MLMichael Lundemann

Michael Lundemann

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

Michael Lundemann has a total of 122 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Hip fracture detection on radiographs using an artificial intelligence–based support tool: a diagnostic accuracy study

    by Nicholas Bonde, Kristian Kjærgaard, Henriette Aunaas, Stine Hangaard, Cecilie Daugaard, Janus Nybing, Mikael Boesen, Rikke Bachmann, Michael Lundemann, Søren Overgaard

    Background: Assessment of subtle hip fractures on radiographs can be difficult, especially among less experienced emergency physicians, which may prolong the diagnosis and ultimately time to surgery. Clinical artificial intelligence (AI) decision support tools have shown great potential in assisting the detection of fractures on radiographs. Purpose: To investigate how a CE-marked AI fracture detection tool affects junior doctors’ diagnostic accuracy in detecting hip fractures on radiographs. Material and Methods: Eight junior doctors with affiliation to the Accident and Emergency (A&E) department read 246 hip radiographic examinations with and without AI support. The reference standard was determined by two musculoskeletal radiologists, to measure sensitivity and specificity for readers without and with support from the AI tool as well as the AI tool's standalone performance. Results: Mean sensitivity in detecting hip fractures increased significantly from 0.89 (95% confidence interval [CI] = 0.85–0.93) without AI support to 0.94 (95% CI = 0.92–0.97) (χ 2 = 9.27; P = 0.002) with AI support and the false-negative cases was thereby reduced by 49%. There was no significant change in mean specificity 0.90 (95% CI = 0.86–0.93) to 0.91 (95% CI = 0.88–0.94) (χ 2 = 0.34; P = 0.56). The AI standalone performance was 0.99 (95% CI = 0.99–1.00) and 0.73 (95% CI = 0.67–0.80) in sensitivity and specificity, respectively. Conclusion: Out of eight junior doctors, seven detected more fractures with AI assistance than without. The applied performance gain for readers highlights the value of the product.


  • Open Access

    Retrospective Evaluation of an AI System to Classify Negative Musculoskeletal Trauma Radiographs

    by Steffen Czolbe, Jonas Christophersen, Rikke Bachmann, Janitha Mudannayake, Andreas Nexmann, Fabizio Perria, Nikolaus Knauer, Pavel Lisouski, Michael Lundemann

    Study objectives: To evaluate an AI system for musculoskeletal (MSK) radiography in confidently identifying examinations without injury-related pathologies to support AI-guided discharge of patients without traumatic findings. Methods: We retrospectively sampled radiographic examinations, including one or more radiographs and a radiological report, of suspected MSK trauma from > 1,000 clinical sites in two countries. Medically trained professionals independently classified all exams. When disagreements between the classification and the original radiological report arose, adjudication was performed by a third professional. Annotators were also asked to record their confidence level of each classification. The AI system analyzed all exams and assigned them to the categories AI Positive, AI Negative , and AI Negative (very high confidence) . Performance of the AI system was assessed using error rate, false negative rate (FNR), and qualitative review of misclassified exams. Results: A total of 2,962 exams were included. The AI classified 27.6% (818/2,961; 95% CI: 26.0–29.3) of exams as AI Negative (very high confidence) . Of all exams, 0.7% (21/2,962; 95% CI: 0.0–1.0) were falsely classified as highconfidence negatives, corresponding to a false negative rate (FNR) of 2.0% (21/1,026; 95% CI: 0.0–2.9). Qualitative review of false negatives showed that the majority had no clinical consequence if correctly diagnosed during routine follow-up the following day, and no clearly high-risk exams were missed. Conclusion: The AI system identified over one-quarter of MSK trauma radiographs as confidently negative with a very low rate of false negatives, performing on par or better than the reported standard-of-care. These results suggest potential for safe, AI-driven decision support and workflow optimization for the discharge of patients with clearly negative examinations.

  • Open Access

    Protocol for Radiographer x AI led discharge

    by Rikke Bachmann, Andreas Nexmann, Janitha Mudannayake, Janni Jensen, Sigurd S. Lauvsnes, Michael Lundemann

    Introduction: Emergency Department (ED) overcrowding, often exacerbated by prolonged patient length of stay (LOS), is a global challenge. Patients presenting with suspected fractures—many of whom are triaged as low-acuity, contribute significantly to the ED burden. Radiographer-led discharge (RLD), supported by artificial intelligence (AI), presents a potential strategy to streamline care, reduce LOS, and maintain diagnostic safety. Methods: This multi-centre, retrospective study evaluates whether diagnostic radiographers, assisted by the radiological AI decision support tool RBfracture 2 . 6 , can safely discharge patients with no acute skeletal or joint injury. Fifteen radiographers from three countries will independently assess 340 retrospective radiographic examinations (300 consecutive and 40 enriched with rare findings). Referral notes and AI predictions are available. Reference standards are established by consensus among three MSK radiologists/reporting radiographers. Primary outcomes include ED workload reduction (true negatives) and false negative rate. Secondary outcomes will assess AI standalone performance and inter-country comparison. Results: The primary object is to evaluate whether diagnostic radiographers from three different countries, assisted by the AI tool Rbfracture 2.6, can safely reduce the emergency department (ED) workload by discharging patients without acute skeletal or joint injuries, specifically those referred with suspected bone fracture or joint dislocation. A secondary objective is to validate the performance of RBfracture 2.6 in detecting fractures, joint dislocations, elbow effusions, knee effusions, and knee lipohemarthrosis.

  • Open Access

    Improving traumatic fracture detection on radiographs with artificial intelligence support: a multi-reader study

    by Rikke Bachmann, Gozde Gunes, Stine Hangaard, Andreas Nexmann, Pavel Lisouski, Mikael Boesen, Michael Lundemann, Scott G Baginski

    The aim of this study was to evaluate the diagnostic performance of nonspecialist readers with and without the use of an artificial intelligence (AI) support tool to detect traumatic fractures on radiographs of the appendicular skeleton. The design was a retrospective, fully crossed multi-reader, multi-case study on a balanced dataset of patients (≥2 years of age) with an AI tool as a diagnostic intervention. Fifteen readers assessed 340 radiographic exams, with and without the AI tool in 2 different sessions and the time spent was automatically recorded. Reference standard was established by 3 consultant radiologists. Sensitivity, specificity, and false positives per patient were calculated. Patient-wise sensitivity increased from 72% to 80% (P < .05) and patient-wise specificity increased from 81% to 85% (P < .05) in exams aided by the AI tool compared to the unaided exams. The increase in sensitivity resulted in a relative reduction of missed fractures of 29%. The average rate of false positives per patient decreased from 0.16 to 0.14, corresponding to a relative reduction of 21%. There was no significant difference in average reading time spent per exam. The largest gain in fracture detection performance, with AI support, across all readers, was on nonobvious fractures with a significant increase in sensitivity of 11 percentage points (pp) (60%-71%). The diagnostic performance for detection of traumatic fractures on radiographs of the appendicular skeleton improved among nonspecialist readers tested AI fracture detection support tool showed an overall reader improvement in sensitivity and specificity when supported by an AI tool. Improvement was seen in both sensitivity and specificity without negatively affecting the interpretation time. The division and analysis of obvious and nonobvious fractures are novel in AI reader comparison studies like this.


  • Improving Traumatic Fracture Detection on Radiographs with AI Support: A Multi-Reader Study

    by Rikke Bachmann, Pavel Lisouski, Andreas Nexmann, Michael Lundemann, Gozde Gunes, Stine Hangaard, Scott Baginski, Mikael P. Boesen

    Text Abstract Objectives The aim of this study was to evaluate the diagnostic performance of readers using an AI support tool to detect | POSTER

  • Open Access

    Minimal Hip Joint Space Width Measured on X-rays by an Artificial Intelligence Algorithm—A Study of Reliability and Agreement

    by Anne Mathilde Andersen, Benjamin S. B. Rasmussen, Ole Graumann, Søren Overgaard, Michael Lundemann, Martin Haagen Haubro, Claus Varnum, Janne Rasmussen, Janni Jensen

    Minimal joint space width (mJSW) is a radiographic measurement used in the diagnosis of hip osteoarthritis. A large variance when measuring mJSW highlights the need for a supporting diagnostic tool. This study aimed to estimate the reliability of a deep learning algorithm designed to measure the mJSW in pelvic radiographs and to estimate agreement between the algorithm and orthopedic surgeons, radiologists, and a reporting radiographer. The algorithm was highly consistent when measuring mJSW with a mean difference at 0.00. Human readers, however, were subject to variance with a repeatability coefficient of up to 1.31. Statistically, although not clinically significant, differences were found between the algorithm’s and all readers’ measurements with mean measured differences ranging from −0.78 to −0.36 mm. In conclusion, the algorithm was highly reliable, and the mean measured difference between the human readers combined and the algorithm was low, i.e., −0.5 mm bilaterally. Given the consistency of the algorithm, it may be a useful tool for monitoring hip osteoarthritis.


  • Open Access

    Protocol for the AutoRayValid-RBfracture Study: Evaluating the efficacy of an AI fracture detection system

    by Huib Ruitenbeek, Liv Egnell, Katharina Ziegeler, Mathias Willadsen Brejnebøl, Janus Uhd Nybing, Anders Lensskjold, Pavel Klastrup Lisouski, Michael Lundemann, Kay Geert A. Hermann, Mikael Boesen, et al.

    Background Rapidly diagnosing fractures in appendicular skeletons is vital in the ED, where junior physicians often interpret initial radiographs. However, missed fractures remain a concern, prompting AI-assisted detection exploration. Yet, existing studies lack clinical context. We propose a multi-center retrospective study evaluating the AI aid RBfracture™ v.1, aiming to assess AI’s impact on diagnostic thinking by analyzing consecutive cases with clinical data, providing insights into fracture detection and clinical decision-making. Objectives To provide new insights on the potential value of AI tools across borders and different healthcare systems. We will evaluate the performance of the AI aid to detect fractures on conventional x-ray images and how its use could affect handling of these cases in a healthcare setting. In order to explore if the use of a trained and certified AI tool on clinical data exposes new challenges, a daily practice clinical scenario will be approached by minimising selection criteria and using consecutive cases. A multicenter, retrospective, diagnostic accuracy cross-sectional design incorporates clinical context. Methods The multicenter study spans three European sites without onsite hardware. AI system RBfracture™ v.1 maintains consistent sensitivity and specificity thresholds. Eligibility involves age ≥21 with x-ray indications for appendicular fractures. Exclusions include casts, follow-up x-rays, nearby hardware. AI aids retrospective fracture detection. Reader sessions include radiology and emergency care residents and trainees reading with and without AI. Fractures are marked, rated, with expert-established reference standards. Data Sequential patient studies at three sites yield 500 cases per site. Data includes anatomy, referral notes, radiology reports, and radiographic images. Expert readers use annotations, clinical context for standards. Statistical methods include dichotomized confidence ratings, sensitivity, specificity calculations, site-based analysis and subgroup considerations. Reference Standard Two experienced readers annotate fractures; if their annotations overlap by 25% or more, the common area is the reference. Discrepancies are resolved by a local expert. Individual fractures are labelled.


  • Open Access

    A Deep Learning Algorithm for Radiographic Measurements of the Hip in Adults—A Reliability and Agreement Study

    by Janni Jensen, Ole Graumann, Søren Overgaard, Oke Gerke, Michael Lundemann, Martin Haagen Haubro, Claus Varnum, Lene Bak, Janne Rasmussen, Lone B. Olsen, et al.

    Hip dysplasia (HD) is a frequent cause of hip pain in skeletally mature patients and may lead to osteoarthritis (OA). An accurate and early diagnosis may postpone, reduce or even prevent the onset of OA and ultimately hip arthroplasty at a young age. The overall aim of this study was to assess the reliability of an algorithm, designed to read pelvic anterior-posterior (AP) radiographs and to estimate the agreement between the algorithm and human readers for measuring (i) lateral center edge angle of Wiberg (LCEA) and (ii) Acetabular index angle (AIA). The algorithm was based on deep-learning models developed using a modified U-net architecture and ResNet 34. The newly developed algorithm was found to be highly reliable when identifying the anatomical landmarks used for measuring LCEA and AIA in pelvic radiographs, thus offering highly consistent measurement outputs. The study showed that manual identification of the same landmarks made by five specialist readers were subject to variance and the level of agreement between the algorithm and human readers was consequently poor with mean measured differences from 0.37 to 9.56° for right LCEA measurements. The algorithm displayed the highest agreement with the senior orthopedic surgeon. With further development, the algorithm may be a good alternative to humans when screening for HD.


  • Open Access

    Protocol for the AutoRayValid-RBknee Study: a Retrospective, Multicenter, Fully-crossed, Multi-reader, Multi-case Study Investigating the Effect of a Knee Osteoarthritis Severity Classification Model on Reader Diagnostic Accuracy

    by Mathias Willadsen Brejnebøl, Liv Egnell, Michael Lundemann, Anders Lenskjold, Janus Uhd Nybing, Huib Ruitenbeek, Katharina Ziegeler, Jacob Visser, Kay Geert A. Hermann, Edwin H.G. Oei, et al.

    Background: Radiographic evaluation of knee osteoarthritis (KOA) commonly supports clinical findings. Ground truth is difficult to establish and concerns exist on the inter-and intrarater agreement of the findings. RBknee™ is a CE-marked and FDA-cleared AI tool for automatic assessment and reporting of radiographic KOA on standard projection radiographs. Objectives: To investigate how the use of an AI tool affects the accuracy among human readers across three European hospitals in grading the severity of osteoarthritis and associated individual radiographic features. In addition, the performance of the AI tool will also be compared to reference standards established by experts in a stand-alone validation. Methods: In this retrospective multicenter, fully-crossed, multi-reader, multi-case (MRMC) study, the AI support tool RBknee is introduced as a diagnostic intervention. Four Index Readers from each site (two orthopaedic surgeons and two radiologists) will read all studies twice in two runs separated by a washout period of at least four weeks. In both runs, the experiment will be arranged so that the AI-aid will be available for half of the images in the first session and for the second half of the images in the second session. The order of the images will be randomised in order to minimise temporal effects and biases. The primary endpoint is the difference in diagnostic test accuracy for radiographic KOA grading without and with the aid of the AI tool and will be measured as the ordinal weighted accuracy. Data: The data includes radiographic images from 225 studies (unique patients, retrospective data) with weight-bearing bilateral PA/AP and LAT projections of the symptomatic knee(s). Each site contributes to the cohort with 75 studies of which 70 will be consecutive and 5 will be selected to balance the prevalence of radiographic KOA severity. Reference standard: The reference standard will be established based on independent grading by three KOA Reference Experts and adjudicated by majority vote. Where impossible to resolve by majority voting, adjudication will be established by consensus. Index test, AI tool (stand-alone validation): The diagnostic accuracy of RBknee will be tested against the reference standard. Index test, Index Readers: The 12 readers will grade KL on the PA/AP projection and patellar osteophytes on the lateral projection. Administrative information: Title: The trial is titled “AutoRayValid-RBknee”. Protocol version: Revision History Funding: This project has received funding from the European Union’s Horizon 2020 research and innovation programme under grant agreement No 954221 for the EIC SME Instrument project AutoRay. The work only reflects the authors’ view and the European Commission is not responsible for any use that may be made from the information it contains. Roles and responsibilities: Authors’ contributions: MWB, MB, EHGO, JV, and KGH initiated the study design and AL, JUN, KZ and HR helped with implementation. MWB, LE and MJL provided statistical expertise in clinical study design and MWB is conducting the primary statistical analysis. MWB, MJL and LE prepared the manuscript and all authors contributed to its refinement and approved the final manuscript. Sponsor Contact information: Trial Sponsor: Radiobotics ApS Contact name: Liv Egnell Address: Esplanaden 8C, 1263 Copenhagen K, Denmark Email: liv@radiobotics.com