Zihan Lin

Zihan Lin

MS Robotics | BS Mechanical Engineering

Sitemap

A list of all the posts and pages found on the site. For you robots out there, there is an XML version available for digesting as well.

Pages

Posts

Smart Glove System Design for Tumor Detection and 3D Visualization

1 minute read

Published:

This post outlines a conceptual design for a smart glove system that combines tactile sensing, AI analysis, and 3D visualization to assist in detecting subcutaneous abnormalities such as tumors.

1. System Architecture

Smart Glove Hardware

  • Tactile Sensors: Embedded in the glove to detect pressure, hardness, and tissue morphology
  • Position Tracking System: Micro IMU (Inertial Measurement Unit) or optical sensors on the glove to record movement trajectories on the body
  • Data Transmission Module: Real-time data transfer via Bluetooth/Wi-Fi to a computer or mobile device

AI Analysis and Modeling Software

  • Tactile Data Analysis: AI model converts sensor data into hardness maps indicating potential tumor locations
  • 3D Model Generation: Real-time trajectory recording combined with 3D modeling algorithms to construct tissue models of the examined area
  • Tumor Annotation: Highlights high-hardness regions in the 3D model with size, location, and depth information

Display Device

  • Real-time visualization on AR glasses, tablets, or screens showing the 3D model and annotated tumor locations

2. Technical Core

Tactile Sensors and Data Acquisition

  • Pressure Sensors: Detect pressure distribution and generate hardness data
  • Flexible Sensors: Conform to body contours and capture subtle tactile variations

3D Modeling and Reconstruction

  • SLAM (Simultaneous Localization and Mapping): Generates real-time 3D models using glove position data
  • Point Cloud Generation: Accumulates touch points into point clouds, then applies surface fitting techniques to create smooth 3D models

AI Model Analysis

  • Hardness Classification Model: Machine learning (e.g., CNN) analyzes tactile data to detect hardness anomalies
  • Segmentation and Annotation: Marks abnormal regions in the 3D model and calculates specific tumor parameters (diameter, depth)

3D Visualization

  • Tools like Unity or Blender generate 3D tissue models
  • Real-time rendering of examined areas with highlighted tumor regions

Future Work

This design concept could be extended to include:

  • Integration with medical imaging data (ultrasound, MRI) for validation
  • Haptic feedback for the examiner
  • Clinical trials for accuracy assessment

Chinese Version

设计方案

  1. 系统架构 智能手套硬件：

触觉传感器：嵌入手套，用于检测压力、硬度和组织形态。 位置跟踪系统：手套上的微型 IMU（惯性测量单元）或光学传感器，用于记录手套在人体上的移动轨迹。 数据传输模块：通过蓝牙/Wi-Fi 将触觉数据实时传输到计算机或移动设备。 AI 分析与建模软件：

触觉数据分析：AI 模型将传感器数据转换为硬度图（表征肿块可能性）。 3D 模型生成：通过实时轨迹记录，利用 3D 建模算法构建触摸区域的人体组织模型。 肿块标注：在 3D 模型中标注高硬度区域，并提供尺寸、位置和深度信息。 显示设备：

AR 眼镜、平板或屏幕上实时显示触摸区域的 3D 模型和肿块位置。

  1. 技术核心 触觉传感器和数据采集：

压力传感器：检测压力分布，生成硬度数据。 柔性传感器：适应人体曲线，捕捉细微的触感变化。 3D 建模与重建：

SLAM（同步定位与建图）算法：通过手套的位置数据实时生成3D模型。 点云生成：将触摸点累积为点云，并通过表面拟合技术生成光滑的3D模型。 AI 模型分析：

硬度分类模型：使用机器学习（例如 CNN）分析触觉数据，检测硬度异常。 分割与标注：在3D模型中标注异常区域，并计算肿块的具体参数（如直径和深度）。 3D 可视化：

使用 Unity 或 Blender 等工具生成人体组织的3D模型。 实时渲染触摸区域，并在模型中高亮肿块区域。

projects

publications

Indenting a thick gel with a solid spherical inclusion

Published in European Journal of Mechanics - A/Solids, 2024

This paper investigates the mechanical behavior of soft gel with embedded spherical inclusions under indentation loading.

Recommended citation: Wen, D., Lin, Z., Tjiptowidjojo, Y., & Wan, K. (2024). Indenting a thick gel with a solid spherical inclusion. European Journal of Mechanics - A/Solids, 106, 105343.
Download Paper

talks

teaching