A system for endobronchial video analysis

Patrick D. Byrnes, William E. Higgins

Research output: Chapter in Book/Report/Conference proceedingConference contribution

1 Scopus citations

Abstract

Image-guided bronchoscopy is a critical component in the treatment of lung cancer and other pulmonary disorders. During bronchoscopy, a high-resolution endobronchial video stream facilitates guidance through the lungs and allows for visual inspection of a patient's airway mucosal surfaces. Despite the detailed information it contains, little effort has been made to incorporate recorded video into the clinical workow. Follow-up procedures often required in cancer assessment or asthma treatment could significantly benefit from effectively parsed and summarized video. Tracking diagnostic regions of interest (ROIs) could potentially better equip physicians to detect early airway-wall cancer or improve asthma treatments, such as bronchial thermoplasty. To address this need, we have developed a system for the postoperative analysis of recorded endobronchial video. The system first parses an input video stream into endoscopic shots, derives motion information, and selects salient representative key frames. Next, a semi-automatic method for CT-video registration creates data linkages between a CT-derived airway-tree model and the input video. These data linkages then enable the construction of a CT-video chest model comprised of a bronchoscopy path history (BPH) | defining all airway locations visited during a procedure | and texture-mapping information for rendering registered video frames onto the airway-tree model. A suite of analysis tools is included to visualize and manipulate the extracted data. Video browsing and retrieval is facilitated through a video table of contents (TOC) and a search query interface. The system provides a variety of operational modes and additional functionality, including the ability to define regions of interest. We demonstrate the potential of our system using two human case study examples.

Original languageEnglish (US)
Title of host publicationMedical Imaging 2017
Subtitle of host publicationImage-Guided Procedures, Robotic Interventions, and Modeling
EditorsRobert J. Webster, Baowei Fei
PublisherSPIE
ISBN (Electronic)9781510607156
DOIs
StatePublished - Jan 1 2017
EventMedical Imaging 2017: Image-Guided Procedures, Robotic Interventions, and Modeling - Orlando, United States
Duration: Feb 14 2017Feb 16 2017

Publication series

NameProgress in Biomedical Optics and Imaging - Proceedings of SPIE
Volume10135
ISSN (Print)1605-7422

Other

OtherMedical Imaging 2017: Image-Guided Procedures, Robotic Interventions, and Modeling
CountryUnited States
CityOrlando
Period2/14/172/16/17

All Science Journal Classification (ASJC) codes

  • Electronic, Optical and Magnetic Materials
  • Biomaterials
  • Atomic and Molecular Physics, and Optics
  • Radiology Nuclear Medicine and imaging

Fingerprint Dive into the research topics of 'A system for endobronchial video analysis'. Together they form a unique fingerprint.

Cite this