• Type of project: Research Project/Hauptseminar/Thesis (literature review + coding)
  • Contact: han.wang@tu-ilmenau.de

Background

In many real-world applications, high-performance computational imaging from incomplete or compressed measurements plays a crucial role. Deep learning has emerged as a powerful tool for computational imaging, offering data-driven reconstruction methods that can learn optimal priors and enhance image quality beyond traditional approaches. This project focuses on developing deep learning-based reconstruction models, particularly using convolutional neural networks (CNNs) such as U-Net, to recover high-fidelity ultrasound images from compressed data.

Task

  • Literature review on deep learning-based denoising and reconstruction methods.
  • Implement a deep neural network architecture and train the model on simulated and/or experimental ultrasound data.
  • Evaluate algorithm performance using appropriate metrics, compare with SOTA/baseline algorithms, and perform the ablation analysis.

Reference

  1. Ronneberger O, Fischer P, Brox T. U-Net: Convolutional Networks for Biomedical Image Segmentation[C]//International Conference on Medical Image Computing and Computer-Assisted Intervention. Springer, Cham, 2015: 234-241.
  2. Cao H, Wang Y, Chen J, et al. Swin-unet: Unet-like pure transformer for medical image segmentation[C]//European conference on computer vision. Cham: Springer Nature Switzerland, 2022: 205-218.