Marc-André Weber’s research while affiliated with University Medical Center Rostock and other places

The musculoskeletal region is the main area in terms of easily missed pathologies in the emergency radiology setting, because the majority of diagnoses missed in the emergency setting are fractures. A review of the literature was performed by searching the PubMed and ScienceDirect databases, using the keywords (‘missed injuries’ or ‘missed fractures’) and (‘emergency radiology’ or ‘emergency room’) and (‘musculoskeletal’ or ‘bone’ or ‘skeleton’) for the title and abstract query. The inclusion criteria were scientific papers presented in the English and German languages. Among the 347 relevant hits between 1980 and 2024 as identified by the author of this review article, there were 114 relevant articles from the years between 2018 and 2024. Based on this literature search and the author’s personal experience, this study presents useful information for reducing the number of missed pathologies in the musculoskeletal system in the emergency radiology setting. Predominant factors that make up the majority of missed fractures are ‘subtle but still visible fractures’ and ‘radiographically imperceptible fractures’. Radiologists are able to minimize the factors contributing to fractures being missed. For example, implementing a ‘four-eyes principle’, i.e., two readers read the radiographs, would help to overcome the missing of ‘subtle but still visible fractures’ and the additional use of cross-sectional imaging would help to overcome the missing of ‘radiographically imperceptible fractures’. Knowledge of what is commonly missed and evaluation of high-risk areas with utmost care also increase the diagnostic performance of radiologists.

Subchondral insufficiency fracture (SIF) represents a potentially severe condition that can advance to osteoarthritis, with collapse of the articular surface. SIF manifests as a fracture in bone weakened by non-tumorous disease, precipitated by repetitive physiological stress, without a clear history of major trauma. It is observed along the central weight-bearing region of the femoral condyle, with a higher incidence in the medial femoral condyle, but also in other large weight-bearing synovial joints, such as the femoral head, tibial plateau, or talus. A review of the literature from the past six years was performed by searching PubMed and ScienceDirect databases, using the keywords “subchondral insufficiency fracture” and “spontaneous osteonecrosis of the knee”. The inclusion criteria were scientific papers presented in the English language that reported on the magnetic resonance imaging (MRI) aspects of SIF of the lower limb. Detecting SIF at the level of the hip, knee, and ankle may present challenges both clinically and radiologically. The MRI appearance is dominated by a bone marrow edema-like signal and subchondral bone changes that can sometimes be subtle. Subchondral abnormalities are more specific than the pattern of bone marrow edema-like signal and are best shown on T2-weighted and proton-density-weighted MR images. MRI plays an important role in accurately depicting even subtle subchondral fractures at the onset of the disease and proves valuable in follow-up, prognosis, and the differentiation of SIF from other conditions.

Background As the number of elderly people is rapidly increasing, we are faced with the challenge that diagnostic services are demanded more frequently. At the same time, the number of medical centers and available experts remains almost constant. Tools for diagnostic assistance are urgently needed to improve the effectiveness of healthcare. In four externally funded projects, we will investigate aspects and strategies of how machine learning prototype systems can be translated into functional healthcare applications. Method In the ongoing project “ ExplAInation ” funded by the German research foundation (DFG), we are developing an artificial neural network framework to generate visual and textual explanations to improve the comprehensibility and interpretability of deep learning models. This effort includes participatory research with clinical users. In the complementary project “ TESIComp ” funded by the Federal Ministry of Education and Research (BMBF), we will investigate ethical and social aspects of the emerging field of computational psychiatry. Patients, caregivers, and medical doctors will be interviewed, which will be qualitatively analyzed. Within the German Medical Informatics Initiative, we will contribute to the project “ Open Medical Inference ” ( OMI ), which will develop a network of distributed machine learning evaluation services. We also participate in the international “ Clinical AI‐based Diagnostics ” ( CAIDX ) consortium, which receives funding from the European Interreg Baltic Sea Region program. Result We developed a deep learning application for the detection of dementia atrophy patterns in brain MRI scans. Derived relevance maps highlight diagnostically important brain areas for further evaluation by the radiologists. Interviews with clinicians will provide information on expectations, key requirements and the functional utility of machine learning‐based assistance. Interviews with patients and caregivers will elucidate future changes and challenges in the doctor’s role and responsibilities. The OMI network will allow hospitals to use distributed machine learning evaluation services remotely, without the need of operating all the tools locally. In CAIDX , we will develop best‐practice guidelines for the overarching process of integrating machine learning prototypes and commercial tools into the hospital. Conclusion In “Healthcare 3.0”, the digital transformation will change current diagnostic procedures and roles. Our activities focus on the involved stakeholders, regulatory aspects and implementation strategies to better steer this process.

Background As the number of elderly people is rapidly increasing, we are faced with the challenge that diagnostic services are demanded more frequently. At the same time, the number of medical centers and available experts remains almost constant. Tools for diagnostic assistance are urgently needed to improve the effectiveness of healthcare. In four externally funded projects, we will investigate aspects and strategies of how machine learning prototype systems can be translated into functional healthcare applications. Method In the ongoing project “ExplAInation” funded by the German research foundation (DFG), we are developing an artificial neural network framework to generate visual and textual explanations to improve the comprehensibility and interpretability of deep learning models. This effort includes participatory research with clinical users. In the complementary project “TESIComp” funded by the Federal Ministry of Education and Research (BMBF), we will investigate ethical and social aspects of the emerging field of computational psychiatry. Patients, caregivers, and medical doctors will be interviewed, which will be qualitatively analyzed. Within the German Medical Informatics Initiative, we will contribute to the project “Open Medical Inference” (OMI), which will develop a network of distributed machine learning evaluation services. We also participate in the international “Clinical AI‐based Diagnostics” (CAIDX) consortium, which receives funding from the European Interreg Baltic Sea Region program. Result We developed a deep learning application for the detection of dementia atrophy patterns in brain MRI scans. Derived relevance maps highlight diagnostically important brain areas for further evaluation by the radiologists. Interviews with clinicians will provide information on expectations, key requirements and the functional utility of machine learning‐based assistance. Interviews with patients and caregivers will elucidate future changes and challenges in the doctor’s role and responsibilities. The OMI network will allow hospitals to use distributed machine learning evaluation services remotely, without the need of operating all the tools locally. In CAIDX, we will develop best‐practice guidelines for the overarching process of integrating machine learning prototypes and commercial tools into the hospital. Conclusion In “Healthcare 3.0”, the digital transformation will change current diagnostic procedures and roles. Our activities focus on the involved stakeholders, regulatory aspects and implementation strategies to better steer this process.

Background: Posttreatment recurrence is an unpredictable complication after liver transplant for hepatocellular carcinoma (HCC) that is associated with poor survival. Biomarkers are needed to estimate recurrence risk before organ allocation. Objective: To conduct a proof-of-concept study evaluating use of machine learning (ML) to predict recurrence from pretreatment laboratory, clinical, and MRI data in patients with early-stage HCC initially eligible for liver transplant. Methods: This retrospective study included 120 patients (88 men, 32 women; median age, 60.0 years) diagnosed between June 2005 and March 2018 with early-stage HCC who were initially eligible for liver transplant and underwent treatment by transplant, resection, or thermal ablation. Patients underwent pretreatment MRI and posttreatment imaging surveillance. Imaging features were extracted from postcontrast phases of pretreatment MRI examinations using a pre-trained convolutional neural network. Pretreatment clinical characteristics (including laboratory data) and extracted imaging features were integrated to develop three ML models (clinical model, imaging model, combined model) for predicting recurrence within six time frames ranging from 1 through 6 years after treatment. Kaplan-Meier analysis with time to recurrence as the endpoint was used to assess clinical relevance of model predictions. Results: Tumor recurred in 44/120 (36.7%) patients during follow-up. The three models predicted recurrence with AUCs across the six time frames of 0.60-0.78 (clinical model), 0.71-0.85 (imaging model), and 0.62-0.86 (combined model). Mean AUC was higher for the imaging model than the clinical model (0.76 vs. 0.68, respectively; p=.03), but was not significantly different between the clinical and combined, or between the imaging and combined, models (p>.05). Kaplan-Meier curves were significantly different between patients predicted to be at low- and high-risk by all three models for 2-, 3-, 4-, 5-, and 6-year time frames (p<.05). Conclusion: The findings suggest that ML-based models can predict recurrence before therapy allocation in patients with early-stage HCC initially eligible for liver transplant. Adding MRI data as model input improved predictive performance over clinical parameters alone. The combined model did not surpass the imaging model's performance. Clinical Impact: ML-based models applied to currently underutilized imaging features may help design more reliable criteria for organ allocation and liver transplant eligibility.