Virtual Library

Start Your Search

Lyndsey Pickup



Author of

  • +

    EX04 - Mini Oral Abstract Session - MA08.06, MA18.02, MA19.02, MA20.11 (ID 1006)

    • Event: WCLC 2018
    • Type: Exhibit Showcase
    • Track: Advanced NSCLC
    • Presentations: 1
    • +

      EX04.04 - Automatic Nodule Size Measurements Can Improve Prediction Accuracy Within a Brock Risk Model (ID 14018)

      10:10 - 10:15  |  Author(s): Lyndsey Pickup

      • Abstract
      • Slides

      Background

      The intrinsic variance of manually measured nodule diameters may limit the predictive accuracy of the Brock University Cancer Prediction Model, especially given the relative weight of its coefficient within the model. Size measurements that are automatically derived may improve this prediction performance. This study aims to examine whether automatic nodule segmentation can improve the predictive efficacy of the Brock model.

      a9ded1e5ce5d75814730bb4caaf49419 Method

      A retrospective analysis was performed on all images from the third annual screening study (T2) of the US National Lung Cancer Screening Trial (NLST). Following BTS guidelines, all nodules >=5mm were selected with no other exclusions whether on type, attenuation margin, or otherwise. This resulted in 7551 benign and 314 malignant nodules from 5373 patients with mean age 62±5 years and 3180 males. An automatic segmentation method, based on Deep Learning, was used to segment the nodule volume using a single point placed within the nodule on the CT image as an initialization. The nodule volume, V, was calculated from the segmentation and then converted to an equivalent spherical diameter, Dsph=∛(6/π V).

      We evaluated four implementations: 1) the original Brock model using as input the manual nodule sizes provided in the NLST dataset (BaselineManual); 2) the original Brock model using as input automatically calculated equivalent spherical diameter, Dsph, (BaselineAuto); 3) a Brock model re-fitted to the NLST dataset, using the manual nodule sizes (OptimManual); 4) a Brock model re-fitted to the NLST dataset, using the automatic nodule sizes (OptimAuto).

      Statistics were computed by bootstrapping across 1000 draws without replacement with 70%/30% training/testing per-patient splits. The performance of each combination was measured using Area-Under-the-Receiver-Operating-Curve (AUC-ROC) of predicting nodule malignancy. The relative weighting of size coefficients was also compared.

      4c3880bb027f159e801041b1021e88e8 Result

      The AUC-ROC was 86.5% (95% confidence interval (CI): 83.2, 89.7) for BaselineManual, 87.4% for OptManual (84.3, 90.5), 88.5% for BaselineAuto (85.7, 91.4), and 89.0% for OptAuto (86.2, 91.8). The relative absolute weight of the size coefficient is 0.46 in OptManual (95% CI: 0.42,0.49) increasing to 0.53 (0.49, 0.57) in OptAuto, an increase of 0.07 (0.06, 0.09)

      8eea62084ca7e541d918e823422bd82e Conclusion

      Automatic segmentation appears to improve the prediction accuracy of both the original Brock model and the NLST optimized version. The benefits of repeatable measurements from automatic segmentations are apparent even as direct replacements within the original model, i.e. without re-fitting. However, re-fitting the parameters increases the relative weight of the nodule size coefficient and further improves performance.

      6f8b794f3246b0c1e1780bb4d4d5dc53

      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.

  • +

    MA20 - Implementation of Lung Cancer Screening (ID 923)

    • Event: WCLC 2018
    • Type: Mini Oral Abstract Session
    • Track: Screening and Early Detection
    • Presentations: 1
    • Moderators:
    • Coordinates: 9/25/2018, 15:15 - 16:45, Room 206 F
    • +

      MA20.10 - Lung Cancer Prediction Using Deep Learning Software: Validation on Independent Multi-Centre Data (ID 14022)

      16:20 - 16:25  |  Author(s): Lyndsey Pickup

      • Abstract
      • Presentation
      • Slides

      Background

      Artificial Intelligence software has shown promise in predicting malignancy in indeterminate CT detected pulmonary nodules. This study aimed to assess the accuracy of a convolutional neural network (CNN) based lung cancer prediction software on an independent dataset of indeterminate incidentally detected nodules in a retrospective European multicentre trial.

      a9ded1e5ce5d75814730bb4caaf49419 Method

      The software was trained using the US National Lung Screening Trial (NLST) dataset which was manually curated, such that each reported nodule and cancer was located, contoured and diagnostically characterised (9310 benign nodule patients; 1058 cancer patients). From this complete dataset, a training set was built by selecting all patients with solid and part-solid lesions of 6mm and above, where benign nodules and cancers could be confidently identified by clinicians (5972 patients, of which 575 were cancer patients). A CNN classifier was trained using Deep Learning on this data to produce a malignancy score per nodule. We defined a benign nodule rule-out test by calculating thresholds on the malignancy score that achieve 100% and 99.5% sensitivity on the NLST data.

      The study was set up so that a malignancy score for each nodule was generated. Overall performance was evaluated using Area-Under-the-ROC-Curve analysis (AUC) and rule-out performance measured the specificity at the two thresholds, i.e. the proportion of benign nodules correctly stratified at each threshold.

      There were 2201 nodules, measuring between 5-15mm from 1719 patients from three tertiary referral centres in the UK, Germany and Netherlands. The CT data included heterogeneous scan parameters, scanner manufacturers and clinical indications. Diagnostic ground-truth was established according to Fleischner or British Thoracic Society guidelines. The dataset contained 222 unique cancers from 215 patients.

      4c3880bb027f159e801041b1021e88e8 Result

      AUC on all-site data was 0.92 (95%CI = 0.89-0.93) and broken down per-site the AUC was 0.97 (Netherlands, n=883, 26 cancers), 0.93 (UK, n= 698, 51 cancers), and 0.84 (Germany, 620, 145 cancers).

      The score thresholds used for the target sensitivity of 100% and 99.5% were the same and achieved an overall sensitivity on the data of 99.1% with a specificity of 25.0%. Per-site results were 25.6% (Netherlands), 27.8% (UK) and 20.6% (Germany) specificity with 100%, 100% and 98.6% sensitivity respectively.

      8eea62084ca7e541d918e823422bd82e Conclusion

      Performance of the AI software on independent European multicentre data was comparable to that achieved on the NLST training data, although there was some variability in the performance of the system across the three centres, potentially providing an opportunity for further optimisation.

      6f8b794f3246b0c1e1780bb4d4d5dc53

      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.

  • +

    P1.11 - Screening and Early Detection (Not CME Accredited Session) (ID 943)

    • Event: WCLC 2018
    • Type: Poster Viewing in the Exhibit Hall
    • Track:
    • Presentations: 1
    • Moderators:
    • Coordinates: 9/24/2018, 16:45 - 18:00, Exhibit Hall
    • +

      P1.11-08 - AI Based Malignancy Prediction of Indeterminate Pulmonary Nodules: Robustness to CT Contrast Media (ID 13904)

      16:45 - 18:00  |  Author(s): Lyndsey Pickup

      • Abstract

      Background

      Artificial Intelligence (AI) based malignancy prediction of indeterminate pulmonary nodules has been previously demonstrated to perform well on screen-detected nodules imaged with low-dose, non-contrast CT. This study aimed to assess the impact of contrast media on the classification performance of such a system.

      a9ded1e5ce5d75814730bb4caaf49419 Method

      A Convolutional Neural Network (CNN) was trained on the US National Lung Screening Trial (NLST), which contained only low-dose non-contrast screening images, selecting all nodules 6mm and greater in size (14761 benign nodules from 5972 patients; 932 cancer from 575 patients). A CNN classifier was trained using Deep Learning on this data to produce a malignancy score per nodule.

      For validation, an independent retrospective dataset of incidentally detected solid nodules was used. None of the patients had a cancer diagnosis within the past 5 years, and all had fewer than 5 nodules. The dataset contained 571 nodules from 505 patients, including 42 cancer from 39 patients.

      A CT was considered to be contrasted if the mode HU value within an ROI placed at the aortic arch was greater than 60HU. This resulted in two groups: the non-contrast group had 313 nodules from 276 patients (16 cancer from 14 patients); the contrast group had 258 nodules from 229 patients (26 cancer from 25 patients). The overall efficacy was assessed using Area-Under-the-ROC-Curve analysis (AUC) for each group.

      4c3880bb027f159e801041b1021e88e8 Result

      The AUC on the non-contrast CT group was 0.96 (95% CI 0.93 to 0.98) and 0.95 (95% CI 0.90 to 0.99) on the contrast CT group. Further analysis revealed that excluding high contrast cases, where the HU in the aortic arch was greater than 300HU (40 nodules from 38 patients; 4 cancer), resulted in an AUC of 0.97 (95% CI 0.92 to 1.00).

      8eea62084ca7e541d918e823422bd82e Conclusion

      The CNN classifier seems to be robust to the presence of contrast media with only a moderate reduction in performance. Excluding cases with high contrast restored the performance, although, with only 38 nodules excluded by this, the result may be not statistically significant. These results indicate that a CNN developed to predict pulmonary nodule malignancy, that has been trained on low-dose, non-contrast enhanced CT images, may be used with CT images with moderate levels of contrast without retraining.

      6f8b794f3246b0c1e1780bb4d4d5dc53