Browsing by Author "Clausdorff Fiedler, Hans Jurgen"

Now showing 1 - 4 of 4
  • Thumbnail Image
    Item
    Improving the Generalizability and Performance of an Ultrasound Deep Learning Model Using Limited Multicenter Data for Lung Sliding Artifact Identification
    (2024) Wu, Derek; Smith, Delaney; VanBerlo, Blake; Roshankar, Amir; Lee, Hoseok; Li, Brian; Ali, Faraz; Rahman, Marwan; Basmaji, John; Tschirhart, Jared; Ford, Alex; VanBerlo, Bennett; Durvasula, Ashritha; Vannelli, Claire; Dave, Chintan; Deglint, Jason; Ho, Jordan; Chaudhary, Rushil; Clausdorff Fiedler, Hans Jurgen; Prager, Ross; Millington, Scott; Shah, Samveg; Buchanan, Brian; Arntfield, Robert
    Deep learning (DL) models for medical image classification frequently struggle to generalize to data from outside institutions. Additional clinical data are also rarely collected to comprehensively assess and understand model performance amongst subgroups. Following the development of a single-center model to identify the lung sliding artifact on lung ultrasound (LUS), we pursued a validation strategy using external LUS data. As annotated LUS data are relatively scarce—compared to other medical imaging data—we adopted a novel technique to optimize the use of limited external data to improve model generalizability. Externally acquired LUS data from three tertiary care centers, totaling 641 clips from 238 patients, were used to assess the baseline generalizability of our lung sliding model. We then employed our novel Threshold-Aware Accumulative Fine-Tuning (TAAFT) method to fine-tune the baseline model and determine the minimum amount of data required to achieve predefined performance goals. A subgroup analysis was also performed and Grad-CAM++ explanations were examined. The final model was fine-tuned on one-third of the external dataset to achieve 0.917 sensitivity, 0.817 specificity, and 0.920 area under the receiver operator characteristic curve (AUC) on the external validation dataset, exceeding our predefined performance goals. Subgroup analyses identified LUS characteristics that most greatly challenged the model’s performance. Grad-CAM++ saliency maps highlighted clinically relevant regions on M-mode images. We report a multicenter study that exploits limited available external data to improve the generalizability and performance of our lung sliding model while identifying poorly performing subgroups to inform future iterative improvements. This approach may contribute to efficiencies for DL researchers working with smaller quantities of external validation data.
  • Thumbnail Image
    Item
    Interrater Agreement of Physicians Identifying Lung Sliding Artifact on B-Mode And M-Mode Point of Care Ultrasound (POCUS)
    (2025) Prager, Ross; Clausdorff Fiedler, Hans Jurgen; Smith, Delaney; Wu, Derek; Arntfield, Robert
    Background: Chest point of care ultrasound (POCUS) is a first-line diagnostic test to identify lung sliding, an important artifact to diagnose or rule out pneumothorax. Despite enthusiastic adoption of this modality, the interrater reliability forphysicians to identify lung sliding is unknown. Additionally, the relative diagnostic performance of physicians interpreting B-mode and M-mode ultrasound is unclear. We sought to determine the interrater reliability of physicians to detect lung sliding on B-mode and M-mode POCUS. Methods: We performed a cross-sectional interrater agreement study surveying acute care physicians on their interpretation of 20 B-mode and M-mode POCUS clips. Two experienced clinicians determined the reference standard diagnosis. Respondents reported their interpretation of each POCUS B-mode clip or M-mode image. The primary outcome was the interrater agreement, determined by an intra-class correlation coefficient (ICC). Results: From September to November 2023, there were 20 survey respondents. Fourteen (70%) respondents were resident physicians. Respondents were confident or very confident in their skill performing chest POCUS in 14 (70%) cases, with 19 (90%) performing chest POCUS every week or more frequently. The ICC on B-mode was 0.44 and for M-mode was 0.43, indicating moderate agreement. There were no significant differences in interrater reliability between subgroups of confidence or experience. Conclusion: There is only moderate interrater reliability between clinicians to diagnose lung sliding. Clinicians have superior accuracy on B-mode compared to M-mode clips.
  • No Thumbnail Available
    Item
    Point-of-Care Ultrasound stratified by the Wells Score for the diagnosis of proximal deep vein thrombosis: A Prospective Study
    (2025) Rojas Muñoz, Nicolás; Clausdorff Fiedler, Hans Jurgen; Riquelme Morales, Felipe Ignacio; Vidal Zamorano, Victor Alejandro; Seydewitz Osses, María Francisca; Rivera Gonzalez, Sofía Viviana; Basaure Verdejo, Carlos Eugenio
    Background Deep vein thrombosis (DVT) affects 1 in 1000 people, with complications associated both in under and over diagnosis. Duplex ultrasound is the gold standard but its use in emergency settings is limited. Two-point Point-of-Care ultrasound protocol performed by emergency physicians can foster its diagnosis. However, 6 % of cases can be missed and its performance stratified by clinical pre-test probability is unknown. Objective To evaluate the diagnostic performance of an extended compression ultrasound (ECUS) protocol performed by emergency physicians when stratified by Wells score. Methods We conducted a prospective diagnostic accuracy study. Adult patients (≥18 years) with suspected DVT were stratified by Wells score (low, intermediate, high risk) and underwent ECUS by trained emergency physicians or residents. Results were compared to complete duplex ultrasound (CDUS) performed by radiologists within 24 h. Results Among 194 patients analyzed (54 % female, mean age 61 ± 18 years), the overall prevalence of proximal DVT was 17 %. The ECUS protocol demonstrated a global sensitivity of 97 % (95 % CI: 84.2–99.9), specificity of 94.4 % (95 % CI: 89.7–97.4), positive predictive value of 78.6 % (95 % CI: 63.2–89.4), and negative predictive value of 99.3 % (95 % CI: 96.4–100). In the low-risk group, sensitivity was 100 % (95 % CI: 29.2–100) with a negative predictive value of 100 % (95 % CI: 90.7–100). Conclusion The combined use of Wells score stratification and ECUS can reliably exclude proximal DVT in low and intermediate-risk patients, potentially optimizing emergency department resources and facilitating timely clinical decisions. In low-risk patients, this strategy may yield results comparable to comprehensive Doppler ultrasound.
  • Thumbnail Image
    Item
    Utility analysis of an adapted Mini-CEX WebApp for clinical practice assessment in physiotherapy undergraduate students
    (2023) Fuentes Cimma, Javiera Carolina; Fuentes López, Eduardo; Isbej Esposito, Lorena Pilar; De La Fuente, Cancino Carlos Ignacio; Riquelme Pérez, Arnoldo Javier; Clausdorff Fiedler, Hans Jurgen; Torres Riveros, Gustavo Andrés; Villagrán Gutiérrez, Ignacio Andrés
    Clinical workplace-based learning is essential for undergraduate health professions, requiring adequate training and timely feedback. While the Mini-CEX is a well-known tool for workplace-based learning, its written paper assessment can be cumbersome in a clinical setting. We conducted a utility analysis to assess the effectiveness of an adapted Mini-CEX implemented as a mobile device WebApp for clinical practice assessment. We included 24 clinical teachers from 11 different clinical placements and 95 undergraduate physical therapy students. The adapted Mini-CEX was tailored to align with the learning outcomes of clinical practice requirements and made accessible through a WebApp for mobile devices. To ensure the validity of the content, we conducted a Delphi panel. Throughout the semester, the students were assessed four times while interacting with patients. We evaluated the utility of the adapted Mini-CEX based on validity, reliability, acceptability, cost, and educational impact. We performed factor analysis and assessed the psychometric properties of the adapted tool. Additionally, we conducted two focus groups and analyzed the themes from the discussions to explore acceptability and educational impact. The adapted Mini-CEX consisted of eight validated items. Our analysis revealed that the tool was unidimensional and exhibited acceptable reliability (0.78). The focus groups highlighted two main themes: improving learning assessment and the perceived impact on learning. Overall, the eight-item Mini-CEX WebApp proved to be a valid, acceptable, and reliable instrument for clinical practice assessment in workplace-based learning settings for undergraduate physiotherapy students. We anticipate that our adapted Mini-CEX WebApp can be easily implemented across various clinical courses and disciplines.