Abstracts

RETURN TO ABSTRACT LISTING


2163. A Novel CNN Algorithm for Pathological Complete Response Prediction Using the I-SPY TRIAL Breast MRI Database
Authors * Denotes Presenting Author
  1. Shawn Sun *; Columbia University Irving Medical Center/New York Presbyterian
  2. Simuyaki Mutasa; Thomas Jefferson University/Sidney Kimmel Medical College
  3. Michael Liu; Columbia University Irving Medical Center/New York Presbyterian
  4. Allison Borowski; Columbia University Irving Medical Center/New York Presbyterian
  5. Maham Siddique; Columbia University Irving Medical Center/New York Presbyterian
  6. Elise Desperito; Columbia University Irving Medical Center/New York Presbyterian
  7. Richard Ha; Columbia University Irving Medical Center/New York Presbyterian
Objective:
To predict neoadjuvant chemotherapy (NAC) response using a novel convolutional neural network (CNN) algorithm on an external breast MRI dataset from the I-SPY TRIAL (Investigation of Serial Studies to Predict Your Therapeutic Response With Imaging and Molecular Analysis).

Materials and Methods:
The ISPY TRIAL protocol was HIPAA-compliant and the informed consent process were approved by the American College of Radiology Institutional Review Board and local-site institutional review boards. Women with invasive breast cancer of 3 cm or greater undergoing NACT with an anthracycline-based regimen, with or without a Taxane, were enrolled between May 2002 and March 2006. MRI data was collected as described by Hylton et al (1). From the ISPY TRIAL Breast MRI public database, 131 cases collected from 9 different institutions in the United States were successfully downloaded for this study. This multi-institution breast cancer study has made breast MRI imaging dataset publicly available for research. First post-contrast MRI images were used for 3D segmentation using 3D slicer. Similar to our previously developed CNN algorithm (2-4), CNN was implemented entirely of 3 x 3 convolutional kernels and linear layers. The convolutional kernels consisted of 6 residual layers, totaling 12 convolutional layers. Dropout with a 0.5 keep probability and L2 normalization was utilized. Training was implemented by using the Adam optimizer. A 5-fold cross validation was used for performance evaluation. Software code was written in Python using the TensorFlow module on a Linux workstation with one NVidia Titan X GPU.

Results:
Of 131 patients, 40 patients achieved pathological complete response (pCR) following NAC (group 1) and 91 patients did not achieve pCR following NAC (group 2). Diagnostic accuracy of our CNN two classification model distinguishing patients with pCR vs non-pCR was 72.5% (SD± 8.4), with sensitivity 65.5% (SD± 28.1) and specificity of 78.9% (SD±15.2). The area under a ROC Curve (AUC) was 0.72 (SD± 0.08).

Conclusion:
It is feasible to use a CNN algorithm to predict NAC response in patients using a multi-institution dataset. Larger dataset will likely improve our model.