Validation of Segmented Brain Tumor from MRI Images Using 3D Printingthe

Document Type : Research Articles

Authors

1 Department of Computer Science and Engineering, Manipal Institute of Technology, Manipal Academy of Higher Education, Manipal, 576104, India.

2 Research and Development, RTWO Healthcare Private LLP, Mahalakshmipuram, Bengaluru, 560086, India.

Abstract

Background: Early diagnosis of a brain tumor is important for improving the treatment possibilities. Manually segmenting the tumor from the volumetric data is time-consuming, and the visualization of the tumor is rather challenging. Methods: This paper proposes a user-guided brain tumour segmentation from MRI (Magnetic Resonance Imaging) images developed using Medical Imaging Interaction Toolkit (MITK) and printing the segmented object using the 3D printer for tumour quantification. The proposed method includes segmenting the tumour interactively using connected threshold method, then printing the physical object from the segmented volume of interest. Then the distance between two voxels was measured using electronic callipers on the 3D volume in a specific direction. And next, the same distance was measured in the same direction on the 3D printed object. Results: The technique was tested with n=5 samples (20 readings) of brain MRI images from RIDER Neuro MRI dataset of National Cancer Institute. MITK provides various tools that enable image visualization, registration, and contouring. We were able to achieve the same measurements using both the approaches and this has been tested statistically with paired t-test method. Through this and the observer’s opinion, the accuracy of the segmentation was proved. Conclusion: When the difference in measurement of tumor volume through the electronic calipers and with 3D printed object equates to zero, proves that the segmentation technique is accurate. This helps to delineate the tumor more accurately during radio therapy.

Keywords

Main Subjects

Asian Pacific Journal of Cancer Prevention
Volume 22, Issue 2
February 2021
Pages 523-530

Files

History

  • Receive Date: 17 December 2020
  • Revise Date: 13 January 2021
  • Accept Date: 05 February 2021

Share

How to cite

Statistics