Developing the new diagnostic model by integrating bioinformatics and machine learning for osteoarthritis DOI Creative Commons

Jian Du,

Tian Zhou, Wei Zhang

и другие.

Journal of Orthopaedic Surgery and Research, Год журнала: 2024, Номер 19(1)

Опубликована: Дек. 18, 2024

Osteoarthritis (OA) is a common cause of disability among the elderly, profoundly affecting quality life. This study aims to leverage bioinformatics and machine learning develop an artificial neural network (ANN) model for diagnosing OA, providing new avenues early diagnosis treatment. From Gene Expression Omnibus (GEO) database, we first obtained OA synovial tissue microarray datasets. Differentially expressed genes (DEGs) associated with were identified through utilization Limma package weighted gene co-expression analysis (WGCNA). Subsequently, protein-protein interaction (PPI) employed identify most relevant potential feature ANN diagnostic receiver operating characteristic (ROC) curve constructed evaluate performance model. In addition, expression levels verified using real-time quantitative polymerase chain reaction (qRT-PCR). Finally, immune cell infiltration was performed CIBERSORT algorithm explore correlation between cells. The WGCNA total 72 DEGs related which 12 up-regulated 60 down-regulated. Then, PPI 21 hub genes, three algorithms finally screened four (BTG2, CALML4, DUSP5, GADD45B). based on these genes. AUC training set 0.942, validation 0.850. qRT-PCR results demonstrated significant downregulation BTG2, GADD45 mRNA in samples compared normal samples, while CALML4 level exhibited upregulation. Immune revealed B cells memory, T gamma delta, naive, Plasma cells, CD4 memory resting, NK abnormal activated may be progression OA. GADD45B as good developed, perspective personalized treatment

Язык: Английский

Interpretable machine learning and radiomics in hip MRI diagnostics: comparing ONFH and OA predictions to experts DOI Creative Commons
Tariq Alkhatatbeh,

Ahmad Alkhatatbeh,

Qin Guo

и другие.

Frontiers in Immunology, Год журнала: 2025, Номер 16

Опубликована: Янв. 29, 2025

Purpose Distinguishing between Osteonecrosis of the femoral head (ONFH) and Osteoarthritis (OA) can be subjective vary users with different backgrounds expertise. This study aimed to construct evaluate several Radiomics-based machine learning models using MRI differentiate those two disorders compare their efficacies medical experts. Methods 140 scans were retrospectively collected from electronic records. They split into training testing sets in a 7:3 ratio. Handcrafted radiomics features harvested following careful manual segmentation regions interest (ROI). After thoroughly selecting these features, various have been constructed. The evaluation was carried out receiver operating characteristic (ROC) curves. Then NaiveBayes (NB) selected establish our final Radiomics-model as it performed best. Three expertise diagnosed labeled dataset either OA or ONFH. Their results compared Radiomics-model. Results amount handcrafted 1197 before processing; after selection, only 12 key retained used. User 1 had an AUC 0.632 (95% CI 0.4801-0.7843), 2 recorded 0.565 0.4102-0.7196); while 3 on top 0.880 0.7753-0.9843). On other hand, Radiomics model attained 0.971 0.9298-1.0000); showing greater efficacy than all users. It also demonstrated sensitivity 0.937 specificity 0.885. DCA (Decision Curve Analysis displayed that radiomics-model clinical benefit differentiating Conclusion We successfully constructed evaluated interpretable radiomics-based could distinguish method has ability aid both junior senior professionals precisely diagnose take prompt treatment measures.

Язык: Английский

Процитировано

1
Progress in multi-omics studies of osteoarthritis DOI Creative Commons
Yuanyuan Wei, Qian He, Xiaoyu Zhang

и другие.

Biomarker Research, Год журнала: 2025, Номер 13(1)

Опубликована: Фев. 11, 2025

Язык: Английский

Процитировано

1
Integration of longitudinal load-bearing tissue MRI radiomics and neural network to predict knee osteoarthritis incidence DOI Creative Commons
Tianyu Chen,

Jian Chen,

Hao Liu

и другие.

Journal of Orthopaedic Translation, Год журнала: 2025, Номер 51, С. 187 - 197

Опубликована: Март 1, 2025

Язык: Английский

Процитировано

1
Innovative Diagnostic Approaches for Predicting Knee Cartilage Degeneration in Osteoarthritis Patients: A Radiomics-Based Study DOI Creative Commons
Francesca Angelone, Federica Kiyomi Ciliberti,

Giovanni Paolo Tobia

и другие.

Information Systems Frontiers, Год журнала: 2024, Номер unknown

Опубликована: Сен. 12, 2024

Язык: Английский

Процитировано

6
Substantiation of the Preclinical Stage of Gonarthritis. Prospects for Early Chondroprotective Therapy DOI Creative Commons
Е. В. Аршин, DANIL A. GORSHKOV, MIKHAIL A RADOSHCHEKIN

и другие.

Ural Medical Journal, Год журнала: 2025, Номер 24(1), С. 39 - 49

Опубликована: Март 2, 2025

The relevance of the problem . Late diagnosis gonarthritis (GA) based on radiological criteria determines a decrease in effectiveness chondroprotective drugs (CD). aim is to identify early changes hyaline cartilage knee joints and evaluate therapy at an stage disease. Materials methods 186 patients with high risk GA were examined. All signed informed consent. 119 took CD for two years, 67 did not receive therapy. control group consisted 31 healthy people without factors. Initially 2 years later, everyone underwent ultrasound examination knees. dynamics minimum thickness (HC) was evaluated. Results After HC height decreased (2.84±0.16) mm had no statistically significant differences from initial value. In comparison which take CD, by (0.24±0.15) mm, 2.7 times more than receiving 4.8 ( p = 0.01). (0.09±0.12) comparable indicators 0.49). Conclusions GA, initially low determined, its intensive loss noted, compared control. use prevents preclinical stage.

Язык: Английский

Процитировано

0
Radiomics as a new frontier in modern rheumatology: Chest pathology visualization advances and prospects DOI Creative Commons
Т. V. Beketova, Е. Л. Насонов,

M. A. Alekseev

и другие.

Rheumatology Science and Practice, Год журнала: 2025, Номер 63(1), С. 24 - 36

Опубликована: Март 2, 2025

The article discusses the modern trends in development of digital technologies medicine, exemplified by rheumatology, especially, significance radiomics, which combines radiology, mathematical modeling, and deep machine learning. Texture analysis computed tomography images other imaging methods provides a more deeply characterization pathophysiological features tissues can be considered as non-invasive “virtual biopsy”. It is shown that radiomics enhances quality diagnostic predictive modeling. potential application radiomic models for studying predicting chest organ lesions various pathological conditions, including immune mediated inflammatory diseases, systemic vasculitis. Progress diagnosis treatment rheumatic diseases may facilitated integration omics technologies. era, opens up vast prospects advancements will undoubtedly require complex solutions to new technical, legal, ethical challenges.

Язык: Английский

Процитировано

0
A radiomics nomogram based on MRI for differentiating vertebral osteomyelitis from vertebral compression fractures DOI Creative Commons
Hao Xing, Zhe Liu, Zheng Li

и другие.

European Journal of Radiology, Год журнала: 2025, Номер unknown, С. 112106 - 112106

Опубликована: Апрель 1, 2025

This study aims to investigate the value of a radiomics nomogram based on magnetic resonance imaging (MRI) in distinguishing vertebral compression fractures (VCFs) from osteomyelitis (VOs). We conducted retrospective analysis clinical data 100 patients with VCFs and VOs, respectively at our hospital. The cases were randomly divided into training (n = 140) testing sets 60) 7:3 ratio. Two experienced radiologists outlined regions interest (ROI) MRI images using T2-weighted fat suppression (T2WI-FS) extracted radiomic features. Least Absolute Shrinkage Selection Operator (Lasso) algorithm was used select reduce features establish model (Model 1), Logistic Regression construct score. A multivariable logistic regression 2). combined (radiomics nomogram, Model 3) built score independent factors. diagnostic performance Models 1, 2, 3 validated Area Under Curve (AUC) Decision Analysis (DCA). included 68/72 32/28 respectively. There no statistically significant differences characteristics such as age, sex, body mass index (BMI), CRP levels, ESR, lesion stage between (P > 0.05). total 873 6 extracted. After screening, 10 optimal selected build while 5 2. 1 2 create plotted. All three models constructed algorithms. achieved higher AUC than for both sets: 0.946 0.904 0.871 (training) 0.900 0.854 0.818 (testing), Additionally, DCA indicated that had better utility Our features, provides guidance spinal osteomyelitis.

Язык: Английский

Процитировано

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

Journal of Computer Assisted Tomography, Год журнала: 2025, Номер unknown

Опубликована: Апрель 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

Язык: Английский

Процитировано

0
Impact of age on degenerative joint disease of the temporomandibular joint: A systematic review and meta-analysis DOI Creative Commons

Zhiyuan Wu,

Yi‐Sheng Lin,

You-Lai Lin

и другие.

Medicine, Год журнала: 2025, Номер 104(17), С. e41915 - e41915

Опубликована: Апрель 25, 2025

Background: It is unclear that the influence of age on degenerative joint disease (DJD) temporomandibular (TMJ). Methods: Relevant literature was retrieved from PubMed, Elsevier, Web Science, and Google Scholar. EndNote 21 used to consolidate these databases. Key information were extracted included studies, statistical analysis performed using Stata 15.0. The quality studies evaluated cross-sectional study evaluation criteria recommended by Agency for Healthcare Research Quality. Results: A total 11 involving 2832 participants (1099 males, 1744 females) included. incidence DJD TMJ approximately 35% among individuals aged 20 39, 43% those 40 59, 54% 60–69. Conclusion: Age progression a key risk factor development TMJ. increases progressively across different groups, with significant rise observed in middle older groups.

Язык: Английский

Процитировано

0
Biomarker-Guided Imaging and AI-Augmented Diagnosis of Degenerative Joint Disease DOI Creative Commons
Rahul Kumar, Kyle Sporn,

Aryan Borole

и другие.

Diagnostics, Год журнала: 2025, Номер 15(11), С. 1418 - 1418

Опубликована: Июнь 3, 2025

Degenerative joint disease remains a leading cause of global disability, with early diagnosis posing significant clinical challenge due to its gradual onset and symptom overlap other musculoskeletal disorders. This review focuses on emerging diagnostic strategies by synthesizing evidence specifically from studies that integrate biochemical biomarkers, advanced imaging techniques, machine learning models relevant osteoarthritis. We evaluate the utility cartilage degradation markers (e.g., CTX-II, COMP), inflammatory cytokines IL-1β, TNF-α), synovial fluid microRNA profiles, how they correlate quantitative readouts T2-mapping MRI, ultrasound elastography, dual-energy CT. Furthermore, we highlight recent developments in radiomics AI-driven image interpretation assess space narrowing, osteophyte formation, subchondral bone changes high fidelity. The integration these datasets using multimodal approaches offers novel phenotypes stratify patients stage risk progression. Finally, explore implementation tools point-of-care diagnostics, including portable devices rapid biomarker assays, particularly aging underserved populations. By presenting unified pipeline, this article advances future detection personalized monitoring degeneration.

Язык: Английский

Процитировано

0