Incidental pulmonary embolism in patients with cancer: prevalence, underdiagnosis and evaluation of an AI algorithm for automatic detection of pulmonary embolism

Wiklund, Peder; Medson, Koshiar; Elf, Johan

doi:10.1007/s00330-022-09071-0

Cited by 18 publications

(3 citation statements)

References 18 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This may likely be due to sub-segmental pulmonary embolus or due to non-optimized protocol for the pulmonary arteries as non-CT Pulmonary Angiogram (non-CTPA) studies were also processed by the algorithms; there may be flow artefacts or perivascular infiltrates rendering the image algorithms to struggle. 18 , 19 This was reflected by the 50% sensitivity at 98% specificity, for which there was a potential risk of over-diagnosis of false positives. Similar algorithms for chest CTPA 9 showed AI-assisted identification of incidental pulmonary embolism reduced the miss rate from 44.8% (without AI) to 2.6% which was remarkable compared to the static pulmonary embolism algorithm used in our study.…”

Section: Discussionmentioning

confidence: 99%

A real-world evaluation of the diagnostic accuracy of radiologists using positive predictive values verified from deep learning and natural language processing chest algorithms deployed retrospectively

Bhatia,

Morlese,

Yusuf

et al. 2023

BJR|Open

View full text Add to dashboard Cite

Objectives This diagnostic study assessed the accuracy of radiologists retrospectively, using the deep learning and natural language processing chest algorithms implemented in Clinical Review version 3.2 for: pneumothorax, rib fractures in digital chest X-ray radiographs (CXR); aortic aneurysm, pulmonary nodules, emphysema, and pulmonary embolism in CT images. Methods The study design was double-blind (artificial intelligence [AI] algorithms and humans), retrospective, non-interventional, and at a single NHS Trust. Adult patients (≥18 years old) scheduled for CXR and CT were invited to enroll as participants through an opt-out process. Reports and images were de-identified, processed retrospectively, and AI-flagged discrepant findings were assigned to two lead radiologists, each blinded to patient identifiers and original radiologist. The radiologist’s findings for each clinical condition were tallied as a verified discrepancy (true positive) or not (false positive). Results The missed findings were: 0.02% rib fractures, 0.51% aortic aneurysm, 0.32% pulmonary nodules, 0.92% emphysema, and 0.28% pulmonary embolism. The positive predictive values (PPVs) were: pneumothorax (0%), rib fractures (5.6%), aortic dilatation (43.2%), pulmonary emphysema (46.0%), pulmonary embolus (11.5%), and pulmonary nodules (9.2%). The PPV for pneumothorax was nil owing to lack of available studies that were analysed for outpatient activity. Conclusions The number of missed findings was far less than generally predicted. The chest algorithms deployed retrospectively were a useful quality tool and AI augmented the radiologists’ workflow. Advances in knowledge The diagnostic accuracy of our radiologists generated missed findings of 0.02% for rib fractures CXR, 0.51% for aortic dilatation, 0.32% for pulmonary nodule, 0.92% for pulmonary emphysema, and 0.28% for pulmonary embolism for CT studies, all retrospectively evaluated with AI used as a quality tool to flag potential missed findings. It is important to account for prevalence of these chest conditions in clinical context and use appropriate clinical thresholds for decision-making, not relying solely on AI.

show abstract

Section: Discussionmentioning

confidence: 99%

A real-world evaluation of the diagnostic accuracy of radiologists using positive predictive values verified from deep learning and natural language processing chest algorithms deployed retrospectively

Bhatia,

Morlese,

Yusuf

et al. 2023

BJR|Open

View full text Add to dashboard Cite

show abstract

“…Further use of bilateral Doppler ultrasound of the limb veins or CT venography of the deep leg veins is then needed to confirm or exclude DVT and to inform the pathway for subsequent treatment [ 26 , 27 ]. In addition, novel artificial intelligence algorithms, such as Aidoc, may become increasingly useful in assisting diagnosis due to their high specificity and sensitivity in the detection of incidental PE [ 28 ].…”

Section: Reviewmentioning

confidence: 99%

Perioperative Management of Incidental Pulmonary Embolisms on Trauma CT Scans: A Narrative Review

Rama¹,

Adeosun²,

Thahir³

et al. 2023

Cureus

View full text Add to dashboard Cite

Unsuspected pulmonary embolism (PE) may be identified on an initial trauma computed tomography (CT) scan. The clinical importance of these incidental PEs remains to be elucidated. In patients who require surgery, careful management is needed. We sought to investigate the optimal perioperative management of such patients, including the use of pharmacological and mechanical thromboprophylaxis, possible thrombolytic therapy, and inferior vena cava (IVC) filters.A literature search was conducted, and all relevant articles were identified, investigated, and included. Medical guidelines were also consulted where appropriate. Pharmacological thromboprophylaxis is the mainstay of preoperative treatment, and low-molecular-weight heparins, fondaparinux, and unfractionated heparin may all be used. It has been suggested that prophylaxis should be administered as soon as possible after trauma. Such agents may be contraindicated in patients with significant bleeding, and mechanical prophylaxis and inferior vena cava filters may be favoured in these patients. Therapeutic anticoagulation and thrombolytic therapies may be considered but are associated with an increased risk of haemorrhage. Delaying surgery might help to minimise the risk of recurrent venous thromboembolism, and any interruption of prophylaxis must be strategically planned. Recommendations for postoperative care include a continuation of prophylaxis and therapeutic anticoagulation, with follow-up clinical evaluation within six months.Incidental PE is a common finding on trauma CT scans. Although the clinical significance is unknown, careful management of the balance between anticoagulation and bleeding is needed, especially in trauma patients and even more so in trauma patients requiring surgery.

show abstract

“…AI has been introduced and put into practical use in the United States, primarily in radiological imaging [ 5 , 6 ]. This method utilizes deep learning, where the AI is trained based on supervised images.…”

Section: Introductionmentioning

confidence: 99%

Construction and Validation of an Image Discrimination Algorithm to Discriminate Necrosis from Wounds in Pressure Ulcers

Sakakibara

Takekawa

et al. 2023

JCM

View full text Add to dashboard Cite

Artificial intelligence (AI) in medical care can raise diagnosis accuracy and improve its uniformity. This study developed a diagnostic imaging system for chronic wounds that can be used in medically underpopulated areas. The image identification algorithm searches for patterns and makes decisions based on information obtained from pixels rather than images. Images of 50 patients with pressure sores treated at Kobe University Hospital were examined. The algorithm determined the presence of necrosis with a significant difference (p = 3.39 × 10−5). A threshold value was created with a luminance difference of 50 for the group with necrosis of 5% or more black pixels. In the no-necrosis group with less than 5% black pixels, the threshold value was created with a brightness difference of 100. The “shallow wounds” were distributed below 100, whereas the “deep wounds” were distributed above 100. When the algorithm was applied to 24 images of 23 new cases, there was 100% agreement between the specialist and the algorithm regarding the presence of necrotic tissue and wound depth evaluation. The algorithm identifies the necrotic tissue and wound depth without requiring a large amount of data, making it suitable for application to future AI diagnosis systems for chronic wounds.

show abstract

Incidental pulmonary embolism in patients with cancer: prevalence, underdiagnosis and evaluation of an AI algorithm for automatic detection of pulmonary embolism

Cited by 18 publications

References 18 publications

A real-world evaluation of the diagnostic accuracy of radiologists using positive predictive values verified from deep learning and natural language processing chest algorithms deployed retrospectively

A real-world evaluation of the diagnostic accuracy of radiologists using positive predictive values verified from deep learning and natural language processing chest algorithms deployed retrospectively

Perioperative Management of Incidental Pulmonary Embolisms on Trauma CT Scans: A Narrative Review

Construction and Validation of an Image Discrimination Algorithm to Discriminate Necrosis from Wounds in Pressure Ulcers

Contact Info

Product

Resources

About