CoChat
Sign Up
Cited by 10Osteoarthritis Treatment and MechanismsHip disorders and treatmentsBone and Joint Diseases

Biomarker-Guided Imaging and AI-Augmented Diagnosis of Degenerative Joint Disease

Open access
Kumar, Rahul, Sporn, Kyle, Borole, Aryan, Khanna, Akshay, Gowda, Chirag, Paladugu, Phani, Ngo, Alex, Jagadeesan, Ram +2 more·2025·Diagnostics
DOI 10.3390/diagnostics15111418PMC PMC12154452
Open this paper in CoChat — free accountOpen PDF

Abstract

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

AI snapshot

Written from the paper’s full text.

TL;DR

This review synthesizes evidence on integrating biochemical biomarkers, advanced imaging, and AI for diagnosing degenerative joint disease, especially osteoarthritis. It proposes a multimodal systems-medicine pipeline for earlier detection, patient stratification, and point-of-care monitoring, while emphasizing standardization, validation, and implementation barriers.

Key findings

  • Cartilage degradation markers CTX-II and COMP are highlighted: CTX-II is elevated in urine of knee OA patients, higher in Kellgren-Lawrence grades 3–4 vs. 2, and correlates with MRI-detected structural abnormalities (osteophytes R = 0.330, bone marrow lesions R = 0.252, cartilage degradation R = 0.218, p < 0.05); COMP is a 524 kDa glycoprotein increased in OA and correlated with age, BMI, pain, and IL-1β, with negative association with disease duration.
  • Inflammatory cytokines IL-1β, TNF-α, and IL-6 are described as reinforcing catabolic and inflammatory circuits; IL-1β signals via NF-κB and induces MMP-3/MMP-13 and iNOS/NO, while TNF-α and IL-6 contribute to MMP/ADAMTs production, RANKL/OPG effects, pain, synovial proliferation, and cartilage degradation.
  • Advanced imaging (T2 mapping MRI, ultrasound elastography, dual-energy CT) and AI/radiomics/multimodal learning are reported to capture cartilage composition, mechanical properties, and bone marrow edema-like signal; multimodal T2 mapping MRI plus dual-energy CT showed enhanced capability over single-mode evaluation, though AI limitations include reliance on annotated datasets, bias, and opacity.
  • Point-of-care tools are described, including miniaturized ultrasound (5–15 MHz broadband transducers, 20–30 Hz frame rates), low-field MRI (0.1–0.5 Tesla), and rapid biomarker assays (lateral flow results in 10–20 min with nanogram-range sensitivity; microfluidic and multiplexed biosensors).

Methods

  • Systematic literature search across PubMed, Scopus, and Web of Science for studies published 1 January 2013 to 15 March 2025, combining MeSH and free-text terms for disease, diagnostics, and computational approaches, plus backward citation tracking and expert consultation.
  • Inclusion criteria included original studies or systematic reviews reporting diagnostic biomarkers or imaging findings relevant to osteoarthritis/degenerative joint disease, diagnostic accuracy or biomarker-imaging correlations or AI-assisted image analysis, and human clinical or translational studies; exclusions included non-English publications, conference abstracts/editorials/opinion pieces without primary data, animal-only studies unless highly translational, and therapeutic-only articles. Two reviewers independently screened titles/abstracts, with a third resolving discrepancies.
  • Data extraction covered study design, sample size, diagnostic modality, biomarker/imaging data, diagnostic performance metrics, validation methods, and clinical integration outcomes; AI studies were additionally reviewed for algorithm type, feature extraction, validation strategy, interpretability, and external generalizability. Due to heterogeneity, findings were synthesized narratively into molecular, tissue, imaging, and computational integration themes.

Limitations

  • Stated in the paper: technical standardization of biomarker measurement and imaging acquisition, clinical validation showing incremental value over traditional methods, workflow integration, cost-effectiveness, accessibility, training needs, regulatory pathways, AI generalizability/bias/opacity, and the need for external validation, explainable AI, and regulatory oversight.
  • Stated in the paper: biomarker adoption is limited by inter-laboratory standardization, age/sex/ethnicity-stratified reference ranges, preanalytical variability, low circulating cytokine levels, and assay cross-reactivity/matrix effects; imaging reproducibility depends on standardized acquisition parameters, fitting algorithms, and magnetic field strength.
  • Evident from the text: as a narrative review, it does not report a primary dataset, meta-analysis, or pooled effect sizes; conclusions rely on heterogeneous selected studies, so diagnostic accuracy claims are not independently verified within the paper.

Model-written summary. Check it against the paper before relying on it.

Integrity signals

Registry

Crossref · 10.3390/diagnostics15111418

Cites

167 of 171 references verified3 citation errors · 1 not checkable

Statistics

Nothing to recompute.

Claims the paper makes

70 of 134 cited sources checked · 6 supported · 2 flagged · accuracy not yet measured

  • Premisecitation

    CTX-II correlates with disease severity, being higher in Kellgren-Lawrence grades 3-4 than grade 2.

    “correlating with disease severity (higher in Kellgren-Lawrence grades 3–4 vs. 2)”
    Quote in the paperSubstantial claim2 of 3 sources checked
    • cites [23] Styrkarsdottir, 2023, Cartilage Acidic Protein 1 in Plasma Associates with Prevalent Osteoarthritis and Predicts Future Risk as Well as Progression to Joint Replacements:Results from the UK Biobank Resource, Arthritis Rheumatol. source
    • cites [24] Szilagyi, 2023, Plasma Proteomics Identifies CRTAC1 as a Biomarker for Osteoarthritis Severity and Progression, Rheumatology source

      The Introduction says urine CTX-II is associated with osteoarthritis prevalence and progression, but does not address its relationship with disease severity. The Methods and Results use Kellgren–Lawrence scores to assess severity for other proteins; they do not report CTX-II levels in grades 3–4 versus grade 2 (searched: CTX-II and the grade comparison). (LLM, conf=0.76)

      Experimental — accuracy not yet measured
    • cites [25] Joseph, 2018, Associations Between Molecular Biomarkers and MR-based Cartilage Composition and Knee Joint Morphology: Data from the Osteoarthritis Initiative, Osteoarthr. Cartil. source

      The Results (Subject Characteristics) report no significant association between uCTXII and Kellgren–Lawrence grade. The study included grades 0–3, so it could not establish that CTX-II is higher in grades 3–4 than in grade 2. (LLM, conf=0.92)

      Checked by a reasoning model — may be wrong.

      Experimental — accuracy not yet measured
  • Premisecitation

    Early diagnosis of degenerative joint disease poses a significant clinical challenge because of its gradual onset and symptom overlap with other musculoskeletal disorders.

    “Degenerative joint disease remains a leading cause of global disability, with early diagnosis posing a significant clinical challenge due to its gradual onset and symptom overlap with other musculoskeletal disorders.”
    Quote in the paperSubstantial claim1 of 3 sources checked
    • cites [5] Guermazi, 2009, Plain Radiography and Magnetic Resonance Imaging Diagnostics in Osteoarthritis: Validated Staging and Scoring, J. Bone Jt. Surg. Am. source
    • cites [6] Chu, 2014, Osteoarthritis: From Palliation to Prevention: AOA Critical Issues, J. Bone Jt. Surg. Am. source
    • cites [7] Felson, 2013, Osteoarthritis as a Disease of Mechanics, Osteoarthr. Cartil. source

      The Introduction (page unavailable) says patients with osteoarthritis typically develop symptoms and present for treatment only after considerable joint pathology exists, which bears on the difficulty of recognizing disease early. It does not say that gradual onset or symptom overlap with other musculoskeletal disorders makes early diagnosis a clinical challenge. (LLM, conf=0.76)

      Experimental — accuracy not yet measured
  • Premisecitation

    CTX-II is elevated in urine of knee OA patients.

    “and is elevated in urine of knee OA patients”
    Quote in the paperSubstantial claim2 of 3 sources checked
    • cites [23] Styrkarsdottir, 2023, Cartilage Acidic Protein 1 in Plasma Associates with Prevalent Osteoarthritis and Predicts Future Risk as Well as Progression to Joint Replacements:Results from the UK Biobank Resource, Arthritis Rheumatol. source
    • cites [24] Szilagyi, 2023, Plasma Proteomics Identifies CRTAC1 as a Biomarker for Osteoarthritis Severity and Progression, Rheumatology source

      The Introduction states that urinary CTX-II levels are associated with OA prevalence and progression at the population level. It does not specify knee OA or state that urinary CTX-II is elevated in knee OA patients relative to a comparison group. (LLM, conf=0.84)

      Experimental — accuracy not yet measured
    • cites [25] Joseph, 2018, Associations Between Molecular Biomarkers and MR-based Cartilage Composition and Knee Joint Morphology: Data from the Osteoarthritis Initiative, Osteoarthr. Cartil. source

      The Methods identify CTX-II as a urine biomarker measured in this study, but the copy does not report that its level is elevated in knee OA patients relative to people without OA. The Results (Subject Characteristics; page unavailable) report no significant association between urinary CTX-II and KL grade; the reported association with patellar cartilage T2 does not establish the claimed elevation. (LLM, conf=0.72)

      Experimental — accuracy not yet measured
  • Premisecitation

    A systems medicine approach to degenerative joint disease diagnosis combines many data streams to provide a thorough evaluation of joint health across many biological levels.

    “From molecular pathways to tissue-level alterations, a systems medicine approach to degenerative joint disease diagnosis combines many data streams to provide a thorough evaluation of joint health across many biological levels [8].”
    Quote in the paperSubstantial claim1 of 1 sources checked
    • cites [8] Mobasheri, 2017, The Role of Metabolism in the Pathogenesis of Osteoarthritis, Nat. Rev. Rheumatol. source

      The Conclusions (page unavailable) propose combining metabolomics, proteomics, lipidomics and bioinformatics to investigate osteoarthritis pathways, and describe changes in cartilage, bone and synovium. They do not describe a systems-medicine diagnostic approach that integrates molecular and tissue-level data to provide a comprehensive evaluation of joint health. (LLM, conf=0.84)

      Experimental — accuracy not yet measured
  • Premisecitation

    CTX-II and COMP are biochemical consequences of cartilage breakdown.

    “C-terminal cross-linked telopeptides of type II collagen (CTX-II) and cartilage oligomeric matrix protein (COMP) are among the many biochemical consequences of cartilage breakdown [21].”
    Quote in the paperSubstantial claim1 of 1 sources checked
    • cites [21] Hao, 2019, Cartilage Oligomeric Matrix Protein, C-terminal Cross-linking Telopeptide of Type II Collagen, and Matrix Metalloproteinase-3 as Biomarkers for Knee and Hip Osteoarthritis Diagnosis: A Systematic Review and Meta-analysis, Osteoarthr. Cartil. source

      The Introduction identifies CTX-II as a byproduct of articular cartilage degradation. It describes COMP as a protein found primarily in cartilage and studied as an OA biomarker, but does not say that COMP itself is a consequence of cartilage breakdown; a reference title in the paper describes COMP fragments as markers of degradation. (LLM, conf=0.82)

      Experimental — accuracy not yet measured
  • May combine two claims. 36
  • Classified as probably not a substantial claim. 31
  • The quote only partly states this claim. 1

  • May combine two claims.

    Premisecitation

    CTX-II correlates with MRI-detected structural abnormalities: osteophytes R = 0.330, bone marrow lesions R = 0.252, and cartilage degradation R = 0.218, p < 0.05.

    “and MRI-detected structural abnormalities (osteophytes R = 0.330, bone marrow lesions R = 0.252, cartilage degradation R = 0.218, p < 0.05)”
    • Also stated in: § 4.2. Multi-Modal Data Integration · hedged
      “Significant correlations have been seen, for example, between urine CTX-II levels and MRI-detected structural abnormalities such as osteophytes, bone marrow lesions, and cartilage degradation, suggesting that molecular cartilage degradation products reflect wider joint disease [23].”
    Quote in the paperSubstantial claim2 of 3 sources checked
    • cites [23] Styrkarsdottir, 2023, Cartilage Acidic Protein 1 in Plasma Associates with Prevalent Osteoarthritis and Predicts Future Risk as Well as Progression to Joint Replacements:Results from the UK Biobank Resource, Arthritis Rheumatol. source
    • cites [24] Szilagyi, 2023, Plasma Proteomics Identifies CRTAC1 as a Biomarker for Osteoarthritis Severity and Progression, Rheumatology source

      The Introduction (page unavailable) says urinary CTX-II is associated with osteoarthritis prevalence and progression at the population level. It does not report the claimed MRI correlations or their R and p values. The Methods describe radiographic, rather than MRI, measurements. (LLM, conf=0.82)

      Experimental — accuracy not yet measured
    • cites [25] Joseph, 2018, Associations Between Molecular Biomarkers and MR-based Cartilage Composition and Knee Joint Morphology: Data from the Osteoarthritis Initiative, Osteoarthr. Cartil. source

      The Results (Joint Morphology; page unavailable) say that the only significant biomarker association with bone marrow abnormalities was for sMMP3, not uCTXII, and that no primary biomarker was significantly associated with cartilage WORMS scores. The Results (Cartilage MRI T2; page unavailable) report a different uCTXII association—patellar T2, r=0.19, p=0.04. The cited correlations of 0.330, 0.252, and 0.218 are not reported in these passages. (LLM, conf=0.88)

      Checked by a reasoning model — may be wrong.

      Experimental — accuracy not yet measured

  • May combine two claims.

    Premisecitation

    The present diagnostic paradigm is not sensitive enough to identify early molecular and cellular changes that precede macroscopic tissue degradation, delaying diagnosis and treatment during the most therapeutically sensitive period.

    “Moreover, the present diagnostic paradigm is not sensitive enough to identify early molecular and cellular changes that precede macroscopic tissue degradation, therefore delaying diagnosis and treatment during the most therapeutically sensitive period of illness [6,7].”
    Quote in the paperSubstantial claim1 of 2 sources checked
    • cites [6] Chu, 2014, Osteoarthritis: From Palliation to Prevention: AOA Critical Issues, J. Bone Jt. Surg. Am. source
    • cites [7] Felson, 2013, Osteoarthritis as a Disease of Mechanics, Osteoarthr. Cartil. source

      The Introduction (page unavailable) says that patients typically present for treatment only after considerable joint pathology exists, supporting the limited point that treatment often begins late. It does not say that current diagnostics miss early molecular or cellular changes, that diagnostic insensitivity causes the delay, or that the missed period is the most therapeutically sensitive. (LLM, conf=0.76)

      Experimental — accuracy not yet measured

  • May combine two claims.

    Premisecitation

    CTX-II is a fragment of type II collagen, which constitutes 90% of cartilage collagen, and is cleaved by collagenases.

    “Specifically, CTX-II is a fragment of type II collagen (90% of cartilage collagen) cleaved by collagenases”
    Quote in the paperNot a substantial claim2 of 3 sources checked
    • cites [23] Styrkarsdottir, 2023, Cartilage Acidic Protein 1 in Plasma Associates with Prevalent Osteoarthritis and Predicts Future Risk as Well as Progression to Joint Replacements:Results from the UK Biobank Resource, Arthritis Rheumatol. source
    • cites [24] Szilagyi, 2023, Plasma Proteomics Identifies CRTAC1 as a Biomarker for Osteoarthritis Severity and Progression, Rheumatology source

      The Introduction mentions urinary CTX-II as an osteoarthritis biomarker, but does not define it as a type II collagen fragment or give the 90% figure. The Discussion mentions collagenases in relation to MMP-10, not CTX-II. Searches for CTX-II cleavage and cartilage collagen composition found no passage addressing the claim. (LLM, conf=0.76)

      Experimental — accuracy not yet measured
    • cites [25] Joseph, 2018, Associations Between Molecular Biomarkers and MR-based Cartilage Composition and Knee Joint Morphology: Data from the Osteoarthritis Initiative, Osteoarthr. Cartil. source

      The Methods section (Serum and Urine Biomarkers; page not specified) identifies uCTXII as carboxy-terminal telopeptides of type II collagen, supporting its identification as a collagen fragment. Our copy does not state that type II collagen constitutes 90% of cartilage collagen or that collagenases cleave CTX-II. (LLM, conf=0.86)

      Experimental — accuracy not yet measured

  • May combine two claims.

    Premisecitation

    An integrated framework synthesizing circulating biomarkers, synovial fluid analysis, quantitative imaging parameters, and clinical evaluation can enable clinicians to detect disease earlier, stratify patients by disease subtypes, predict progression trajectories, and monitor therapeutic responses with greater precision.

    “By synthesizing information from circulating biomarkers, synovial fluid analysis, quantitative imaging parameters, and clinical evaluation, an integrated framework can help facilitate holistic orthopedic care and enable clinicians to detect disease earlier, stratify patients according to disease subtypes, predict progression trajectories, and monitor therapeutic responses with greater precision [9”
    Quote in the paperSubstantial claim2 of 3 sources checked
    • cites [9] Roemer, 2011, Advances in Imaging of Osteoarthritis and Cartilage, Radiology source
    • cites [10] Kraus, 2011, Application of Biomarkers in the Development of Drugs Intended for the Treatment of Osteoarthritis, Osteoarthr. Cartil. source

      Section 3.5 (Diagnostic) describes potential early detection and identification of OA subtypes; section 3.3 (Prognostic) describes predicting progression and treatment response. Section 6.1 (General Overview) suggests that combining biomarkers with imaging may eventually inform patient diagnosis, treatment, and monitoring. The paper does not establish that an integrated framework using all four named inputs already enables clinicians to deliver these benefits with greater precision; it presents these applications largely as prospective, particularly in the context of drug development. (LLM, conf=0.84)

      Experimental — accuracy not yet measured
    • cites [11] Bauer, 2006, Classification of Osteoarthritis Biomarkers: A Proposed Approach, Osteoarthr. Cartil. source

      Split into 2 atoms — 0 confirmed, 2 partially supported, 0 disputed, 0 misattributed, 0 unverifiable. Expand the claim to see per-atom verdicts.

      Experimental — accuracy not yet measured

  • May combine two claims.

    Premisecitation

    CTX-II and COMP are helpful as diagnostic-prognostic biomarkers and are quantifiable by ELISA.

    “As diagnostic-prognostic biomarkers and quantifiable by enzyme-linked immunosorbent tests (ELISAs), they are particularly helpful [22].”
    Quote in the paperSubstantial claim1 of 1 sources checked
    • cites [22] Lorenzo, 2017, Quantification of Cartilage Oligomeric Matrix Protein (COMP) and a COMP Neoepitope in Synovial Fluid of Patients with Different Joint Disorders by Novel Automated Assays, Osteoarthr. Cartil. source

      The Introduction (page unavailable) describes COMP as a biomarker for identifying patients at risk of rapid joint destruction, and the Discussion (page unavailable) describes diagnostic and prognostic potential for the COMP assays. The paper does not address CTX-II. Its assay protocols describe automated chemiluminescent immunoassays, not ELISAs, so it does not support the claim about ELISA quantification. (LLM, conf=0.87)

      Experimental — accuracy not yet measured

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation

    Ultrasound elastography measures tissue stiffness with strain or shear wave approaches.

    “By measuring tissue stiffness with strain or shear wave approaches, ultrasound elastography advances this [68].”
    Quote in the paperNot a substantial claim1 of 1 sources checked
    • cites [68] Ishibashi, K., Sasaki, E., Ota, S., Chiba, D., Yamamoto, Y., Tsuda, E., Yoshikuni, S., Ihara, K., and Ishibashi, Y. (2020). Detection of synovitis in early knee osteoarthritis by MRI and serum biomarkers in Japanese general population. Sci. Rep., 10. source

      The Materials and methods section describes measuring synovitis with MRI, not ultrasound elastography. Searches for elastography, tissue stiffness, strain, and shear-wave approaches found no passage addressing how ultrasound elastography works. (LLM, conf=0.75)

      Experimental — accuracy not yet measured

  • May combine two claims.

    Premisecitation

    IL-1β induces iNOS expression, generating more nitric oxide and aggravating cartilage disintegration.

    “By means of chondrocyte mortality, MMP activation, and proteoglycan synthesis inhibition, IL-1β induces iNOS expression, hence generating more nitric oxide (NO) and aggravating cartilage disintegration [41].”
    Quote in the paperNot a substantial claim1 of 1 sources checked
    • cites [41] Wojdasiewicz, 2014, The Role of Inflammatory and Anti-inflammatory Cytokines in the Pathogenesis of Osteoarthritis, Mediat. Inflamm. source

      Section 2.1 reports that IL-1β affects iNOS, which generates nitric oxide, and separately describes IL-1β-driven cartilage deterioration, MMP production, impaired matrix synthesis, and chondrocyte apoptosis (page not available). It does not say that those effects are the means by which IL-1β induces iNOS, or explicitly attribute cartilage disintegration to the resulting nitric oxide. (LLM, conf=0.84)

      Experimental — accuracy not yet measured

  • Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

    Premisecitation

    IL-6 amplifies the cytokine network through classical and trans-signaling involving IL-6R and gp130, activating JAK-STAT pathways.

    “Through classical and trans-signaling involving IL-6R and gp130, IL-6 further amplifies this cytokine network, hence activating JAK-STAT pathways controlling inflammatory gene expression and cellular survival [47].”
    Quote in the paperNot a substantial claim1 of 1 sources checked
    • cites [47] Hunter, 2006, Osteoarthritis, BMJ source

      The Clinical review’s sections on epidemiology and intra-articular steroids discuss osteoarthritis and joint inflammation, but do not address IL-6 signaling. Searches for IL-6R, gp130, classical or trans-signaling, and JAK-STAT found no passage supporting the attributed mechanism in our copy. (LLM, conf=0.76)

      Experimental — accuracy not yet measured

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation

    Multiple cell types, including synoviocytes, chondrocytes, and invading immune cells, synthesize IL-1β in osteoarthritic joints.

    “Multiple cell types, including synoviocytes, chondrocytes, and invading immune cells, synthesize IL-1β in osteoarthritic joints, thereby building a complex network of paracrine and autocrine signals [35].”
    Quote in the paperNot a substantial claim1 of 1 sources checked
    • cites [35] Robinson, 2013, Mechanistic Biomarkers for Clinical Decision Making in Rheumatic Diseases, Nat. Rev. Rheumatol. source

      The Cytokines and chemokines section discusses IL-1 in autoinflammatory diseases, not its production in osteoarthritic joints. Searches for osteoarthritis, IL-1β, synoviocytes, and chondrocytes found no passage identifying the claimed cell types as IL-1β producers or describing the attributed signaling network. (LLM, conf=0.75)

      Experimental — accuracy not yet measured

  • May combine two claims.

    Premisecitation

    High cartilage T2 values predict disease progression and match osteoarthritis risk factors.

    “Clinically, high cartilage T2 values predict disease progression and match osteoarthritis risk factors [61].”
    Quote in the paperSubstantial claim1 of 1 sources checked
    • cites [61] Liebl, 2015, Early T2 Changes Predict Onset of Radiographic Knee Osteoarthritis: Data from the Osteoarthritis Initiative, Ann. Rheum. Dis. source

  • May combine two claims.

    Premisecitation

    COMP correlates with age, BMI, pain, and IL-1β, shows gender differences, and has negative associations with disease duration.

    “correlates with age, BMI, pain, and IL-1β, gender differences, and has negative associations with disease duration [27].”
    Quote in the paperNot a substantial claim1 of 1 sources checked
    • cites [27] Tseng, 2009, Cartilage Oligomeric Matrix Protein (COMP): A Biomarker of Arthritis, Biomark. Insights source

      The Biological Variations of COMP section (page unavailable) reports associations of serum COMP with age and BMI, a lower-limb pain score in OA patients, and gender differences. Our copy does not report a correlation with IL-1β or a negative association with disease duration; searches for those terms found no supporting passage. (LLM, conf=0.88)

      Experimental — accuracy not yet measured

  • May combine two claims.

    Premisecitation

    Mass spectrometry provides greater accuracy for cytokine measurement but requires sophisticated procedures.

    “Although it requires sophisticated procedures, mass spectrometry provides greater accuracy [53].”
    Quote in the paperNot a substantial claim1 of 1 sources checked
    • cites [53] Boffa, A., Merli, G., Andriolo, L., Lattermann, C., Salzmann, G.M., and Filardo, G. (2021). Synovial Fluid Biomarkers in Knee Osteoarthritis: A Systematic Review and Meta-Analysis. J. Clin. Med., 10. source

      The Results discuss cytokines as synovial-fluid biomarkers, and the Discussion calls for improved biomarker assays, but neither passage assesses mass spectrometry. Searches for mass spectrometry, measurement accuracy, and procedural complexity found no passage supporting the claimed trade-off. (LLM, conf=0.76)

      Experimental — accuracy not yet measured

  • May combine two claims.

    Premisecitation

    COMP is released into blood during cartilage turnover and displays increased levels in OA patients.

    “Released into blood during cartilage turnover, it displays increased levels in OA patients”
    Quote in the paperNot a substantial claim1 of 1 sources checked
    • cites [27] Tseng, 2009, Cartilage Oligomeric Matrix Protein (COMP): A Biomarker of Arthritis, Biomark. Insights source

  • May combine two claims.

    Premisecitation

    IL-6 levels in OA fluid correlate with pain, synovial proliferation, and cartilage degradation indicators and control cartilage and bone metabolism.

    “By MMP induction and RANKL/OPG axis control, IL-6 levels in OA fluid correlate with pain, synovial proliferation, and cartilage degradation indicators and control cartilage and bone metabolism [48,49].”
    Quote in the paperNot a substantial claim1 of 2 sources checked
    • cites [48] Pearle, 2007, Elevated High-Sensitivity C-Reactive Protein Levels Are Associated with Local Inflammatory Findings in Patients with Osteoarthritis, Osteoarthr. Cartil. source

      The IL-6 and IL-1β results section (page not available) reports a trend toward correlation between synovial-fluid IL-6 and inflammatory-cell infiltration in a subset of patients with advanced OA. The Discussion links pain and cartilage loss to synovitis or hsCRP, not directly to IL-6 levels. It does not establish correlations with synovial proliferation or cartilage-degradation indicators, or show IL-6 controlling cartilage and bone metabolism through MMPs or the RANKL/OPG axis. (LLM, conf=0.87)

      Experimental — accuracy not yet measured
    • cites [49] Spector, 1997, Low-Level Increases in Serum C-Reactive Protein Are Present in Early Osteoarthritis and Increase with Disease Progression, Arthritis Rheum. source

  • May combine two claims.

    Premisecitation

    Multivariate analysis highlights osteophyte load and synovitis as predictors of CTX-II, suggesting contributions from bone and synovium.

    “with multivariate analysis highlighting osteophyte load and synovitis as predictors, suggesting contributions from bone and synovium [23,24,25].”
    Quote in the paperNot a substantial claim2 of 3 sources checked
    • cites [23] Styrkarsdottir, 2023, Cartilage Acidic Protein 1 in Plasma Associates with Prevalent Osteoarthritis and Predicts Future Risk as Well as Progression to Joint Replacements:Results from the UK Biobank Resource, Arthritis Rheumatol. source
    • cites [24] Szilagyi, 2023, Plasma Proteomics Identifies CRTAC1 as a Biomarker for Osteoarthritis Severity and Progression, Rheumatology source

      The Results (Knee OA) link osteophyte formation to CRTAC1 levels, not CTX-II. The Discussion states that uCTX-II data were unavailable, so the paper could not assess the claimed CTX-II predictors; the Introduction mentions synovial inflammation only as a general feature of osteoarthritis. No page numbers are provided in our copy. (LLM, conf=0.78)

      Experimental — accuracy not yet measured
    • cites [25] Joseph, 2018, Associations Between Molecular Biomarkers and MR-based Cartilage Composition and Knee Joint Morphology: Data from the Osteoarthritis Initiative, Osteoarthr. Cartil. source

      The Results (Cartilage MRI T2; page unavailable) report an association between uCTXII and patellar cartilage T2, not osteophyte load or synovitis. The Methods (Image Analysis; page unavailable) describe scoring cartilage, meniscal and bone-marrow lesions; searches for osteophytes, synovitis and multivariate predictors found no passage supporting the claimed bone and synovium contributions. (LLM, conf=0.78)

      Experimental — accuracy not yet measured

  • May combine two claims.

    Premisecitation

    COMP is a 524 kDa non-collagenous glycoprotein that can be a beneficial biomarker.

    “COMP, a 524 kDa non-collagenous glycoprotein that can also be a beneficial biomarker [26].”
    Quote in the paperNot a substantial claim1 of 1 sources checked
    • cites [26] Kraus, 2001, Serum Cartilage Oligomeric Matrix Protein Reflects the Presence of Clinically Diagnosed Synovitis in Patients with Knee Osteoarthritis, Osteoarthr. Cartil. source

      The Introduction (page unavailable) calls COMP a non-collagenous glycoprotein of high molecular weight (>500 kDa) and a promising candidate biomarker for osteoarthritis. It does not give the specific molecular weight of 524 kDa. The Discussion (page unavailable) also cautions that serum COMP cannot discriminate between individual patients with and without synovitis. (LLM, conf=0.88)

      Experimental — accuracy not yet measured

  • May combine two claims.

    Interpretationcitation

    Together, these approaches enable systems-level insights into disease mechanisms, improve prognostic stratification, and support the development of standardized, clinically actionable biomarkers for degenerative joint disease management.

    “Together, these approaches enable systems-level insights into disease mechanisms, improve prognostic stratification, and support the development of standardized, clinically actionable biomarkers for degenerative joint disease management, despite challenges like technical complexity, data integration, and evolving assay technologies [79].”
    Quote in the paperSubstantial claim1 of 1 sources checked
    • cites [79] Hirvasniemi, 2022, MRI-Based Radiomic Features of the Tibial Bone Can Discriminate Between Knees Without and With Osteoarthritis: Data from the Osteoarthritis Initiative, J. Orthop. Res. source

      The Discussion (page unavailable) says tibial-bone radiomic features could serve as quantitative imaging biomarkers in future studies and could contribute to a more comprehensive view of osteoarthritis-related tissue changes. It does not establish prognostic stratification, standardized clinically actionable biomarkers, or systems-level insight into disease mechanisms; it says prognostic value still needs study and the link between features and biological processes is not always clear. The cited technical, data-integration, and assay-technology challenges are also not supported by these passages. (LLM, conf=0.88)

      Experimental — accuracy not yet measured

  • The quote only partly states this claim.

    Premisecitation

    Portable imaging technologies and fast biomarker assays are revolutionizing diagnosis by bringing point-of-care capabilities into primary care environments, rural clinics, and community screening programs.

    “When combined with mobile health platforms and telemedicine, portable imaging technologies and fast biomarker assays are revolutionizing the way we diagnose degenerative joint disease by directly bringing point-of-care capabilities into primary care environments, rural clinics, and even community screening programs [129].”
    Quote in the paperSubstantial claim1 of 1 sources checked
    • cites [129] Zhang, 2007, EULAR evidence based recommendations for the management of hand osteoarthritis: Report of a Task Force of the EULAR Standing Committee for International Clinical Studies Including Therapeutics (ESCISIT), Ann. Rheum. Dis. source

      The Discussion (page unavailable) identifies concerns about diagnosing hand osteoarthritis but says those issues will be examined later; this paper presents recommendations for management. Searches for portable imaging, rapid biomarker assays, telemedicine, and point-of-care diagnosis found no passage addressing the claimed changes in primary care, rural clinics, or community screening. (LLM, conf=0.76)

      Experimental — accuracy not yet measured

  • May combine two claims.

    Premisecitation

    Single-cell RNA sequencing, ATAC-seq, proteomics, and spatial transcriptomics provide unprecedented resolution of cellular heterogeneity, revealing distinct cell subpopulations and molecular signatures in joint tissues that drive pathogenesis.

    “Single-cell RNA sequencing, ATAC-seq, proteomics, and spatial transcriptomics provide an unprecedented resolution of cellular heterogeneity, revealing distinct cell subpopulations and molecular signatures in joint tissues that drive pathogenesis”
    Quote in the paperNot a substantial claim1 of 2 sources checked
    • cites [159] Zhang, 2024, Spatial Transcriptomics Reveals Cellular Heterogeneity in Osteoarthritic Synovial Tissue, Arthritis Rheumatol.

      Couldn’t be matched to a paper in Crossref, so it can’t be checked.

    • cites [160] Gu, 2023, Single-cell RNA sequencing in osteoarthritis, Cell Prolif. source

      Split into 2 atoms — 1 confirmed, 1 partially supported, 0 disputed, 0 misattributed, 0 unverifiable. Expand the claim to see per-atom verdicts.

      Experimental — accuracy not yet measured

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Interpretationreasoning

    Single-cell technologies offer potential cellular biomarkers for precise disease classification and targeted therapies.

    “offering potential cellular biomarkers for precise disease classification and targeted therapies [159,160].”
    Quote in the paperNot a substantial claim1 of 2 sources checked
    • cites [159] Zhang, 2024, Spatial Transcriptomics Reveals Cellular Heterogeneity in Osteoarthritic Synovial Tissue, Arthritis Rheumatol.

      Couldn’t be matched to a paper in Crossref, so it can’t be checked.

    • cites [160] Gu, 2023, Single-cell RNA sequencing in osteoarthritis, Cell Prolif. source

  • May combine two claims.

    Premisecitation

    A machine learning study for early-stage knee osteoarthritis used XGBoost for clinical features, convolutional neural networks for radiographic features, and demographic data to regularize intercorrelations, and performed better than single-modal strategies.

    “A machine learning study for early-stage knee osteoarthritis used XGBoost to extract clinical features, convolutional neural networks for radiographic features, and demographic data to regularize intercorrelations among many features using L1-norm-based optimization [116]. This method performed better than single-modal strategies, highlighting the need for integrated evaluation [117].”
    Quote in the paperSubstantial claim1 of 2 sources checked
    • cites [116] Tiulpin, A., Klein, S., Bierma-Zeinstra, S.M.A., Thevenot, J., Rahtu, E., van Meurs, J., Oei, E.H.G., and Saarakkala, S. (2019). Multimodal Machine Learning-based Knee Osteoarthritis Progression Prediction from Plain Radiographs and Clinical Data. Sci. Rep., 9. source

      Split into 2 atoms — 1 confirmed, 1 partially supported, 0 disputed, 0 misattributed, 0 unverifiable. Expand the claim to see per-atom verdicts.

      Experimental — accuracy not yet measured
    • cites [117] Almhdie-Imjabbar, A., Toumi, H., and Lespessailles, E. (2024). Machine Learning-Based Classification of Osteoporotic Vertebral Fractures Using Radiomics Features from CT Images. Diagnostics, 14.

      Couldn’t be matched to a paper in Crossref, so it can’t be checked.

  • Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

    Premisecitation

    Federated learning runs via distributed training and safe model change aggregation and is often complemented with methods such as homomorphic encryption or differential privacy.

    “It runs via distributed training and safe model change aggregation and is often complemented with methods such as homomorphic encryption or differential privacy [181].”
    Quote in the paperNot a substantial claim1 of 1 sources checked
    • cites [181] Kaissis, 2020, Secure, Privacy-Preserving and Federated Machine Learning in Medical Imaging, Nat. Mach. Intell. source

  • May combine two claims.

    Premisecitation

    Optical imaging excels in preclinical models but faces clinical translation challenges due to limited tissue penetration.

    “while optical imaging excels in preclinical models but faces clinical translation challenges due to limited tissue penetration [167].”
    Quote in the paperNot a substantial claim1 of 1 sources checked
    • cites [167] Kazemi, 2025, Raman Spectroscopic Probe Provides Optical Biomarkers of Cartilage Composition Predictive of Tissue Function, Osteoarthr. Cartil. source

      The Discussion (page unavailable) describes potential preclinical uses and acknowledges challenges in moving ex-vivo Raman measurements into surgical settings. The Methods report approximately 270 μm of tissue penetration, but the paper does not say that limited penetration causes clinical translation challenges for optical imaging generally, or that optical imaging excels in preclinical models. (LLM, conf=0.83)

      Experimental — accuracy not yet measured

  • Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

    Interpretationcitation

    Together with real-time analytics enabled by GPU/FPGAs, model compression, and edge computing, these developments form a transforming ecosystem in degenerative joint disease management that allows timely, individualized, clinically significant AI interventions.

    “Together with real-time analytics enabled by GPU/FPGAs, model compression, and edge computing, these developments form a transforming ecosystem in degenerative joint disease management that allows timely, individualized, clinically significant AI interventions [184] (Table 2).”
    Quote in the paperNot a substantial claim1 of 1 sources checked
    • cites [184] Ng, 2021, Federated Learning: A Collaborative Effort to Achieve Better Medical Imaging Models for Individual Sites That Have Small Labelled Datasets, Quant. Imaging Med. Surg. source

      The paper’s conclusion (page not available in our copy) discusses federated learning as a way to improve medical-imaging AI models across institutions. Searches for GPU/FPGAs, model compression, edge computing, and individualized interventions found no passage addressing real-time analytics or degenerative joint disease management. (LLM, conf=0.76)

      Experimental — accuracy not yet measured

  • May combine two claims.

    Premisecitation

    Federated learning preserves privacy and increases model resilience through cooperative model training across institutions without exposing raw patient data.

    “Through cooperative model training across institutions without exposing raw patient data, federated learning preserves privacy and increases model resilience, hence improving explainable artificial intelligence [179].”
    Quote in the paperNot a substantial claim1 of 1 sources checked
    • cites [179] Rieke, 2020, The Future of Digital Health with Federated Learning, NPJ Digit. Med. source

      The Introduction and ‘The promise of federated efforts’ describe collaborative training across institutions without moving patient data and a degree of privacy protection. They do not establish an unqualified increase in model resilience: ‘Data heterogeneity’ says broader data may reduce some bias but can also cause federated strategies to fail. ‘Privacy and security’ cautions that model updates can leak private information, so privacy is not guaranteed. (LLM, conf=0.86)

      Experimental — accuracy not yet measured

  • Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

    Interpretationcitation

    Asynchronous updates and local model customization might help address technological difficulties from statistical heterogeneity and communication in federated learning.

    “Although technological difficulties arise from statistical heterogeneity and communication, asynchronous updates and local model customization might help to address these issues [182].”
    Quote in the paperNot a substantial claim1 of 1 sources checked
    • cites [182] Sheller, M.J., Edwards, B., Reina, G.A., Martin, J., Pati, S., Kotrotsou, A., Milchenko, M., Xu, W., Marcus, D., and Colen, R.R. (2020). Federated learning in medicine: Facilitating multi-institutional collaborations without sharing patient data. Sci. Rep., 10. source

      The Introduction (page unavailable) says non-IID institutional data can affect learning, and Methods > Final model selection (page unavailable) discusses communication costs. But the paper describes parallel federated rounds and local model selection, not asynchronous updates or local model customization as ways to address those difficulties. (LLM, conf=0.83)

      Experimental — accuracy not yet measured

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation

    Explainable artificial intelligence (XAI) techniques have become indispensable instruments in medical artificial intelligence systems to solve the intrinsic opacity of deep learning algorithms.

    “Particularly for clinical and regulatory adoption, explainable artificial intelligence (XAI) techniques have become indispensable instruments in medical artificial intelligence systems to solve the intrinsic opacity of deep learning algorithms [174].”
    Quote in the paperNot a substantial claim1 of 1 sources checked
    • cites [174] de Vries, B.M., Zwezerijnen, G.J.C., Burchell, G.L., van Velden, F.H.P., Menke-van der Houven van Oordt, C.W., and Boellaard, R. (2023). Explainable artificial intelligence (XAI) in radiology and nuclear medicine: A literature review. Front. Med., 10. source

      The Introduction says deep-learning algorithms lack transparency and that XAI can mitigate this problem by explaining model decisions. But the Discussion says current XAI methods may not suffice for clinical deployment, and the Conclusion says no single method addresses all concerns. The source therefore supports XAI’s potential importance, not the claim that it has become indispensable or solves opacity. (LLM, conf=0.88)

      Experimental — accuracy not yet measured

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Interpretationcitation

    In joint imaging, attention mechanisms could emphasize radiographic features linked with osteoarthritis.

    “In joint imaging, for instance, attention mechanisms could emphasize radiographic features linked with osteoarthritis”
    Quote in the paperNot a substantial claim2 of 2 sources checked
    • cites [175] Reyes, 2020, On the Interpretability of Artificial Intelligence in Radiology: Challenges and Opportunities, Radiol. Artif. Intell. source

      The Introduction (Triaging; page unavailable) says visualization methods could help check whether radiologic clinical correlates, rather than spurious features, drive an AI system. It does not discuss joint imaging, osteoarthritis, or attention mechanisms emphasizing osteoarthritis-related features (searched: osteoarthritis, joint imaging, attention mechanisms). (LLM, conf=0.82)

      Experimental — accuracy not yet measured
    • cites [176] Zhang, Y., Weng, Y., and Lund, J. (2022). Applications of Explainable Artificial Intelligence in Diagnosis and Surgery. Diagnostics, 12. source

      Section 1.2 lists attention mechanisms as a category of explainability methods, but does not describe their use in joint imaging. Section 3.1 describes LIME emphasizing regions in Parkinson’s SPECT images, not attention mechanisms highlighting osteoarthritis features. Searches for osteoarthritis, joint radiographs, and attention mechanisms found no passage addressing the claimed example; page numbers were not available in our copy. (LLM, conf=0.74)

      Experimental — accuracy not yet measured

  • May combine two claims.

    Premisecitation

    Multiplex immunoassays permit more general profiling but suffer from cross-reactivity and matrix effects, whereas ELISAs offer sensitivity and specificity for particular cytokines.

    “Multiplex immunoassays permit more general profiling but suffer from cross-reactivity and matrix effects, whereas ELISAs offer sensitivity and specificity for particular cytokines [52].”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [52] Brzustewicz, 2015, The role of cytokines in the pathogenesis of rheumatoid arthritis—Practical and potential application of cytokines as biomarkers and targets of personalized therapy, Cytokine source

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation

    IL-1β signaling through NF-κB regulates transcription of inflammatory genes including MMPs, COX-2, and iNOS.

    “Released NF-κB first goes to the nucleus to regulate the transcription of numerous genes associated with inflammation, including those encoding matrix metalloproteinases (MMPs), cyclooxygenase-2 (COX-2), and inducible nitric oxide synthase (iNOS) [38,39].”
    Quote in the paperNot a substantial claim0 of 2 sources checked
    • cites [38] Liu, F., Ding, W., Fang, Y., Yang, J., Zhang, Y., Ding, C., Huang, B., and Wang, K. (2024). Inflammatory Molecular Cluster in Knee Osteoarthritis Derived from Transcriptome and Machine Learning Analysis. Int. J. Mol. Sci., 25.

      Couldn’t be matched to a paper in Crossref, so it can’t be checked.

    • cites [39] Felson, 1988, Obesity and Knee Osteoarthritis. The Framingham Study, Ann. Intern. Med. source

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Interpretationnone given

    The opacity of deep learning algorithms challenges clinical adoption because many models lack interpretable outputs that clinicians can trust.

    “The opacity of deep learning algorithms also challenges clinical adoption, as many models lack interpretable outputs that clinicians can trust.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation

    Feature selection helps mitigate the curse of dimensionality.

    “Using filter methods (correlation, mutual information), wrapper methods (recursive feature elimination, sequential selection), embedded methods (LASSO, elastic net, ridge regression), or dimensionality reduction (PCA, LDA, t-SNE) to maximize predictive performance and lower overfitting, feature selection helps to mitigate the curse of dimensionality [82].”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [82] Jamshidi, 2019, Machine-Learning-Based Patient-Specific Prediction Models for Knee Osteoarthritis, Nat. Rev. Rheumatol. source

  • Classified as probably not a substantial claim.

    Study design and conductnone given

    A systematic literature search was performed across PubMed, Scopus, and Web of Science for studies published from 1 January 2013 to 15 March 2025.

    “A systematic literature search was performed across PubMed, Scopus, and Web of Science databases for studies published from 1 January 2013 to 15 March 2025.”
    Quote in the paperSubstantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing), S4 (based).

    Interpretationnone given

    Model performance can degrade in diverse populations due to bias in training data.

    “Moreover, model performance can degrade in diverse populations due to bias in training data.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Premisecitation

    Longitudinal biomarker studies track molecular changes over time through repeated sampling, using advanced statistical models like mixed-effects modeling and trajectory analysis to capture dynamic patterns related to disease progression and treatment response.

    “Longitudinal biomarker studies track molecular changes over time through repeated sampling, using advanced statistical models like mixed-effects modeling and trajectory analysis to capture dynamic patterns related to disease progression and treatment response”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [161] Thudium, 2018, Development and use of biochemical markers in osteoarthritis: Current update, Curr. Opin. Rheumatol. source

  • Classified as probably not a substantial claim.

    Study design and conductreasoning

    Because of heterogeneity in study designs and outcome metrics, findings were synthesized narratively and organized into thematic categories corresponding to biological scales.

    “Given the heterogeneity in study designs and outcome metrics, findings were synthesized narratively and organized into thematic categories corresponding to biological scales (molecular, tissue, imaging, and computational integration).”
    Quote in the paperSubstantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

    Study design and conductnone given

    Two reviewers independently screened titles and abstracts for eligibility, and discrepancies were resolved by a third reviewer.

    “Two reviewers independently screened titles and abstracts for eligibility. A full-text review was then conducted for shortlisted articles. Discrepancies were resolved by a third reviewer.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing), S4 (based). May combine two claims.

    Interpretationnone given

    Most current AI models rely on high-quality, annotated datasets that may not reflect real-world clinical variability, potentially limiting generalizability.

    “Most current models rely on high-quality, annotated datasets that may not reflect real-world clinical variability, potentially limiting generalizability.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Premisecitation

    PET uses radiopharmaceuticals like 18F-fluorodeoxyglucose to quantify inflammation and 18F-sodium fluoride for bone remodeling.

    “with PET using radiopharmaceuticals like 18F-fluorodeoxyglucose to quantify inflammation, 18F-sodium fluoride for bone remodeling, and others for specific joint degeneration processes [166].”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [166] Teoh, Y.J., Lai, K.W., Usman, J., Goh, S.L., Mohafez, H., Hasikin, K., Su, K., Chu, H.C., Wu, X., and Dhanalakshmi, S. (2024). Exploring the Biomechanical and Biochemical Signatures of Knee Osteoarthritis: Towards Personalized Diagnostics and Targeted Interventions. Diagnostics, 14.

      Couldn’t be matched to a paper in Crossref, so it can’t be checked.

  • May combine two claims.

    Premisecitation

    Ultra-high field MRI at 7 Tesla and above offers improved resolution and new contrast mechanisms, permitting early identification of cartilage and synovial changes.

    “ultra-high field MRI at 7 Tesla and above offers improved resolution and new contrast mechanisms, therefore permitting the early identification of cartilage and synovial changes [171].”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [171] Juras, V., Szomolanyi, P., Serfaty, J.M., Chang, G., Regatte, R.R., and Trattnig, S. (2019). Sodium Imaging at Ultra-High Field MRI: A New Frontier in Musculoskeletal Imaging. NMR Biomed., 32.

      Couldn’t be matched to a paper in Crossref, so it can’t be checked.

  • Probably not a substantial claim: may not pass S2 (checkable), S4 (based).

    Interpretationnone given

    Clinical validation studies have to show the incremental value of advanced diagnostics against traditional methods, proving significant changes in patient outcomes or management choices that support the extra complexity and resource use.

    “Clinical validation studies have to show the incremental value of advanced diagnostics against traditional methods, proving significant changes in patient outcomes or management choices that support the extra complexity and resource use.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Interpretationreasoning

    Artificial intelligence applied to these intricate, multimodal datasets improves diagnostic capabilities, possibly identifying disease signatures undetectable by conventional analysis.

    “By means of automated feature extraction, pattern recognition, and integration of several information sources, artificial intelligence applied to these intricate, multimodal datasets improves diagnostic capabilities, so possibly identifying disease signatures undetectable by conventional analysis.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S2 (checkable), S4 (based).

    Interpretationreasoning

    Integrated diagnostic systems provide a complete characterization of disease state and trajectory by including complementary information from molecular, cellular, tissue, and organ-level assessments rather than depending just on isolated biomarkers or imaging findings.

    “Integrated diagnostic systems provide a complete characterization of disease state and trajectory by including complementary information from molecular, cellular, tissue, and organ-level assessments rather than depending just on isolated biomarkers or imaging findings.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S2 (checkable).

    Interpretationnone given

    Widespread application of advanced diagnostic techniques depends on the technical standardization of both biomarker measurement and imaging acquisition to guarantee result repeatability in many healthcare environments.

    “Widespread application depends on the technical standardization of both biomarker measurement and imaging acquisition, so guarantee result repeatability in many healthcare environments.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S2 (checkable), S3 (load-bearing), S4 (based).

    Interpretationnone given

    Continuous technological innovation, including molecular imaging approaches, ultra-high field MRI, explainable artificial intelligence methods, and federated learning frameworks, promises further refinement of diagnostic capabilities.

    “While structured implementation science will enable efficient translation from research to clinical practice, continuous technological innovation, including molecular imaging approaches, ultra-high field MRI, explainable artificial intelligence methods, and federated learning frameworks, promises further refinement of diagnostic capabilities.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S2 (checkable), S3 (load-bearing), S4 (based).

    Interpretationnone given

    Artificial intelligence algorithms could help to carefully negotiate regulatory paths for multimodal diagnostic approaches, including biomarker testing and imaging assessments, to ensure safety, efficacy, and appropriate clinical application.

    “Artificial intelligence algorithms could help to carefully negotiate regulatory paths for multimodal diagnostic approaches, including biomarker testing and imaging assessments, to ensure safety, efficacy, and appropriate clinical application.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S4 (based).

    Interpretationnone given

    Complementing these molecular markers with new biomarkers like microRNAs and metabolomic signatures provides hope for identifying disease-related changes before structural changes show up on traditional imaging.

    “Complementing these molecular markers with new biomarkers like microRNAs and metabolomic signatures provides hope for identifying disease-related changes before structural changes show up on traditional imaging.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Interpretationreasoning

    By integrating biochemical biomarkers, advanced imaging modalities, and artificial intelligence, clinicians can comprehensively approach degenerative joint disease diagnosis, enabling earlier detection, more precise characterization, and personalized monitoring of disease progression.

    “In all, by integrating biochemical biomarkers, advanced imaging modalities, and artificial intelligence, clinicians can comprehensively approach degenerative joint disease diagnosis, enabling earlier detection, more precise characterization, and personalized monitoring of disease progression.”
    Quote in the paperSubstantial claim

  • Probably not a substantial claim: may not pass S2 (checkable), S4 (based). May combine two claims.

    Interpretationreasoning

    Advancing biomarker-guided imaging and AI-augmented diagnosis of degenerative joint disease will help to provide early intervention, more accurate therapeutic targeting, and finally better outcomes for patients.

    “Advancing biomarker-guided imaging and AI-augmented diagnosis of degenerative joint disease will help to provide early intervention, more accurate therapeutic targeting, and finally better outcomes for the millions of patients afflicted by these crippling diseases worldwide.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S4 (based).

    Premisenone given

    Advanced imaging technologies including T2 mapping MRI, ultrasonic elastography, and dual-energy CT enable the visualization and quantification of subtle tissue compositional changes, providing spatial information that balances the systemic or local concentration measurements of biochemical indicators.

    “Concurrently, advanced imaging technologies including T2 mapping MRI, ultrasonic elastography, and dual-energy CT enable the visualization and quantification of subtle tissue compositional changes, so providing spatial information that balances the systemic or local concentration measurements of biochemical indicators.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Interpretationcitation

    Advanced imaging techniques correlate molecular, functional, and anatomical data and promise earlier detection, improved patient stratification, and enhanced therapeutic monitoring, guiding individualized treatment strategies for degenerative joint diseases.

    “Although they are mostly research tools, these advanced imaging techniques—which correlate molecular, functional, and anatomical data—promise earlier detection, improved patient stratification, and enhanced therapeutic monitoring, so guiding individualized treatment strategies for degenerative joint diseases despite obstacles including increased costs, technical complexity, and radiation exposure”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites [173] Subburaj, 2015, Quantitative Evaluation of Bone and Cartilage Changes Using Multispectral MR Imaging Following Bone Marrow Stimulation in a Porcine Model, J. Magn. Reson. Imaging

      Couldn’t be matched to a paper in Crossref, so it can’t be checked.

  • Classified as probably not a substantial claim.

    Premisereasoning

    Osteoarthritis is a whole-joint condition involving interrelated pathological processes across cartilage, bone, synovium, and supporting structures, rather than a simple cartilage disease.

    “This multiscale view fits the present knowledge of osteoarthritis as a whole-joint condition involving interrelated pathological processes across cartilage, bone, synovium, and supporting structures, rather than a simple cartilage disease.”
    Quote in the paperSubstantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing), S4 (based).

    Premisenone given

    Inflammatory cytokines, including IL-1β and TNF-α, reflect the important function of inflammation in disease pathogenesis.

    “While inflammatory cytokines, including IL-1β and TNF-α, reflect the important function of inflammation in disease pathogenesis”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S4 (based).

    Premisenone given

    Biochemical markers such as CTX-II and COMP offer molecular insights into cartilage degradation mechanisms.

    “biochemical markers such as CTX-II and COMP offer molecular insights into cartilage degradation mechanisms.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Premisecitation

    Functional imaging including blood oxygen level-dependent MRI, diffusion-weighted imaging, and dynamic contrast-enhanced MRI captures dynamic physiological processes including perfusion, tissue microstructure, and oxygenation, providing quantitative biomarkers for disease activity and therapeutic response.

    “Dynamic physiologically processes, including perfusion, tissue microstructure, and oxygenation, are captured by functional imaging, including blood oxygen level-dependent MRI, diffusion-weighted imaging, and dynamic contrast-enhanced MRI, so providing quantitative biomarkers for disease activity and therapeutic response [172].”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites [172] Menendez, 2018, Dynamic Contrast-Enhanced MRI in Murine Arthritis Models: Evaluation of Perfusion and Inflammation, J. Magn. Reson. Imaging

      Couldn’t be matched to a paper in Crossref, so it can’t be checked.

  • May combine two claims.

    Premisecitation

    MRI with targeted nanoparticles combines excellent soft tissue contrast with molecular specificity.

    “MRI with targeted nanoparticles combines excellent soft tissue contrast with molecular specificity [168].”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [168] Aicher, 2014, The Spatial Organisation of Joint Surface Chondrocytes: Review of Its Potential Roles in Tissue Functioning, Disease and Early, Preclinical Diagnosis of Osteoarthritis, Ann. Rheum. Dis. source

  • Probably not a substantial claim: may not pass S2 (checkable), S4 (based).

    Interpretationreasoning

    The diagnosis of degenerative joint disease will be shaped in the future by increasingly tailored evaluations based on individual patient traits and particular clinical questions, together with standardized core measurements.

    “The diagnosis of degenerative joint disease will be shaped in the future by increasingly tailored evaluations based on individual patient traits and particular clinical questions, together with standardized core measurements.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Premisecitation

    Federated models have increased generalizability, ongoing improvement from far-off data sources, and adherence to data security rules like HIPAA and GDPR.

    “Increased generalizability, ongoing improvement from far-off data sources, and adherence to data security rules like HIPAA and GDPR are features of federated models [183].”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [183] Pfitzner, 2021, Federated Learning in a Medical Context: A Systematic Literature Review, ACM Trans. Internet Technol. source

  • Probably not a substantial claim: may not pass S2 (checkable), S4 (based). May combine two claims.

    Interpretationnone given

    Point-of-care applications using portable imaging technologies, fast biomarker assays, and mobile health platforms potentially extend advanced diagnostic capabilities beyond specialized centers to primary care and community settings, improving accessibility, particularly for aging and underprivileged populations.

    “Particularly for aging and underprivileged populations, point-of-care applications using portable imaging technologies, fast biomarker assays, and mobile health platforms potentially extend advanced diagnostic capabilities beyond specialized centers to primary care and community settings, so improving accessibility.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation

    Emerging technologies like photoacoustic imaging and multiparametric MRI further expand capabilities.

    “while emerging technologies like photoacoustic imaging and multiparametric MRI further expand capabilities [169,170].”
    Quote in the paperNot a substantial claim0 of 2 sources checked
    • cites [169] Caravan, 2007, Collagen-Targeted MRI Contrast Agent for Molecular Imaging of Fibrosis, Angew. Chem. Int. Ed. Engl. source
    • cites [170] Lee, 2022, Photoacoustic Imaging for Non-Invasive Assessment of Knee Osteoarthritis: A Preliminary Study, Photoacoustics

      Couldn’t be matched to a paper in Crossref, so it can’t be checked.

60 of 60

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

Each contradiction between two passages listed here was confirmed by an independent check.

  • Telemedicine description cites a retinal-image algorithm

    Minor

    The paper cites [139] for remote assessments by video or messaging, but its own reference entry identifies [139] as a diabetic-retinopathy image-detection study.

  • CT diagnostic claim cites a Parkinson's MRI paper

    Minor

    The cited reference [76] is titled as a study of MRI for Parkinson's disease, not deep learning applied to CT or radiomics for joint diagnosis.

  • Image-registration description cites a macrophage study

    Minor

    Reference [91] is titled as a study of monocyte recruitment and macrophages in a mouse model, rather than spatial registration of images.

3 of 3

Cites

Flagged first, then by citations

7 of the 178 references couldn’t be matched to a known paper.

Research works printed but not linked: 20

  1. Jayadev, 2012, The Influence of Preanalytical Variables on Soluble CD40 Ligand Levels in Patients with Knee Osteoarthritis, Int. J. Rheum. Dis., 15, 191
  2. Valdes, 2014, Large Scale Meta-Analysis of Urinary C-Telopeptide of Type II Collagen (uCTX-II) as a Biomarker of Osteoarthritis and Joint Diseases, Osteoarthr. Cartil., 22, S218
  3. Baumgartner, 2022, Advanced Imaging of Osteoarthritis: Radiography, CT, and Dual Energy CT, Radiol. Clin. N. Am., 60, 613
  4. Wirth, 2017, Layer-Specific Analysis of Femorotibial Cartilage T2 Relaxation Time in Knees with and without Early Osteoarthritis, Osteoarthr. Cartil., 25, 1106
  5. Peuna, 2024, Complexity of Knee Osteoarthritis Pathogenesis: A Comprehensive Review of Clinical, Molecular, and Imaging Insights with a Call for Interdisciplinary Research, Arthritis Res. Ther., 26, 155
  6. Hirvasniemi, 2022, MRI-Based Radiomic Features of the Tibial Bone Can Discriminate Between Knees Without and With Osteoarthritis: Data from the Osteoarthritis Initiative, J. Orthop. Res., 40, 2257
  7. Oei, 2024, Multimodal Machine Learning-Based Knee Osteoarthritis Progression Prediction: Results from the OAI Study, J. Orthop. Res., 42, 1108
  8. Krustev, 2021, A Protocol for the Use of DMM/PTX-Induced Mouse Models of Osteoarthritis and Rheumatoid Arthritis, Curr. Protoc., 1, e288
  9. Rastegar, 2024, Multimodal Machine Learning for Modeling Knee Osteoarthritis Progression Using Radiographic, Metabolic, and Clinical Data, Arthritis Res. Ther., 26, 103
  10. Chang, 2024, Combining Multimodal Imaging to Improve Diagnosis and Prognosis in Knee Osteoarthritis, Osteoarthr. Cartil., 32, 1388
  11. Xu, 2024, Multimodal Machine Learning Model Based on FLAG and 3D-CNN for Diagnosis and Prognosis Prediction in Knee Osteoarthritis, Arthritis Res. Ther., 26, 138
  12. Deveza, 2019, Phenotypes and Biomarkers in Osteoarthritis: Where Are We Now?, Curr. Opin. Rheumatol., 31, 74
  13. Guan, 2024, Clustering Analysis of Knee Osteoarthritis Patients Based on Synovial Fluid Multi-Omics Data, Arthritis Res. Ther., 26, 160
  14. Thudium, 2024, Prediction of Knee Osteoarthritis Progression Using Epidemiological and Molecular Profiles: A Machine Learning Approach, Arthritis Res. Ther., 26, 59
  15. Zoad, 2024, Transforming Healthcare Cybersecurity: A Prescription for Resilience, Cureus, 16, e67249
  16. Zhang, 2024, Spatial Transcriptomics Reveals Cellular Heterogeneity in Osteoarthritic Synovial Tissue, Arthritis Rheumatol., 76, 1087
  17. Xie, 2023, Radiomics Analysis of Cartilage and Subchondral Bone to Differentiate Knees Predisposed to Posttraumatic Osteoarthritis, Osteoarthr. Cartil., 31, 1596
  18. Lee, 2022, Photoacoustic Imaging for Non-Invasive Assessment of Knee Osteoarthritis: A Preliminary Study, Photoacoustics, 26, 100346
  19. Menendez, 2018, Dynamic Contrast-Enhanced MRI in Murine Arthritis Models: Evaluation of Perfusion and Inflammation, J. Magn. Reson. Imaging, 47, 1081
  20. Subburaj, 2015, Quantitative Evaluation of Bone and Cartilage Changes Using Multispectral MR Imaging Following Bone Marrow Stimulation in a Porcine Model, J. Magn. Reson. Imaging, 41, 1669

Cited by

Sorted by citations

Related

Authors
Kumar, R., Sporn, K., Borole, A., Khanna, A., Gowda, C., Paladugu, P., Ngo, A., Jagadeesan, R., Zaman, N., & Tavakkoli, A.
Affiliations
  • University of Miami (US)
  • SUNY Upstate Medical University (US)
  • Department of Medicine, Robert Wood Johnson Medical School, New Brunswick, NJ 08901, USA
  • Rutgers, The State University of New Jersey (US)
  • Cisco Systems (United States) (US)
  • University of Nevada, Reno (US)
Corresponding author
Kumar, Rahul
ORCID
Venue
Diagnostics
Publisher
MDPI AG
Published
2025-06-03
Volume
15
Issue
11
Pages
1418
ISSN
2075-4418 (ISSN-L)
DOI
10.3390/diagnostics15111418
PMC
PMC12154452
OpenAlex
W4410997804
Semantic Scholar
d1817e7b40bceea98adcd2232fc781b42b803eec
Cited by
10
References
178
Citations by year
2026: 8 · 2025: 2
FWCI
6.36
Type
journal article
Language
en
Open access
Licence
Locations
  • Diagnostics (publishedVersion, cc-by)
  • Preprints.org (acceptedVersion, cc-by)
  • PubMed (publishedVersion)
  • PubMed Central (submittedVersion, cc-by)
  • The Medicine Forum (submittedVersion, cc-by)
  • DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
Topics
  • Health Sciences › Medicine › Rheumatology › Osteoarthritis Treatment and Mechanisms
  • Health Sciences › Medicine › Surgery › Hip disorders and treatments
  • Health Sciences › Medicine › Orthopedics and Sports Medicine › Bone and Joint Diseases
Keywords
osteoarthritis, biomarkers, CTX-II, COMP, inflammatory cytokines, ultrasound elastography, dual-energy CT, radiomics, multimodal learning, joint space narrowing, osteophytes, subchondral bone changes, point-of-care testing, early diagnosis, disease progression risk stratification

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Claims read by deepseek-v4.1-flash under definition version 3 (extract-v3) on 2026-10-12.

Classified by jev-1.13-20260917 (gold-guideline-v0.2) on 2026-10-12.

Cited sources are checked by gpt-6-sol.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review