Sift-生物力学数据分析软件

Sift 专为处理大型生物力学数据集的用户量身定制，是无需导出即可实现无缝数据分析的解决方案。同时加载多个 CMZ 库，简化数据管理，并轻松征服笨拙的数据库。使用各种可视化工具、尖部分析技术以及将 Visual3D 引擎无缝集成到这个独立的应用程序中，直接深入研究您的数据集。从整合 Visual3D 管道指令到执行统计参数映射、主成分分析或异常值验测算法，Sift 使您能够在单个应用程序中完成全方面的生物力学分析。立即体验 Sift 的强大功能，释放数据的真正潜力！

Sift 允许用户：加载多个 C3D 文件，手动或通过统计算法验测和删除异常值，根据独特条件对信号进行分组，通过内置的主成分分析、曲线配准、统计参数映射和步态测量完成高 级分析，并生成定制的可视化结果。

Sift 旨在与 Visual3D 无缝集成。Visual3D 主要是一种基于会话的工具，用于处理运动捕捉数据，而 Sift 则允许研究人员获取 Visual3D 结果并在研究级别进行处理。用户可以利用 Visual3D 引擎的强大功能，直接通过 Sift 的界面在加载的 CMZ 库上运行管道指令。

从本质上讲，Sift 旨在帮助研究人员完成知识发现过程：在传达结果之前收集、清理、塑造和分析他们的数据。

收集数据：Sift 允许您加载包含与您的研究关联的所有 C3D 文件的 CMZ 文件。Sift 使您能够使用 Sift 界面中的原始 C3D 文件构建 CMZ 库。此外，您可以将 Theia3D 文件无缝集成并合并到 Sift 中。

清理数据：您的数据可以很容易地可视化为单个跟踪、工作区平均值或组平均值。您可以单击特定数据跟踪并选择将其从分析中排除。早期数据可视化和正式的异常值验测技术可帮助您确保仅使用有效数据进行分析。通过执行 Dynamic Time Warping 和 QA 技术来更进一步，以验证您在一次彻底扫描中没有遗漏任一异常轨迹或主题。

塑造数据：一项研究可以包含多个问题，每个问题都以不同的方式查看基础数据。Sift 可以自动为您对信号进行分组，或者您可以根据标签、事件或表达式定义自己的自定义查询。常见的信号组包括 Left 和 Right 信号、Affected side 和 Unaffected side 以及 Pitching vs. Non-pitching side。

执行分析：Sift 实施了常见的数据分析技术，例如汇总统计计算、主成分分析、统计参数映射、曲线配准、动态时间扭曲、步态测量和聚类算法。

传达结果：分析结果可以导出为多种不同的文本格式，包括 Visual3D、ASCII、P2D 和 SPSS。Sift 还配备了一个高 级可视化工具箱，允许您绘制信号-时间、信号-信号、公制和公制-公制样式，同时使用 Sift 的可视化控制面板自定义颜色、字体和绘图样式。

准备好体验 Sift 的强大功能并释放数据的真正潜力了吗？开始吧！

安装

目前，Sift软件未针对 Windows 以外的系统进行编译。Sift与大多数软件一样，计算机的容量越低，程序的反应时间就越慢。Sift 可以在运行 Windows Vista、Windows 7、8 和 10 的计算机上运行。Sift 设计为在 64 位计算机上运行。

【英文介绍】

Tailored for users handling large biomechanical datasets, Sift is your solution for seamless data analysis with no exporting necessary. Load multiple CMZ libraries simultaneously, simplify data management, and conquer unwieldy databases effortlessly. Dive straight into your datasets with a variety of visualization tools, cutting-edge analysis techniques, and seamless integration of the Visual3D engine into this stand alone application. From incorporating Visual3D pipeline commands to performing Statistical Parametric Mapping, Principal Component Analysis, or outlier detection algorithms, Sift empowers you to complete comprehensive biomechanics analyses within a single application. Experience the power of Sift and unlock the true potential of your data today!

Sift allows users to: load multiple C3D files, detect and remove outliers manually or through statistical algorithms, group signals based on unique conditions, complete advanced analyses through built in Principal Component Analysis, curve registration, Statistical Parametric Mapping, and gait measures, and produce customized visualizations.

Sift is designed to integrate seamlessly with Visual3D. Where Visual3D is primarily a session-based tool for processing motion capture data, Sift lets researchers take their Visual3D results and process them at study-level. Users can leverage the power of the Visual3D engine by running pipeline commands on loaded CMZ libraries directly through Sift's interface.

At its heart, Sift is all about helping researchers through the knowledge discovery process: collecting, cleaning, shaping, and analysing their data before communicating their results.

Collecting data: Sift lets you load CMZ files containing all of the C3D files associated with your study. Sift enables you to build CMZ libraries using raw C3D files from the Sift interface. Additionally, you can seamlessly integrate and merge Theia3D files into Sift.

Cleaning data: Your data can be visualized easily as individual traces, workspace means, or group means. You can click on specific data traces and choose to exclude them from analysis. Early data visualization and formal outlier detection techniques help you ensure that only valid data is used for your analysis. Take this a step further by performing Dynamic Time Warping and QA techniques to verify that you didn't miss any outlier traces or subjects in one thorough sweep.

Shaping data: A single study can contain multiple questions, each looking at the underlying data in different ways. Sift can automatically group signals for you or you can define your own custom queries based on tags, events, or expressions. Common signal groups include Left and Right signals, Affected and Unaffected sides, and Pitching vs. Non-pitching sides.

Performing analysis: Sift implements common data analysis techniques such as Summary Statistics calculation, Principal Component Analysis, Statistical Parametric Mapping, Curve Registration, Dynamic Time Warping, Gait Measures, and Clustering algorithms.

Communicating results: Analysis results can be exported to a number of different text formats including Visual3D ASCII, P2D, and SPSS. Sift is also equipped with an advanced visualization toolbox that allows you to plot Signal-Time, Signal-Signal, Metric, and Metric-Metric styles, all while customizing colours, fonts, and plot styles with Sift's visualization control panel.

Ready to experience the power of Sift and unlock the true potential of your data? Get Started!