Patient-specific hyperparameter learning for optimization-based CT image reconstruction.

Phys Med Biol

Department of Radiology and Imaging Sciences, University of Utah, United States of America.

Published: September 2021

We propose a hyperparameter learning framework that learnshyperparameters for optimization-based image reconstruction problems for x-ray CT applications. The framework consists of two functional modules: (1) a hyperparameter learning module parameterized by a convolutional neural network, (2) an image reconstruction module that takes as inputs both the noisy sinogram and the hyperparameters from (1) and generates the reconstructed images. As a proof-of-concept study, in this work we focus on a subclass of optimization-based image reconstruction problems with exactly computable solutions so that the whole network can be trained end-to-end in an efficient manner. Unlike existing hyperparameter learning methods, our proposed framework generates patient-specific hyperparameters from the sinogram of the same patient. Numerical studies demonstrate the effectiveness of our proposed approach compared to bi-level optimization.

Download full-text PDF

Source
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC8584383PMC
http://dx.doi.org/10.1088/1361-6560/ac0f9aDOI Listing

Publication Analysis

Top Keywords

hyperparameter learning
16
image reconstruction
16
optimization-based image
12
reconstruction problems
8
patient-specific hyperparameter
4
learning
4
learning optimization-based
4
image
4
reconstruction
4
reconstruction propose
4

Similar Publications

Background: This study aims to investigate the associations between signal-level physical activity (PA) features derived from wrist accelerometry data and cognitive status in older adults, and to evaluate their potential predictive value when combined with demographics.

Methods: We analyzed PA data from 3,363 older adults (NHATS: n = 747; NHANES: n = 2,616), with each participant contributing a complete 3-day continuous activity sequence. We extracted the most relevant PA features associated with cognitive function using feature engineering and recursive feature elimination.

View Article and Find Full Text PDF

Cancer, as a shocking disease, is one of the most common malignant tumors among women, posing a huge threat to the physical health and safety of women worldwide. With the continuous development of science and technology, more and more high and new technologies are involved in the diagnosis and prediction of breast cancer. In recent years, intelligent medical assistants supported by data mining and machine learning algorithms have provided necessary support for doctors' diagnosis.

View Article and Find Full Text PDF

Enhancement of security, personalization, and safety in advanced transportation systems depends on driver identification. In this context, this work suggests a new method to find drivers by means of a Random Forest model optimized using the osprey optimization algorithm (OOA) for feature selection and the salp swarm optimization (SSO) for hyperparameter tuning based on driving behavior. The proposed model achieves an accuracy of 92%, a precision of 91%, a recall of 93%, and an F1-score of 92%, significantly outperforming traditional machine learning models such as XGBoost, CatBoost, and Support Vector Machines.

View Article and Find Full Text PDF

Efficient optimisation of physical reservoir computers using only a delayed input.

Commun Eng

January 2025

Laboratoire d'Information Quantique CP224, Université libre de Bruxelles (ULB), Av. F. D. Roosevelt 50, 1050, Bruxelles, Belgium.

Reservoir computing is a machine learning algorithm for processing time dependent data which is well suited for experimental implementation. Tuning the hyperparameters of the reservoir is a time-consuming task that limits is applicability. Here we present an experimental validation of a recently proposed optimisation technique in which the reservoir receives both the input signal and a delayed version of the input signal.

View Article and Find Full Text PDF

Machine learning analysis of rivaroxaban solubility in mixed solvents for application in pharmaceutical crystallization.

Sci Rep

January 2025

Department of Oral & Maxillofacial Surgery and Diagnostic Sciences, Faculty of Dentistry, Taif University, 21944, Taif, Saudi Arabia.

This study investigates the use of machine learning models to predict solubility of rivaroxaban in binary solvents based on temperature (T), mass fraction (w), and solvent type. Using a dataset with over 250 data points and including solvents encoded with one-hot encoding, four models were compared: Gradient Boosting (GB), Light Gradient Boosting (LGB), Extra Trees (ET), and Random Forest (RF). The Jellyfish Optimizer (JO) algorithm was applied to tune hyperparameters, enhancing model performance.

View Article and Find Full Text PDF

Want AI Summaries of new PubMed Abstracts delivered to your In-box?

Enter search terms and have AI summaries delivered each week - change queries or unsubscribe any time!