Virtual Library

Start Your Search

Kevin Zhang



Author of

  • +

    P42 - Screening and Early Detection - Risk Modelling and Artificial Intelligence (ID 177)

    • Event: WCLC 2020
    • Type: Posters
    • Track: Screening and Early Detection
    • Presentations: 1
    • Moderators:
    • Coordinates: 1/28/2021, 00:00 - 00:00, ePoster Hall
    • +

      P42.05 - Investigating the Accuracy of Clinical Mathematical Models for Estimating the Probability of Malignancy in Patients With Pulmonary Nodules (ID 1396)

      00:00 - 00:00  |  Presenting Author(s): Kevin Zhang

      • Abstract
      • Slides

      Introduction

      Pulmonary Nodules (PNs) have become increasingly common with the increased use of computed tomography (CT). Over the years, multiple models have been developed for the evaluation of PNs. However, a consensus has not been made on the diagnostic value of these models.

      Methods

      PubMed, Embase, Cochrane Library and Web of Science were searched for studies reporting mathematical models for PN evaluation until March 2020. Eligible models were summarized, and network meta-analysis was performed on externally validated models (PROSPEROdatabase CRD42020154731).

      Results

      Twenty-three original models were proposed in 42 included articles. Age and nodule size were most commonly used in the models, while PET-CT results were used when collected. The Mayo model was validated in 28 studies. The area under the curve values of 4 models (PKU, Brock, Mayo, VA) with sufficient external validation (at least 5 independent cohorts) were 0.830, 0.785, 0.743, and 0.750. Surgical intervention group (SI) models demonstrated better results in SI patients with a pooled sensitivity and specificity of 0.80 (95%CI 0.73-0.86) and 0.75 (95%CI 0.71-0.79) and 0.68 (95%CI 0.50-0.84) and 0.71 (95%CI 0.38-0.93) in screening and outpatient group (SO) patients. For SO models, the pooled specificity decreased from 0.80 (95%CI 0.67-0.90) to 0.68 (95%CI 0.59-0.76), while the sensitivity changed from 0.74 (95%CI 0.65-0.82) to 0.72 (95%CI 0.63-0.80) when the study population changed from SO to SI patients. Artificial intelligence-based models show promising preliminary results.

      Conclusion

      The Mayo model is the most widely used and validated model, while the PKU model yields the best results with certain external validation.A suitable model should be used on appropriate cohorts, but their clinical application needs further investigation.

      Only Active Members that have purchased this event or have registered via an access code will be able to view this content. To view this presentation, please login or select "Add to Cart" and proceed to checkout.