University of Twente Student Theses
Scannerless MRI Generation Using Generative Adversarial Networks With Multiple Surrogate Signals
Damme, Koen (2024) Scannerless MRI Generation Using Generative Adversarial Networks With Multiple Surrogate Signals.
|
PDF
11MB |
| Abstract: | Respiratory-induced motion (RIM) presents a challenge in targeting liver tumors during medical procedures, as it causes the tumor to shift position within the body. Motion models can track the position of a liver tumor based on a surrogate signal, compensating for RIM to enable more accurate ablation and biopsy procedures. However, interpreting tumor position as an XYZ-coordinate would be challenging for clinicians. This study presents a conditional progressively growing generative adversarial network (cProGAN) that can generate scannerless MR-images using one or multiple surrogate signals for guidance during liver interventions. We compared three signals: a heat camera measuring the airflow, and an ultrasound transducer and external markers, to capture the internal and external abdominal motion, respectively. This study is validated in seven human subject experiments, where MR-images and the three surrogate signals are simultaneously collected while each subject is following a specific breathing protocol. The quality of the scannerless images is assessed by the structural similarity index measure (SSIM) and by extracting the superior-inferior movement of the liver border in the real and scannerless images and comparing the resulting waveform using the mean absolute error (MAE, in millimeters and as a percentage of the average liver movement) and the coefficient of determination metrics. The model trained on external markers generated images with the most accurate liver positions during breathing (MAE of 5.02 ± 3.74 mm) and breath holds (MAE of 9.14 ± 1.31 mm). The highest SSIM was for the combined model during breathing (51.42%) and for the external marker model during breath holds (36.47%). Models using the other surrogate signals resulted in a significantly higher MAE and lower SSIM. These results suggest that external marker tracking provides the most accurate respiratory motion modeling for scannerless MRI generation, though further research is needed to improve image quality. The proposed solution can potentially be expanded by adding more sources of motion and generating entire 3D volumes and we believe that this could greatly improve the precision of percutaneous procedures in the liver and make them easier to perform |
| Item Type: | Essay (Master) |
| Faculty: | EEMCS: Electrical Engineering, Mathematics and Computer Science |
| Subject: | 54 computer science |
| Programme: | Computer Science MSc (60300) |
| Link to this item: | https://purl.utwente.nl/essays/104047 |
| 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