AI Article Synopsis
- This study aims to create reliable K parametric images from a shorter [F]FDG total-body PET scan using a self-supervised neural network algorithm, avoiding the need for extensive full-dynamic PET training.
- The researchers developed a self-supervised algorithm (SN-Patlak) that combines a neural network with the Patlak method, using a modified fitting approach to handle limitations in data from short scans.
- Results show that K images generated from a 10-minute PET scan using SN-Patlak are comparable to those from a standard method over a 40-minute scan, indicating the algorithm's strong reliability and effectiveness for clinical use.
Article Abstract
Purpose: The objective of this study is to generate reliable K parametric images from a shortened [F]FDG total-body PET for clinical applications using a self-supervised neural network algorithm.
Methods: We proposed a self-supervised neural network algorithm with Patlak graphical analysis (SN-Patlak) to generate K images from shortened dynamic [F]FDG PET without 60-min full-dynamic PET-based training. The algorithm deeply integrates neural network architecture with a Patlak method, employing the fitting error of the Patlak plot as the neural network's loss function. As the 0-60 min blood time activity curve (TAC) required by the standard Patlak plot is unobtainable from shortened dynamic PET scans, a population-based "normalized time" (integral-to-instantaneous blood concentration ratio) was used for the linear fitting of Patlak plot of t* to 60 min, and the modified Patlak plot equation was then incorporated into the neural network. K images were generated by minimizing the difference between the input layer (measured tissue-to-blood concentration ratios) and the output layer (predicted tissue-to-blood concentration ratios). The effects of t* (20 to 50 min post injection) on the K images generated from the SN-Patlak and standard Patlak was evaluated using the normalized mean square error (NMSE), and Pearson's correlation coefficient (Pearson's r).
Results: The K images generated by the SN-Patlak are robust to the dynamic PET scan duration, and the K images generated by the SN-Patlak from just a 10-minute (50-60 min post-injection) dynamic [F]FDG total-body PET scan are comparable to those generated by the standard Patlak method from 40-min (20-60 min post injection) with NMSE = 0.15 ± 0.03 and Pearson's r = 0.93 ± 0.01.
Conclusions: The SN-Patlak parametric imaging algorithm is robust and reliable for quantification of 10-min dynamic [F]FDG total-body PET.
Download full-text PDF |
Source |
|---|---|
| http://dx.doi.org/10.1007/s00259-024-07008-x | DOI Listing |
Publication Analysis
Top Keywords
Similar Publications
Nf-Root: A Best-Practice Pipeline for Deep-Learning-Based Analysis of Apoplastic pH in Microscopy Images of Developmental Zones in Plant Root Tissue.
Quant Plant Biol
December 2024
Quantitative Biology Center (QBiC), University of Tübingen, Tübingen, Germany.
Hormonal mechanisms associated with cell elongation play a vital role in the development and growth of plants. Here, we report Nextflow-root (nf-root), a novel best-practice pipeline for deep-learning-based analysis of fluorescence microscopy images of plant root tissue from A. thaliana.
View Article and Find Full Text PDFAssessment of human emotional reactions to visual stimuli "deep-dreamed" by artificial neural networks.
Front Psychol
December 2024
Department of Computer Science and Engineering, University of Notre Dame, Notre Dame, IN, United States.
Introduction: While the fact that visual stimuli synthesized by Artificial Neural Networks (ANN) may evoke emotional reactions is documented, the precise mechanisms that connect the strength and type of such reactions with the ways of how ANNs are used to synthesize visual stimuli are yet to be discovered. Understanding these mechanisms allows for designing methods that synthesize images attenuating or enhancing selected emotional states, which may provide unobtrusive and widely-applicable treatment of mental dysfunctions and disorders.
Methods: The Convolutional Neural Network (CNN), a type of ANN used in computer vision tasks which models the ways humans solve visual tasks, was applied to synthesize ("dream" or "hallucinate") images with no semantic content to maximize activations of neurons in precisely-selected layers in the CNN.
Computational modeling of fear and stress responses: validation using consolidated fear and stress protocols.
Front Syst Neurosci
December 2024
Universidade Federal de Goias, School of Electrical, Mechanical and Computer Engineering, Goiânia, Brazil.
Dysfunction in fear and stress responses is intrinsically linked to various neurological diseases, including anxiety disorders, depression, and Post-Traumatic Stress Disorder. Previous studies using in vivo models with Immediate-Extinction Deficit (IED) and Stress Enhanced Fear Learning (SEFL) protocols have provided valuable insights into these mechanisms and aided the development of new therapeutic approaches. However, assessing these dysfunctions in animal subjects using IED and SEFL protocols can cause significant pain and suffering.
View Article and Find Full Text PDFAn objective quantitative diagnosis of depression using a local-to-global multimodal fusion graph neural network.
Patterns (N Y)
December 2024
Medical Robot Research Institute, School of Biomedical Engineering, Shanghai Jiao Tong University, Shanghai 200240, China.
This study developed an artificial intelligence (AI) system using a local-global multimodal fusion graph neural network (LGMF-GNN) to address the challenge of diagnosing major depressive disorder (MDD), a complex disease influenced by social, psychological, and biological factors. Utilizing functional MRI, structural MRI, and electronic health records, the system offers an objective diagnostic method by integrating individual brain regions and population data. Tested across cohorts from China, Japan, and Russia with 1,182 healthy controls and 1,260 MDD patients from 24 institutions, it achieved a classification accuracy of 78.
View Article and Find Full Text PDFClinical-grade human dental pulp stem cells improve adult hippocampal neural regeneration and cognitive deficits in Alzheimer's disease.
Theranostics
January 2025
Center of Regenerative Medicine, Department of Stomatology, Renmin Hospital of Wuhan University, Wuhan University, Wuhan, China.
Disrupted hippocampal functions and progressive neuronal loss represent significant challenges in the treatment of Alzheimer's disease (AD). How to achieve the improvement of pathological progression and effective neural regeneration to ameliorate the intracerebral dysfunctional environment and cognitive impairment is the goal of the current AD therapy. We examined the therapeutic potential of clinical-grade human derived dental pulp stem cells (hDPSCs) in cognitive function and neuropathology in AD.
View Article and Find Full Text PDFWant 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!