A deep learning approach for automated diagnosis of pulmonary embolism on computed tomographic pulmonary angiography

Abstract

Background Computed tomographic pulmonary angiography (CTPA) is the diagnostic standard for confirming pulmonary embolism (PE). Since PE is a life-threatening condition, early diagnosis and treatment are critical to avoid PE-associated morbidity and mortality. However, PE remains subject to misdiagnosis. Methods We retrospectively identified 251 CTPAs performed at a tertiary care hospital between January 2018 to January 2021. The scans were classified as positive (n = 55) and negative (n = 196) for PE based on the annotations made by board-certified radiologists. A fully anonymized CT slice served as input for the detection of PE by the 2D segmentation model comprising U-Net architecture with Xception encoder. The diagnostic performance of the model was calculated at both the scan and the slice levels. Results The model correctly identified 44 out of 55 scans as positive for PE and 146 out of 196 scans as negative for PE with a sensitivity of 0.80 [95% CI 0.68, 0.89], a specificity of 0.74 [95% CI 0.68, 0.80], and an accuracy of 0.76 [95% CI 0.70, 0.81]. On slice level, 4817 out of 5183 slices were marked as positive for the presence of emboli with a specificity of 0.89 [95% CI 0.88, 0.89], a sensitivity of 0.93 [95% CI 0.92, 0.94], and an accuracy of 0.89 [95% CI 0.887, 0.890]. The model also achieved an AUROC of 0.85 [0.78, 0.90] and 0.94 [0.936, 0.941] at scan level and slice level, respectively for the detection of PE. Conclusion The development of an AI model and its use for the identification of pulmonary embolism will support healthcare workers by reducing the rate of missed findings and minimizing the time required to screen the scans.

Deep Learning-Based Algorithm for Automatic Detection of Pulmonary Embolism in Chest CT Angiograms

“…They tested the model performance on a single external dataset consisting of CTAs from 251 patients of whom 55 presented PE. Sensitivity and specificity were 80% and 74% respectively with an accuracy of 0.76 [ 25 ]. Huang et al developed and evaluated an end-to-end DL model capable of detecting PE using the entire volumetric CTA imaging examination.…”

“…The radiologist’s sensitivity for detecting PE has been measured in the range of 67 to 87% with a specificity of 89 to 99% [ 17 , 18 , 19 ]. Studies have reported promising results in applying deep learning (DL) models to automate PE diagnosis on CTA [ 20 , 21 , 22 , 23 , 24 , 25 ]. In a systematic review with the meta-analysis of DL-based AI algorithms developed to detect PE on CTA, published in 2021, five studies provided enough data to calculate the test accuracies [ 26 ].…”

Machine learning in cancer-associated thrombosis: hype or hope in untangling the clot

“…38 Given that pulmonary embolism can be clinically misdiagnosed or missed in up to one-fourth of patients, 39 several groups have worked on ML-based automatic detection models for this clinical event. 38,40,41 A deep learning model (PENet) for automatic detection of pulmonary embolism from volumetric computed tomography (CT) pulmonary angiograms was developed that achieved an AUROC of 0.85 [95% CI 0.81-0.87] on an external dataset. 42 Such tools can be envisioned to serve as secondary reading tools and also prioritize scans in radiologist review queues to prevent delays in diagnosis.…”

Harnessing Artificial Intelligence for Automated Diagnosis

“…In another article [18], a 2D segmentation model of the U-Net-based architecture classification algorithm was developed and tested targeting the diagnosis of pulmonary embolism. The researchers used a clinical dataset containing 251 computerized tomography pulmonary angiograms (CTPAs).…”

“…The researchers used a clinical dataset containing 251 computerized tomography pulmonary angiograms (CTPAs). Setting a calibration that aimed at minimizing false negative rates, they calculated that blood clots in the main pulmonary and segmental arteries could be detected with 93% sensitivity and 89% specificity (Pranav Ajmeral et al, 2022) [18] (Table 1). Another study that utilized CAD for the detection of pulmonary embolism by means of multiaxial segmentation [19] indicated an obvious superiority of the 2.5D training method over the 2D network architectures but no clear difference between the 2.5D and 3D method.…”