Evan Michael Porter1, Nicholas K Myziuk2, Thomas J Quinn3, Daniela Lozano4, Avery B Peterson5, Duyen M Quach4, Zaid A Siddiqui6, Thomas M Guerrero2. 1. Medical Physics, Wayne State University, 4201 St. Antoine Blvd, Detroit, Michigan, 48202-3489, UNITED STATES. 2. Radiation Oncology, Beaumont Health System, Royal Oak, Michigan, UNITED STATES. 3. Research Institute, Beaumont Health, Royal Oak, Michigan, UNITED STATES. 4. Oakland University William Beaumont School of Medicine, Rochester, Michigan, UNITED STATES. 5. Medical Physics, Wayne State University, Detroit, Michigan, UNITED STATES. 6. Radiation Oncology, University of Pittsburgh School of Medicine, Pittsburgh, Pennsylvania, UNITED STATES.
Abstract
PURPOSE: To develop and evaluate the performance of a deep learning model to generate synthetic pulmonary perfusion images from clinical 4DCT images for patients undergoing radiotherapy for lung cancer. METHODS: A clinical data set of 58 pre- and post-radiotherapy 99mTc-labelled MAA-SPECT perfusion studies (32 patients) each with contemporaneous 4DCT studies was collected. Using the inhale and exhale phases of the 4DCT, a 3D-Residual Network was trained to create synthetic perfusion images utilizing the MAA-SPECT as ground truth. The training process was repeated for a 50-imaging study, five-fold validation with twenty model instances trained per fold. The highest performing model instance from each fold was selected for inference upon the eight-study test set. A manual lung segmentation was used to compute correlation metrics constrained to the voxels within the lungs. From the pre-treatment test cases (N=5), 50th-percentile contours of well-perfused lung were generated from both the clinical and synthetic perfusion images and the agreement was quantified. RESULTS: Across the hold-out test set, our deep learning model predicted perfusion with a Spearman correlation coefficient of 0.70 (IQR: 0.61 - 0.76) and a Pearson correlation coefficient of 0.66 (IQR: 0.49 - 0.73). The agreement of the functional avoidance contour pairs was Dice of 0.803 (IQR: 0.750 - 0.810) and average surface distance of 5.92mm (IQR: 5.68 - 7.55). CONCLUSION: We demonstrate that from 4DCT alone, a deep learning model can generate synthetic perfusion images with potential application in functional avoidance treatment planning.
PURPOSE: To develop and evaluate the performance of a deep learning model to generate synthetic pulmonary perfusion images from clinical 4DCT images for patients undergoing radiotherapy for lung cancer. METHODS: A clinical data set of 58 pre- and post-radiotherapy 99mTc-labelled MAA-SPECT perfusion studies (32 patients) each with contemporaneous 4DCT studies was collected. Using the inhale and exhale phases of the 4DCT, a 3D-Residual Network was trained to create synthetic perfusion images utilizing the MAA-SPECT as ground truth. The training process was repeated for a 50-imaging study, five-fold validation with twenty model instances trained per fold. The highest performing model instance from each fold was selected for inference upon the eight-study test set. A manual lung segmentation was used to compute correlation metrics constrained to the voxels within the lungs. From the pre-treatment test cases (N=5), 50th-percentile contours of well-perfused lung were generated from both the clinical and synthetic perfusion images and the agreement was quantified. RESULTS: Across the hold-out test set, our deep learning model predicted perfusion with a Spearman correlation coefficient of 0.70 (IQR: 0.61 - 0.76) and a Pearson correlation coefficient of 0.66 (IQR: 0.49 - 0.73). The agreement of the functional avoidance contour pairs was Dice of 0.803 (IQR: 0.750 - 0.810) and average surface distance of 5.92mm (IQR: 5.68 - 7.55). CONCLUSION: We demonstrate that from 4DCT alone, a deep learning model can generate synthetic perfusion images with potential application in functional avoidance treatment planning.