An Instant Tumor Detection Method for Ex-vivo Tissue Palpation Utilizing Self-learning High-density Flexible Tactile Sensor Array Based on Attention Mechanism Neural Network.

Journal: Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Annual International Conference
PMID:

Abstract

This paper reports an innovative and instant method for tumor detection. We delineate the tactile signal matrixes of tumor tissue with varying stiffness utilizing self-learning high-density flexible tactile sensor array, and depict the morphological features of the tumor at the fingertip tactile perception with an attention mechanism neural network. A novel algorithm to instantaneously recognize the shape and volume of tumor tissue, incorporating an attention mechanism model for autonomous learning of pressure signal characteristics from each sensing unit, is developed in the ex-vivo experiments, for the first time, by delivering the dense sensing array that enables intimate location-stable contact with skin. After an extensive series of experiments, our self-learning tactile sensor array has achieved a accuracy of 97.2% in identifying the precise depths of tumors. Additionally, the recognition rate for the dimensional aspects of tumors reached 97.8%. Furthermore, the sensor array exhibits the capability to detect tumors buried at depths of up to 5mm. Notably, the system demonstrates exceptional sensitivity to alterations in tumors with diameters exceeding 2mm when buried. The proposed method provides a promising technological avenue for cancer incipient screening in tissue palpation, surgical robots, and robot-assisted minimally invasive surgery.

Authors

  • Fang Wang
    Key Laboratory of Intelligent Computing and Information Processing of Ministry of Education and Hunan Key Laboratory for Computation and Simulation in Science and Engineering, Xiangtan University, Xiangtan, China.
  • Heng Yang
    State Key Laboratory of Transducer Technology, Shanghai Institute of Microsystem and Information Technology, Chinese Academy of Sciences, Shanghai, 200050, China.
  • Yue He
    Department of Breast Surgery, Hunan Cancer Hospital, Changsha, Hunan, China.
  • Ke Sun
  • Yi Sun
    Department of Environmental and Occupational Health, Program in Public Health, University of California, Irvine, CA, USA.
  • Xinxin Li
    School of Artificial Intelligence, Hebei University of Technology, Tianjin 300130, China.
  • Xikun Zheng
  • Jingqing Hu