Review Article

The Role of Non-invasive Tests in Pulmonary Embolism

Register or Login to View PDF Permissions
Permissions× For commercial reprint enquiries please contact Springer Healthcare: ReprintsWarehouse@springernature.com.

For permissions and non-commercial reprint enquiries, please visit Copyright.com to start a request.

For author reprints, please email rob.barclay@radcliffe-group.com.
Information image

  • Views: 57

  • Likes: 0

  • Downloads: 11

  • Read time: 21m 42s
Average (ratings)
No ratings
Your rating

Abstract

Pulmonary embolism represents a major cause of morbidity and mortality worldwide, and remains a diagnostic challenge due to its highly nonspecific clinical presentation. Early recognition is critical because timely diagnosis not only improves prognosis but also guides risk stratification, which is essential for therapeutic decision-making. Non-invasive diagnostic modalities – including clinical prediction rules, biomarkers, imaging techniques and bedside assessments – have become indispensable tools for rapid and accurate identification of patients with suspected pulmonary embolism. These methods enable clinicians to stratify risk, predict outcomes and tailor treatment strategies to individual patients, reducing both underdiagnosis and overtreatment. Despite their significant impact, limitations persist, such as access disparities, overdiagnosis of subsegmental events and interpretation challenges in special populations. Ongoing advances, including artificial intelligence and novel biomarkers, hold promise for refining diagnostic accuracy and personalised risk assessment. Ultimately, the integration of non-invasive tests into structured algorithms ensures earlier detection, better prognostic evaluation and improved clinical outcomes for patients with acute pulmonary embolism.

Keywords

Pulmonary embolism, diagnostic imaging, echocardiography, CT pulmonary angiography, biomarkers, risk assessment, artificial intelligence,

Received:

Accepted:

Published online:

DOI: https://doi.org/10.15420/icr.2025.07

Disclosure: DT has received honoraria from Inari Medical. AJ has received grants from EIT Health, Shockwave and Medtronic, honoraria from Abbott, Boston, Shockwave and Philips, and travel support from Abbott and Izasa, and is on advisory boards for Medtronic, Boston, SMT, Shockwave and Philips. AT has no conflicts of interest to declare. The authors have used AI assistance to optimise the translation and refine certain parts of the text.

Correspondence: Daniel Tébar, La Paz University Hospital, Paseo de la Castellana, 261, 28046 Madrid, Spain. E: daniel.tebar.m@gmail.com

Copyright:

© The Author(s). This work is open access and is licensed under CC-BY-NC 4.0. Users may copy, redistribute and make derivative works for non-commercial purposes, provided the original work is cited correctly.

Venous thromboembolic disease, with clinical presentation as deep vein thrombosis (DVT) or pulmonary embolism (PE), is the third leading acute cardiovascular syndrome worldwide after acute MI and stroke.1 PE is caused by obstruction of the pulmonary arteries, usually due to migration of a thrombus from a deep venous source. Despite advances in therapeutic strategies, PE remains a major cause of morbidity and mortality globally.2

In recent years, there has been a trend of increasing incidence rates, possibly highlighting the overdiagnosis of PE in the modern era.3 At the same time, there has been a decrease in case fatality rates for acute PE, possibly due to better use of more effective therapies and interventions, together with improved adherence to therapy.4,5

Early diagnosis is essential for establishing prognosis and assessing risk, which, in turn, guide therapeutic decisions. However, achieving an early diagnosis is often challenging due to the non-specific clinical presentation of PE. This has underscored the importance of non-invasive diagnostic tools, which allow clinicians to rapidly and accurately identify and stratify patients with suspected PE, while minimising unnecessary risks.

This review explores the current role of non-invasive testing in PE, detailing its diagnostic accuracy, clinical integration and limitations.

The Diagnostic Challenge of Pulmonary Embolism

Clinical Presentation

The clinical presentation of PE is very diverse because signs and symptoms are non-specific. Patients may present with dyspnoea, chest pain, presyncope, syncope, haemoptysis or haemodynamic instability.6–8 Extracardiac manifestations, including abdominal pain and epileptic seizures, may also be observed. In some cases, the condition may be asymptomatic or diagnosed incidentally. It is also important to take into account the presence of predisposing factors that increase the clinical likelihood of PE.9

In terms of signs, hypoxaemia and hypocapnia are common.10,11 Chest X-ray is usually abnormal, and is used for differential diagnosis of other clinical conditions.12 The ECG reflects right ventricular overload with findings, such as inverted T waves in leads V1–V4, QR pattern in V1, S1Q3T3 pattern, complete or incomplete right bundle branch block, in more severe PE, or tachycardia in milder cases.13 Data on serial ECG changes in patients with acute PE are scarce. The evolution of these changes has been described during hospitalisation, regardless of haemodynamic status, although with varying incidence. In general, resolution of the S1Q3T3 pattern is followed by normalisation of right bundle branch block, whether complete or incomplete, and subsequently by resolution of T-wave inversions in the precordial leads.14 Furthermore, the persistence of these ECG changes appears to have prognostic significance, which is not currently addressed in clinical guidelines. Atrial arrhythmias, such as AF, may also be associated with acute PE. Accurate diagnosis is critical, as untreated PE can lead to right ventricular failure, cardiogenic shock and death. However, overdiagnosis carries its own risks, exposing patients to unnecessary anticoagulation and imaging.

Non-invasive diagnostic tests have emerged as indispensable in managing PE. These include clinical scoring systems, biomarkers, imaging modalities and point-of-care assessments. Their integration into structured diagnostic algorithms has improved diagnostic accuracy, reduced reliance on invasive procedures and optimised resource usage.

Clinical Prediction Rules: Wells and Geneva Rules

To classify patients with suspected PE into different clinical or pretest probability categories, we use standardised clinical prediction models, such as the revised Geneva rule and the Wells rule.15 These models are available in simplified and validated versions, which facilitate their implementation in clinical practice.16,17

The Wells rule incorporates clinical signs, symptoms of DVT and likelihood of alternative diagnoses. Meanwhile, the Geneva rule, designed for an objective evaluation, relies solely on clinical and demographic data.

There are also exclusion criteria for PE, used in the emergency department to select patients with such a low probability of developing PE that it is not necessary to initiate a diagnostic work-up.18–20 These include eight clinical variables, such as age <50 years, pulse <100 BPM, oxygen saturation >94%, no unilateral leg swelling, no haemoptysis, no recent trauma or surgery, no history of venous thromboembolism and no use of oral hormones. However, they are not being widely used.

D-dimer Testing

In cases of acute thrombosis, D-dimer levels are elevated due to the concomitant activity of coagulation and fibrinolysis. D-dimer has a high negative predictive value; therefore, a normal D-dimer level makes PE or acute DVT unlikely. However, it has a low positive predictive value and does not act as a confirmatory test. D-dimer is elevated in a variety of circumstances, such as hospitalised or cancer patients, infections, inflammatory diseases and in pregnancy.

The YEARS model, which combines D-dimer levels adjusted for clinical probability, helps to rule out PE.21 It includes signs of DVT, haemoptysis and whether an alternative diagnosis is less likely than PE. According to this model, PE can be excluded in patients with no clinical items and D-dimer levels <1,000 mg/l, or in those with one or more clinical items and D-dimer <500 mg/l. Therefore, in low-risk patients, a negative D-dimer test effectively excludes PE without the need for imaging.22–24 Conversely, in high-risk patients, immediate imaging is warranted regardless of D-dimer levels. In contrast, its specificity decreases with age. Consequently, the latest guidelines recommend the use of the age-adjusted D-dimer cut-off value (age × 10 mg/l, in patients aged >50 years) as an alternative to the fixed cut-off value.25,26

Imaging Modalities

CT Pulmonary Angiography

CT pulmonary angiography (CTPA) is the gold standard for imaging patients with suspected PE. It allows adequate visualisation of the pulmonary arteries down to the subsegmental level and enables the assessment of secondary signs, such as right ventricle (RV) dilatation and interventricular septal shift.27–29 Sensitivity and specificity are high (sensitivity of 96–100 % and specificity of 97–98 % have been reported), varying according to clinical probability.30 Thus, the available data suggest that a negative CTPA result allows adequate exclusion of PE in patients with low or intermediate clinical probability. However, a negative result in patients with high clinical probability implies that further testing should be performed.

Advances in CT technology have improved diagnostic accuracy while reducing radiation exposure. However, CTPA is not without limitations. Renal insufficiency and contrast allergies are contraindications, and the modality may be less accessible in resource-limited settings. Furthermore, incidental findings can complicate management, leading to unnecessary investigations.

An emerging CT technology is photon-counting CT. This has the advantage of increased spatial resolution, reduced noise and contrast enhancement, reduced radiation exposure, and the use of alternative contrast agents.31 The potential of photon-counting CT needs to be confirmed in future prospective treatment outcome studies.

Dual-energy CT or CT Pulmonary Angiography with Perfusion Images

Iodine maps in dual-energy CT detects further segmental and sub-segmental PEs in CTPA. Even dual-energy CT enables thrombus characterisation. In addition, subtraction CT can be used to assess lung perfusion in CTPA.32

The addition of CT perfusion imaging combined with dual-energy CT or CTPA improves the detection of acute PE.33 Identification of areas of reduced pulmonary perfusion downstream of emboli allows correlation with functional impairment and possible haemodynamic consequences.

Ventilation/Perfusion Scanning: Single-photon Emission CT

Planar ventilation/perfusion scan (V/Q lung scan) is an accepted diagnostic test for suspected PE.34–36 Perfusion scans are combined with ventilation studies, thus increasing specificity. It is especially indicated for outpatients with a low clinical probability and a normal chest X-ray, for young patients (particularly women), for pregnant women, for patients with a history of contrast-induced anaphylaxis and for patients with severe renal insufficiency. A normal V/Q scan effectively excludes PE, but intermediate-probability results require further investigation. An important limitation is the high frequency of non-diagnostic scans, which require further diagnostic testing. Therefore, it is considered that data acquisition in single-photon emission CT (SPECT) imaging, with or without low-dose CT, can reduce the proportion of non-diagnostic scans to 0–5%.37–40 However, SPECT studies and diagnostic criteria are highly variable. Spatial resolution and contrast increases both from planar V/Q to SPECT and from SPECT to SPECT/CT.

Pulmonary Angiography

For a long time, pulmonary angiography was considered the diagnostic method of choice to confirm or rule out PE; however, nowadays, most patients undergo CTPA, as it offers better resolution while being a less invasive test. Notably, pulmonary angiography is essential for guiding catheter-based therapies and may be particularly useful in patients who, due to their severe hemodynamic or clinical instability, cannot undergo a CT scan.

MRI

Cardiac MRI is an emerging modality for PE diagnosis, particularly in pregnancy and paediatric populations, where radiation exposure is a concern. MRI can provide detailed anatomical and functional insights into the pulmonary vasculature and RV performance. Magnetic resonance angiography is difficult to apply in clinical practice due to its low sensitivity, high cost, lengthy acquisition times, high frequency of inconclusive studies and limited availability in emergency departments.41,42

Magnetic resonance direct thrombus imaging is a non-invasive technique that does not require the administration of intravenous contrast and is based on the formation of methaemoglobin in a fresh thrombus. This differentiates between acute and chronic PE. Magnetic resonance direct thrombus imaging has been mostly evaluated for the diagnosis of DVT. At present, it is not a routine test in the diagnosis of acute PE.32

ECG

ECG plays a dual role in PE, serving both diagnostic and prognostic purposes. While not a first-line test for PE confirmation, ECG can detect secondary pressure overload and RV dysfunction. However, it has a negative predictive value of 40–50%, so a negative result cannot exclude this aetiology.

Furthermore, due to the irregular geometry of the RV, there is no single individual ECG parameter that provides complete information on function and size. ECG findings include RV dilatation, and the more specific combination with high positive predictive value of a pulmonary ejection acceleration time (measured at the RV outflow tract) <60 ms, with a peak tricuspid valve systolic gradient <60 mmHg (‘60/60’ sign), or with depressed RV free wall contractility with apical sparing (McConnell’s sign).43,44 However, these findings were seen in only 12 and 20% of unselected PE patients.40 Other findings in this setting include decreased tricuspid annulus plane systolic excursion, ECG parameters of RV function derived from tissue Doppler imaging and wall strain assessment.

Mobile thrombi in the right heart are detected in only <4% of patients.45,46 In haemodynamically stable patients, it is not mandatory in routine work-up.47 In contrast, in a haemodynamically compromised patient with suspected PE, unequivocal signs of RV pressure overload may justify emergency reperfusion therapy for PE in the appropriate clinical setting.48 Point-of-care ultrasound has gained traction as a bedside tool for detecting signs of RV strain, particularly in emergency settings. Although operator-dependent, its rapid availability and ease of use make it an attractive option in critically ill patients.

Compression Ultrasonography

PE usually originates from DVT. Detection of DVT was demonstrated in 70% of patients with proven PE.49 A positive proximal compression ultrasonography result has a high positive predictive value for PE, with high specificity (96%) and low sensitivity (41%).50,51 The likelihood of a positive proximal ultrasound scan increases in those patients with signs and symptoms related to leg veins. Compression ultrasonography should explore four examination sites (bilateral groin and popliteal fossa). In addition, we should consider compression ultrasonography as an alternative for diagnosis in patients with contraindications to CT.

CT Venography

In the same acquisition as CTPA, deep veins of the lower extremity could be studied. This approach is not validated and increases the radiation dose.52

Risk Stratification

To establish the appropriate therapeutic management, we must perform risk stratification for those patients diagnosed with acute PE. In this initial risk stratification, we focus on symptoms and signs of haemodynamic instability. In most haemodynamically stable patients, further risk stratification requires clinical, imaging and laboratory prognostic criteria for severity of PE, mainly related to the presence of right ventricular dysfunction; and in turn, the presence of comorbidities and any other worsening conditions.

Clinical Parameters

Acute RV failure has prognostic value.53 In turn, the presence of signs, such as tachycardia, low systolic blood pressure, respiratory failure (tachypnoea and/or low oxygen saturation) and syncope, are associated with unfavourable short-term prognosis in acute PE.

Imaging of the Right Ventricle

Echocardiography

Echocardiographic parameters for assessing RV morphology and function are a valuable tool for prognostic assessment of normotensive patients with acute PE. We found a higher prognostic association in the case of a RV/left ventricle (LV) diameter ratio ≥1.0 and a tricuspid annulus plane systolic excursion <16 mm.54 Similarly, the identification of a patent foramen ovale with right-to-left shunt and the presence of thrombus in the right heart has been associated with higher mortality in patients with acute PE.55,56

CT Pulmonary Angiography

CTPA identifies different parameters associated with early risk stratification of patients with PE. Outstanding is RV enlargement, assessed by RV end-diastolic diameter and RV/LV diameter ratio measured in the transverse or four-chamber projection.57 The prognostic value of an enlarged right ventricle is validated, especially RV/LV ratios >1.0 (instead of 0.9) to indicate poor prognosis.58 This was associated with a 2.5-fold higher risk of all-cause mortality or adverse events, and a fivefold higher risk of PE-related mortality. In addition, volumetric analysis of the cardiac chambers and assessment of contrast reflux to the inferior vena cava have prognostic significance.55,59–61

Quantification of Thrombotic Burden with CT Pulmonary Angiography

Quantification of thrombus burden by CTPA is a novel and increasingly valuable tool for stratifying the severity and prognosis of PE.62 In contrast to traditional imaging parameters, thrombus burden quantification provides a direct assessment of the clot burden within the pulmonary vasculature, offering additional diagnostic information and aiding therapeutic decisions in each case. This refers to the measurement of the volume, extent or distribution of emboli within the pulmonary arteries using advanced imaging techniques. These measurements are obtained from CTPA and analysed manually, semi-automatically or by automated algorithms.

Qanadli, Mastora and Ghanima have developed scores with quantifiable and reproducible data from the assessment of thrombotic burden observed in CTPA studies.63–65

The Qanadli score (Obstruction Index) is a semi-quantitative method assessing embolism severity by scoring the degree of vascular obstruction and thrombus location in the pulmonary arterial tree. It established correlations between the severity of pulmonary arterial obstruction and data obtained by ECG and angiography. The formula includes the sum of the product of the proximal thrombus value (determined by the number of segmental arteries, from 1 to 20) and the degree of obstruction (between 0 and 2).63

The Mastora score is similar to Qanadli, but incorporates vessel diameter, providing a more detailed evaluation of vascular obstruction. Arterial lumen occlusions are divided into five levels (<25%, 25–49%, 50–74%, 75–99% and 100%). It established central, peripheral or global severity according to measurements of mediastinal, lobar and segmental arteries.64

The Ghanima score defines four zones within the pulmonary arterial tree– the subsegmental, segmental, lobar and main. It holds that more central locations correlate with greater severity, regardless of the degree of obstruction.65

Quantification of thrombus burden is particularly meaningful in haemodynamically stable patients with intermediate-risk PE, where therapy strategies are still under debate. Patients with a higher thrombus burden may benefit from closer follow-up or treatment escalation. In this way, we could establish personalised therapy. Thus, high thrombus burden may lead to catheter-directed thrombolysis or thrombectomy in patients unsuitable for systemic thrombolysis.

Thrombus Composition

The composition of a thrombus can influence therapeutic decisions. Its composition depends on its age and structural organisation. The efficacy of pharmacological interventions, such as thrombolysis, and mechanical procedures, such as thrombectomy, depends on the density and composition of the thrombus. The fragmentation of the thrombus during removal also varies. Therefore, treating older and more complex thrombi requires more aggressive therapies.

Recently, thrombus texture analysis has been assessed as a possible prognostic marker in PE.66 A number of CTPA-derived pulmonary emboli texture features were associated with mortality, intensive care unit admission, and clinical and serological parameters in the setting of acute PE. This potential clinical benefit needs further studies for validation.

Laboratory Biomarkers

Troponin

Elevated troponin (Tn) in patients with PE is due to RV overload, ischaemia and necrosis caused by PE itself. In several studies, elevated troponin levels are associated with RV dysfunction, short-term mortality, PE mortality and serious adverse events.67–70

The advantage of this parameter is that it is a simple and rapid, relatively low-cost, widely available method of analysis. Furthermore, it can be performed in normal laboratories or emergency departments. In contrast, it has a number of limitations. There are four types of troponin, such as cardiac Tn I, cardiac Tn T, high-sensitivity cardiac Tn I or high-sensitivity cardiac Tn T. It is important to note that there are different detection methods and different thresholds. Therefore, various studies use different criteria, so there is a lack of uniformity and consensus. In addition, their values may be influenced by other factors or conditions (myocarditis, heart failure, atherosclerosis, etc.).

High-sensitivity troponin assays allow detection of troponin levels even within the normal reference range, with greater accuracy at lower concentrations. Although high-sensitivity troponin assays have revolutionised the diagnosis of MI, their role in PE is an evolving area of research. High-sensitivity troponin T assays allow for improved risk stratification for PE. It has demonstrated excellent sensitivity and negative predictive value (both 100%).71 Even age-adjusted cut-off values of Tn T (14 pg/ml in patients aged <75 years, 45 pg/ml in patients aged >75 years) improve the negative predictive value of this biomarker.72 However, this high sensitivity with detection of minimal troponin elevations suggests that it is not accompanied by clinical significance compared with conventional Tn I.73 This could suggest an overestimation of risk in stable PE, necessitating integration with imaging and other clinical parameters.

The different guidelines underline the importance of troponin measurement in this scenario.25,74 However, they also indicate that it should be combined with other parameters for more accurate and complete stratification. For example, combined use of echocardiography and cardiac Tn levels may more accurately identify and change the management of a certain subgroup of intermediate-high-risk patients.

Natriuretic Peptides

Plasma levels of natriuretic peptides (B-type natriuretic peptide [BNP] and N-terminal BNP) increase as a consequence of myocardial stretch secondary to right ventricular pressure overload due to acute PE. Their elevation has been shown to increase the risk of early death and of an adverse clinical outcome.75 However, elevated BNP or N-terminal BNP concentrations have low specificity and positive predictive value for early mortality in normotensive patients with PE, but low BNP or N-terminal BNP levels are able to exclude an unfavourable early clinical outcome, with high sensitivity and negative predictive value.76,77

Others

Heart-type fatty acid-binding protein is an early and sensitive marker of myocardial injury that provides prognostic information. It has been associated with adverse short-term outcomes and all-cause mortality.78

Other laboratory parameters have shown prognostic associations. These include lactate, elevated serum creatinine levels and decreased glomerular filtration rate, elevated neutrophil gelatinase-associated lipocalin and cystatin C, hyponatraemia, and elevated levels of copeptin (a surrogate marker for vasopressin).

Scores

Prognostic assessment of patients without haemodynamic instability requires scores combining clinical, imaging and laboratory parameters described above, as these parameters individually are insufficient. Therefore, different tools and methods exist to identify and stratify individuals according to the likelihood of developing clinical deterioration or suffering a PE-related death.

The Pulmonary Embolism Severity Index (PESI) also integrates parameters related to aggravating conditions and comorbidity. The original scale includes 11 weighted variables.79

The Simplified PESI is a streamlined version of the original PESI, designed to simplify risk stratification in patients with acute pulmonary embolism. It assesses six variables: age >80 years, history of cancer, chronic cardiopulmonary disease, elevated heart rate, decreased systolic blood pressure and reduced arterial oxygen saturation – each assigned 1 point.80 This simplified approach maintains prognostic accuracy comparable to the original index while improving usability in clinical practice. The Simplified PESI accurately identifies patients at low risk of 30-day mortality, and is widely used to rapidly select low-risk candidates for outpatient management and to guide treatment decisions. It is practical and highly validated, making it the preferred tool in many clinical settings. Its use in the acute phase is recommended by the European guidelines (IIaB).25

The Bova score includes four variables: systolic blood pressure, elevated troponins, right ventricular dysfunction on imaging (CT or ECG) and elevated heart rate.81,82 Its scores range from 0 to 7 points, and is focused specifically on early mortality and clinical deterioration. It is particularly useful in intermediate-risk patients, helping to identify those who may benefit from escalated therapies, such as thrombolysis or catheter-directed interventions.

FAST is another prognostic tool that includes the combination of elevated heart-type fatty acid-binding protein levels, heart rate >110 BPM and syncope.83 If heart-type fatty acid-binding protein measurement is not available, an age-adjusted troponin (modified FAST score) could be used.

Other criteria for risk identification are the PEITHO criteria.84 Variables included are myocardial injury, RV dysfunction and more than one of the following: low systolic blood pressure, decreased oxygen saturation and history of chronic heart failure.

New scores have been developed in recent years, such as Calgary Acute Pulmonary Embolism (CAPE), SHIeLD, Pulmonary Embolism Mortality Score (PEMS), pulmonary embolism short-term clinical outcomes risk estimation (PE-SCORE) and Functional Assessment Staging Tool, without validation or comparative studies.85–89 These scores are a composite of variables with additional risk factors, such as serum lactate, pH, diastolic blood pressure, central embolism and creatinine.

Risk Stratification Strategy

From the initial stages of the diagnostic evaluation, risk assessment should be performed. Identification of high-risk patients (previously defined as massive PE) with haemodynamic instability is critical, as they require emergency diagnosis and reperfusion therapy. In patients without haemodynamic instability, further risk stratification for PE is recommended, as it has implications for the choice of therapeutic strategy, early discharge or hospitalisation and patient follow-up.

In accordance with a variety of clinical, imaging and laboratory parameters or different comorbidities or conditions, different risk groups are distinguished, such as high risk, intermediate risk and low risk. The American Heart Association, American College of Chest Physicians and European Society of Cardiology/European Respiratory Society guidelines differ in the subdivision of this heterogeneous group.25,90,91

According to the latter, patients in the intermediate-risk group who show evidence of both RV dysfunction (on ECG or CT angiography) and elevated troponin levels are classified in the intermediate-high-risk category.92 Patients in whom the RV appears normal and/or have normal levels of cardiac biomarkers belong to the low-intermediate-risk category. This approach will determine the therapeutic management to be carried out. Table 1 describes the similarities and differences in risk stratification between Europe and the US.

Table 1: Similarities and Differences in Risk Stratification between the European and US Guidelines

Article image
Download

Artificial Intelligence

Artificial intelligence (AI) has a promising role in the PE setting. Its potential use lies in diagnosis, improving work lists, and quantifying thrombus burden and prognostic value with the integration of clinical data.

Currently, decisions to order investigations for PE assessment are based on clinical scores and D-dimer. AI can also help in this decision. The machine learning model of development and performance of pulmonary embolism outcome prognosis was developed for this purpose.93 The model achieved an area under the receiver operating characteristic curve for predicting a subsequent positive PE CT of 0.90 in the validation cohort and 0.71 in the outpatient cohort. In the outpatient cohorts, the model achieved a better area under the receiver operating characteristic curve than our commonly used clinical scoring systems. AI-supported decisions could reduce costs, radiation exposure and overdiagnosis.

AI may also have relevance for image classification, leading to prioritisation of the worklist and faster PE diagnosis.94

Several AI models and algorithms have been developed for the detection of PE in CTPA.32 The nnU-Net algorithm has been evaluated for the detection of PE, particularly central PE, and for measuring patients’ blood clot volume in automated severity stratification.95 The results suggest that blood clot volume is a significant indicator for detecting PE and central PE, as well as for assessing RV overload. This is an indication of its potential to improve efficiency in the diagnosis and treatment of PE. The results of the various studies are variable in diagnostic benefit. Therefore, more evidence from prospective studies is needed.

The use of AI may contribute to risk stratification of patients with acute PE. Clot burden quantitatively measured with a deep learning convolutional neural network method correlated with clot burden assessed with Qanadli and Mastora scores, and with functional parameters of the RV in CTPA.96 Manual analysis of the RV/LV ratio has also been correlated with automated calculation of the RV/LV diameter ratio in CTPA.97

The incorporation of AI could be helpful in establishing both the diagnosis and prognosis of EP. Current radiological practice with AI needs to be evaluated, as well as the challenge of developing and implementing AI in our routine practice.

Limitations and Challenges

Despite the advantages, non-invasive testing is not without limitations. Routine use of CTPA has led to overdiagnosis of subsegmental PE, raising questions about overtreatment. In addition, non-specific biomarkers, such as D-dimer, may lead to false positive results, especially in hospitalised or postoperative patients.

Access to advanced imaging continues to be a major challenge in resource-limited settings, where dependence on clinical judgement and basic tests can undermine diagnostic accuracy. In contrast, incorrect use of non-invasive tests can lead to unnecessary costs, diagnostic delays and patient anxiety. Therefore, advances in technology require clinicians trained in the proper selection and interpretation of tests.

Future Directions

The future of PE diagnosis lies in leveraging AI and machine learning to enhance diagnostic accuracy. AI algorithms can analyse imaging data and clinical parameters to identify patterns indicative of PE, potentially reducing interobserver variability. Its implementation can significantly improve the efficiency and accuracy of PD diagnosis, aiding and reducing radiologists’ workload, and leading to faster decisions and better patient outcomes. The integration of AI into our routine practice is of increasing importance to improve healthcare delivery.

New biomarkers and imaging techniques are emerging that could improve risk stratification and treatment monitoring. For example, molecular imaging targeting specific thrombus markers could provide real-time information on clot burden and composition.

Telemedicine and portable imaging devices are poised to improve access to diagnostic tools in remote and underserved areas, bridging the gap in global healthcare disparities.

Conclusion

The management of PE has undergone significant advances in both diagnosis and treatment. Non-invasive tests are the cornerstone of modern PE diagnosis and management. By enabling rapid, accurate and safe evaluation, these tools have revolutionised patient care, reducing reliance on invasive procedures and improving outcomes. Their integration into evidence-based diagnostic algorithms ensures that patients receive timely and appropriate interventions while minimising unnecessary risks.

Advances in diagnostic tools with clinical, laboratory and imaging parameters have allowed for more accurate risk stratification. This highlights the uniqueness of each case, requiring personalised decisions tailored to the individual needs of each patient. As technology and biomarkers continue to be developed, non-invasive diagnostics will play an increasingly pivotal role in the fight against this potentially fatal condition.

References

  1. Raskob GE, Angchaisuksiri P, Blanco AN, et al. Thrombosis: a major contributor to global disease burden. Arterioscler Thromb Vasc Biol 2014;34:2363–71. 
    Crossref | PubMed
  2. Heit JA, Spencer FA, White RH. The epidemiology of venous thromboembolism. J Thromb Thrombolysis 2016;41:3–14. 
    Crossref | PubMed
  3. Wiener RS, Schwartz LM, Woloshin S. Time trends in pulmonary embolism in the United States: evidence of overdiagnosis. Arch Intern Med 2011;171:831–7. 
    Crossref | PubMed
  4. Roy PM, Meyer G, Vielle B, et al. Appropriateness of diagnostic management and outcomes of suspected pulmonary embolism. Ann Intern Med 2006;144:157–64. 
    Crossref | PubMed
  5. Jimenez D, Bikdeli B, Barrios D, et al. Management appropriateness and outcomes of patients with acute pulmonary embolism. Eur Respir J 2018;51:1800445. 
    Crossref | PubMed
  6. Pollack CV, Schreiber D, Goldhaber SZ, et al. Clinical characteristics, management, and outcomes of patients diagnosed with acute pulmonary embolism in the emergency department: initial report of EMPEROR (Multicenter Emergency Medicine Pulmonary Embolism in the Real World Registry). J Am Coll Cardiol 2011;57:700–6. 
    Crossref | PubMed
  7. Miniati M, Prediletto R, Formichi B, et al. Accuracy of clinical assessment in the diagnosis of pulmonary embolism. Am J Respir Crit Care Med 1999;159:864–71. 
    Crossref | PubMed
  8. Wells PS, Ginsberg JS, Anderson DR, et al. Use of a clinical model for safe management of patients with suspected pulmonary embolism. Ann Intern Med 1998;129:997–1005. 
    Crossref | PubMed
  9. White RH. The epidemiology of venous thromboembolism. Circulation 2003;107(Suppl 1):I4–8. 
    Crossref | PubMed
  10. Rodger MA, Carrier M, Jones GN, et al. Diagnostic value of arterial blood gas measurement in suspected pulmonary embolism. Am J Respir Crit Care Med 2000;162:2105–8. 
    Crossref | PubMed
  11. Stein PD, Goldhaber SZ, Henry JW, Miller AC. Arterial blood gas analysis in the assessment of suspected acute pulmonary embolism. Chest 1996;109:78–81. 
    Crossref | PubMed
  12. Elliott CG, Goldhaber SZ, Visani L, DeRosa M. Chest radiographs in acute pulmonary embolism. Results from the international cooperative pulmonary embolism registry. Chest 2000;118:33–8. 
    Crossref | PubMed
  13. Shopp JD, Stewart LK, Emmett TW, Kline JA. Findings from 12-lead electrocardiography that predict circulatory shock from pulmonary embolism: systematic review and meta-analysis. Acad Emerg Med 2015;22:1127–37. 
    Crossref | PubMed
  14. Zuin M, Rigatelli G, Bilato C, et al. Prognostic role of serial electrocardiographic changes in patients with acute pulmonary embolism. Data from the Italian pulmonary embolism registry. Thromb Res 2022;217:15–21. 
    Crossref | PubMed
  15. Wells PS, Anderson DR, Rodger M, et al. Derivation of a simple clinical model to categorize patients probability of pulmonary embolism: increasing the models utility with the SimpliRED D-dimer. Thromb Haemost 2000;83:416–20. 
    Crossref | PubMed
  16. Klok FA, Mos ICM, Nijkeuter M, et al. Simplification of the revised Geneva score for assessing clinical probability of pulmonary embolism. Arch Intern Med 2008;168:2131–6. 
    Crossref | PubMed
  17. Gibson NS, Sohne M, Kruip MJHA, et al. Further validation and simplification of the Wells clinical decision rule in pulmonary embolism. Thromb Haemost 2008;99:229–34. 
    Crossref | PubMed
  18. Kline JA, Mitchell AM, Kabrhel C, et al. Clinical criteria to prevent unnecessary diagnostic testing in emergency department patients with suspected pulmonary embolism. J Thromb Haemost 2004;2:1247–55. 
    Crossref | PubMed
  19. Penaloza A, Soulie C, Moumneh T, et al. Pulmonary embolism rule-out criteria (PERC) rule in European patients with low implicit clinical probability (PERCEPIC): a multicentre, prospective, observational study. Lancet Haematol 2017;4:e615–21. 
    Crossref | PubMed
  20. Freund Y, Cachanado M, Aubry A, et al. Effect of the pulmonary embolism rule-out criteria on subsequent thromboembolic events among low-risk emergency department patients: the PROPER randomized clinical trial. JAMA 2018;319:559–66. 
    Crossref | PubMed
  21. van der Hulle T, Cheung WY, Kooij S, et al. Simplified diagnostic management of suspected pulmonary embolism (the YEARS study): a prospective, multicentre, cohort study. Lancet 2017;390:289–97. 
    Crossref | PubMed
  22. Perrier A, Roy PM, Sanchez O, et al. Multidetector-row computed tomography in suspected pulmonary embolism. N Engl J Med 2005;352:1760–8. 
    Crossref | PubMed
  23. Perrier A, Roy PM, Aujesky D, et al. Diagnosing pulmonary embolism in outpatients with clinical assessment, D-dimer measurement, venous ultrasound, and helical computed tomography: a multicenter management study. Am J Med 2004;116:291–9. 
    Crossref | PubMed
  24. Wells PS, Anderson DR, Rodger M, et al. Excluding pulmonary embolism at the bedside without diagnostic imaging: management of patients with suspected pulmonary embolism presenting to the emergency department by using a simple clinical model and D-dimer. Ann Intern Med 2001;135:98–107. 
    Crossref | PubMed
  25. Konstantinides SV, Meyer G, Becattini C, et al. 2019 ESC Guidelines for the diagnosis and management of acute pulmonary embolism developed in collaboration with the European Respiratory Society (ERS). Eur Heart J 2020;41:543–603. 
    Crossref | PubMed
  26. Righini M, Van Es J, Den Exter PL, et al. Age-adjusted Ddimer cutoff levels to rule out pulmonary embolism: the ADJUST-PE study. JAMA 2014;311:1117–24. 
    Crossref | PubMed
  27. Patel S, Kazerooni EA, Cascade PN. Pulmonary embolism: optimization of small pulmonary artery visualization at multi-detector row CT. Radiology 2003;227:455–60. 
    Crossref | PubMed
  28. Ghaye B, Szapiro D, Mastora I, et al. Peripheral pulmonary arteries: how far in the lung does multi-detector row spiral CT allow analysis? Radiology 2001;219:629–36. 
    Crossref | PubMed
  29. Carrier M, Righini M, Wells PS, et al. Subsegmental pulmonary embolism diagnosed by computed tomography: incidence and clinical implications. A systematic review and meta-analysis of the management outcome studies. J Thromb Haemost 2010;8:1716–22. 
    Crossref | PubMed
  30. Stein PD, Fowler SE, Goodman LR, et al. Multidetector computed tomography for acute pulmonary embolism. N Engl J Med 2006;354:2317–27. 
    Crossref | PubMed
  31. Pannenbecker P, Huflage H, Grunz JP, et al. Photon-counting CT for diagnosis of acute pulmonary embolism: potential for contrast medium and radiation dose reduction. Eur Radiol 2023;33:7830–9. 
    Crossref | PubMed
  32. de Jong CMM, Kroft LJM, van Mens TE, et al. Modern imaging of acute pulmonary embolism. Thromb Res 2024;238:105–16. 
    Crossref | PubMed
  33. Grob D, Smit E, Prince J, et al. Iodine maps from subtraction CT or dual-energy CT to detect pulmonary emboli with CT angiography: a multiple-observer study. Radiology 2019;292:197–205. 
    Crossref | PubMed
  34. Reid JH, Coche EE, Inoue T, et al. Is the lung scan alive and well? Facts and controversies in defining the role of lung scintigraphy for the diagnosis of pulmonary embolism in the era of MDCT. Eur J Nucl Med Mol Imaging 2009;36:505–21. 
    Crossref | PubMed
  35. Anderson DR, Kahn SR, Rodger MA, et al. Computed tomographic pulmonary angiography vs ventilation-perfusion lung scanning in patients with suspected pulmonary embolism: a randomized controlled trial. JAMA 2007;298:2743–53. 
    Crossref | PubMed
  36. Waxman AD, Bajc M, Brown M, et al. Appropriate use criteria for ventilation-perfusion imaging in pulmonary embolism: summary and excerpts. J Nucl Med 2017;58:13N–5N.
    PubMed
  37. Bajc M, Olsson B, Palmer J, Jonson B. Ventilation/perfusion SPECT for diagnostics of pulmonary embolism in clinical practice. J Intern Med 2008;264:379–87. 
    Crossref | PubMed
  38. Gutte H, Mortensen J, Jensen CV, et al. Detection of pulmonary embolism with combined ventilation-perfusion SPECT and low-dose CT: head-to-head comparison with multidetector CT angiography. J Nucl Med 2009;50:1987–92. 
    Crossref | PubMed
  39. Reinartz P, Wildberger JE, Schaefer W, et al. Tomographic imaging in the diagnosis of pulmonary embolism: a comparison between V/Q lung scintigraphy in SPECT technique and multislice spiral CT. J Nucl Med 2004;45:1501–8.
    PubMed
  40. Collart JP, Roelants V, Vanpee D, et al. Is a lung perfusion scan obtained by using single photon emission computed tomography able to improve the radionuclide diagnosis of pulmonary embolism? Nucl Med Commun 2002;23:1107–13. 
    Crossref | PubMed
  41. Revel MP, Sanchez O, Couchon S, et al. Diagnostic accuracy of magnetic resonance imaging for an acute pulmonary embolism: results of the “IRM-EP” study. J Thromb Haemost 2012;10:743–50. 
    Crossref | PubMed
  42. Stein PD, Chenevert TL, Fowler SE, et al. Gadolinium-enhanced magnetic resonance angiography for pulmonary embolism: a multicenter prospective study (PIOPED III). Ann Intern Med 2010;152:434–43. 
    Crossref | PubMed
  43. Kurnicka K, Lichodziejewska B, Goliszek S, et al. Echocardiographic pattern of acute pulmonary embolism: analysis of 511 consecutive patients. J Am Soc Echocardiogr 2016;29:907–13. 
    Crossref | PubMed
  44. Kurzyna M, Torbicki A, Pruszczyk P, et al. Disturbed right ventricular ejection pattern as a new Doppler echocardiographic sign of acute pulmonary embolism. Am J Cardiol 2002;90:507–11. 
    Crossref | PubMed
  45. Casazza F, Becattini C, Guglielmelli E, et al. Prognostic significance of free-floating right heart thromboemboli in acute pulmonary embolism: results from the Italian Pulmonary Embolism Registry. Thromb Haemost 2014;111:53–7. 
    Crossref | PubMed
  46. Torbicki A, Galié N, Covezzoli A, et al. Right heart thrombi in pulmonary embolism: results from the International Cooperative Pulmonary Embolism Registry. J Am Coll Cardiol 2003;41:2245–51. 
    Crossref | PubMed
  47. Roy PM, Colombet I, Durieux P, et al. Systematic review and meta-analysis of strategies for the diagnosis of suspected pulmonary embolism. BMJ 2005;331:259. 
    Crossref | PubMed
  48. Dresden S, Mitchell P, Rahimi L, et al. Right ventricular dilatation on bedside echocardiography performed by emergency physicians aids in the diagnosis of pulmonary embolism. Ann Emerg Med 2014;63:16–24. 
    Crossref | PubMed
  49. Hull RD, Hirsh J, Carter CJ, et al. Pulmonary angiography, ventilation lung scanning, and venography for clinically suspected pulmonary embolism with abnormal perfusion lung scan. Ann Intern Med 1983;98:891–9. 
    Crossref | PubMed
  50. Le Gal G, Righini M, Sanchez O, et al. H.A positive compression ultrasonography of the lower limb veins is highly predictive of pulmonary embolism on computed tomography in suspected patients. Thromb Haemost 2006;95:963–6. 
    Crossref | PubMed
  51. Da Costa Rodrigues J, Alzuphar S, Combescure C, et al. Diagnostic characteristics of lower limb venous compression ultrasonography in suspected pulmonary embolism: a meta-analysis. J Thromb Haemost 2016;14:1765–72. 
    Crossref | PubMed
  52. Righini M, Le Gal G, Aujesky D, et al. Diagnosis of pulmonary embolism by multidetector CT alone or combined with venous ultrasonography of the leg: a randomised non-inferiority trial. Lancet 2008;371:1343–52. 
    Crossref | PubMed
  53. Harjola VP, Mebazaa A, Celutkiene J, et al. Contemporary management of acute right ventricular failure: a statement from the Heart Failure Association and the Working Group on Pulmonary Circulation and Right Ventricular Function of the European Society of Cardiology. Eur J Heart Fail 2016;18:226–41. 
    Crossref | PubMed
  54. Pruszczyk P, Goliszek S, Lichodziejewska B, et al. Prognostic value of echocardiography in normotensive patients with acute pulmonary embolism. JACC Cardiovasc Imaging 2014;7:553–60. 
    Crossref | PubMed
  55. Konstantinides S, Geibel A, Kasper W, et al. Patent foramen ovale is an important predictor of adverse outcome in patients with major pulmonary embolism. Circulation 1998;97:1946–51. 
    Crossref | PubMed
  56. Barrios D, Rosa-Salazar V, Morillo R, et al. Prognostic significance of right heart thrombi in patients with acute symptomatic pulmonary embolism: systematic review and meta-analysis. Chest 2017;151:409–16. 
    Crossref | PubMed
  57. Becattini C, Agnelli G, Vedovati MC, et al. Multidetector computed tomography for acute pulmonary embolism: diagnosis and risk stratification in a single test. Eur Heart J 2011;32:1657–63. 
    Crossref | PubMed
  58. Meinel FG, Nance JW Jr, Schoepf UJ, et al. Predictive value of computed tomography in acute pulmonary embolism: systematic review and meta-analysis. Am J Med 2015;128:747–59.e2. 
    Crossref | PubMed
  59. Aviram G, Soikher E, Bendet A, et al. Prediction of mortality in pulmonary embolism based on left atrial volume measured on CT pulmonary angiography. Chest 2016;149:667–75. 
    Crossref | PubMed
  60. Aviram G, Sirota-Cohen C, Steinvil A, et al. Automated volumetric analysis of four cardiac chambers in pulmonary embolism: a novel technology for fast risk stratification. Thromb Haemost 2012;108:384–93. 
    Crossref | PubMed
  61. Bach AG, Nansalmaa B, Kranz J, et al. CT pulmonary angiography findings that predict 30-day mortality in patients with acute pulmonary embolism. Eur J Radiol 2015;84:332–7. 
    Crossref | PubMed
  62. Diaz-Lorenzo I, Alonso-Burgos A, Friera Reyes A, et al. Current role of CT pulmonary angiography in pulmonary embolism: a state-of-the-art review. J Imaging 2024;10:323. 
    Crossref | PubMed
  63. Qanadli SD, El Hajjam M, Vieillard-Baron A, et al. New CT index to quantify arterial obstruction in pulmonary embolism: comparison with angiographic index and echocardiography. AJR 2001;176:1415–20. 
    Crossref | PubMed
  64. Mastora I, Remy-Jardin M, Masson P, et al. Severity of acute pulmonary embolism: evaluation of a new spiral CT angiographic score in correlation with echocardiographic data. Eur Radiol 2003;13:29–35. 
    Crossref | PubMed
  65. Ghanima W, Abdelnoor M, Holmen LO, et al. The association between the proximal extension of the clot and the severity of pulmonary embolism (PE): a proposal for a new radiological score for PE. J Intern Med 2007;261:74–81. 
    Crossref | PubMed
  66. Leonhardi J, Bailis N, Lerche M, et al. Computed tomography embolus texture analysis as a prognostic marker of acute pulmonary embolism. Angiology 2023;74:461–71. 
    Crossref | PubMed
  67. Konstantinides S, Geibel A, Olschewski M, et al. Importance of cardiac troponins I and T in risk stratification of patients with acute pulmonary embolism. Circulation 2002;106:1263–8. 
    Crossref | PubMed
  68. Daquarti G, March Vecchio N, Mitrione CS, et al. High-sensitivity troponin and right ventricular function in acute pulmonary embolism. Am J Emerg Med 2016;34:1579–82. 
    Crossref | PubMed
  69. Becattini C, Vedovati MC, Agnelli G. Prognostic value of troponins in acute pulmonary embolism: a meta-analysis. Circulation 2007;116:427–33. 
    Crossref | PubMed
  70. Bajaj A, Saleeb M, Rathor P, et al. Prognostic value of troponins in acute nonmassive pulmonary embolism: a meta-analysis. Heart Lung 2015;44:327–34. 
    Crossref | PubMed
  71. Lankeit M, Friesen D, Aschoff J, et al. Highly sensitive troponin T assay in normotensive patients with acute pulmonary embolism. Eur Heart J 2010;31:1836–44. 
    Crossref | PubMed
  72. Kaeberich A, Seeber V, Jiménez D, et al. Age-adjusted high-sensitivity troponin T cut-off value for risk stratification of pulmonary embolism. Eur Respir J 2015;45:1323–31. 
    Crossref | PubMed
  73. Bikdeli B, Muriel A, Rodríguez C, et al. High-sensitivity vs conventional troponin cutoffs for risk stratification in patients with acute pulmonary embolism. JAMA Cardiol 2024;9:64–70. 
    Crossref | PubMed
  74. Rivera-Lebron B, McDaniel M, Ahrar K, et al. Diagnosis, treatment and follow up of acute pulmonary embolism: consensus practice from the PERT consortium. Clin Appl Thromb Hemost 2019;25:1076029619853037. 
    Crossref | PubMed
  75. Klok FA, Mos ICM, Huisman MV. Brain-type natriuretic peptide levels in the prediction of adverse outcome in patients with pulmonary embolism: a systematic review and meta-analysis. Am J Respir Crit Care Med 2008;178:425–30. 
    Crossref | PubMed
  76. Kucher N, Goldhaber SZ. Cardiac biomarkers for risk stratification of patients with acute pulmonary embolism. Circulation 2003;108:2191–4. 
    Crossref | PubMed
  77. Coutance G, Cauderlier E, Ehtisham J, et al. The prognostic value of markers of right ventricular dysfunction in pulmonary embolism: a meta-analysis. Crit Care 2011;15:R103. 
    Crossref | PubMed
  78. Bajaj A, Rathor P, Sehgal V, et al. Risk stratification in acute pulmonary embolism with heart-type fatty acid-binding protein: a meta-analysis. J Crit Care 2015;30:1151.e1–7. 
    Crossref | PubMed
  79. Aujesky D, Obrosky DS, Stone RA, et al. Derivation and validation of a prognostic model for pulmonary embolism. Am J Respir Crit Care Med 2005;172:1041–6. 
    Crossref | PubMed
  80. Jiménez D, Aujesky D, Moores L, et al. Simplification of the pulmonary embolism severity index for prognostication in patients with acute symptomatic pulmonary embolism. Arch Intern Med 2010;170:1383–9. 
    Crossref | PubMed
  81. Bova C, Sanchez O, Prandoni P, et al. Identification of intermediate-risk patients with acute symptomatic pulmonary embolism. Eur Respir J 2014;44:694–703. 
    Crossref | PubMed
  82. Bova C, Vanni S, Prandoni P, et al. A prospective validation of the Bova score in normotensive patients with acute pulmonary embolism. Thromb Res 2018;165:107–11. 
    Crossref | PubMed
  83. Lankeit M, Friesen D, Schäfer K, et al. A simple score for rapid risk assessment of non-high-risk pulmonary embolism. Clin Res Cardiol 2013;102:73–80. 
    Crossref | PubMed
  84. Barco S, Vicaut E, Klok FA, et al. Improved identification of thrombolysis candidates amongst intermediate-risk pulmonary embolism patients: implications for future trials. Eur Respir J 2018;51:1701775. 
    Crossref | PubMed
  85. Solverson K, Humphreys C, Liang Z, et al. Rapid prediction of adverse outcomes for acute normotensive pulmonary embolism: derivation of the Calgary Acute Pulmonary Embolism score. ERJ Open Res 2021;7:00879-2020. 
    Crossref | PubMed
  86. Freitas P, Santos AR, Ferreira AM, et al. Derivation and external validation of the SHIeLD score for predicting outcome in normotensive pulmonary embolism. Int J Cardiol 2019;281:119–24. 
    Crossref | PubMed
  87. Surov A, Akritidou M, Bach AG, et al. A new index for the prediction of 30-day mortality in patients with pulmonary embolism: the pulmonary embolism mortality score (PEMS). Angiology 2021;72:787–93. 
    Crossref | PubMed
  88. Weekes AJ, Raper JD, Lupez K, et al. Development and validation of a prognostic tool: pulmonary embolism short-term clinical outcomes risk estimation (PE-SCORE). PLoS One 2021;16:e0260036. 
    Crossref | PubMed
  89. Dellas C, Tschepe M, Seeber V, et al. A novel H-FABP assay and a fast prognostic score for risk assessment of normotensive pulmonary embolism. Thromb Haemost 2014;111:996–1003. 
    Crossref | PubMed
  90. Jaff MR, McMurtry MS, Archer SL, et al. Management of massive and submassive pulmonary embolism, iliofemoral deep vein thrombosis, and chronic thromboembolic pulmonary hypertension: a scientific statement from the American Heart Association. Circulation 2011;123:1788–830. 
    Crossref | PubMed
  91. Kearon C, Akl EA, Ornelas J, et al. Antithrombotic therapy for VTE disease: CHEST guideline and expert panel report. Chest 2016;149:315–52. 
    Crossref | PubMed
  92. Steering Committee. Single-bolus Tenecteplase plus heparin compared with heparin alone for normotensive patients with acute pulmonary embolism who have evidence of right ventricular dysfunction and myocardial injury: rationale and design of the Pulmonary Embolism Thrombolysis (PEITHO) trial. Am Heart J 2012;163:33–38.e1. 
    Crossref | PubMed
  93. Banerjee I, Sofela M, Yang J, et al. Development and performance of the pulmonary embolism result forecast model (PERFORM) for computed tomography clinical decision support. JAMA Netw Open 2019;2:e198719. 
    Crossref | PubMed
  94. Soffer S, Klang E, Shimon O, et al. Deep learning for pulmonary embolism detection on computed tomography pulmonary angiogram: a systematic review and meta-analysis. Sci Rep 2021;11:15814. 
    Crossref | PubMed
  95. Lanza E, Ammirabile A, Francone M. nnU-Net-based deep-learning for pulmonary embolism: detection, clot volumen quantification, and severity correlation in the RSPECT dataset. Eur J Radiol 2024;177:111592. 
    Crossref | PubMed
  96. Liu W, Liu M, Guo X, et al. Evaluation of acute pulmonary embolism and clot burden on CTPA with deep learning. Eur Radiol 2020;30:3567–75. 
    Crossref | PubMed
  97. Foley RW, Glenn-Cox S, Rossdale J, et al. Automated calculation of the right ventricle to left ventricle ratio on CT for the risk stratification of patients with acute pulmonary embolism. Eur Radiol 2021;31:6013–20. 
    Crossref | PubMed
Previous Volume Article Next Volume Article