Vision-based Sensing Mechanism for Soft Tissue Stiffness Estimation

Student thesis: Doctoral ThesisDoctor of Philosophy


During the last few decades, advanced instrumentations have been widely explored for research and diagnostic applications in medical and life sciences. Traditionally, these instruments have been primarily found in laboratory environments. Yet, as the technology finds its way from research facilities to the point of care, diagnostic tools are increasingly used in hospitals and other near-patient settings.

In medical diagnosis, physicians palpate the anatomical surfaces with their fingertips to assess variations in stiffness across the surface of organs. Hence, they locate abnormally stiff areas by relying on their sense of touch. However, palpation is a highly sophisticated manual skill, which can only be performed in areas that are accessible to the hands, therefore it cannot be employed in minimally invasive procedures. Several instruments have been developed with the aim of reproducing the physician’s “sense of touch”. These devices should be inexpensive and thus affordable to point-of-care providers. They should
also be versatile so that multiple tests can be performed to improve efficiency and, at the same time, reduce costs. Moreover, since operators may not be trained technicians, these devices need to be intuitive and easy to use. Furthermore, if designed for minimally invasive procedures, they should be small enough to fit through a standard trocar port. This thesis presents a novel vision-based sensor for soft tissue stiffness estimation, measuring multiple tool-tissue interaction forces in parallel employing a set of deformable elements. By means of an analysis of the differences between the measured forces the
stiffness of the material the sensor is interacting with can be computed. The motion of the deformable elements whilst in contact with the environment are captured by a camera sensor and related to force and, subsequently, to stiffness exploiting knowledge of the spring constant of the deformable elements.

The developed system provides quantitative measurements of the tissue stiffness which can be used to diagnose abnormalities. The proposed sensing principle has been used to develop a hand-held stiffness probe for tumour identification. In order to retrieve the “sense of touch” in minimally invasive procedures, a clip-on stiffness sensor has been developed that can be integrated with a surgical endoscopic camera. The new, integrated system extends the sensing capabilities of the camera: the system can be used to visualise
the anatomical areas inside the human body as well as measure the stiffness of these abdominal structures. Adding a purely mechanical device to an endoscopic camera, proves to be a cost-effective way to introduce remote stiffness measurement capabilities to minimally invasive intervention. The experimental results demonstrate the effectiveness and accuracy of the proposed system, successfully discriminating soft tissues over a wide range of tissue stiffness values. To make the system intuitive, a colour-coded stiffness map that is generated in real-time is used to visualise the stiffness distribution of the
examined soft object surface. Moreover, the sensory mechanism can be manufactured at a low price, does not use any electronic components; it is easy to use and does not require any calibration. The measurement range and resolution can be easily customised.

Date of Award2016
Original languageEnglish
Awarding Institution
  • King's College London
SupervisorThrishantha Nanayakkara (Supervisor) & Kaspar Althoefer (Supervisor)

Cite this