Implementation of Machine Learning and Deep Learning algorithms for Respiratory Motion Estimation Models

MSc assignment

Keywords: Motion models, machine learning, deep learning

Project Description:

Percutaneous liver biopsy and tumor ablation are two procedures widely implemented in the diagnosis and treatment of hepatic lesions. Respiratory motion limits the accuracy of abdominal/thoracic percutaneous procedures. 

Figure 1. Respiratory motion during liver biopsy needle insertion [1]. 

Respiratory motion models offer an attractive approach to accurately estimate the real-time position of the tumor using external signals. Simple supervised machine learning algorithms can be implemented to create motion models. However, we aim to determine whether more complex machine learning or deep learning algorithms could improve the accuracy of our current approach. This research aims to create different respiratory motion models based on machine learning and deep learning algorithms and compare and evaluate their performance. Additionally, the motion models would be tested for:

  • Inter-cycle variation:  data from the same experiment but different breathing modality.
  • Inter-fraction variation: data from different experiments. 
  • Cross-population: data from different participants. 


[1]  Fahmi, S., Simonis, F. F., & Abayazid, M. (2018). Respiratory motion estimation of the liver with abdominal motion as a surrogate. IJMRCAS.