1
|
Ayobi A, Chang PD, Chow DS, Weinberg BD, Tassy M, Franciosini A, Scudeler M, Quenet S, Avare C, Chaibi Y. Performance and clinical utility of an artificial intelligence-enabled tool for pulmonary embolism detection. Clin Imaging 2024; 113:110245. [PMID: 39094243 DOI: 10.1016/j.clinimag.2024.110245] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 05/14/2024] [Revised: 07/25/2024] [Accepted: 07/27/2024] [Indexed: 08/04/2024]
Abstract
PURPOSE Diagnosing pulmonary embolism (PE) is still challenging due to other conditions that can mimic its appearance, leading to incomplete or delayed management and several inter-observer variabilities. This study evaluated the performance and clinical utility of an artificial intelligence (AI)-based application designed to assist clinicians in the detection of PE on CT pulmonary angiography (CTPA). PATIENTS AND METHODS CTPAs from 230 US cities acquired on 57 scanner models from 6 different vendors were retrospectively collected. Three US board certified expert radiologists defined the ground truth by majority agreement. The same cases were analyzed by CINA-PE, an AI-driven algorithm capable of detecting and highlighting suspected PE locations. The algorithm's performance at a per-case and per-finding level was evaluated. Furthermore, cases with PE not mentioned in the clinical report but correctly detected by the algorithm were analyzed. RESULTS A total of 1204 CTPAs (mean age 62.1 years ± 16.6[SD], 44.4 % female, 14.9 % positive) were included in the study. Per-case sensitivity and specificity were 93.9 % (95%CI: 89.3 %-96.9 %) and 94.8 % (95%CI: 93.3 %-96.1 %), respectively. Per-finding positive predictive value was 89.5 % (95%CI: 86.7 %-91.9 %). Among the 196 positive cases, 29 (15.6 %) were not mentioned in the clinical report. The algorithm detected 22/29 (76 %) of these cases, leading to a reduction in the miss rate from 15.6 % to 3.8 % (7/186). CONCLUSIONS The AI-based application may improve diagnostic accuracy in detecting PE and enhance patient outcomes through timely intervention. Integrating AI tools in clinical workflows can reduce missed or delayed diagnoses, and positively impact healthcare delivery and patient care.
Collapse
Affiliation(s)
- Angela Ayobi
- Avicenna.AI, 375 Avenue du Mistral, 13600 La Ciotat, France
| | - Peter D Chang
- Department of Radiological Sciences, University of California Irvine, Irvine, CA 92697, USA; Center for Artificial Intelligence in Diagnostic Medicine, University of California Irvine, Irvine, CA 92697, USA
| | - Daniel S Chow
- Department of Radiological Sciences, University of California Irvine, Irvine, CA 92697, USA; Center for Artificial Intelligence in Diagnostic Medicine, University of California Irvine, Irvine, CA 92697, USA
| | - Brent D Weinberg
- Department of Radiology and Imaging Sciences, Emory University, Atlanta, GA 30322, USA
| | - Maxime Tassy
- Avicenna.AI, 375 Avenue du Mistral, 13600 La Ciotat, France
| | | | | | - Sarah Quenet
- Avicenna.AI, 375 Avenue du Mistral, 13600 La Ciotat, France
| | | | - Yasmina Chaibi
- Avicenna.AI, 375 Avenue du Mistral, 13600 La Ciotat, France
| |
Collapse
|
2
|
Robb CL, Marquis KM, Steinbrecher KL, Hammer MM, Henry T, Bhalla S, Raptis CA. Evaluation and Utilization of Flow Artifacts at CT. Radiographics 2024; 44:e230134. [PMID: 38662588 DOI: 10.1148/rg.230134] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 05/29/2024]
Abstract
Flow artifacts are commonly encountered at contrast-enhanced CT and can be difficult to discern from true pathologic conditions. Therefore, radiologists must be comfortable distinguishing flow artifacts from true pathologic conditions. This is of particular importance when evaluating the pulmonary arteries and aorta, as a flow artifact may be mistaken for a pulmonary embolism or dissection flap. Understanding the mechanics of flow artifacts and how these artifacts are created can help radiologists in several ways. First, this knowledge can help radiologists appreciate how the imaging characteristics of flow artifacts differ from true pathologic conditions. This information can also help radiologists better recognize the clinical conditions that predispose patients to flow artifacts, such as pneumonia, chronic lung damage, and altered cardiac output. By understanding when flow artifacts may be confounding the interpretation of an examination, radiologists can then know when to pursue other troubleshooting methods to assist with the diagnosis. In these circumstances, the radiologist can consider several troubleshooting methods, including adjusting the imaging protocols, recommending when additional imaging may be helpful, and suggesting which imaging study would be the most beneficial. Finally, flow artifacts can also be used as a diagnostic tool when evaluating the vascular anatomy, examples of which include the characterization of shunts, venous collaterals, intimomedial flaps, and alternative patterns of blood flow, as seen in extracorporeal membrane oxygenation circuits. ©RSNA, 2024 Test Your Knowledge questions for this article are available in the supplemental material.
Collapse
Affiliation(s)
- Caroline L Robb
- From the Department of Radiology, Mallinckrodt Institute of Radiology, Washington University, 510 S Kingshighway Blvd, CB 8131, St Louis, MO 63110 (C.L.R., K.M.M., K.L.S., S.B., C.A.R.); Department of Radiology, Brigham and Women's Hospital, Boston, Mass (M.M.H.); and Department of Radiology, Duke University, Durham, NC (T.H.)
| | - Kaitlin M Marquis
- From the Department of Radiology, Mallinckrodt Institute of Radiology, Washington University, 510 S Kingshighway Blvd, CB 8131, St Louis, MO 63110 (C.L.R., K.M.M., K.L.S., S.B., C.A.R.); Department of Radiology, Brigham and Women's Hospital, Boston, Mass (M.M.H.); and Department of Radiology, Duke University, Durham, NC (T.H.)
| | - Kacie L Steinbrecher
- From the Department of Radiology, Mallinckrodt Institute of Radiology, Washington University, 510 S Kingshighway Blvd, CB 8131, St Louis, MO 63110 (C.L.R., K.M.M., K.L.S., S.B., C.A.R.); Department of Radiology, Brigham and Women's Hospital, Boston, Mass (M.M.H.); and Department of Radiology, Duke University, Durham, NC (T.H.)
| | - Mark M Hammer
- From the Department of Radiology, Mallinckrodt Institute of Radiology, Washington University, 510 S Kingshighway Blvd, CB 8131, St Louis, MO 63110 (C.L.R., K.M.M., K.L.S., S.B., C.A.R.); Department of Radiology, Brigham and Women's Hospital, Boston, Mass (M.M.H.); and Department of Radiology, Duke University, Durham, NC (T.H.)
| | - Travis Henry
- From the Department of Radiology, Mallinckrodt Institute of Radiology, Washington University, 510 S Kingshighway Blvd, CB 8131, St Louis, MO 63110 (C.L.R., K.M.M., K.L.S., S.B., C.A.R.); Department of Radiology, Brigham and Women's Hospital, Boston, Mass (M.M.H.); and Department of Radiology, Duke University, Durham, NC (T.H.)
| | - Sanjeev Bhalla
- From the Department of Radiology, Mallinckrodt Institute of Radiology, Washington University, 510 S Kingshighway Blvd, CB 8131, St Louis, MO 63110 (C.L.R., K.M.M., K.L.S., S.B., C.A.R.); Department of Radiology, Brigham and Women's Hospital, Boston, Mass (M.M.H.); and Department of Radiology, Duke University, Durham, NC (T.H.)
| | - Constantine A Raptis
- From the Department of Radiology, Mallinckrodt Institute of Radiology, Washington University, 510 S Kingshighway Blvd, CB 8131, St Louis, MO 63110 (C.L.R., K.M.M., K.L.S., S.B., C.A.R.); Department of Radiology, Brigham and Women's Hospital, Boston, Mass (M.M.H.); and Department of Radiology, Duke University, Durham, NC (T.H.)
| |
Collapse
|
3
|
Robb C, Rajput MZ, Raptis D, Bhalla S. Don't skip a beat! Critical findings in imaging studies performed in adults with congenital heart disease. Curr Probl Diagn Radiol 2024; 53:297-307. [PMID: 38272749 DOI: 10.1067/j.cpradiol.2024.01.005] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 03/06/2023] [Revised: 12/14/2023] [Accepted: 01/16/2024] [Indexed: 01/27/2024]
Abstract
With ongoing advances in both medical and surgical management, the population of adults with congenital heart disease (CHD) continues to grow each year and has surpassed the number of pediatric cases. These adult patients will present to adult emergency departments with increasing frequency. Adults with CHD are at increased risk of developing not only cardiovascular complications, such as aortic dissection and thromboemboli, but also abdominopelvic and neurologic processes at younger ages. These individuals are also more likely to develop less urgent but clinically significant complications including end-organ dysfunction, baffle leaks, or bleeding collateral vessels. Ultimately, imaging can play a critical role in determining the triage, diagnosis, and management of adult CHD patients. To accomplish this goal, radiologists must be able to distinguish acute and chronic complications of treated CHD from benign processes, including expected post-surgical changes or imaging artifacts. Radiologists also need to be familiar with the various long-term risks and complications associated with both treated and untreated forms of CHD, particularly those in adults with complex lesions.
Collapse
Affiliation(s)
- Caroline Robb
- Mallinckrodt Institute of Radiology, Washington University in St. Louis, 510 S. Kingshighway Boulevard, Campus Box 8131, St. Louis, MO 63110, USA
| | - M Zak Rajput
- Mallinckrodt Institute of Radiology, Washington University in St. Louis, 510 S. Kingshighway Boulevard, Campus Box 8131, St. Louis, MO 63110, USA.
| | - Demetrios Raptis
- Mallinckrodt Institute of Radiology, Washington University in St. Louis, 510 S. Kingshighway Boulevard, Campus Box 8131, St. Louis, MO 63110, USA
| | - Sanjeev Bhalla
- Mallinckrodt Institute of Radiology, Washington University in St. Louis, 510 S. Kingshighway Boulevard, Campus Box 8131, St. Louis, MO 63110, USA
| |
Collapse
|
4
|
Szabari MV, Ni C, Davila D, Viragh K. Pulmonary Arterial Tumor Embolism From Recurrent Metastatic Renal Cell Carcinoma on FDG PET/CT. Clin Nucl Med 2024; 49:160-161. [PMID: 37976430 DOI: 10.1097/rlu.0000000000004956] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/19/2023]
Abstract
ABSTRACT A 55-year-old man with renal cell carcinoma extending into the renal vein/inferior vena cava (status post nephrectomy and inferior vena cava thrombectomy, pT3bN0M0), and perioperative pulmonary bland thromboembolism (resolved with 3-month of anticoagulation), followed by 3.5 years of complete remission, developed new incidental pulmonary arterial filling defects on a surveillance CT examination (asymptomatic, normal d -dimer, no deep vein thrombosis). Despite anticoagulation, the filling defects not only persisted but also demonstrated intense FDG activity on a restaging PET/CT performed 4 months later for new pulmonary oligometastasis. The FDG activity resolved after systemic immunotherapy, which suggested the retrospective diagnosis of pulmonary arterial tumor emboli, a rare finding.
Collapse
Affiliation(s)
- Margit V Szabari
- From the Department of Radiology, Ronald Reagan UCLA Medical Center, Los Angeles
| | - Chiayi Ni
- Department of Radiology and Nuclear Medicine, Olive View-UCLA Medical Center, Sylmar, CA
| | - Diego Davila
- Department of Radiology and Nuclear Medicine, Olive View-UCLA Medical Center, Sylmar, CA
| | - Karoly Viragh
- Department of Radiology and Nuclear Medicine, Olive View-UCLA Medical Center, Sylmar, CA
| |
Collapse
|
5
|
Pascoe HM, Pascoe D, McCusker MW, Heinze SB. Turbulent flow artefact mimicking central pulmonary emboli in pulmonary hypertension: A report of two cases. J Med Imaging Radiat Oncol 2019; 63:481-483. [PMID: 31021530 DOI: 10.1111/1754-9485.12893] [Citation(s) in RCA: 2] [Impact Index Per Article: 0.3] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 11/16/2018] [Accepted: 03/26/2019] [Indexed: 12/01/2022]
Abstract
Many artefacts can mimic pulmonary emboli (PE) on CT pulmonary angiography (CTPA), most commonly in the peripheral pulmonary arteries (PA's). We describe flow-related artefact mimicking PE in the central PA's of two patients, both with pulmonary arterial hypertension (PAH). To our knowledge, this is the first report of this PE mimic in the central PA's.
Collapse
Affiliation(s)
- Heather M Pascoe
- Department of Radiology, Royal Melbourne Hospital, University of Melbourne, Melbourne, Victoria, Australia
| | - Diane Pascoe
- Department of Radiology, Royal Melbourne Hospital, University of Melbourne, Melbourne, Victoria, Australia.,University of Melbourne, Melbourne, Victoria, Australia
| | - Mark W McCusker
- Department of Radiology, Royal Melbourne Hospital, University of Melbourne, Melbourne, Victoria, Australia.,University of Melbourne, Melbourne, Victoria, Australia
| | - Stefan B Heinze
- Department of Radiology, Royal Melbourne Hospital, University of Melbourne, Melbourne, Victoria, Australia.,University of Melbourne, Melbourne, Victoria, Australia
| |
Collapse
|