KWKlaus Woertler

Klaus Woertler

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article

Previous Organizations / Institutes

The organizations / institutes from the scholar's other articles in the past ten years


Klaus Woertler has a total of 116 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Promises and limitations of deep learning for predicting knee osteoarthritis progression from medical imaging: A systematic review

    by Amna Gillani, Rüdiger von Eisenhart‐Rothe, Klaus Woertler, Michael T. Hirschmann, Daniel Rueckert, Florian Hinterwimmer

    Purpose: To systematically evaluate the performance, methodological quality, and translational barriers of deep learning (DL) models for predicting knee osteoarthritis (KOA) progression from medical imaging. Methods: Following PRISMA guidelines, we searched PubMed, Scopus, and Web of Science (inception to June 2026) for peer‐reviewed studies applying DL to predict KOA progression from medical imaging. Two reviewers independently screened studies, extracted data, and assessed risk of bias using PROBAST‐AI. The primary outcome was area under the receiver operating characteristic curve (AUC). Narrative synthesis was performed due to heterogeneity in imaging modalities, outcome definitions, and prediction horizons. Results: From 1116 records, 33 studies (2019–2026) met the inclusion criteria, encompassing 50 predictive models. Sample sizes ranged from 340 to 52,981 knees (median: 4298). Nine progression definitions were identified and categorised as structural deterioration ( n = 14), symptomatic worsening ( n = 2), surgical endpoints ( n = 8) or combined outcomes ( n = 9). Nearly all studies (97%) used the Osteoarthritis Initiative dataset for training, and only 27% performed external validation. Surgery prediction models showed the highest descriptive median AUC (0.87), compared with structural (0.78) and symptomatic (0.79) progression models, but reflect non‐disease factors including access to care. MRI‐based models showed higher internal median AUCs (0.85) than radiograph‐based models (0.81), but none underwent external validation. Externally validated models showed performance degradation (median AUC: 0.75 vs. 0.81 internal). Combining imaging modalities did not consistently improve predictions. However, studies leveraging longitudinal trajectories reported higher internal performance. Most studies showed low risk of bias in participant selection and outcome assessment, but 76% had unclear or high risk in the analysis domain. Conclusions: DL models demonstrate proof‐of‐concept for predicting KOA progression but require substantial improvements before clinical deployment. Overcoming translational barriers requires standardised progression definitions integrating structural and symptomatic outcomes, rigorous multi‐site validation, and models that effectively leverage multimodal and longitudinal data. Level of Evidence: N/A.


  • Top Cited Papers
    Open Access

    Biological Sample Collection to Advance Research and Treatment: A Fight Osteosarcoma Through European Research and Euro Ewing Consortium Statement

    by Darrell Green, Roelof van Ewijk, Elisa Tirtei, Dimosthenis Andreou, Fredrik Baecklund, Daniel Baumhoer, Stefan S. Bielack, Rajesh Botchu, Kjetil Boye, Bernadette Brennan, et al.

    Osteosarcoma and Ewing sarcoma are bone tumors mostly diagnosed in children, adolescents, and young adults. Despite multimodal therapy, morbidity is high and survival rates remain low, especially in the metastatic disease setting. Trials investigating targeted therapies and immunotherapies have not been groundbreaking. Better understanding of biological subgroups, the role of the tumor immune microenvironment, factors that promote metastasis, and clinical biomarkers of prognosis and drug response are required to make progress. A prerequisite to achieve desired success is a thorough, systematic, and clinically linked biological analysis of patient samples, but disease rarity and tissue processing challenges such as logistics and infrastructure have contributed to a lack of relevant samples for clinical care and research. There is a need for a Europe-wide framework to be implemented for the adequate and minimal sampling, processing, storage, and analysis of patient samples. Two international panels of scientists, clinicians, and patient and parent advocates have formed the Fight Osteosarcoma Through European Research consortium and the Euro Ewing Consortium. The consortia shared their expertise and institutional practices to formulate new guidelines. We report new reference standards for adequate and minimally required sampling (time points, diagnostic samples, and liquid biopsy tubes), handling, and biobanking to enable advanced biological studies in bone sarcoma. We describe standards for analysis and annotation to drive collaboration and data harmonization with practical, legal, and ethical considerations. This position paper provides comprehensive guidelines that should become the new standards of care that will accelerate scientific progress, promote collaboration, and improve outcomes.


  • Feasibility of Dark-Field Radiography to Enhance Detection of Nondisplaced Fractures

    by Florian Schaff, Christoph Jud, Martin Dierolf, Benedikt Günther, Klaus Achterhold, Bernhard Gleich, Andreas Sauter, Klaus Woertler, Johannes Thalhammer, Felix Meurer, et al.

    In this proof-of-concept study, x-ray dark-field radiography, an imaging technique based on x-ray scattering, improved observer confidence in detecting nondisplaced fractures in a porcine rib model compared with attenuation-based radiography alone.


  • Top Cited Papers

    Multitask Deep Learning for Segmentation and Classification of Primary Bone Tumors on Radiographs

    by Claudio E. von Schacky, Nikolas J. Wilhelm, Valerie S. Schäfer, Yannik Leonhardt, Felix G. Gassert, Sarah C. Foreman, Florian T. Gassert, Matthias Jung, Pia M. Jungmann, Maximilian F. Russe, et al.

    A multitask deep learning model that provided classification of primary bone tumors on radiographs demonstrated performance that surpassed that of radiologic residents and was comparable with that ... Background An artificial intelligence model that assesses primary bone tumors on radiographs may assist in the diagnostic workflow. Purpose To develop a multitask deep learning (DL) model for simul...


  • Ten Commandments for the Diagnosis of Bone Tumors

    by Mark Davies, Radhesh Lalam, Klaus Woertler, Johan L. Bloem, Gunnar Åström

    The diagnosis of tumors and tumorlike lesions of bone is a routine part of both general and specialist radiologic practices. The spectrum of disorders ranges from the small incidental lesion to the potentially life-limiting malignancies whether primary or secondary. In this review, authored by experts from several European orthopaedic oncology centers, we present a collection of pieces of advice in the form of 10 commandments. Adherence in daily practice to this guidance should help minimize adverse patient experiences and outcomes.


  • Local Staging of Soft-Tissue Sarcoma: Emphasis on Assessment of Neurovascular Encasement—Value of MR Imaging in 174 Confirmed Cases

    by Konstantin Holzapfel, Jennifer Regler, Thomas Baum, Hans Rechl, Katja Specht, Bernhard Haller, Ruediger von Eisenhart-Rothe, Reiner Gradinger, Ernst J. Rummeny, Klaus Woertler

    To evaluate utility of magnetic resonance (MR) imaging in local staging of soft-tissue sarcoma, with an emphasis on assessment of neurovascular encasement. Institutional review board approval was obtained; informed consent requirement was waived. Preoperative MR images in 174 patients with soft-tissue sarcoma were analyzed by two readers. Tumor staging according to the American Joint Committee on Cancer/Union International Contre le Cancer and Enneking staging systems and analysis of osseous and articular invasion were performed. To assess neurovascular encasement, contact between tumor and arteries, between tumor and veins, and between tumor and nerves was classified (no contact, contact ≤90°, 91°-180°, 181°-270°, >271°). Interobserver agreement was determined; imaging findings were correlated with intraoperative findings and/or histopathologic findings (Pearson correlation coefficient [r] and Cohen κ coefficient). Intraoperative evaluation and/or histopathologic evaluation confirmed osseous, articular, and neurovascular invasion in 8.6%, 2.9%, and 25.3% of patients. Interobserver agreement was excellent for tumor staging (American Joint Committee on Cancer/Union International Contre le Cancer staging, κ = 0.811; Enneking staging, κ = 0.943) and osseous invasion (κ = 1.000). It was substantial for articular invasion (κ = 0.794). Sensitivity and specificity for osseous invasion were 100% and 98.7%, respectively (both readers). For articular invasion, sensitivity was 80% (both readers); specificities were 100% and 98.8% for readers 1 and 2, respectively. Interobserver agreement in quantifying contact between tumor and vessels and between tumor and nerves was excellent for arteries, veins, and nerves (κ = 0.845, 0.892, 0.893, respectively). Receiver operating characteristic analysis revealed optimal threshold of greater than 180° for prediction of arterial and venous encasement (both readers). For neural encasement, optimal threshold was greater than 180° (reader 1) and greater than 270° (reader 2). Sensitivities in diagnosing encasement for arteries, veins, and nerves were 84.6%, 84.6%, and 77.8% (reader 1) and 84.6%, 84.6%, and 72.2% (reader 2). Specificities for encasement of arteries, veins, and nerves, respectively, were 97.5%, 97.5%, and 93.2% (reader 1) and 93.8%, 94.7%, 97.3% (reader 2). MR imaging allows reliable and accurate local staging of soft-tissue sarcoma. Encasement of arteries, veins, and nerves should be diagnosed, if the contact between tumor and vascular or neural circumference exceeds 180°.


  • Alignment does not influence cartilage T2 in asymptomatic knee joints

    by M. Sauerschnig, J. S. Bauer, L. Kohn, S. Hinterwimmer, S. Landwehr, K. Woertler, P. M. Jungmann, W. Koestler, P. Niemeyer, A. B. Imhoff, et al.

    Purpose: To investigate whether the static knee alignment affects articular cartilage ultrastructures when measured using T2 relaxation among asymptomatic subjects.Methods: Both knee joints (n = 96) of 48 asymptomatic volunteers (26 females, 22 males; 25.4 ± 1.7 years; no history of major knee trauma or surgery) were evaluated clinically (Lysholm, Tegner) and by MRI (hip–knee–ankle angle, standard knee protocol, T2 mapping). Group (n = 4) division was as follows: neutral (4° varus) and valgus (2°–4° valgus) deformity with n = 12 subjects/group; n = 24 knees/group. Regions of interest (ROI) for T2 assessment were placed within full‐thickness cartilage across the whole joint surface and were divided respecting compartmental as well as functional joint anatomy.Results: Leg alignment was 0.7° ± 0.5° varus among neutral, 3.0° ± 0.6° varus among mild varus, 5.0° ± 1.1° varus among severe varus and 2.5° ± 0.7° valgus among valgus group subjects and thus significantly different. No differences between the groups emerged from clinical measures. No morphological pathology was detected in any knee joint. Global T2 values (42.3 ± 2.3; 37.7–47.9 ms) of ROIs placed within every knee joint per subject were not different between alignment groups or between genders, respectively.Conclusion: Static frontal plane leg malalignment does not affect cartilage ultrastructure among young, asymptomatic individuals as measured by T2 quantitative imaging.Level of evidence: Cross‐sectional study, Level II‐III.


  • Ganglia of the tarsal sinus: MR imaging features and clinical findings

    by Jan S. Bauer, Dirk Müller, Martin Sauerschnig, Andreas B. Imhoff, H. Rechl, Ernst J. Rummeny, Klaus Woertler


  • Osteolysis after rotator cuff repair with bioabsorbable anchors

    by H. Pilge, J. Spang, T. Rose, H. Wolter, K. Woertler, A. B. Imhoff

    The current study was designed to evaluate the rate of osteolysis and the clinical and functional outcomes in patients who underwent mini-open rotator cuff repairs using first generation bioabsorbable suture anchors. Our hypothesis was that patients with osteolysis noted on post-repair MRI would have an accompanying decrease in functional and clinical patient outcomes when compared with patients who did not have osteolysis evident on post-repair MRI.


  • Patellar Articular Cartilage Lesions: In Vitro MR Imaging Evaluation after Placement in Gadopentetate Dimeglumine Solution

    by Klaus Woertler, Horst Buerger, Jens Moeller, Ernst J. Rummeny

    To evaluate T1-weighted magnetic resonance (MR) imaging after diffusion of gadopentetate dimeglumine for visualization of articular cartilage lesions. MR imaging was performed in eight human cadaveric patella specimens immediately and 4 hours after placement into a vessel filled with gadopentetate dimeglumine solution (2.5 mmol/L). T1-weighted spin-echo and inversion-recovery turbo spin-echo MR sequences with nulled cartilage signal (inversion time of 300 msec) were used. In a total of 128 articular cartilage areas, MR imaging findings were compared with macroscopic and histopathologic findings. Pathologic evaluation was performed by one musculoskeletal pathologist. With knowledge of pathologic observations, MR images were analyzed by one musculoskeletal radiologist with regard to intrinsic signal intensity characteristics and surface abnormalities of articular cartilage. Histopathologic findings demonstrated 67 areas of normal articular cartilage and 66 cartilage lesions (grade 1, n = 19; grade 2, n = 15; grade 3, n = 26; grade 4, n = 6). All grade 3 and 4 lesions could be identified on MR images obtained immediately after submersion and after 4 hours. Ninety-four percent of grade 1 and 2 lesions were identified as areas of predominantly decreased contrast enhancement on delayed MR images obtained with both sequences. MR images obtained immediately after submersion demonstrated abnormal signal intensity in only 9% and 12% of grade 1 and 2 lesions, respectively. T1-weighted MR images obtained in vitro after gadopentetate dimeglumine diffusion allow demonstration of articular cartilage surface lesions and early stages of cartilage degradation.