A 3D convolutional neural network detected hypoperfusion in chronic pulmonary embolism from CT pulmonary angiography with an AUC of 0.87 and MCC of 0.46, outperforming a density-based threshold method.
Observational (n=50)
Yes
Does a 3D convolutional neural network improve the detection of hypoperfusion at CT pulmonary angiography compared to a Hounsfield unit threshold method in patients with chronic pulmonary embolism?
A 3D convolutional neural network algorithm demonstrated superior performance compared to a standard Hounsfield unit threshold method for detecting hypoperfusion on CT pulmonary angiography in patients with chronic pulmonary embolism.
Effect estimate: Difference 0.11 (95% CI 0.05-0.16)
Absolute Event Rate: 0.46% vs 0.35%
BACKGROUND: Chronic pulmonary embolism (CPE) is a life-threatening disease easily misdiagnosed on computed tomography. We investigated a three-dimensional convolutional neural network (CNN) algorithm for detecting hypoperfusion in CPE from computed tomography pulmonary angiography (CTPA). METHODS: Preoperative CTPA of 25 patients with CPE and 25 without pulmonary embolism were selected. We applied a 48%-12%-40% training-validation-testing split (12 positive and 12 negative CTPA volumes for training, 3 positives and 3 negatives for validation, 10 positives and 10 negatives for testing). The median number of axial images per CTPA was 335 (min-max, 111-570). Expert manual segmentations were used as training and testing targets. The CNN output was compared to a method in which a Hounsfield unit (HU) threshold was used to detect hypoperfusion. Receiver operating characteristic area under the curve (AUC) and Matthew correlation coefficient (MCC) were calculated with their 95% confidence interval (CI). RESULTS: The predicted segmentations of CNN showed AUC 0.87 (95% CI 0.82-0.91), those of HU-threshold method 0.79 (95% CI 0.74-0.84). The optimal global threshold values were CNN output probability ≥ 0.37 and ≤ -850 HU. Using these values, MCC was 0.46 (95% CI 0.29-0.59) for CNN and 0.35 (95% CI 0.18-0.48) for HU-threshold method (average difference in MCC in the bootstrap samples 0.11 (95% CI 0.05-0.16). A high CNN prediction probability was a strong predictor of CPE. CONCLUSIONS: We proposed a deep learning method for detecting hypoperfusion in CPE from CTPA. This model may help evaluating disease extent and supporting treatment planning.
Vainio et al. (Fri,) conducted a observational in Chronic pulmonary embolism (n=50). 3D convolutional neural network (CNN) vs. Naïve HU-threshold method was evaluated on Matthews correlation coefficient (MCC) for hypoperfusion detection (Difference 0.11, 95% CI 0.05-0.16). A 3D convolutional neural network detected hypoperfusion in chronic pulmonary embolism from CT pulmonary angiography with an AUC of 0.87 and MCC of 0.46, outperforming a density-based threshold method.