MRI Radiomics-Based Diagnosis of Knee Meniscal Injury DOI
Jing Liao, Ke Yu

Journal of Computer Assisted Tomography, Journal Year: 2025, Volume and Issue: unknown

Published: April 14, 2025

Objective: This study aims to explore a grading diagnostic method for the binary classification of meniscal tears based on magnetic resonance imaging radiomics. We hypothesize that radiomics model can accurately grade injuries in knee joint. By extracting T2-weighted features, was developed distinguish from nontear abnormalities. Materials and Methods: retrospective included data 100 patients at our institution between May 2022 2024. The subjects were with pain or functional impairment, excluding those severe osteoarthritis, infections, cysts, other relevant conditions. randomly allocated training group test 4:1 ratio. Sagittal fat-suppressed sequences utilized extract radiomic features. Feature selection performed using minimum Redundancy Maximum Relevance (mRMR) method, final constructed Least Absolute Shrinkage Selection Operator (LASSO) regression. Model performance evaluated both sets receiver operating characteristic curves, sensitivity, specificity, accuracy. Results: results showed achieved area under curve values 0.95 0.94 sets, respectively, indicating high accuracy distinguishing injury noninjury. In confusion matrix analysis, set 88%, 92%, 87%, while 89%, 82%, 85%, respectively. Conclusions: Our demonstrates abnormalities, providing reliable tool clinical decision-making. Although demonstrated slightly lower specificity set, its overall good capabilities. Future research could incorporate more optimize further improve

Language: Английский

Artificial intelligence-driven radiomics: developing valuable radiomics signatures with the use of artificial intelligence DOI Creative Commons

Konstantinos Vrettos,

Matthaios Triantafyllou,

Kostas Marias

et al.

Deleted Journal, Journal Year: 2024, Volume and Issue: 1(1)

Published: Jan. 1, 2024

Abstract The advent of radiomics has revolutionized medical image analysis, affording the extraction high dimensional quantitative data for detailed examination normal and abnormal tissues. Artificial intelligence (AI) can be used enhancement a series steps in pipeline, from acquisition preprocessing, to segmentation, feature extraction, selection, model development. aim this review is present most AI methods explaining advantages limitations methods. Some prominent architectures mentioned include Boruta, random forests, gradient boosting, generative adversarial networks, convolutional neural transformers. Employing these models process analysis significantly enhance quality effectiveness while addressing several that reduce predictions. Addressing enable clinical decisions wider adoption. Importantly, will highlight how assist overcoming major bottlenecks implementation, ultimately improving translation potential method.

Language: Английский

Citations

4
The value of 2D and 3D MRI texture models in Grade II and III anterior cruciate ligament injuries DOI
Qian Zhang,

Yeyu Xiao,

H. J. Yang

et al.

The Knee, Journal Year: 2025, Volume and Issue: unknown

Published: Feb. 1, 2025

Language: Английский

Citations

0
Combining an improved political optimizer with convolutional neural networks for accurate anterior cruciate ligament tear detection in sports injuries DOI Creative Commons
Wei Hu, Saeid Razmjooy

Scientific Reports, Journal Year: 2025, Volume and Issue: 15(1)

Published: Feb. 26, 2025

A new technique has been developed to identify ACL tears in sports injuries. This method utilizes a Convolutional Neural Network (CNN) combination with modified Political Optimizer (IPO) algorithm, resulting major breakthrough detecting tears. The study provides an innovative approach this type of injury. CNN/IPO surpasses traditional optimization techniques, ensuring precise and timely detection the potential significantly improve treatment results, enabling clinicians intervene promptly effectively, leading enhanced recovery rehabilitation for athletes. integration CNN IPO algorithm unparalleled level accuracy efficiency identifying tears, facilitating more tailored strategies sports-related findings have revolutionize way medical professionals musculoskeletal injuries, enhancing overall well-being athletic performance. research's significance extends beyond medicine, illuminating avenues management paving advancements injury diagnosis treatment.

Language: Английский

Citations

0
Deep Learning Models to Detect Anterior Cruciate Ligament Injury on MRI: A Comprehensive Review DOI Creative Commons
Michele Mercurio,

Federica Denami,

Dimitra Melissaridou

et al.

Diagnostics, Journal Year: 2025, Volume and Issue: 15(6), P. 776 - 776

Published: March 19, 2025

Magnetic resonance imaging (MRI) is routinely used to confirm the suspected diagnosis of anterior cruciate ligament (ACL) injury. Recently, many studies explored role artificial intelligence (AI) and deep learning (DL), a sub-category AI, in musculoskeletal field medical imaging. The aim this study was review current applications DL models detect ACL injury on MRI, thus providing an updated critical synthesis existing literature identifying emerging trends challenges field. A total 23 relevant articles were identified included review. Articles originated from 10 countries, with China having most contributions (n = 9), followed by United State America 4). Throughout article, we analyzed concept tears provided examples how these tools can impact clinical practice patient care. for MRI detection reported high values accuracy, especially helpful less experienced clinicians. Time efficiency also demonstrated. Overall, have proven be valid resource, although still requiring technological developments implementation daily practice.

Language: Английский

Citations

0
MRI Radiomics-Based Diagnosis of Knee Meniscal Injury DOI
Jing Liao, Ke Yu

Journal of Computer Assisted Tomography, Journal Year: 2025, Volume and Issue: unknown

Published: April 14, 2025

Objective: This study aims to explore a grading diagnostic method for the binary classification of meniscal tears based on magnetic resonance imaging radiomics. We hypothesize that radiomics model can accurately grade injuries in knee joint. By extracting T2-weighted features, was developed distinguish from nontear abnormalities. Materials and Methods: retrospective included data 100 patients at our institution between May 2022 2024. The subjects were with pain or functional impairment, excluding those severe osteoarthritis, infections, cysts, other relevant conditions. randomly allocated training group test 4:1 ratio. Sagittal fat-suppressed sequences utilized extract radiomic features. Feature selection performed using minimum Redundancy Maximum Relevance (mRMR) method, final constructed Least Absolute Shrinkage Selection Operator (LASSO) regression. Model performance evaluated both sets receiver operating characteristic curves, sensitivity, specificity, accuracy. Results: results showed achieved area under curve values 0.95 0.94 sets, respectively, indicating high accuracy distinguishing injury noninjury. In confusion matrix analysis, set 88%, 92%, 87%, while 89%, 82%, 85%, respectively. Conclusions: Our demonstrates abnormalities, providing reliable tool clinical decision-making. Although demonstrated slightly lower specificity set, its overall good capabilities. Future research could incorporate more optimize further improve

Language: Английский

Citations

0