State-of-the-Art Pulmonary CT Angiography for Acute Pulmonary Embolism

Albrecht, Moritz H.; Bickford, Matthew W.; Nance, John W.; Zhang, Longjiang; Cecco, Carlo N. De; Wichmann, Julian L.; Vogl, Thomas J.; Schoepf, U. Joseph

doi:10.2214/ajr.16.17202

Cited by 87 publications

(73 citation statements)

References 89 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Subsegmental only PE is felt to be of questionable clinical value, so we removed all subsegmental only PE studies from our dataset. 40 Training data were labeled on a slice level for the presence/absence of a PE. Before feeding into the model, examinations were extracted from Digital Imaging and Communications in Medicine (DICOM) format and scaled to 512 × 512 pixels.…”

Section: Annotations and Image Preprocessingmentioning

confidence: 99%

PENet—a scalable deep-learning model for automated diagnosis of pulmonary embolism using volumetric CT imaging

et al. 2020

View full text Add to dashboard Cite

Pulmonary embolism (PE) is a life-threatening clinical problem and computed tomography pulmonary angiography (CTPA) is the gold standard for diagnosis. Prompt diagnosis and immediate treatment are critical to avoid high morbidity and mortality rates, yet PE remains among the diagnoses most frequently missed or delayed. In this study, we developed a deep learning model-PENet, to automatically detect PE on volumetric CTPA scans as an end-to-end solution for this purpose. The PENet is a 77-layer 3D convolutional neural network (CNN) pretrained on the Kinetics-600 dataset and fine-tuned on a retrospective CTPA dataset collected from a single academic institution. The PENet model performance was evaluated in detecting PE on data from two different institutions: one as a hold-out dataset from the same institution as the training data and a second collected from an external institution to evaluate model generalizability to an unrelated population dataset. PENet achieved an AUROC of 0.84 [0.82-0.87] on detecting PE on the hold out internal test set and 0.85 [0.81-0.88] on external dataset. PENet also outperformed current state-of-the-art 3D CNN models. The results represent successful application of an end-to-end 3D CNN model for the complex task of PE diagnosis without requiring computationally intensive and time consuming preprocessing and demonstrates sustained performance on data from an external institution. Our model could be applied as a triage tool to automatically identify clinically important PEs allowing for prioritization for diagnostic radiology interpretation and improved care pathways via more efficient diagnosis.npj Digital Medicine (2020) 3:61 ; https://doi.

show abstract

Section: Annotations and Image Preprocessingmentioning

confidence: 99%

PENet—a scalable deep-learning model for automated diagnosis of pulmonary embolism using volumetric CT imaging

et al. 2020

View full text Add to dashboard Cite

show abstract

“…Image acquisition was triggered with bolus tracking (80 HU threshold) on the main pulmonary artery. Axial 4‐chamber views were used to measure the maximal diameter of the right and left ventricles, and the RV:LV ratio was calculated …”

Section: Methodsmentioning

confidence: 99%

Cardiopulmonary Exercise Testing in Patients Following Massive and Submassive Pulmonary Embolism

Albaghdadi

Dudzinski

Giordano

et al. 2018

JAHA

View full text Add to dashboard Cite

BackgroundLittle data exist regarding the functional capacity of patients following acute pulmonary embolism. We sought to characterize the natural history of symptom burden, right ventricular (RV) structure and function, and exercise capacity among survivors of massive and submassive pulmonary embolism.Methods and ResultsSurvivors of submassive or massive pulmonary embolism (n=20, age 57±13.3 years, 8/20 female) underwent clinical evaluation, transthoracic echocardiography, and cardiopulmonary exercise testing at 1 and 6 months following hospital discharge. At 1 month, 9/20 (45%) patients had New York Heart Association II or greater symptoms, 13/20 (65%) demonstrated either persistent RV dilation or systolic dysfunction, and 14/20 (70%) had objective exercise impairment as defined by a peak oxygen consumption (trueV˙O2) of <80% of age‐sex predicted maximal values (16.25 [13.4–20.98] mL/kg per minute). At 6 months, no appreciable improvements in symptom severity, RV structure or function, and peak trueV˙O2 (17.45 [14.08–22.48] mL/kg per minute, P=NS) were observed. No patients demonstrated an exercise limitation attributable to either RV/pulmonary vascular coupling, as defined by a VE/VCO 2 slope >33, or a pulmonary mechanical limit to exercise at either time point. Similarly, persistent RV dilation or dysfunction was not significantly related to symptom burden or peak trueV˙O2 at either time point.ConclusionsPersistent symptoms, abnormalities of RV structure and function, and objective exercise limitation are common among survivors of massive and submassive pulmonary embolism. Functional impairment appears to be attributable to general deconditioning rather than intrinsic cardiopulmonary limitation, suggesting an important role for prescribed exercise rehabilitation as a means toward improved patient outcomes and quality of life.

show abstract

“…Computerized tomography pulmonary angiography (CTPA) is the gold standard technique in detecting PE with high sensitivity and specificity [2]. Since CTPA is the first imaging modality in detecting PE in susceptible patients, detecting predictive CTPA indices in identifying high-risk patients would be of high value to reduce the mortality rate [3].…”

Section: Introductionmentioning

confidence: 99%

The assessment of acute pulmonary embolism severity using CT angiography features

et al. 2020

View full text Add to dashboard Cite

Background: This study was conducted to detect the association between radiologic features of CT pulmonary angiography (CTPA) and pulmonary embolism severity index (PESI). Methods: A total of 150 patients with a definite diagnosis of PE entered the study. The CTPA feature including obstruction index, pulmonary trunk size, presence of backwash contrast, septal morphology, right ventricular (RV) and left ventricular (LV) dimensions, and RV/LV ratio were examined. The severity of the PE was estimated using PESI. The association between CTPA indices and PESI was measured. Statistical analysis was conducted using the SPSS software. P value < 0.05 was considered as statistically significant. Results: A positive correlation was detected between the obstruction index and PESI (r = 0.45, P < 0.05). Moreover, PESI was significantly higher in patients with a more dilated pulmonary trunk (r = 0.20, P < 0.05). The backwash contrast and abnormal septal morphology were significantly more common among patients with higher PESI (P < 0.05). However, no significant correlation was detected between RV, LV, RV/LV, and PESI. The most predictor of high-risk PE was dilated pulmonary trunk with an odds ratio of 4.4. Conclusion: Higher Obstruction index, dilated pulmonary trunk, presence of backwash contrast, and an abnormal septal morphology can be associated with a higher PESI.

show abstract

State-of-the-Art Pulmonary CT Angiography for Acute Pulmonary Embolism

Cited by 87 publications

References 89 publications

PENet—a scalable deep-learning model for automated diagnosis of pulmonary embolism using volumetric CT imaging

PENet—a scalable deep-learning model for automated diagnosis of pulmonary embolism using volumetric CT imaging

Cardiopulmonary Exercise Testing in Patients Following Massive and Submassive Pulmonary Embolism

The assessment of acute pulmonary embolism severity using CT angiography features

Contact Info

Product

Resources

About