ICU imaging

Deep Learning-based Diagnosis and Localization of Pneumothorax on Portable Supine Chest X-ray in Intensive and Emergency Medicine: A Retrospective Study

PURPOSE

To develop two deep learning-based systems for diagnosing and localizing pneumothorax on portable supine chest X-rays (SCXRs).

METHODS

For this retrospective study, images meeting the following inclusion criteria were included: (1) patient age ≥ 20 years; (2) portable SCXR; (3) imaging obtained in the emergency department or intensive care unit. Included images were temporally split into training (1571 images, between January 2015 and December 2019) and testing (1071 images, between January 2020 to December 2020) datasets. All images were annotated using pixel-level labels. Object detection and image segmentation were adopted to develop separate systems. For the detection-based system, EfficientNet-B2, DneseNet-121, and Inception-v3 were the architecture for the classification model; Deformable DETR, TOOD, and VFNet were the architecture for the localization model. Both classification and localization models of the segmentation-based system shared the UNet architecture.

RESULTS

In diagnosing pneumothorax, performance was excellent for both detection-based (Area under receiver operating characteristics curve [AUC]: 0.940, 95% confidence interval [CI]: 0.907-0.967) and segmentation-based (AUC: 0.979, 95% CI: 0.963-0.991) systems. For images with both predicted and ground-truth pneumothorax, lesion localization was highly accurate (detection-based Dice coefficient: 0.758, 95% CI: 0.707-0.806; segmentation-based Dice coefficient: 0.681, 95% CI: 0.642-0.721). The performance of the two deep learning-based systems declined as pneumothorax size diminished. Nonetheless, both systems were similar or better than human readers in diagnosis or localization performance across all sizes of pneumothorax.

CONCLUSIONS

Both deep learning-based systems excelled when tested in a temporally different dataset with differing patient or image characteristics, showing favourable potential for external generalizability.

The accuracy of intensive care nurses' interpretation of chest radiographs to recognise misplacement of endotracheal and nasogastric tubes after a single training session and comparison with residents' interpretation

BACKGROUND

Misplacements of endotracheal and nasogastric tubes are frequent encounters in critically ill patients.

OBJECTIVES

The purpose of this study was to assess the effectiveness of a single standardised training session on the ability of intensive care registered nurses (RNs) to recognise the misplacement of endotracheal and nasogastric tubes on bedside chest radiographs of patients in intensive care units (ICUs).

METHODS

In eight French ICUs, RNs received a 110-min standardised teaching on the position of endotracheal and nasogastric tubes on chest radiographs. Their knowledge was evaluated within the subsequent weeks. For 20 chest radiographs, each with an endotracheal and nasogastric tube, RNs had to indicate whether each tube was in the proper or incorrect position. Training success was defined as >90% for the lower bound of the 95% confidence interval (95% CI) of the mean correct response rate (CRR). Residents of the participating ICUs underwent the same evaluation (without prior specific training).

RESULTS

In total, 181 RNs were trained and evaluated and 110 residents were evaluated. The global mean CRR for RNs was 84.6% (95% CI: 83.3-85.9), significantly higher than for residents (81.4% [95% CI: 79.7-83.2]) (P < 0.0001). The mean CRR for RNs and residents was 95.9% (93.9-98.0) and 97.0% (94.7-99.3) for misplaced nasogastric tubes (P = 0.54), 86.8% (85.2-88.5) and 82.6% (79.4-85.7) (P = 0.07) for nasogastric tubes in the correct position, 86.6% (83.8-89.3) and 62.7% (57.9-67.5) for misplaced endotracheal tubes (P < 0.0001), and 79.1% (76.6-81.6) and 84.7% (82.1-87.2) for endotracheal tubes in the correct position (P = 0.01), respectively.

CONCLUSIONS

The ability of trained RNs to detect tube misplacement did not reach the predetermined arbitrary level, indicating training success. Their mean CRR was higher than that for residents and was considered satisfactory for detecting misplaced nasogastric tubes. This finding is encouraging but insufficient to ensure patient safety. Transferring responsibility for reading radiographs to detect the misplacement of endotracheal tubes to intensive care RNs will need a more advanced or more in-depth teaching method.

Imaging in the Intensive Care Unit

Radiology plays an important role in the management of the most seriously ill patients in the hospital. Over the years, continued advances in imaging technology have contributed to an improvement in patient care. However, even with such advances, the portable chest radiograph (CXR) remains one of the most commonly requested radiographic examinations. While they provide valuable information, CXRs remain relatively insensitive at revealing abnormalities and are often nonspecific. Chest computed tomography (CT) can display findings that are occult on CXR and is particularly useful at identifying and characterizing pleural effusions, detecting barotrauma including small pneumothoraces, distinguishing pneumonia from atelectasis, and revealing unsuspected or additional abnormalities which could result in increased morbidity and mortality if left untreated. CT pulmonary angiography is the modality of choice in the evaluation of pulmonary emboli which can complicate the hospital course of the ICU patient. This article will provide guidance for interpretation of CXR and thoracic CT images, discuss some of the invasive devices routinely used, and review the radiologic manifestations of common pathologic disease states encountered in ICU patients. In addition, imaging findings and complications of more specific clinical scenarios in which the incidence has increased in the ICU setting, such as patients who are immunocompromised, have interstitial lung disease, or COVID-19, will also be discussed. Communication between the radiologist and intensivist, particularly on complicated cases, is important to help increase diagnostic accuracy and leads to an improvement in the management of the most critically ill patients.

Chest X-ray in intensive care unit patients: what there is to know about thoracic devices

Critically ill patients admitted to the intensive care unit require continuous monitoring of vital functions as well as mechanical and pharmacological support, provided through different devices. Chest radiographs play a fundamental role in monitoring the conditions of these patients and assessing the intensive-care devices after their insertion; therefore, the radiologist needs to know their normal appearance and their correct position and should be aware of the possible complications that may occur after their placement. This pictorial review illustrates the radiographic appearance of non-cardiological devices commonly used in clinical practice (central venous catheters, tunneled catheters, Swan-Ganz catheters, chest tubes, endotracheal tubes, and nasogastric tubes), their correct position and the most common complications that may occur after their placement.

Deep-Learning-Based Diagnosis of Bedside Chest X-ray in Intensive Care and Emergency Medicine

OBJECTIVES

Validation of deep learning models should separately consider bedside chest radiographs (CXRs) as they are the most challenging to interpret, while at the same time the resulting diagnoses are important for managing critically ill patients. Therefore, we aimed to develop and evaluate deep learning models for the identification of clinically relevant abnormalities in bedside CXRs, using reference standards established by computed tomography (CT) and multiple radiologists.

MATERIALS AND METHODS

In this retrospective study, a dataset consisting of 18,361 bedside CXRs of patients treated at a level 1 medical center between January 2009 and March 2019 was used. All included CXRs occurred within 24 hours before or after a chest CT. A deep learning algorithm was developed to identify 8 findings on bedside CXRs (cardiac congestion, pleural effusion, air-space opacification, pneumothorax, central venous catheter, thoracic drain, gastric tube, and tracheal tube/cannula). For the training dataset, 17,275 combined labels were extracted from the CXR and CT reports by a deep learning natural language processing (NLP) tool. In case of a disagreement between CXR and CT, human-in-the-loop annotations were used. The test dataset consisted of 583 images, evaluated by 4 radiologists. Performance was assessed by area under the receiver operating characteristic curve analysis, sensitivity, specificity, and positive predictive value.

RESULTS

Areas under the receiver operating characteristic curve for cardiac congestion, pleural effusion, air-space opacification, pneumothorax, central venous catheter, thoracic drain, gastric tube, and tracheal tube/cannula were 0.90 (95% confidence interval [CI], 0.87-0.93; 3 radiologists on the receiver operating characteristic [ROC] curve), 0.95 (95% CI, 0.93-0.96; 3 radiologists on the ROC curve), 0.85 (95% CI, 0.82-0.89; 1 radiologist on the ROC curve), 0.92 (95% CI, 0.89-0.95; 1 radiologist on the ROC curve), 0.99 (95% CI, 0.98-0.99), 0.99 (95% CI, 0.98-0.99), 0.98 (95% CI, 0.97-0.99), and 0.99 (95% CI, 0.98-1.00), respectively.

CONCLUSIONS

A deep learning model used specifically for bedside CXRs showed similar performance to expert radiologists. It could therefore be used to detect clinically relevant findings during after-hours and help emergency and intensive care physicians to focus on patient care.

Covid-19 imaging: A narrative review

Background

The 2019 novel coronavirus disease (COVID-19) imaging data is dispersed in numerous publications. A cohesive literature review is to be assembled.

Objective

To summarize the existing literature on Covid-19 pneumonia imaging including precautionary measures for radiology departments, Chest CT's role in diagnosis and management, imaging findings of Covid-19 patients including children and pregnant women, artificial intelligence applications and practical recommendations.

Methods

A systematic literature search of PubMed/med line electronic databases.

Results

The radiology department's staff is on the front line of the novel coronavirus outbreak. Strict adherence to precautionary measures is the main defense against infection's spread. Although nucleic acid testing is Covid-19's pneumonia diagnosis gold standard; kits shortage and low sensitivity led to the implementation of the highly sensitive chest computed tomography amidst initial diagnostic tools. Initial Covid-19 CT features comprise bilateral, peripheral or posterior, multilobar ground-glass opacities, predominantly in the lower lobes. Consolidations superimposed on ground-glass opacifications are found in few cases, preponderantly in the elderly. In later disease stages, GGO transformation into multifocal consolidations, thickened interlobular and intralobular lines, crazy paving, traction bronchiectasis, pleural thickening, and subpleural bands are reported. Standardized CT reporting is recommended to guide radiologists. While lung ultrasound, pulmonary MRI, and PET CT are not Covid-19 pneumonia's first-line investigative diagnostic modalities, their characteristic findings and clinical value are outlined. Artificial intelligence's role in strengthening available imaging tools is discussed.

Conclusion

This review offers an exhaustive analysis of the current literature on imaging role and findings in COVID-19 pneumonia.

•

Chest computed tomography is a highly sensitive Covid −19 pneumonia's diagnostic tool.

•

Initial Covid-19 CT features are bilateral, multifocal, peripheral or posterior ground-glass opacities, mainly in the lower lobes.

•

Multifocal consolidations, bronchiectasis, pleural thickening, and subpleural bands are late disease stages features.

•

Standardized CT reporting is recommended to guide radiologists.

•

Artificial intelligence could strengthen available imaging tools.

Accuracy of Conventional and Machine Learning Enhanced Chest Radiography for the Assessment of COVID-19 Pneumonia: Intra-Individual Comparison with CT

PURPOSE

To evaluate diagnostic accuracy of conventional radiography (CXR) and machine learning enhanced CXR (mlCXR) for the detection and quantification of disease-extent in COVID-19 patients compared to chest-CT.

METHODS

Real-time polymerase chain reaction (rt-PCR)-confirmed COVID-19-patients undergoing CXR from March to April 2020 together with COVID-19 negative patients as control group were retrospectively included. Two independent readers assessed CXR and mlCXR images for presence, disease extent and type (consolidation vs. ground-glass opacities (GGOs) of COVID-19-pneumonia. Further, readers had to assign confidence levels to their diagnosis. CT obtained ≤ 36 h from acquisition of CXR served as standard of reference. Inter-reader agreement, sensitivity for detection and disease extent of COVID-19-pneumonia compared to CT was calculated. McNemar test was used to test for significant differences.

RESULTS

Sixty patients (21 females; median age 61 years, range 38-81 years) were included. Inter-reader agreement improved from good to excellent when mlCXR instead of CXR was used (k = 0.831 vs. k = 0.742). Sensitivity for pneumonia detection improved from 79.5% to 92.3%, however, on the cost of specificity 100% vs. 71.4% (p = 0.031). Overall, sensitivity for the detection of consolidation was higher than for GGO (37.5% vs. 70.4%; respectively). No differences could be found in disease extent estimation between mlCXR and CXR, even though the detection of GGO could be improved. Diagnostic confidence was better on mlCXR compared to CXR (p = 0.013).

CONCLUSION

In line with the current literature, the sensitivity for detection and quantification of COVID-19-pneumonia was moderate with CXR and could be improved when mlCXR was used for image interpretation.

Advances in imaging for lung emphysema

Lung emphysema represents a major public health burden and still accounts for five percent of all deaths worldwide. Hence, it is essential to further understand this disease in order to develop effective diagnostic and therapeutic strategies. Lung emphysema is an irreversible enlargement of the airways distal to the terminal bronchi (i.e., the alveoli) due to the destruction of the alveolar walls. The two most important causes of emphysema are (I) smoking and (II) α1-antitrypsin-deficiency. In the former lung emphysema is predominant in the upper lung parts, the latter is characterized by a predominance in the basal areas of the lungs. Since quantification and evaluation of the distribution of lung emphysema is crucial in treatment planning, imaging plays a central role. Imaging modalities in lung emphysema are manifold: computed tomography (CT) imaging is nowadays the gold standard. However, emerging imaging techniques like dynamic or functional magnetic resonance imaging (MRI), scintigraphy and lately also the implementation of radiomics and artificial intelligence are more and more diffused in the evaluation, diagnosis and quantification of lung emphysema. The aim of this review is to shortly present the different subtypes of lung emphysema, to give an overview on prediction and risk assessment in emphysematous disease and to discuss not only the traditional, but also the new imaging techniques for diagnosis, quantification and evaluation of lung emphysema.

In Vivo Endomicroscopy of Lung Injury and Repair in ARDS: Potential Added Value to Current Imaging

BACKGROUND

Standard clinical imaging of the acute respiratory distress syndrome (ARDS) lung lacks resolution and offers limited possibilities in the exploration of the structure-function relationship, and therefore cannot provide an early and clear discrimination of patients with unexpected diagnosis and unrepair profile. The current gold standard is open lung biopsy (OLB). However, despite being able to reveal precise information about the tissue collected, OLB cannot provide real-time information on treatment response and is accompanied with a complication risk rate up to 25%, making longitudinal monitoring a dangerous endeavor. Intravital probe-based confocal laser endomicroscopy (pCLE) is a developing and innovative high-resolution imaging technology. pCLE offers the possibility to leverage multiple and specific imaging probes to enable multiplex screening of several proteases and pathogenic microorganisms, simultaneously and longitudinally, in the lung. This bedside method will ultimately enable physicians to rapidly, noninvasively, and accurately diagnose degrading lung and/or fibrosis without the need of OLBs.

OBJECTIVES AND METHODS

To extend the information provided by standard imaging of the ARDS lung with a bedside, high-resolution, miniaturized pCLE through the detailed molecular imaging of a carefully selected region-of-interest (ROI). To validate and quantify real-time imaging to validate pCLE against OLB.

RESULTS

Developments in lung pCLE using fluorescent affinity- or activity-based probes at both preclinical and clinical (first-in-man) stages are ongoing-the results are promising, revealing correlations with OLBs in problematic ARDS.

CONCLUSION

It can be envisaged that safe, high-resolution, noninvasive pCLE with activatable fluorescence probes will provide a "virtual optical biopsy" and will provide decisive information in selected ARDS patients at the bedside.

Diagnostic accuracy and added value of dual-energy subtraction radiography compared to standard conventional radiography using computed tomography as standard of reference

PURPOSE

To retrospectively evaluate diagnostic performance of dual-energy subtraction radiography (DESR) for interpretation of chest radiographs compared to conventional radiography (CR) using computed tomography (CT) as standard of reference.

MATERIAL AND METHODS

A total of 199 patients (75 female, median age 67) were included in this institutional review board (IRB)-approved clinical trial. All patients were scanned in posteroanterior and lateral direction with dual-shot DE-technique. Chest CT was performed within ±72 hours. The system provides three types of images: bone weighted-image, soft tissue weighted-image, herein termed as DESR-images, and a standard image, termed CR-image (marked as CR-image). Images were evaluated by two radiologists for presence of inserted life support lines, pneumothorax, pleural effusion, infectious consolidation, interstitial lung changes, tumor, skeletal alterations, soft tissue alterations, aortic or tracheal calcification and pleural thickening. Inter-observer agreement between readers and diagnostic performance were calculated. McNemar's test was used to test for significant differences.

RESULTS

Mean inter-observer agreement throughout the investigated parameters was higher in DESR images compared to CR-images (kDESR = 0.935 vs. kCR = 0.858). DESR images provided significantly increased sensitivity compared to CR-images for the detection of infectious consolidations (42% vs. 62%), tumor (46% vs. 57%), interstitial lung changes (69% vs. 87%) and aortic or tracheal calcification (25 vs. 73%) (p<0.05). There were no significant differences in sensitivity for the detection of inserted life support lines, pneumothorax, pleural effusion, skeletal alterations, soft tissue alterations or pleural thickening (p>0.05).

CONCLUSION

DESR increases significantly the sensibility without affecting the specificity evaluating chest radiographs, with emphasis on the detection of interstitial lung diseases.

Postprocedural chest radiograph: Impact on the management in critical care unit

Postprocedural chest radiograph is done to illustrate the position of endotracheal tubes (ETTs), nasogastric and drainage tubes, indwelling catheters, and intravascular lines or any other lifesaving devices to confirm their position. These devices are intended to save life, but may be life-threatening if in the wrong place. The incidence of malposition and complications ranges from 3% to 14%, respectively. The portable chest radiograph is of tremendous value, inexpensive and can be obtained quickly at the patient's bedside in any location of the hospital. A systemic literature search was performed in PubMed and the Cochranre library by setting up the search using either single text word or combinations. Those studies were also included where the chest radiograph was compared with other imaging modalities. Its clinical efficacy, cost-effectiveness and practicality allow anesthesiologist to evaluate the post-procedural position and complications of ETT, indwelling catheters, and multi lumen intravascular lines. Knowledge of the radiological features of commonly used devices is of utmost importance.

The role of early postmortem CT in the evaluation of support-line misplacement in patients with severe trauma

OBJECTIVE

The purpose of this study was to retrospectively assess the role of early postmortem CT in evaluating support-line misplacement to improve future treatment in the trauma setting.

MATERIALS AND METHODS

We included all postmortem CT examinations that were performed for trauma patients within the 1st hour after declaration of death in our tertiary medical center between August 1, 2008, and August 31, 2013. Correct placement of the following support lines was evaluated: endotracheal tubes (ETTs), chest drains, central venous catheters (CVCs), and nasogastric tubes (NGTs). Prehospital resuscitation efforts were started in all cases.

RESULTS

Early postmortem CT was performed on average 22 minutes after declaration of death in 25 consecutive patients with severe trauma. Overall, 14 subjects (56%) had suboptimal or misplaced support lines. Of ETTs inserted into 18 trauma victims; three (17%) were mislaid in the right main bronchus and five (28%) were near or at the level of the carina. Of chest drains inserted into 13 subjects, 10 were suboptimally positioned (77%). Of CVCs inserted into eight subjects (seven femoral and one brachiocephalic), one femoral CVC (13%) was malpositioned in the soft tissues of the pelvis. Of NGTs inserted in five trauma victims, one was folded within the pharynx.

CONCLUSION

Early postmortem CT for patients who have experienced severe poly-trauma can be of important educational value to radiologists and the trauma teams, providing immediate feedback regarding the location of the support lines and possibly contributing to improved training and command of the learning curve by medical staff.

The Impact of Computed Tomography of the Chest on the Management of Patients in a Medical Intensive Care Unit

PURPOSE

To understand whether chest computed tomographies (CTs) have utility in a medical intensive care unit (MICU) population as previously noted in nonmedical critically ill patients.

PATIENTS AND METHODS

We conducted a retrospective cohort study of patients receiving chest CTs in the MICU at an urban, academic institution. Indications for, findings on, and care changes made after chest CT were obtained. We identified patient characteristics associated with having a care change clearly related to the CT using multivariate regression.

RESULTS

We evaluated 134 patients; 64 (47.8%) had a chest CT with intravenous contrast. Common findings included pulmonary consolidation (46.3%), nonconsolidative pulmonary parenchymal disease (29.1%), and pleural effusion (35.1%). Of the chest CTs, 23.9% were followed by changes in management clearly related to the study. Use of intravenous contrast was associated with increased odds of having a care change (adjusted odds ratio [95% confidence interval [CI] versus noncontrast study: 3.14 [1.18-8.37], P = .022) and having the CT performed 1 or 2 days after ICU admission versus on the day of ICU admission was associated with lower odds of a care change (odds ratio [95% CI]: 0.29 [0.09-0.99], P = .048).

CONCLUSION

Less than one-quarter of chest CTs in the MICU result in management. Intravenous contrast-enhanced CTs and CTs done on the day of ICU admission have increased odds of utility.

Ventilator-Associated Pneumonia Bundle

The ventilator-associated pneumonia (VAP) bundle is a focus of many health care institutions. Many hospitals are conducting process-improvement projects in an attempt to improve VAP rates by implementing the bundle. However, this bundle is controversial in the literature, because the evidence supporting the VAP interventions is weak. In addition, definitions used for surveillance are interpreted differently than definitions used for clinical diagnosis. The variance in definitions has led to lower reported VAP rates, which may not be accurate. Because of the variance in definitions, the Centers for Disease Control and Prevention developed a ventilator-associated event algorithm. Health care institutions are under pressure to reduce the VAP infection rate, but correctly identifying VAP can be very challenging. This article reviews the current evidence related to VAP and provides insight into implementing a suggested revision of the care of patients being treated with mechanical ventilation.

Chest radiography in intensive care: an irreplaceable survey?

PURPOSE

This study evaluated the impact and value of bedside chest X-ray in intensive care units.

MATERIALS AND METHODS

This observational study considered the bedside chest X-rays performed on 258 consecutive patients (160 men, 98 women; mean age, 58 years) admitted to intensive care units. Stratification of patients according to the reason for hospitalisation and analysis of the reasons for chest X-ray examinations were performed to assess the diagnostic efficacy (DE).

RESULTS

DE for chest X-rays was 84.5%, with 15.5% of tests remaining unchanged over time. Patient stratification by disease indicated that the DE was 85.27% in transplant, 90.79% in postoperative care after general surgery, 83.89% in respiratory failure, 82.42% in polytrauma, 90.54% in postoperative care after neurosurgery, 86.6% in postoperative care after vascular surgery, 83.3% in neurological conditions and 93.4% in other diseases.

CONCLUSIONS

Chest X-rays performed at the bedside are the most widely used imaging method in the follow-up of critically ill patients. DE is approximately 84.5%. Radiologists should maintain familiarity with the interpretation of this examination.

Chest radiography in the ICU: Part 1, Evaluation of airway, enteric, and pleural tubes

OBJECTIVE

In this pictorial essay, we discuss and illustrate normal and aberrant positioning of nonvascular support and monitoring devices frequently used in critically ill patients, including endotracheal and tracheostomy tubes, chest tubes, and nasogastric and nasoenteric tubes, as well as their inherent complications.

CONCLUSION

The radiographic evaluation of the support and monitoring devices used in patients in the ICU is important because the potentially serious complications arising from their introduction and use are often not clinically apparent. Familiarity with normal and abnormal radiographic findings is critical for the detection of these complications.