3D reconstruction

Scenario

This scenario is built under real-time tracking conditions.

Particle detectors are used in lab conditions or in a pre-known geometrical condition to be able to resolv geometrical information from the environment.

Under certain conditions have a knowledge the device, the radiation source, the detector and the relative position of the patient, or device under test, can improve the extraction of the sensor information.

This leads to work under non laboratory or well know conditions, this research line tries to:

  • Provide reconstruction algorithms to capture 3D coordinates of all the required devices.
  • Provide benefits in the reconstruction by moving the capture devices, and include real time 3D infrmation to the reconstruction.
  • Improbe by using Machine Learning the reconstruction. This can be done training models that latter can provide an heuristic robust pattern for reconstruction
  • Include this information to provide enhanced or extra information to the detector (density, compton corrections)ls

Works and tools

This is a large colaboration started from 2012, and started from a patent (currently licensed), to start to develop enhanced method, the main work with the colaboration of the University Politécnica de Valencia, latter colaborators are Instituto de Biomecánica de Valencia, and Hospital Politécnico La Fe.

From the begining the main tasks where to provide robust methods to permorm image registration, image registration in placing two diferent imaging devices in the same coordinates. The main idea is perform latter calculations with this information.

Information can arise from, depth cameras, which requiries procesing in large GPU devices a large number of points, visible cameras, which requieres to place the system in diferent locations to provide precise 3D. Last but not least, Machine Learnign can be applied to solve models for specific diseases.

Publications & Scientific outputs

  • Densitometric Radiographic Imaging With Contour Sensors  

    Volume: 7

    DOI: 10.1109/ACCESS.2019.2895925

  • Gamma-ray imaging system for real-time measurements in nuclear waste characterisation  

    Publisher: IOP Publishing

    Volume: 13

    Page: P03016-P03016

    DOI: 10.1088/1748-0221/13/03/p03016

  • Identifying the Best Machine Learning Algorithms for Brain Tumor Segmentation, Progression Assessment, and Overall Survival Prediction in the BRATS Challenge  
  • 3D measurements in conventional X-ray imaging with RGB-D sensors  

    Volume: 42

    Page: 73-79

    DOI: 10.1016/j.medengphy.2017.01.024

  • Evaluation of modern camera calibration techniques for conventional diagnostic X-ray imaging settings  

    Volume: 10

    Page: 68-81

    DOI: 10.1007/s12194-016-0369-y

  • 3D measurements from X-ray images and dense surface mappings  

    DOI: 10.1109/GMEPE-PAHCE.2016.7504638

  • Geometrical Calibration of X-Ray Imaging with RGB Cameras for 3D Reconstruction  

    Volume: 35

    Page: 1952-1961

    DOI: 10.1109/TMI.2016.2540929

  • Joint calibration of RGB and X-ray cameras  

    DOI: 10.1109/GMEPE-PAHCE.2016.7504640

  • Measuring X-ray image quality using a perceptual metric  

    DOI: 10.1109/GMEPE-PAHCE.2016.7504639