Real-Time Optical Vascular Imaging, a new method for the diagnosis and monitoring of oral diseases

P. Bastos*, G. Carpentier, V. Patel, D. Papy-Garcia, T. Watson, R. Cook

*Corresponding author for this work

Research output: Contribution to journalArticlepeer-review

3 Citations (Scopus)

Abstract

Introduction: Real-Time Optical Vascular Imaging (RTOVI) is a technology developed in the Centre for Oral Clinical & Translational Sciences, within the Faculty of Dentistry, Oral & Craniofacial Sciences, King's College London, that allows rapid and preparation free, in vivo imaging of the microvascular anatomy of the human oral cavity. Microvascular changes are known to be related to disease subtypes, in particular cancer. This makes in vivo microvascular examination clinically valuable. However, at present there is lack of any analytical method able to objectively assess microvascular morphology images. Discussion: The assessment of microvascular morphology based on a subjective evaluation was proven to be unreliable. There was a need to develop a software-based analysis for in vivo microvascular images to support the validation of RTOVI. This paper reviews the authors work to develop and test an automated microvascular analysis method for RTOVI based on ImageJ, an open-source software. This allowed to determined which parameters offered a more robust mathematical representation of the microvascular anatomy of the gingival margin, such as the mean area per capillary and mean aspect ratio. However, in vivo microvascular images from elsewhere within the oral cavity posed a bigger challenge to the analysis procedure due to the microvascular architectural complexity and poorer contrast. Angiogenesis Analyzer, a well-known ImageJ plugin used for the quantification of in vitro microvascular images, is under development in collaboration with the University of Paris Est Créteil. The aim of this work is to obtain an automated analysis method for in vivo microvascular images able to offer a solid foundation for the diagnostic potential of RTOVI and subsequent clinical integration of this technology. Conclusion: An automated analysis method for in vivo microvascular images is paramount before any attempt to clinically validate RTOVI. Our initial work of testing a software-based analysis demonstrated the effectiveness of some parameters, which is valuable for future work, and led us to move into a more sophisticated method involving customising the Angiogenesis Analyzer plugin. This is an essential step, aiming to extend the potential of in vivo microscopy with the clinical integration of RTOVI. Lay Description: This article summarises the initial research work done in the field on in vivo microvascular imaging aiming to develop a technique for the diagnosis of oral diseases based on the shape of small blood vessels found just below the surface of the "skin" inside the mouth. This offers the potential to examine lesions without the need to take a sample (biopsy/cutting tissue) to observe it microscopically. This ultimately offers a potential to accelerate diagnostic decision making, avoid unpleasant and often deterrent surgical procedures and reducing diagnostic laboratory time and cost burdens. However, in order to assess images of small blood vessels obtained in clinic, we needed to develop and test a software-based analysis to avoid the subjective human interpretation, known not to work. This article describes the authors journey to achieve an automated and sophisticated analysis method unique in the world for in vivo microvascular images derived from real-time optical vascular imaging.

Original languageEnglish
JournalJournal of Microscopy
DOIs
Publication statusAccepted/In press - 2020

Keywords

  • Angiogenesis Analyzer
  • in vivo
  • microvascular anatomy
  • microvascular imaging
  • oral cancer
  • oral diseases
  • real-time
  • software-based analysis

Fingerprint

Dive into the research topics of 'Real-Time Optical Vascular Imaging, a new method for the diagnosis and monitoring of oral diseases'. Together they form a unique fingerprint.

Cite this