University of Twente Student Theses
Estimation of liver respiratory-induced motion using an infrared camera for minimally invasive surgery
Elgebily, Mostafa (2024) Estimation of liver respiratory-induced motion using an infrared camera for minimally invasive surgery.
|
PDF
5MB |
| Abstract: | Liver respiratory induced motion (RIM) presents a significant challenge during biopsies and minimally invasive hepatic surgeries. A method of battling this RIM is the use of respiratory motion estimation (RME). RME utilises external signals called surrogates to relay internal liver RIM noninvasively. This study presents a novel RME method for liver Superior- Inferior (SI) motion tracking using Infrared (IR) imaging as a surrogate signal and deep learning using convolutional neural networks (CNNs) to directly derive the displacement of the liver from the IR images of the nose/mouth area, as a ROI, in patients. This method utilises the raw image data as an input to the fitting method along with MRI liver dome displacement values as ground truth training values. The training and validation splits were used to train the model on the motion patterns while the test split was fed to the trained model to predict motion data from the thermal images. The model was then validated on three healthy human subjects, each participating in three sessions to acquire displacement and thermal ground truth data using an MRI machine and a thermal camera. This paper, to the best of the author’s knowledge, presented a new method for tracking the liver and proved that it can be a viable option. The R2 scores ranged from 0.75-0.98 throughout the 9 sessions and the MAE ranged from 0.91 mm-3.8 mm. The authors also suggested the use of higher resolution (temporal and spatial) MRI and ultrasound (US) scanners to significantly improve the results and allow the exploration of 3D motion detection instead of only SI. |
| Item Type: | Essay (Master) |
| Faculty: | EEMCS: Electrical Engineering, Mathematics and Computer Science |
| Subject: | 44 medicine, 50 technical science in general, 54 computer science |
| Programme: | Systems and Control MSc (60359) |
| Link to this item: | https://purl.utwente.nl/essays/102885 |
| Export this item as: | BibTeX EndNote HTML Citation Reference Manager |
Show download statistics for this publication
Show download statistics for this publicationRepository Staff Only: item control page