Medical volume data classification uncertainty visualization method based on probability sliding block

Abstract

The invention discloses a medical volume data classification uncertainty visualization method based on a probability sliding block. The method comprises the following steps: 1, carrying out brightness conversion operation on medical volume data to obtain converted medical volume data; 2, after the medical volume data subjected to brightness conversion is given, applying a spatial fuzzy c-mean value SFCM classification method to the medical volume data so as to obtain probability volume data of N classifications, each classification corresponding to one medical material; 3, under the condition that N pieces of classified probability volume data are given, generating color volume data; and 4, giving the color volume data generated in the step 3, and applying a ray casting volume rendering algorithm to the color volume data to obtain a 2D image. The generated result can quantitatively represent the occurrence probability of any medical material, can reveal the extreme condition of any material, so as to be used as a boundary condition for decision making. Therefore, medical experts can make more accurate diagnosis or preoperative decisions.

Description

technical field [0001] This patent relates to the field of visualization and visual analysis, and relates to a method of using probability sliders to visualize uncertainty in the classification of medical volume data. Background technique [0002] Uncertainty in the field of visualization is one of the main challenges in visualization research, and many techniques that contribute to this field have been proposed. Brodlie et al. and Potter et al. present two reviews summarizing and classifying the latest uncertainty visualization techniques (Ref. [1] Brodlie, K., Osorio, R.A., Lopes, A.A Review of Uncertaintyin Data Visualization. Expanding the Frontiers of Visual Analytics and Visualization, pp.81–109. Springer London, 2012. E. Brodlie, K., Osorio, R.A., Lopes, A. A Survey of Uncertainty in Data Visualization. Extending Frontiers of Visual Analytics and Visualization, pp. 81–109. London Springer, 2012) (reference [2] Potter, K., Rosen, P., Johnson, C.R. From Quantification ...

AI Technical Summary

Problems solved by technology

However, there are three main problems with this method: first, the visualization results cannot reflect any quantitative information about the classification uncertainty of different medical materials; second, the final renderings are randomly displayed through the random adjustment of the transfer function TF; Third, the classification task and the optical property assignment task were carried out simultaneously

The existence of these problems will lead to the method not enabling medical experts to make accurate diagnoses or preoperative decisions, let alone provide medical experts with a clear concept of how the medical volume data is classified

Images