1
|
Slieker FJB, Van Gemert JTM, Seydani MG, Farsai S, Breimer GE, Forouzanfar T, de Bree R, Rosenberg AJWP, Van Cann EM. Value of cone beam computed tomography for detecting bone invasion in squamous cell carcinoma of the maxilla. Oral Surg Oral Med Oral Pathol Oral Radiol 2022; 134:102-109. [PMID: 35318943 DOI: 10.1016/j.oooo.2022.01.020] [Citation(s) in RCA: 1] [Impact Index Per Article: 0.5] [Reference Citation Analysis] [Abstract] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 07/28/2021] [Revised: 01/13/2022] [Accepted: 01/28/2022] [Indexed: 12/16/2022]
Abstract
OBJECTIVE To determine the diagnostic value of cone beam computed tomography (CBCT) in detecting bone invasion in maxillary squamous cell carcinoma (MSCC). STUDY DESIGN In this retrospective cohort study, preoperative CBCT scans were independently assessed by a single surgeon in imaging assessment 1 (IA 1) and by 1 surgeon with 2 dentists in consensus (IA 2) for the presence of bone invasion in MSCC. Sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), accuracy, area under the receiver operating characteristic curve (AUC), and Cohen's κ were calculated. Histopathologic results of resection specimens served as the reference standard. RESULTS Of 27 patients, 19 (70%) had proven bone invasion. IA 1 yielded 68.4% sensitivity, 75.0% specificity, 86.7% PPV, 50.0% NPV, 70.4% accuracy, and 0.717 AUC. All results of IA 2 were true-positive and true-negative, resulting in 100% sensitivity, specificity, PPV, NPV, accuracy, and AUC. The assessments differed in 6 cases. Interobserver κ was fair (0.38, 95% CI 0.04-0.72, P = .038). There was a significant association between CBCT detection of bone invasion and extent of surgical treatment (P = .006) CONCLUSIONS: The diagnostic accuracy of CBCT was high but observer-dependent. CBCT examination may be useful in surgical treatment planning.
Collapse
Affiliation(s)
- F J B Slieker
- Department of Head and Neck Surgical Oncology, University Medical Center Utrecht, Utrecht, The Netherlands; Department of Oral and Maxillofacial Surgery, University Medical Center Utrecht, Utrecht, The Netherlands
| | - J T M Van Gemert
- Department of Head and Neck Surgical Oncology, University Medical Center Utrecht, Utrecht, The Netherlands; Department of Oral and Maxillofacial Surgery, University Medical Center Utrecht, Utrecht, The Netherlands
| | - M Ghafoori Seydani
- Department of Head and Neck Surgical Oncology, University Medical Center Utrecht, Utrecht, The Netherlands; Department of Oral and Maxillofacial Surgery/Oral Pathology, VU University Medical Center/Academic Centre for Dentistry Amsterdam (ACTA), Amsterdam, The Netherlands
| | - S Farsai
- Department of Head and Neck Surgical Oncology, University Medical Center Utrecht, Utrecht, The Netherlands; Department of Oral and Maxillofacial Surgery/Oral Pathology, VU University Medical Center/Academic Centre for Dentistry Amsterdam (ACTA), Amsterdam, The Netherlands
| | - G E Breimer
- Department of Pathology, University Medical Center Utrecht, Utrecht, The Netherlands
| | - T Forouzanfar
- Department of Oral and Maxillofacial Surgery/Oral Pathology, VU University Medical Center/Academic Centre for Dentistry Amsterdam (ACTA), Amsterdam, The Netherlands
| | - R de Bree
- Department of Head and Neck Surgical Oncology, University Medical Center Utrecht, Utrecht, The Netherlands
| | - A J W P Rosenberg
- Department of Oral and Maxillofacial Surgery, University Medical Center Utrecht, Utrecht, The Netherlands
| | - E M Van Cann
- Department of Head and Neck Surgical Oncology, University Medical Center Utrecht, Utrecht, The Netherlands; Department of Oral and Maxillofacial Surgery, University Medical Center Utrecht, Utrecht, The Netherlands.
| |
Collapse
|
2
|
Akbar MN, Wang X, Erdogmus D, Dalal S. PENet: Continuous-Valued Pulmonary Edema Severity Prediction On Chest X-ray Using Siamese Convolutional Networks. ANNUAL INTERNATIONAL CONFERENCE OF THE IEEE ENGINEERING IN MEDICINE AND BIOLOGY SOCIETY. IEEE ENGINEERING IN MEDICINE AND BIOLOGY SOCIETY. ANNUAL INTERNATIONAL CONFERENCE 2022; 2022:1834-1838. [PMID: 36086469 DOI: 10.1109/embc48229.2022.9871153] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [MESH Headings] [Track Full Text] [Subscribe] [Scholar Register] [Indexed: 06/15/2023]
Abstract
For physicians to make rapid clinical decisions for patients with congestive heart failure, the assessment of pulmonary edema severity in chest radiographs is vital. Although deep learning has shown promise in detecting the presence or absence or discrete grades of severity, of such edema, prediction of continuous-valued severity yet remains a challenge. Here, we propose PENet: Siamese convolutional neural networks to assess the continuous spectrum of severity of lung edema from chest radiographs. We present different modes of implementing this network and demonstrate that our best model outperforms that of earlier work (mean AUC of 0.91 over 0.87), while using only 1/16-th the dimension of input images and 1/69-th the size of training data, thus also saving expensive computation.
Collapse
|
3
|
Horng S, Liao R, Wang X, Dalal S, Golland P, Berkowitz SJ. Deep Learning to Quantify Pulmonary Edema in Chest Radiographs. Radiol Artif Intell 2021; 3:e190228. [PMID: 33937857 DOI: 10.1148/ryai.2021190228] [Citation(s) in RCA: 11] [Impact Index Per Article: 3.7] [Reference Citation Analysis] [Abstract] [Journal Information] [Subscribe] [Scholar Register] [Received: 12/24/2019] [Revised: 12/07/2020] [Accepted: 12/17/2020] [Indexed: 12/15/2022]
Abstract
Purpose To develop a machine learning model to classify the severity grades of pulmonary edema on chest radiographs. Materials and Methods In this retrospective study, 369 071 chest radiographs and associated radiology reports from 64 581 patients (mean age, 51.71 years; 54.51% women) from the MIMIC-CXR chest radiograph dataset were included. This dataset was split into patients with and without congestive heart failure (CHF). Pulmonary edema severity labels from the associated radiology reports were extracted from patients with CHF as four different ordinal levels: 0, no edema; 1, vascular congestion; 2, interstitial edema; and 3, alveolar edema. Deep learning models were developed using two approaches: a semisupervised model using a variational autoencoder and a pretrained supervised learning model using a dense neural network. Receiver operating characteristic curve analysis was performed on both models. Results The area under the receiver operating characteristic curve (AUC) for differentiating alveolar edema from no edema was 0.99 for the semisupervised model and 0.87 for the pretrained models. Performance of the algorithm was inversely related to the difficulty in categorizing milder states of pulmonary edema (shown as AUCs for semisupervised model and pretrained model, respectively): 2 versus 0, 0.88 and 0.81; 1 versus 0, 0.79 and 0.66; 3 versus 1, 0.93 and 0.82; 2 versus 1, 0.69 and 0.73; and 3 versus 2, 0.88 and 0.63. Conclusion Deep learning models were trained on a large chest radiograph dataset and could grade the severity of pulmonary edema on chest radiographs with high performance.Supplemental material is available for this article.See also the commentary by Auffermann in this issue.© RSNA, 2021.
Collapse
Affiliation(s)
- Steven Horng
- Department of Radiology, Beth Israel Deaconess Medical Center, Harvard Medical School, 330 Brookline Ave, Boston, MA 02215 (S.H., S.J.B.); Department of Electrical Engineering and Computer Science, Massachusetts Institute of Technology, Cambridge, Mass (R.L., P.G.); and Clinical Informatics Solutions and Services, Philips Research, Cambridge, Mass (X.W., S.D.).,S.H. (e-mail: )
| | - Ruizhi Liao
- Department of Radiology, Beth Israel Deaconess Medical Center, Harvard Medical School, 330 Brookline Ave, Boston, MA 02215 (S.H., S.J.B.); Department of Electrical Engineering and Computer Science, Massachusetts Institute of Technology, Cambridge, Mass (R.L., P.G.); and Clinical Informatics Solutions and Services, Philips Research, Cambridge, Mass (X.W., S.D.).,S.H. (e-mail: )
| | - Xin Wang
- Department of Radiology, Beth Israel Deaconess Medical Center, Harvard Medical School, 330 Brookline Ave, Boston, MA 02215 (S.H., S.J.B.); Department of Electrical Engineering and Computer Science, Massachusetts Institute of Technology, Cambridge, Mass (R.L., P.G.); and Clinical Informatics Solutions and Services, Philips Research, Cambridge, Mass (X.W., S.D.).,S.H. (e-mail: )
| | - Sandeep Dalal
- Department of Radiology, Beth Israel Deaconess Medical Center, Harvard Medical School, 330 Brookline Ave, Boston, MA 02215 (S.H., S.J.B.); Department of Electrical Engineering and Computer Science, Massachusetts Institute of Technology, Cambridge, Mass (R.L., P.G.); and Clinical Informatics Solutions and Services, Philips Research, Cambridge, Mass (X.W., S.D.).,S.H. (e-mail: )
| | - Polina Golland
- Department of Radiology, Beth Israel Deaconess Medical Center, Harvard Medical School, 330 Brookline Ave, Boston, MA 02215 (S.H., S.J.B.); Department of Electrical Engineering and Computer Science, Massachusetts Institute of Technology, Cambridge, Mass (R.L., P.G.); and Clinical Informatics Solutions and Services, Philips Research, Cambridge, Mass (X.W., S.D.).,S.H. (e-mail: )
| | - Seth J Berkowitz
- Department of Radiology, Beth Israel Deaconess Medical Center, Harvard Medical School, 330 Brookline Ave, Boston, MA 02215 (S.H., S.J.B.); Department of Electrical Engineering and Computer Science, Massachusetts Institute of Technology, Cambridge, Mass (R.L., P.G.); and Clinical Informatics Solutions and Services, Philips Research, Cambridge, Mass (X.W., S.D.).,S.H. (e-mail: )
| |
Collapse
|
4
|
Li MD, Arun NT, Gidwani M, Chang K, Deng F, Little BP, Mendoza DP, Lang M, Lee SI, O’Shea A, Parakh A, Singh P, Kalpathy-Cramer J. Automated Assessment and Tracking of COVID-19 Pulmonary Disease Severity on Chest Radiographs using Convolutional Siamese Neural Networks. Radiol Artif Intell 2020; 2:e200079. [PMID: 33928256 PMCID: PMC7392327 DOI: 10.1148/ryai.2020200079] [Citation(s) in RCA: 69] [Impact Index Per Article: 17.3] [Reference Citation Analysis] [Abstract] [Grants] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 12/15/2022]
Abstract
PURPOSE To develop an automated measure of COVID-19 pulmonary disease severity on chest radiographs (CXRs), for longitudinal disease tracking and outcome prediction. MATERIALS AND METHODS A convolutional Siamese neural network-based algorithm was trained to output a measure of pulmonary disease severity on CXRs (pulmonary x-ray severity (PXS) score), using weakly-supervised pretraining on ∼160,000 anterior-posterior images from CheXpert and transfer learning on 314 frontal CXRs from COVID-19 patients. The algorithm was evaluated on internal and external test sets from different hospitals (154 and 113 CXRs respectively). PXS scores were correlated with radiographic severity scores independently assigned by two thoracic radiologists and one in-training radiologist (Pearson r). For 92 internal test set patients with follow-up CXRs, PXS score change was compared to radiologist assessments of change (Spearman ρ). The association between PXS score and subsequent intubation or death was assessed. Bootstrap 95% confidence intervals (CI) were calculated. RESULTS PXS scores correlated with radiographic pulmonary disease severity scores assigned to CXRs in the internal and external test sets (r=0.86 (95%CI 0.80-0.90) and r=0.86 (95%CI 0.79-0.90) respectively). The direction of change in PXS score in follow-up CXRs agreed with radiologist assessment (ρ=0.74 (95%CI 0.63-0.81)). In patients not intubated on the admission CXR, the PXS score predicted subsequent intubation or death within three days of hospital admission (area under the receiver operating characteristic curve=0.80 (95%CI 0.75-0.85)). CONCLUSION A Siamese neural network-based severity score automatically measures radiographic COVID-19 pulmonary disease severity, which can be used to track disease change and predict subsequent intubation or death.
Collapse
Affiliation(s)
- Matthew D. Li
- From the Athinoula A. Martinos Center for Biomedical Imaging (M.D.L., N.T.A., M.G., K.C., P.S., J.K.C.), Department of Radiology (F.D., M.L.), Division of Thoracic Imaging and Intervention (B.P.L, D.P.M.), Division of Abdominal Imaging (S.I.L., A.O., A.P.), and MGH and BWH Center for Clinical Data Science (J.K.) of the Massachusetts General Hospital, Harvard Medical School, Boston, MA, USA
| | - Nishanth Thumbavanam Arun
- From the Athinoula A. Martinos Center for Biomedical Imaging (M.D.L., N.T.A., M.G., K.C., P.S., J.K.C.), Department of Radiology (F.D., M.L.), Division of Thoracic Imaging and Intervention (B.P.L, D.P.M.), Division of Abdominal Imaging (S.I.L., A.O., A.P.), and MGH and BWH Center for Clinical Data Science (J.K.) of the Massachusetts General Hospital, Harvard Medical School, Boston, MA, USA
| | - Mishka Gidwani
- From the Athinoula A. Martinos Center for Biomedical Imaging (M.D.L., N.T.A., M.G., K.C., P.S., J.K.C.), Department of Radiology (F.D., M.L.), Division of Thoracic Imaging and Intervention (B.P.L, D.P.M.), Division of Abdominal Imaging (S.I.L., A.O., A.P.), and MGH and BWH Center for Clinical Data Science (J.K.) of the Massachusetts General Hospital, Harvard Medical School, Boston, MA, USA
| | - Ken Chang
- From the Athinoula A. Martinos Center for Biomedical Imaging (M.D.L., N.T.A., M.G., K.C., P.S., J.K.C.), Department of Radiology (F.D., M.L.), Division of Thoracic Imaging and Intervention (B.P.L, D.P.M.), Division of Abdominal Imaging (S.I.L., A.O., A.P.), and MGH and BWH Center for Clinical Data Science (J.K.) of the Massachusetts General Hospital, Harvard Medical School, Boston, MA, USA
| | - Francis Deng
- From the Athinoula A. Martinos Center for Biomedical Imaging (M.D.L., N.T.A., M.G., K.C., P.S., J.K.C.), Department of Radiology (F.D., M.L.), Division of Thoracic Imaging and Intervention (B.P.L, D.P.M.), Division of Abdominal Imaging (S.I.L., A.O., A.P.), and MGH and BWH Center for Clinical Data Science (J.K.) of the Massachusetts General Hospital, Harvard Medical School, Boston, MA, USA
| | - Brent P. Little
- From the Athinoula A. Martinos Center for Biomedical Imaging (M.D.L., N.T.A., M.G., K.C., P.S., J.K.C.), Department of Radiology (F.D., M.L.), Division of Thoracic Imaging and Intervention (B.P.L, D.P.M.), Division of Abdominal Imaging (S.I.L., A.O., A.P.), and MGH and BWH Center for Clinical Data Science (J.K.) of the Massachusetts General Hospital, Harvard Medical School, Boston, MA, USA
| | - Dexter P. Mendoza
- From the Athinoula A. Martinos Center for Biomedical Imaging (M.D.L., N.T.A., M.G., K.C., P.S., J.K.C.), Department of Radiology (F.D., M.L.), Division of Thoracic Imaging and Intervention (B.P.L, D.P.M.), Division of Abdominal Imaging (S.I.L., A.O., A.P.), and MGH and BWH Center for Clinical Data Science (J.K.) of the Massachusetts General Hospital, Harvard Medical School, Boston, MA, USA
| | - Min Lang
- From the Athinoula A. Martinos Center for Biomedical Imaging (M.D.L., N.T.A., M.G., K.C., P.S., J.K.C.), Department of Radiology (F.D., M.L.), Division of Thoracic Imaging and Intervention (B.P.L, D.P.M.), Division of Abdominal Imaging (S.I.L., A.O., A.P.), and MGH and BWH Center for Clinical Data Science (J.K.) of the Massachusetts General Hospital, Harvard Medical School, Boston, MA, USA
| | - Susanna I. Lee
- From the Athinoula A. Martinos Center for Biomedical Imaging (M.D.L., N.T.A., M.G., K.C., P.S., J.K.C.), Department of Radiology (F.D., M.L.), Division of Thoracic Imaging and Intervention (B.P.L, D.P.M.), Division of Abdominal Imaging (S.I.L., A.O., A.P.), and MGH and BWH Center for Clinical Data Science (J.K.) of the Massachusetts General Hospital, Harvard Medical School, Boston, MA, USA
| | - Aileen O’Shea
- From the Athinoula A. Martinos Center for Biomedical Imaging (M.D.L., N.T.A., M.G., K.C., P.S., J.K.C.), Department of Radiology (F.D., M.L.), Division of Thoracic Imaging and Intervention (B.P.L, D.P.M.), Division of Abdominal Imaging (S.I.L., A.O., A.P.), and MGH and BWH Center for Clinical Data Science (J.K.) of the Massachusetts General Hospital, Harvard Medical School, Boston, MA, USA
| | - Anushri Parakh
- From the Athinoula A. Martinos Center for Biomedical Imaging (M.D.L., N.T.A., M.G., K.C., P.S., J.K.C.), Department of Radiology (F.D., M.L.), Division of Thoracic Imaging and Intervention (B.P.L, D.P.M.), Division of Abdominal Imaging (S.I.L., A.O., A.P.), and MGH and BWH Center for Clinical Data Science (J.K.) of the Massachusetts General Hospital, Harvard Medical School, Boston, MA, USA
| | - Praveer Singh
- From the Athinoula A. Martinos Center for Biomedical Imaging (M.D.L., N.T.A., M.G., K.C., P.S., J.K.C.), Department of Radiology (F.D., M.L.), Division of Thoracic Imaging and Intervention (B.P.L, D.P.M.), Division of Abdominal Imaging (S.I.L., A.O., A.P.), and MGH and BWH Center for Clinical Data Science (J.K.) of the Massachusetts General Hospital, Harvard Medical School, Boston, MA, USA
| | - Jayashree Kalpathy-Cramer
- From the Athinoula A. Martinos Center for Biomedical Imaging (M.D.L., N.T.A., M.G., K.C., P.S., J.K.C.), Department of Radiology (F.D., M.L.), Division of Thoracic Imaging and Intervention (B.P.L, D.P.M.), Division of Abdominal Imaging (S.I.L., A.O., A.P.), and MGH and BWH Center for Clinical Data Science (J.K.) of the Massachusetts General Hospital, Harvard Medical School, Boston, MA, USA
| |
Collapse
|
5
|
Li MD, Arun NT, Gidwani M, Chang K, Deng F, Little BP, Mendoza DP, Lang M, Lee SI, O'Shea A, Parakh A, Singh P, Kalpathy-Cramer J. Automated assessment of COVID-19 pulmonary disease severity on chest radiographs using convolutional Siamese neural networks. MEDRXIV : THE PREPRINT SERVER FOR HEALTH SCIENCES 2020. [PMID: 32511570 DOI: 10.1101/2020.05.20.20108159] [Citation(s) in RCA: 16] [Impact Index Per Article: 4.0] [Reference Citation Analysis] [Abstract] [Track Full Text] [Subscribe] [Scholar Register] [Indexed: 11/24/2022]
Abstract
Purpose To develop an automated measure of COVID-19 pulmonary disease severity on chest radiographs (CXRs), for longitudinal disease evaluation and clinical risk stratification. Materials and Methods A convolutional Siamese neural network-based algorithm was trained to output a measure of pulmonary disease severity on anterior-posterior CXRs (pulmonary x-ray severity (PXS) score), using weakly-supervised pretraining on ~160,000 images from CheXpert and transfer learning on 314 CXRs from patients with COVID-19. The algorithm was evaluated on internal and external test sets from different hospitals, containing 154 and 113 CXRs respectively. The PXS score was correlated with a radiographic severity score independently assigned by two thoracic radiologists and one in-training radiologist. For 92 internal test set patients with follow-up CXRs, the change in PXS score was compared to radiologist assessments of change. The association between PXS score and subsequent intubation or death was assessed. Results The PXS score correlated with the radiographic pulmonary disease severity score assigned to CXRs in the COVID-19 internal and external test sets (ρ=0.84 and ρ=0.78 respectively). The direction of change in PXS score in follow-up CXRs agreed with radiologist assessment (ρ=0.74). In patients not intubated on the admission CXR, the PXS score predicted subsequent intubation or death within three days of hospital admission (area under the receiver operator characteristic curve=0.80 (95%CI 0.75-0.85)). Conclusion A Siamese neural network-based severity score automatically measures COVID-19 pulmonary disease severity in chest radiographs, which can be scaled and rapidly deployed for clinical triage and workflow optimization.
Collapse
|
6
|
Dobbe L, Rahman R, Elmassry M, Paz P, Nugent K. Cardiogenic Pulmonary Edema. Am J Med Sci 2019; 358:389-397. [PMID: 31813466 DOI: 10.1016/j.amjms.2019.09.011] [Citation(s) in RCA: 16] [Impact Index Per Article: 3.2] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 03/28/2019] [Revised: 09/15/2019] [Accepted: 09/27/2019] [Indexed: 12/20/2022]
Abstract
The initial events in cardiogenic pulmonary edema involve hemodynamic pulmonary congestion with high capillary pressures. This causes increased fluid transfer out of capillaries into the interstitium and alveolar spaces. High capillary pressures can also cause barrier disruption which increases permeability and fluid transfer into the interstitium and alveoli. Fluid in alveoli alters surfactant function and increases surface tension. This can lead to more edema formation and to atelectasis with impaired gas exchange. Patients with barrier disruption have increased levels of surfactant protein B in the circulation, and these levels often remain high after the initial clinical improvement. Routine clinical assessment may not identify patients with increased extravascular fluid in the lungs; pulmonary ultrasound can easily detect pulmonary edema in patients with acute decompensation and in patients at risk for decompensation. Studies using serial pulmonary ultrasound could help characterize patients with cardiogenic pulmonary edema and help identify subgroups who need alternative management. The conventional management of cardiogenic pulmonary edema usually involves diuresis, afterload reduction and in some cases noninvasive ventilation to reduce the work of breathing and improve oxygenation. Patients with persistent symptoms, abnormal chest x-rays and diuretic resistance might benefit from alternative approaches to management. These could include beta agonists and pentoxifylline which warrant more study in patients with cardiogenic pulmonary edema.
Collapse
Affiliation(s)
- Logan Dobbe
- Department of Graduate Medical Education, Madigan Army Medical Center, Tacoma, Washington
| | - Rubayat Rahman
- Department of Internal Medicine, Texas Tech University Health Sciences Center, Lubbock, Texas
| | - Mohamed Elmassry
- Department of Internal Medicine, Texas Tech University Health Sciences Center, Lubbock, Texas
| | - Pablo Paz
- Department of Internal Medicine, Texas Tech University Health Sciences Center, Lubbock, Texas
| | - Kenneth Nugent
- Department of Internal Medicine, Texas Tech University Health Sciences Center, Lubbock, Texas.
| |
Collapse
|
7
|
Touw HR, Schuitemaker AE, Daams F, van der Peet DL, Bronkhorst EM, Schober P, Boer C, Tuinman PR. Routine lung ultrasound to detect postoperative pulmonary complications following major abdominal surgery: a prospective observational feasibility study. Ultrasound J 2019; 11:20. [PMID: 31523784 PMCID: PMC6745303 DOI: 10.1186/s13089-019-0135-6] [Citation(s) in RCA: 10] [Impact Index Per Article: 2.0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 03/09/2019] [Accepted: 08/16/2019] [Indexed: 12/28/2022] Open
Abstract
Background Postoperative pulmonary complications after major abdominal surgery are associated with adverse outcome. The diagnostic accuracy of chest X-rays (CXR) to detect pulmonary disorders is limited. Alternatively, lung ultrasound (LUS) is an established evidence-based point-of-care diagnostic modality which outperforms CXR in critical care. However, its feasibility and diagnostic ability for postoperative pulmonary complications following abdominal surgery are unknown. In this prospective observational feasibility study, we included consecutive patients undergoing major abdominal surgery with an intermediate or high risk developing postoperative pulmonary complications according to the Assess Respiratory risk In Surgical patients in CATalonia (ARISCAT) score. LUS was routinely performed on postoperative days 0–3 by a researcher blinded for CXR or other clinical findings. Then, reports were drawn up for LUS concerning feasibility and detection rates of postoperative pulmonary complications. CXRs were performed on demand according to daily clinical practice. Subsequently, we compared LUS and CXR findings. Results A total of 98 consecutive patients with an ARISCAT score of 41 (34–49) were included in the study. LUS was feasible in all patients. In 94 (95%) patients, LUS detected one or more postoperative pulmonary complications during the first four postoperative days. On day 0, LUS detected 31 out of 43 patients (72.1%) with one or more postoperative pulmonary complications, compared to 13 out of 36 patients (36.1%) with 1 or more postoperative pulmonary complications detected with CXR RR 2.0 (95 CI [1.24–3.20]) (p = 0.004). The number of discordant observations between both modalities was high for atelectasis 23 (43%) and pleural effusion 29 (54%), but not for pneumothorax, respiratory infection and pulmonary edema 8 (15%), 3 (5%), and 5 (9%), respectively. Conclusions This study shows that LUS is highly feasible and frequently detects postoperative pulmonary complications after major abdominal surgery. Discordant observations in atelectasis and pleural effusions for LUS and CXR can be explained by a superior diagnostic ability of LUS in detecting these conditions. The effects of LUS as primary imaging modality on patient outcome should be evaluated in future studies.
Collapse
Affiliation(s)
- H R Touw
- Department of Anaesthesiology, Amsterdam UMC, Vrije Universiteit Amsterdam, Amsterdam Cardiovascular Sciences, De Boelelaan 1117, 1081 HV, Amsterdam, The Netherlands. .,Department of Intensive Care Medicine, Radboud University Medical Center, Geert Grooteplein Zuid 10, 6525 GA, Nijmegen, The Netherlands.
| | - A E Schuitemaker
- Department of Anaesthesiology, Amsterdam UMC, Vrije Universiteit Amsterdam, Amsterdam Cardiovascular Sciences, De Boelelaan 1117, 1081 HV, Amsterdam, The Netherlands
| | - F Daams
- Department of Surgery, Amsterdam UMC, Vrije Universiteit Amsterdam, Amsterdam Cardiovascular Sciences, De Boelelaan 1117, 1081 HV, Amsterdam, The Netherlands
| | - D L van der Peet
- Department of Surgery, Amsterdam UMC, Vrije Universiteit Amsterdam, Amsterdam Cardiovascular Sciences, De Boelelaan 1117, 1081 HV, Amsterdam, The Netherlands
| | - E M Bronkhorst
- Department of Health Evidence, Radboud University Medical Center, Geert Grooteplein Zuid 10, 6525 GA, Nijmegen, The Netherlands
| | - P Schober
- Department of Anaesthesiology, Amsterdam UMC, Vrije Universiteit Amsterdam, Amsterdam Cardiovascular Sciences, De Boelelaan 1117, 1081 HV, Amsterdam, The Netherlands
| | - C Boer
- Department of Anaesthesiology, Amsterdam UMC, Vrije Universiteit Amsterdam, Amsterdam Cardiovascular Sciences, De Boelelaan 1117, 1081 HV, Amsterdam, The Netherlands
| | - P R Tuinman
- Department of Intensive Care Medicine, Amsterdam UMC, Vrije Universiteit Amsterdam, Amsterdam Cardiovascular Sciences, De Boelelaan 1117, 1081 HV, Amsterdam, The Netherlands
| |
Collapse
|
8
|
Phosgene-induced lung edema: Comparison of clinical criteria for increased extravascular lung water content with postmortem lung gravimetry and lavage-protein in rats and dogs. Toxicol Lett 2019; 305:32-39. [DOI: 10.1016/j.toxlet.2019.01.006] [Citation(s) in RCA: 7] [Impact Index Per Article: 1.4] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 09/29/2018] [Revised: 12/13/2018] [Accepted: 01/17/2019] [Indexed: 11/22/2022]
|
9
|
Wu L, Hou Q, Lu Y, Bai J, Sun L, Huang Y, Zhang M, Zheng J. Feasibility of lung ultrasound to assess pulmonary overflow in congenital heart disease children. Pediatr Pulmonol 2018; 53:1525-1532. [PMID: 30251402 DOI: 10.1002/ppul.24169] [Citation(s) in RCA: 13] [Impact Index Per Article: 2.2] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Received: 04/18/2018] [Accepted: 09/05/2018] [Indexed: 12/11/2022]
Abstract
BACKGROUND Pulmonary overflow (PO) is one of the most common complications in congenital heart disease (CHD) children with an incidence of 48-60% approximately. This study explored the feasibility of using lung ultrasound (LUS) to assess pulmonary overcirculation in CHD children and compare the diagnostic performance of LUS and chest radiography (CXR) for the detection of pulmonary overcirculation. METHOD The upper anterior area, lower anterior area, upper lateral area, and lower posterior area, in each hemithorax were scanned in 59 children in the supine position. A-lines, B-lines in each scanned region were recorded, and the worst LUS abnormality in the video clip was considered to characterize the examined region. Sensitivity, specificity, positive and negative predictive values, and diagnostic accuracy of LUS and CXR were compared using computed tomography (CT) as a standard criterion. RESULTS PO was diagnosed in 53% (31/59), 63% (37/59), and 51% (30/59) children with CT, CXR, and LUS, respectively. The sensitivity, specificity, and diagnostic accuracy of PO were 96%, 94%, and 95% for LUS and 74%, 50%, and 63% for CXR. The percentage of mild, moderate, and severe PO diagnosed via LUS were 31% (18/59), 19% (11/59), and 2% (1/59), respectively. Furthermore, the PO incidence diagnosed by LUS in CHD children less than 1 year old were significantly higher than those beyond 1 year old. CONCLUSION LUS is a noninvasive and useful tool for the detection and assessment of PO in CHD children at the operating room, and is better than CXR in sensitivity and specificity, comparable to CT.
Collapse
Affiliation(s)
- Lei Wu
- Department of Anesthesiology, Shanghai Children's Medical Center Affiliated to School of Medicine, Shanghai Jiao Tong University, Shanghai, China
| | - Qiaoru Hou
- Diagnostic imaging Center, Shanghai Children's Medical Center Affiliated to School of Medicine, Shanghai Jiao Tong University, Shanghai, China
| | - Yingying Lu
- Department of Radiology, Renji Hospital Affiliated to School of Medicine, Shanghai Jiao Tong University, Shanghai, China
| | - Jie Bai
- Department of Anesthesiology, Shanghai Children's Medical Center Affiliated to School of Medicine, Shanghai Jiao Tong University, Shanghai, China
| | - Liping Sun
- Department of Anesthesiology, Shanghai Children's Medical Center Affiliated to School of Medicine, Shanghai Jiao Tong University, Shanghai, China
| | - Yue Huang
- Department of Anesthesiology, Shanghai Children's Medical Center Affiliated to School of Medicine, Shanghai Jiao Tong University, Shanghai, China
| | - Mazhong Zhang
- Department of Anesthesiology and Pediatric Clinical Pharmacology Laboratory, Shanghai Children's Medical Center Affiliated to School of Medicine, Shanghai Jiao Tong University, Shanghai, China
| | - Jijian Zheng
- Department of Anesthesiology and Pediatric Clinical Pharmacology Laboratory, Shanghai Children's Medical Center Affiliated to School of Medicine, Shanghai Jiao Tong University, Shanghai, China
| |
Collapse
|
10
|
Cantinotti M, Ait Ali L, Scalese M, Giordano R, Melo M, Remoli E, Franchi E, Clemente A, Moschetti R, Festa P, Haxiademi D, Gargani L. Lung ultrasound reclassification of chest X-ray data after pediatric cardiac surgery. Paediatr Anaesth 2018; 28:421-427. [PMID: 29575312 DOI: 10.1111/pan.13360] [Citation(s) in RCA: 27] [Impact Index Per Article: 4.5] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Accepted: 02/16/2018] [Indexed: 12/01/2022]
Abstract
INTRODUCTION Lung ultrasound is gaining consensus for the diagnosis of some pulmonary conditions. Pulmonary complications are common in pediatric cardiac surgery. However, its use remains limited in this setting. Our aim was to test the feasibility of lung ultrasound following pediatric cardiac surgery and to compare lung ultrasound and chest X-ray findings, assessing whether lung ultrasound may provide additional information. METHODS One hundred and thirty-eight lung ultrasound examinations were performed in 79 children (median age 9.3 months) at different time points after surgery. For each hemithorax, 3 areas (anterior/lateral/posterior) have been evaluated in the upper and lower halves of the chest (for a total of 6 scanning sites per side). Pleural effusion, atelectasis, and the number of B-lines were investigated. RESULTS Lung ultrasound was feasible in all cases in at least 1 of the 3 areas. Feasibility was different for the lateral, posterior, and anterior areas (100%, 90%, and 78%, respectively). The posterior areas were more sensitive than anterior and lateral ones in the diagnosis of effusion/atelectasis. In 81 cases, lung ultrasound allowed reclassification of chest X-ray findings, including 40 new diagnoses (diagnosis of effusion/atelectasis with negative chest X-ray reports) and 41 changes in diagnosis (effusions reclassified as atelectasis/severe congestion or vice versa). Although new diagnosis of small-to-moderate effusion/atelectasis was of limited clinical value, in 29 cases the new diagnosis changed the therapeutic approach. CONCLUSION Lung ultrasound is feasible and accurate for the diagnosis of common pulmonary conditions after pediatric cardiac surgery, allowing reclassification of chest X-ray findings in a significant number of patients.
Collapse
Affiliation(s)
| | - Lamia Ait Ali
- Institute of Clinical Physiology, National Research Council, Pisa, Italy
| | - Marco Scalese
- Institute of Clinical Physiology, National Research Council, Pisa, Italy
| | | | - Manuel Melo
- Fondazione Regione Toscana G. Monasterio, Massa, Italy
| | - Ettore Remoli
- Fondazione Regione Toscana G. Monasterio, Massa, Italy
| | | | | | | | | | | | - Luna Gargani
- Institute of Clinical Physiology, National Research Council, Pisa, Italy
| |
Collapse
|
11
|
Santos TM, Franci D, Gontijo-Coutinho CM, Ozahata TM, de Araújo Guerra Grangeia T, Matos-Souza JR, Carvalho-Filho MA. Inflammatory lung edema correlates with echocardiographic estimation of capillary wedge pressure in newly diagnosed septic patients. J Crit Care 2017; 44:392-397. [PMID: 29304490 DOI: 10.1016/j.jcrc.2017.11.036] [Citation(s) in RCA: 4] [Impact Index Per Article: 0.6] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 07/20/2017] [Revised: 11/08/2017] [Accepted: 11/29/2017] [Indexed: 01/10/2023]
Abstract
PURPOSE Lung ultrasound is an accurate and accessible tool to quantify lung edema. Furthermore, left ventricle filling pressures (LVFP) can be assessed with transthoracic echocardiography (TTE) by the E/e' ratio (E/e'). The present study aimed to assess the correlation between E/e' and lung edema quantified by a simplified lung ultrasound score (LUS) in newly admitted septic patients. MATERIALS AND METHODS In this prospective observational cohort, septic adult patients admitted at the emergency department of a tertiary hospital were included. LUS consisted of four different patterns of lung edema (from normal aeration to parenchymal consolidation). To compare lung edema with LVFP, E/e' was calculated immediately before or within 5min of fluid therapy. RESULTS Fifty patients were enrolled in 3months. The LUS correlated with E/e' (r=0.58, P<0.0001). The LUS also increased among E/e' quartiles (Q) (Q1: E/e'≤4.49; Q2: 4.49<E/e'≤5.49; Q3: 5.49<E/e'≤7.11; Q4: >7.11; P=0.0003 for Q1 and 4; 2 and 4); and LUS was significantly higher in abnormal (≥8) vs. normal (<8) values of E/e' (11.29 vs 8.49, P=0.007). CONCLUSION In newly admitted septic patients, lung edema is positively correlated with LVFP prior to fluid therapy. This finding might help find future targets for fluid resuscitation in sepsis.
Collapse
Affiliation(s)
- Thiago M Santos
- Discipline of Emergency Medicine, Hospital of the University of Campinas (Unicamp), 126 Tessália Vieira de Camargo St., Cidade Universitária "Zeferino Vaz", Postal Code 13083-887 Campinas, SP, Brazil.
| | - Daniel Franci
- Discipline of Emergency Medicine, Hospital of the University of Campinas (Unicamp), 126 Tessália Vieira de Camargo St., Cidade Universitária "Zeferino Vaz", Postal Code 13083-887 Campinas, SP, Brazil
| | - Carolina M Gontijo-Coutinho
- Discipline of Emergency Medicine, Hospital of the University of Campinas (Unicamp), 126 Tessália Vieira de Camargo St., Cidade Universitária "Zeferino Vaz", Postal Code 13083-887 Campinas, SP, Brazil
| | - Tatiana Mirabetti Ozahata
- Discipline of Emergency Medicine, Hospital of the University of Campinas (Unicamp), 126 Tessália Vieira de Camargo St., Cidade Universitária "Zeferino Vaz", Postal Code 13083-887 Campinas, SP, Brazil
| | - Tiago de Araújo Guerra Grangeia
- Discipline of Emergency Medicine, Hospital of the University of Campinas (Unicamp), 126 Tessália Vieira de Camargo St., Cidade Universitária "Zeferino Vaz", Postal Code 13083-887 Campinas, SP, Brazil
| | - José R Matos-Souza
- Discipline of Cardiology, Hospital of the University of Campinas (Unicamp), 126 Tessália Vieira de Camargo St., Cidade Universitária "Zeferino Vaz", Postal Code 13083-887 Campinas, SP, Brazil
| | - Marco A Carvalho-Filho
- Discipline of Emergency Medicine, Hospital of the University of Campinas (Unicamp), 126 Tessália Vieira de Camargo St., Cidade Universitária "Zeferino Vaz", Postal Code 13083-887 Campinas, SP, Brazil
| |
Collapse
|
12
|
Chen S, Zhong S, Yao L, Shang Y, Suzuki K. Enhancement of chest radiographs obtained in the intensive care unit through bone suppression and consistent processing. Phys Med Biol 2016; 61:2283-301. [PMID: 26930386 DOI: 10.1088/0031-9155/61/6/2283] [Citation(s) in RCA: 9] [Impact Index Per Article: 1.1] [Reference Citation Analysis] [Abstract] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/11/2022]
Abstract
Portable chest radiographs (CXRs) are commonly used in the intensive care unit (ICU) to detect subtle pathological changes. However, exposure settings or patient and apparatus positioning deteriorate image quality in the ICU. Chest x-rays of patients in the ICU are often hazy and show low contrast and increased noise. To aid clinicians in detecting subtle pathological changes, we proposed a consistent processing and bone structure suppression method to decrease variations in image appearance and improve the diagnostic quality of images. We applied a region of interest-based look-up table to process original ICU CXRs such that they appeared consistent with each other and the standard CXRs. Then, an artificial neural network was trained by standard CXRs and the corresponding dual-energy bone images for the generation of a bone image. Once the neural network was trained, the real dual-energy image was no longer necessary, and the trained neural network was applied to the consistent processed ICU CXR to output the bone image. Finally, a gray level-based morphological method was applied to enhance the bone image by smoothing other structures on this image. This enhanced image was subtracted from the consistent, processed ICU CXR to produce a soft tissue image. This method was tested for 20 patients with a total of 87 CXRs. The findings indicated that our method suppressed bone structures on ICU CXRs and standard CXRs, simultaneously maintaining subtle pathological changes.
Collapse
Affiliation(s)
- Sheng Chen
- School of Optical-Electrical and Computer Engineering & Engineering Research Center of Optical Instrument and System, Ministry of Education, University of Shanghai for Science and Technology, Shanghai, People's Republic of China
| | | | | | | | | |
Collapse
|
13
|
Laight NS, Levin AI. Transcardiopulmonary Thermodilution-Calibrated Arterial Waveform Analysis: A Primer for Anesthesiologists and Intensivists. J Cardiothorac Vasc Anesth 2015; 29:1051-64. [PMID: 26279223 DOI: 10.1053/j.jvca.2015.03.013] [Citation(s) in RCA: 6] [Impact Index Per Article: 0.7] [Reference Citation Analysis] [Key Words] [Track Full Text] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Received: 10/31/2014] [Indexed: 02/07/2023]
Affiliation(s)
- Nicola S Laight
- Department of Anesthesiology and Critical Care, University of Stellenbosch, Tygerberg Hospital, Cape Town, South Africa
| | - Andrew I Levin
- Department of Anesthesiology and Critical Care, University of Stellenbosch, Tygerberg Hospital, Cape Town, South Africa.
| |
Collapse
|