Article
Computer Science, Interdisciplinary Applications
Liqun Yi, Yuxia Sheng, Li Chai, Jingxin Zhang
Summary: In this paper, a novel denoising method based on spectral graph wavelet transform (SGWT) is proposed for improving the quality of dynamic PET images while preserving edge details. Experimental results show that the proposed method outperforms other common denoising methods and has comparable performance to deep learning-based methods but with lower computational complexity.
MEDICAL & BIOLOGICAL ENGINEERING & COMPUTING
(2023)
Article
Computer Science, Information Systems
Hao Sun, Lihong Peng, Hongyan Zhang, Yuru He, Shuangliang Cao, Lijun Lu
Summary: In this study, a novel deep learning denoising framework, DeepRED denoising, was proposed to enhance the quantitative accuracy of PET images. Compared to conventional methods, DeepRED denoising shows significant improvements in both visual and quantitative accuracy, with and without prior images.
Article
Computer Science, Information Systems
Runxi Cui, Zhigang Chen, Jia Wu, YanLin Tan, GengHua Yu
Summary: The study proposed a new automatic multiprocessing scheme for PET image pre-screening, noise reduction, segmentation, and lesion partitioning. The methods showed good results and better performance in noise reduction, segmentation, and lesion partitioning compared with state-of-the-art methods. The scheme includes pre-screening for reducing time cost, denoising for enhancing quality, and clustering for instance segmentation.
IEEE JOURNAL OF BIOMEDICAL AND HEALTH INFORMATICS
(2021)
Article
Biochemical Research Methods
Palak Wadhwa, Kris Thielemans, Nikos Efthimiou, Kristen Wangerin, Nicholas Keat, Elise Emond, Timothy Deller, Ottavia Bertolli, Daniel Deidda, Gaspar Delso, Michel Tohme, Floris Jansen, Roger N. Gunn, William Hallett, Charalampos Tsoumpas
Summary: This research demonstrates successful computational and physical modelling of the PET-MR system for image acquisition, generating images comparable to those from the manufacturer; the new software developments will be integrated into STIR, providing researchers worldwide with opportunities to establish and expand their image reconstruction methods; by modelling all effects within the system matrix, PET images showing the metabolic uptake of administered radiopharmaceuticals were reconstructed accurately.
Article
Computer Science, Information Systems
Kishore Krishnagiri Manoj Doss, Pei En Mion, Yu-Chieh Jill Kao, Tsung-Ter Kuo, Jyh-Cheng Chen
Summary: This study evaluated the performance of the Bruker sequential micro-PET/MRI scanner according to the NEMA NU 4-2008 standards. The results showed that the scanner had high sensitivity, good spatial resolution, low scatter fraction, and good image quality, indicating its suitability for preclinical imaging studies.
Article
Computer Science, Artificial Intelligence
Jiadong Zhang, Zhiming Cui, Caiwen Jiang, Shanshan Guo, Fei Gao, Dinggang Shen
Summary: This article proposes a learning-based method to reconstruct high-dose positron emission tomography (PET) images from low-dose PET images and corresponding total-body computed tomography (CT) images. The proposed hierarchical framework can consistently improve the performance of all body parts and outperforms the state-of-the-art methods in single-photon emission computed tomography (SPET) image reconstruction, with a peak signal-to-noise ratio (PSNR) of 30.6 dB for total-body PET images.
IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS
(2023)
Article
Engineering, Electrical & Electronic
Pablo Galve, Jose Manuel Udias, Alejandro Lopez-Montes, Fernando Arias-Valcayo, Juan Jose Vaquero, Manuel Desco, Joaquin L. Herraiz
Summary: A novel method using super-iterations to exceed the resolution-noise limits in PET imaging was proposed. Improvement of approximately 10% in resolution and recovery coefficient was achieved while maintaining the same noise level. Qualitative results confirmed the enhancement in image quality from the proposed method.
IEEE TRANSACTIONS ON COMPUTATIONAL IMAGING
(2021)
Article
Medicine, General & Internal
Julian M. M. Rogasch, Frank Hofheinz, Lutz van Heek, Conrad-Amadeus Voltin, Ronald Boellaard, Carsten Kobe
Summary: This review provides a comprehensive summary of important factors influencing quantification and interpretation in positron emission tomography (PET), with a focus on recent developments in PET technology. The study is of significant importance for reassessing the clinical value and interpretation of PET imaging in routine clinical practice.
Article
Computer Science, Interdisciplinary Applications
Robert Twyman, Simon Arridge, Zeljko Kereta, Bangti Jin, Ludovica Brusaferri, Sangtae Ahn, Charles W. Stearns, Brian F. Hutton, Irene A. Burger, Fotis Kotasidis, Kris Thielemans
Summary: Penalised PET image reconstruction algorithms can be improved by using a relaxed step size sequence and stochastic variance reduction gradient algorithms (SAGA and SVRG) for faster convergence. Numerical studies show that SAGA and SVRG outperform BSREM in terms of voxel value variations and step size selection. These algorithms can achieve faster convergence to the penalised maximum likelihood solution, especially in low count data.
IEEE TRANSACTIONS ON MEDICAL IMAGING
(2023)
Article
Radiology, Nuclear Medicine & Medical Imaging
Scott D. Wollenweber, Adam M. Alessio, Paul E. Kinahan
Summary: This work aims to describe a methodology for comparing small lesion detectability between PET imaging systems and reconstruction algorithms. The proposed approach includes the use of different sized phantoms and a multi-scan data collection method. The results show that system sensitivity and TOF technology have a positive impact on small lesion detectability.
Article
Engineering, Biomedical
Seung Kwan Kang, Jae Sung Lee
Summary: This study proposes a Bowsher prior based on the l(1)-norm and an iteratively reweighting scheme for Anatomy-Guided regularized PET image reconstruction, which overcomes the limitations of the original Bowsher method and improves small lesion detection and contrast enhancement.
PHYSICS IN MEDICINE AND BIOLOGY
(2021)
Editorial Material
Hematology
Andrew Wirth, N. George Mikhaeel
Summary: This study demonstrates that in patients with DLBCL, those who are EOT PET-negative may be spared from radiotherapy, while selected patients with EOT PET-positive sites can achieve outcomes comparable to PET-negative patients when receiving selective treatment.
Article
Engineering, Biomedical
Bao Yang, Long Zhou, Ling Chen, Lijun Lu, Huafeng Liu, Wentao Zhu
Summary: The study presents a hybrid iterative reconstruction method implemented through cycle-consistent learning. By utilizing backprojection and a neural network, the proposed method achieves improved reconstruction accuracy and reduced noise without sacrificing time efficiency.
PHYSICS IN MEDICINE AND BIOLOGY
(2022)
Article
Computer Science, Interdisciplinary Applications
Jinyi Qi, Bangyan Huang
Summary: Positron emission tomography (PET) is widely used in clinical and preclinical applications. This study proposes a new image reconstruction method to generate high-resolution positronium lifetime images using existing time-of-flight (TOF) PET scanners, overcoming the challenge of low spatial resolution. The proposed method allows for better understanding of tissue microenvironment and facilitates the study of disease mechanisms and treatment selection.
IEEE TRANSACTIONS ON MEDICAL IMAGING
(2022)
Article
Computer Science, Information Systems
Joaquin Rives Gambin, Mojtaba Jafari Tadi, Jarmo Teuho, Riku Klen, Juhani Knuuti, Juho Koskinen, Antti Saraste, Eero Lehtonen
Summary: Artifacts in PET scans can affect diagnosis accuracy. PET gating reduces motion blurring but increases noise. The study proposes a deep learning denoising approach using U-Net for cardiac gated PET images. Results show successful noise reduction while maintaining original resolution.
Article
Oncology
Nellie Moshkovich, Humberto J. Ochoa, Binwu Tang, Howard H. Yang, Yuan Yang, Jing Huang, Maxwell P. Lee, Lalage M. Wakefield
Article
Computer Science, Information Systems
Kaue Salvador, Diogo Harmel, Luiz Oliveira, Sergio Cabral, Hugo Almaguer
IEEE LATIN AMERICA TRANSACTIONS
(2020)
Article
Computer Science, Information Systems
M. Guevara, V Cruz, O. Vergara, M. Nandayapa, H. Ochoa, H. Sossa
Summary: Nine new activation functions based on combinations of classical functions such as ReLU and sigmoid were proposed and studied for their effects on CNN performance. Experimental results showed that some of the proposed activation functions outperform classical ones, and using the new functions could increase CNN accuracy by 1.18%.
IEEE LATIN AMERICA TRANSACTIONS
(2021)
Article
Computer Science, Artificial Intelligence
Felipe Arias del Campo, Maria Cristina Guevara Neri, Osslan Osiris Vergara Villegas, Vianey Guadalupe Cruz Sanchez, Humberto de Jesus Ochoa Dominguez, Vicente Garcia Jimenez
Summary: Time Series Classification (TSC) is a complex problem with researchers proposing interesting solutions. This paper introduces a new approach to Multilayer Perceptron (MLP) for TSC, adjusting hyperparameters automatically based on the nature of the time series. Empirical study shows competitive accuracy compared to state-of-the-art methods.
EXPERT SYSTEMS WITH APPLICATIONS
(2021)
Article
Computer Science, Information Systems
W. Valderrama, A. Magadan, O. Vergara, J. Ruiz, R. Pinto, G. Reyes
Summary: This paper presents a methodology to address the issue of facial spoofing attacks in distance education. By utilizing facial biometrics, the proposed method detects spoofing behavior in real-time. It combines the Extended Local Binary Patterns (ELBP) descriptor and YCbCr, HSV color models to enhance detection accuracy. Experimental results demonstrate the feasibility and cost-effectiveness of the approach in uncontrolled camera environments.
IEEE LATIN AMERICA TRANSACTIONS
(2022)
Article
Computer Science, Artificial Intelligence
Ruben Castruita Rodriguez, Carlos Mendoza Carlos, Osslan Osiris Vergara Villegas, Vianey Guadalupe Cruz Sanchez, Humberto de Jesus Ochoa Dominguez
Summary: This paper proposes a new methodology for automatic detection and classification of Mexican traffic signs using deep learning. The proposed method achieves high accuracy and robustness in various scenarios, outperforming existing approaches.
EXPERT SYSTEMS WITH APPLICATIONS
(2022)
Article
Computer Science, Information Systems
Maria Cristina Guevara Neri, Osslan Osiris Vergara Villegas, Vianey Guadalupe Cruz Sanchez, Humberto de Jesus Ochoa Dominguez, Manuel Nandayapa, Juan Humberto Sossa Azuela
Summary: This paper proposes a methodology for the automatic character recognition and revision of solving procedures for linear equations with one unknown. The method involves using a camera to acquire images, preprocessing, character recognition with a convolutional neural network, and applying a comparison rule for revision. Experimental results show a 99% accuracy rate in recognizing handwritten characters and an 86.66% efficiency in revising solving procedures.
INFORMATION PROCESSING & MANAGEMENT
(2023)
Article
Computer Science, Artificial Intelligence
Leandro Jose Rodriguez Hernandez, Humberto de Jesus Ochoa Dominguez, Osslan Osiris Vergara Villegas, Vianey Guadalupe Cruz Sanchez, Juan Humberto Sossa Azuela, Javier Polanco Gonzalez
Summary: In this study, a residual three-dimensional (3D) and convolutional neural network (CNN) was proposed to enhance PET sinograms acquired from a small-animal PET scanner. The results showed that the proposed network improved the spillover ratio by up to 4.5% and the uniformity by 55% compared to the U-Net. The network was tested using NEMA phantom data in a simulation environment and validated on real data from a mouse, yielding visually sharper images.
PATTERN RECOGNITION LETTERS
(2023)
Proceedings Paper
Computer Science, Artificial Intelligence
Leandro Jose Rodriguez Hernandez, Humberto de Jesus Ochoa Dominguez, Osslan Osiris Vergara Villegas, Vianey Guadalupe Cruz Sanchez, Juan Humberto Sossa Azuela, Javier Polanco Gonzalez
Summary: In this study, a 3D convolutional neural network was proposed to enhance sinograms acquired from a small-animal PET scanner. After training and prediction, the network improved the spillover ratio and uniformity of the standard phantom.
PATTERN RECOGNITION, MCPR 2022
(2022)
Article
Computer Science, Information Systems
Felipe Arias Del Campo, Osslan Osiris Vergara Villegas, Vianey Guadalupe Cruz Sanchez, Humberto De Jesus Ochoa Dominguez, Manuel Nandayapa
Summary: This paper proposes the use of a digital camera and radial basis function neural network to measure the color quality of images displayed on an LCD, showing experimental results that are similar to measurements obtained with a spectroradiometer.
Article
Computer Science, Information Systems
Alma Guadalupe Rodriguez Ramirez, Manuel Nandayapa, Osslan Osiris Vergara Villegas, Francesco Garcia Luna