Latent diffusion model is a successful generative model in the modern computer vision researches. Modeling the generative process as image denoising, the diffusion models can generate realistic images in high quality and shows superior ability as the GAN-based models. In medical imaging, computed tomography (CT) is a well researched imaging modality and also widely applied in clinics. In this project, we will investigate the feasibility of modern diffusion models for the task of CT synthesis.
In recent years, deep learning has emerged as a transformative force in the realm of image processing, particularly in addressing inverse problems such as denoising and artifact reduction in medical imaging. This research aims to systematically investigate the impact of various loss functions on deep learning-based solutions for inverse problems, with a focus on low-dose Computed Tomography (CT) imaging.
Low-dose CT, while beneficial in reducing radiation exposure, often suffers from increased noise and artifacts, adversely affecting image quality and diagnostic reliability. Traditional denoising techniques, although effective to some extent, struggle to maintain a balance between noise reduction and the preservation of crucial image details. Deep learning, especially Convolutional Neural Networks (CNNs), has shown promising results in surpassing these traditional methods, offering enhanced image reconstruction with remarkable fidelity.
However, the choice of loss function in training deep learning models is critical and often dictates the quality of the reconstructed images. Commonly used loss functions like Mean Squared Error (MSE) or Structural Similarity Index (SSIM) have their limitations and may not always align well with human perceptual quality. This research proposes to explore and compare a variety of loss functions, including novel and hybrid formulations, to evaluate their efficacy in enhancing image quality, reducing noise, and eliminating artifacts in low-dose CT images.
This project focuses on reimplementing the Detectability Index for evaluating individual CT projections, with the goal of improving the performance and adaptability of existing Python-based algorithms using PyTorch. The selected candidate will delve into the current code, identify performance bottlenecks, and propose innovative solutions to optimize efficiency. The goal is to minimize package dependencies to ensure code longevity and maintainability.
The following questions should be considered:
- How can the existing Python code be improved with PyTorch for better performance and adaptability?
- Where do the current code’s performance bottlenecks lie, and how can these be addressed?
- How can the usage of external packages be minimized to ensure the code’s longevity?
- What innovative approaches can be implemented to enhance the Detectability Index calculation?
- How can the updated algorithm be validated for effectiveness and efficiency?
Candidates should possess strong skills in Python and PyTorch, with the ability to quickly understand and improve upon existing code. A background in computational imaging or related fields, along with a problem-solving mindset, is essential.
For your application, please send your transcript of record.
In the realm of computer vision, significant research is currently dedicated to object detection
and recognition. Research groups and developers are actively striving to enhance machine
learning solutions, aiming to boost the accuracy of image detection and recognition in
accordance with specific use cases. The German AI-driven Monuments Detection System is an
innovative project aimed at providing tourists with an enhanced experience by leveraging
artificial intelligence (AI) to recognize and provide historical information about prominent
monuments in Germany. This document presents the project’s methodology, results, and
implications. With this machine, a user can scan the historical place and view its historical
details. There are five categories used in this model. Python is used as a programming
language with the TensorFlow framework.
Use a new open source framework generating X-ray images for deep learning model training.
Requirement: Python, CT reconstruction
Please attach your CV and transcripts to email@example.com
In this project, we perform a computational domain transfer to introduce cone-beam artifacts to the training data. We evaluate its impact on the results of supervised training for the segmentation of the lungs. For this, already labeled CT volumes are reconstructed to artificial CBCT volumes without a complex deep learning-based method, like introduced by Jia X et al.,5 but rather by computational reconstruction. The purpose is to have a network for stable segmentation on real CBCT volumes. A major advantage of our approach is that the artificial
CBCT volumes can not only be computed easily from thoracic CT volumes but also the pixel-wise segmentation can be re-used without putting in the great effort of labeling. This allows for supervised training.
Realistic Simulation of Collimated X-Ray images for Collimator Edge Segmentation using Deep Learning
Collimator detection in X-ray systems has long posed a challenge, particularly when information about the detector’s position relative to the source is either unreliable or completely unavailable. In this paper , we introduce a physically motivated image processing pipeline designed to simulate the intricate characteristics of collimator shadows in X-ray images. The primary objective of this pipeline is to address the scarcity of training data for deep neural networks, which are increasingly promising for collimator detection. By applying the pipeline to deep networks initially limited by small datasets, our approach equips them with the necessary information to learn and generalize effectively.
Our pipeline is a comprehensive solution that leverages several key components to
generate realistic collimator images. Employing randomized labels to describe collimator shapes and their respective locations ensures diversity and representativeness. In addition, we integrate a convolution kernel based scattered radiation simulation mechanism, which is a crucial factor in real-world X-ray imaging. To complete the simulation process, we introduce Poisson noise to replicate the inherent characteristics of collimator shadows in X-ray images.
Comparing the simulated data with real collimator shadows demonstrates the authenticity of our approach and its potential to bridge the gap between synthetic and real-world data. Moreover, incorporating simulated data into our deep learning framework not only serves as a valid substitute for real collimators but also significantly improves generalization in real-world applications, holding great promise for the field of collimator detection.
This work was presented at the DALI workshop at the MICCAI conference in Vancouver, Canada and was published in the proceedings:
1. El-Zein B, Eckert D, Weber T, Rohleder M, Ritschl L, Kappler S et al. A Realistic Collimated
X-Ray Image Simulation Pipeline. Data Augmentation, Labelling, and Imperfections – Third
MICCAIWorkshop, DALI 2023, Held in Conjunction with MICCAI 2023, Vancouver, October
12, 2023, Proceedings. Springer Nature. 2023, (in press)