Virtual Library
Start Your Search
Calum Macaulay
Author of
-
+
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
-
+
P1.11-10 - Optimizing Radiomics Features by Minimizing Boundary Effects and Normalizing with Opposite Lung Tissue Characteristics (ID 14062)
16:45 - 18:00 | Author(s): Calum Macaulay
- Abstract
Background
For wide adoption of LDCT screening it is thought that CAD will likely by necessary. We hypothesize that CAD features that minimizes perimeter effects and normalizes nodule CT features using the lung parenchyma from the opposite lung will improve the ability to determine nodule malignancy.
a9ded1e5ce5d75814730bb4caaf49419 Method
We have developed a CAD system that includes lung tissue segmentation, nodule detection and feature extraction from the segmented nodule, the segmented nodule minus the perimeter transition pixels (Core), and the opposite lung parenchymal tissue. (See Figure 1).
We use the Mann-Whitney U test to compare teh discriminating ability of individual features and combinations of features) extracted from the nodule, nodule core and nodule normalized by mirror region features.
4c3880bb027f159e801041b1021e88e8 Result
In total, 34 early small suspect baseline nodules detected as part of the PanCan screening trial were used, these include 17 nodules proven to be cancer and 17 nodules that resolved on follow-up scans. The comparison of classification ability of features from nodules without edge vs. nodules with edge pixels reveals that the core features show better classification ability for 76 out of the 136 calculated features.
Performing a leave-one-out LDA classifier cross-validation approach in using core features, gives an accuracy of 76% with only 1 feature through 3 features, and 82% with 4 features. However, repeating the same experiment for core plus edge features, shows accuracy of 67% with only 1 feature, 73% with 3 features, 79% with 4 features. Normalizing the core texture features by the texture features of the mirrored region in the opposite lung shows an improved classification ability for 52 out of the 89 texture features.
8eea62084ca7e541d918e823422bd82e Conclusion
In this study, the results suggest using the nodule core improves feature classification as does normalizing of the nodule by the mirrored region in opposite lung.
6f8b794f3246b0c1e1780bb4d4d5dc53
-
+
P3.09 - Pathology (Not CME Accredited Session) (ID 975)
- Event: WCLC 2018
- Type: Poster Viewing in the Exhibit Hall
- Track:
- Presentations: 1
- Moderators:
- Coordinates: 9/26/2018, 12:00 - 13:30, Exhibit Hall
-
+
P3.09-11 - Genomic Organization at Large Scales (GOALS) within Nuclei and Cell Sociology for Predicting Lung Cancer Outcomes (ID 14160)
12:00 - 13:30 | Presenting Author(s): Calum Macaulay
- Abstract
Background
Accurate prediction of the biological aggressiveness of lung cancers in patients from limited material could have utility with respect to patient treatment planning. We examined the hypothesis that the quantification of large scale DNA organization or GOALS within the nucleus combined with tumour microenvironment as quantified by cell sociology (which cells types are adjacent which cell types) could predict patient outcomes.
a9ded1e5ce5d75814730bb4caaf49419 Method
Tissue (TMA cores) from patients with poor outcomes (lung cancer mortality within 24 months) and good outcomes (5+ year survivors) was stained using a stochiometric DNA stain (Feulgen-Thionin). High resolution brightfield imaging of 29 cores was performed using multiple wavelengths and spectral unmixing used to retrieve the DNA concentration at every pixel. In house software was used to automatically segment all the nuclei within each core, calculate 100+ features for each nucleus and classify each nucleus as epithelial, stromal or immune in origin. Further each nucleus was scored as coming from a patient with poor or good outcome.
For each TMA core the percentage of these cell categories was tabulated as well as cell-cell association frequencies (for example the frequently an epithelial cell predicted to have come from a patient with good outcome was found next to a stromal cell predicted to come from a patient with poor outcome). Cell percentages and cell-cell interaction frequencies were used to predict patient outcome.
4c3880bb027f159e801041b1021e88e8 Result
8eea62084ca7e541d918e823422bd82e Conclusion
Many of the individually calculated features had a statistically significant association with patient outcome. Four+ features could predict outcome with an 79% accuracy, 15+ different pairs of features could predict patient outcome with greater than 85% accuracy. Epithelial- stromal cell interactions and stromal cell – immune cell interactions were particularly predictive of outcome suggesting that microenvironment cell - tumour cell interactions predict future biological activity.
This pilot study suggests that GOALS and cell sociology could predict patient outcomes.
6f8b794f3246b0c1e1780bb4d4d5dc53
