Virtual Library

Start Your Search

M. Ye



Author of

  • +

    MINI 23 - Lung Cancer Risk: Genetic Susceptibility and Airway Biology (ID 135)

    • Event: WCLC 2015
    • Type: Mini Oral
    • Track: Screening and Early Detection
    • Presentations: 1
    • +

      MINI23.09 - Clinical Application of Computer Assistant Diagnostic System in Probe-Based Confocal Laser Endomicroscopy (pCLE) for Pulmonary Diseases (ID 2408)

      16:45 - 18:15  |  Author(s): M. Ye

      • Abstract
      • Presentation
      • Slides

      Background:
      Probe-based confocal laser endomicroscopy (pCLE) allows for real-time noninvasive histological imaging via bronchoscopy. Interpreting pCLE images and correlating with pulmonary disease remains challenging. We performed an in vivo study to evaluate the correlation between pathological diagnosis and pCLE imaging of pulmonary disease.

      Methods:
      We sequentially enrolled the patients with undiagnosed lung lesion, and randomly grouped into control group (TBLB and peripheral EBUS) and pCLE group (TBLB + pCLE and peripheral EBUS + pCLE). pCLE was performed with Cellvizio system (Mauna Kea Technologies, Paris, France). All patients were consent to the procedure. Pathologists and pulmonologists reviewed the images by the Columbus Classification (CC). Questionnaires were applied post the procedure to collect patients’ condition. We developed a computer assistant diagnostic (CAD) system to calculate alveolar diameter, vessel diameter and optical density percentage and compare the CAD diagnostic accuracy with CC standard. The CAD system involved image processing methods to calculate the diameters in pixel domain and then transformed them into the real value. Pseudo-color processing was used to show the density percentage of different tissues. And the histogram was also calculated to figure out the distribution alone gray scale.

      Results:
      258 patients enrolled in the study, 98 under pCLE examination, while 160 under control group. Among them 128 lesions were diagnosed as malignant tumor by pathological diagnosis, 87 cases were diagnosed as benign disease. Primary features were observed in the samples using pCLE in the lesion of cancer: The normal alveolar in malignant nodules is smaller than benign nodules. While, the vessel in the malignant nodules is thicker than the benign ones. The cellular structure and vessel domination in various subtypes of lung cancer is different. There was no significance on procedure time between control and pCLE group, as well as patients’ secretion, tolerance and willing for repeat examination.

      Conclusion:
      pCLE can identify lung carcinoma in in vivo procedure with well tolerance and with limit procedure time. As a non-invasive method, pCLE could improve accuracy and avoid unnecessary biopsy. The Computer Assist Diagnosis system could help pulmonologists to better acquire the right image and to differentiate diseases on the site.

      Only 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, select "Add to Cart" and proceed to checkout. If you would like to become a member of IASLC, please click here.

      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.