University of Twente Student Theses

Login

Towards intra-operative feedback on oncological colorectal tissue type prediction

Romeijn, Stephan (2019) Towards intra-operative feedback on oncological colorectal tissue type prediction.

Full text not available from this repository.

Full Text Status:Access to this publication is restricted
Abstract:During surgery, it is challenging for surgeons to differentiate pathological tissue from healthy tissue due to the corresponding visual characteristics. The goal of this study is to overcome the first challenges in implementing real-time tissue classification with multispectral imaging during surgery. In addition, the first challenges will be researched in the possibilities of using a projector to make a backprojection of the tissue type prediction on the patient. The extended vision box, a newly developed medical imaging instrument from a collaboration between Philips and Antoni van Leeuwenhoek, containing a multispectral camera, depth camera and a projector is used for this research. The final algorithm was able to distinguish healthy, fat and tumor tissue with an accuracy of 0.85. The sensitivity/specificity of healthy, fat and tumor were 0.70/0.90, 0.95/0.97 and 0.80/0.92 respectively. The Backprojection showed an overall error of 1.8mm. A proof of concept was given of the backprojection of classified data on a colorectal specimen. Future research should focus on a new algorithm design to increase the classification accuracy, mainly focusing on the classification of healthy vs tumor classification and optimise the overall system error and response time of the backprojection.
Item Type:Essay (Master)
Faculty:TNW: Science and Technology
Subject:33 physics, 44 medicine
Programme:Technical Medicine MSc (60033)
Link to this item:https://purl.utwente.nl/essays/79363
Export this item as:BibTeX
EndNote
HTML Citation
Reference Manager

 

Repository Staff Only: item control page