurately diagnose abnormalities from capsule endoscopic images. The commercialization potential of the proposed system and technology is significant. PUBLIC HEALTH RELEVANCE: The research of this SBIR project will result in an exciting new 2D/3D hybrid technique and software system that significantly reduces the assessment time and increases the detection accuracy in video capsule based endoscopy. The proposed approach will help doctors to perceive anatomical landmarks and the color, shape and size of lesions, polyps and other abnormalities in patient's gastrointestinal tract and to more accurately diagnose the severity of the disease. As the first company to provide both 2D and 3D solutions for capsule endoscopic imaging, we believe that the market potential of our product is significant.