MDCT for automated detection and measurement of pneumothoraces in trauma patients

DAWTran: dynamic adaptive windowing transformer network for pneumothorax segmentation with implicit feature alignment

Objective. This study aims to address the significant challenges posed by pneumothorax segmentation in computed tomography images due to the resemblance between pneumothorax regions and gas-containing structures such as the trachea and bronchus.Approach. We introduce a novel dynamic adaptive windowing transformer (DAWTran) network incorporating implicit feature alignment for precise pneumothorax segmentation. The DAWTran network consists of an encoder module, which employs a DAWTran, and a decoder module. We have proposed a unique dynamic adaptive windowing strategy that enables multi-head self-attention to effectively capture multi-scale information. The decoder module incorporates an implicit feature alignment function to minimize information deviation. Moreover, we utilize a hybrid loss function to address the imbalance between positive and negative samples.Main results. Our experimental results demonstrate that the DAWTran network significantly improves the segmentation performance. Specifically, it achieves a higher dice similarity coefficient (DSC) of 91.35% (a larger DSC value implies better performance), showing an increase of 2.21% compared to the TransUNet method. Meanwhile, it significantly reduces the Hausdorff distance (HD) to 8.06 mm (a smaller HD value implies better performance), reflecting a reduction of 29.92% in comparison to the TransUNet method. Incorporating the dynamic adaptive windowing (DAW) mechanism has proven to enhance DAWTran's performance, leading to a 4.53% increase in DSC and a 15.85% reduction in HD as compared to SwinUnet. The application of the implicit feature alignment (IFA) further improves the segmentation accuracy, increasing the DSC by an additional 0.11% and reducing the HD by another 10.01% compared to the model only employing DAW.Significance. These results highlight the potential of the DAWTran network for accurate pneumothorax segmentation in clinical applications, suggesting that it could be an invaluable tool in improving the precision and effectiveness of diagnosis and treatment in related healthcare scenarios. The improved segmentation performance with the inclusion of DAW and IFA validates the effectiveness of our proposed model and its components.

Evaluation and management of traumatic pneumothorax: A Western Trauma Association critical decisions algorithm

ABSTRACT

This is a recommended algorithm of the Western Trauma Association for the management of a traumatic pneumothorax. The current algorithm and recommendations are based on available published prospective cohort, observational, and retrospective studies and the expert opinion of the Western Trauma Association members. The algorithm and accompanying text represents a safe and reasonable approach to this common problem. We recognize that there may be variability in decision making, local resources, institutional consensus, and patient-specific factors that may require deviation from the algorithm presented. This annotated algorithm is meant to serve as a basis from which protocols at individual institutions can be developed or serve as a quick bedside reference for clinicians.

LEVEL OF EVIDENCE

Consensus algorithm from the Western Trauma Association, Level V.

Deep learning detection and quantification of pneumothorax in heterogeneous routine chest computed tomography

BACKGROUND

Automatically detecting and quantifying pneumothorax on chest computed tomography (CT) may impact clinical decision-making. Machine learning methods published so far struggle with the heterogeneity of technical parameters and the presence of additional pathologies, highlighting the importance of stable algorithms.

METHODS

A deep residual UNet was developed and evaluated for automated, volume-level pneumothorax grading (i.e., labelling a volume whether a pneumothorax was present or not), and pixel-level classification (i.e., segmentation and quantification of pneumothorax), on a retrospective series of routine chest CT data. Ground truth annotations were provided by radiologists. The fully automated pixel-level pneumothorax segmentation method was trained using 43 chest CT scans and evaluated on 9 chest CT scans with pixel-level annotation basis and 567 chest CT scans on a volume-level basis.

RESULTS

This method achieved a receiver operating characteristic area under the curve (AUC) of 0.98, an average precision of 0.97, and a Dice similarity coefficient (DSC) of 0.94. This segmentation performance resulted to be similar to the inter-rater segmentation accuracy of two radiologists, who achieved a DSC of 0.92. The comparison of manual and automated pneumothorax quantification yielded a Pearson correlation coefficient of 0.996. The volume-level pneumothorax grading accuracy was evaluated on 567 chest CT scans and yielded an AUC of 0.98 and an average precision of 0.95.

CONCLUSIONS

We proposed a deep learning method for the detection and quantification of pneumothorax in heterogeneous routine clinical data that may facilitate the automated triage of urgent examinations and enable treatment decision support.

Trends in oncologic hybrid imaging

Hybrid imaging plays a central role in the diagnosis and management of a wide range of malignancies at all stages. In this article, we review the most pertinent historical developments, emerging clinical applications of novel radiotracers and imaging technologies, and potential implications for training and practice. This includes an overview of novel tracers for prostate, breast, and neuroendocrine tumors, assessment of tumor heterogeneity, the concept of image-guided ‘biologically relevant dosing’, and theranostic applications. Recent technological advancements, including time-of-flight PET, PET/MRI, and ‘one-minute whole-body PET’, are also covered. Finally, we discuss how these rapidly evolving applications might affect current training curricula and how imaging-derived big data could be harnessed to the benefit of our patients.

Retroperitoneal Laparoscopy in Dogs: Access Technique, Working Space, and Surgical Anatomy

Objective

To develop and describe a laparoscopic retroperitoneal access technique, investigate working space establishment, and describe the surgical anatomy in the retroperitoneal space as an initial step for clinical application of retroperitoneal laparoscopy in dogs.

Study Design

Cadaveric and experimental study.

Animals

Cadaveric (n=8) and healthy (n=6) adult dogs.

Methods

The retroperitoneal access technique was developed in 3 cadavers based on the human technique and transperitoneal observation. Its application and working space establishment with carbon dioxide (CO₂) insufflation alone was evaluated in 5 cadavers by observing with a transperitoneal telescope and in 6 live dogs by repeated computed tomography (CT) scans at pressure of 0, 5, 10, and 15 mmHg. Recordings of retroperitoneoscopy as well as working space volume and linear dimensions measured on CT images were analyzed.

Results

Retroperitoneal access and working space establishment with CO₂ insufflation alone were successfully performed in all 6 live dogs. The only complication observed was in 1 dog that developed subclinical pneumomediastinum. As pressure increased, working space was established from the ipsilateral to the contralateral side, and peritoneal tearing eventually developed. Working space volume increased significantly from 5 mmHg and linear dimensions increased significantly from 0 to 10 mmHg. With pneumo‐retroperitoneum above 5 mmHg, retroperitoneal organs, including kidneys and adrenal glands, were easily visualized.

Conclusion

The retroperitoneal access technique and working space establishment with CO₂ insufflation starting with 5 mmHg and increasing to 10 mmHg provided adequate working space and visualization of retroperitoneal organs, which may allow direct access for retroperitoneal laparoscopy in dogs.

Chest X-ray findings in late-onset congenital diaphragmatic hernia, a rare emergency easily misdiagnosed as hydropneumothorax: a case report

INTRODUCTION

Late-onset congenital diaphragmatic hernia is a rare anomaly with misleading symptoms and signs.

CASE PRESENTATION

We describe the case of a 7-year-old Middle Eastern girl who presented with acute nonspecific abdominal symptoms and respiratory distress of 2 days' duration after sustaining a blunt trauma on her left chest wall on a background of chronic ill-defined left chest pain of 2 weeks' duration. Her initial chest radiograph showed an air-fluid level, which was thought to be a hydropneumothorax, so a chest tube was inserted and was shown to be positioned between the chest wall and the air collection; therefore, a nasogastric tube was inserted and it was positioned in the left chest cavity so the diagnosis of late-onset congenital diaphragmatic hernia was made. On retrospective analysis of the first abdominal X-ray, it showed a subtle lucent area that was triangular in shape and continued with the chest cavity, which indicates a sign of diaphragmatic hernia. In addition, the next unusual point was the nonvisualization of the diaphragm, which should be reported in any abdominal X-ray. An exploratory laparotomy was performed on our patient using a left-sided subcostal incision; the operative findings revealed a very small posterior rim of the diaphragm and a hypoplastic left lung. Her stomach, spleen, and left colon with the omentum were in the left side of her chest. She made an uneventful recovery postoperatively and was discharged after 1 week.

CONCLUSIONS

Gastric and intestinal gas shadow distribution provides an important marker in the diagnosis of late-onset congenital diaphragmatic hernia and should be sought for in every case of suspected congenital diaphragmatic hernia in addition to noting the position of the nasogastric tube in the chest cavity.

Optimizing working space in laparoscopy: CT measurement of the influence of small body size in a porcine model

INTRODUCTION

In our continuing research into the determinants of laparoscopic working space, the influence of small body size was investigated.

METHODS

In eight 6-kg pigs, the effects of intraabdominal CO2 pneumoperitoneum pressure (IAP), prestretching of the abdominal wall, and neuromuscular blockade (NMB) on laparoscopic working space volume and distances were studied. Computed tomography was used to measure working space during two stepwise abdominal insufflation-runs up to an IAP of 15mm Hg. Results were compared with data from earlier experiments in 20-kg pigs.

RESULTS

Cardiorespiratory parameters were stable up to an IAP of 8-10mm Hg. In 6-kg pigs working-space dimensions were five times smaller than in 20-kg pigs. Working-space volume, anteroposterior (AP) diameter and symphysis-diaphragm distance increased linearly up to an IAP of 8mm Hg. Above 8mm Hg, compliance decreased. Eighty percent of the total volume (618ml) and of AP diameter (3cm) at 15mm Hg had been achieved at an IAP of 10mm Hg. Prestretching by a first insufflation resulted in a statistically significant increase in working space volume and in AP diameter during the second insufflation. This effect was significantly larger than in 20-kg pigs. Neuromuscular blockade did not have a significant effect on working-space.

CONCLUSIONS

Working space in growing individuals is very limited. Eighty percent of the working space created by an IAP of 15mm Hg was already achieved at 10mm Hg, while cardiorespiratory side effects at an IAP of 8-10mm Hg seem acceptable. Prestretching of the abdominal wall significantly increased working space, even more so than in 20-kg pigs. As in 20-kg pigs, NMB had no significant effect on laparoscopic working space. Prestretching of the abdominal wall is a promising cheap, safe and easy strategy to increase laparoscopic working space, lessening the need for prolonged high-pressure pneumoperitoneum.

Optimizing working space in laparoscopy: CT-measurement of the effect of neuromuscular blockade and its reversal in a porcine model

OBJECTIVE

The objective of this paper was to determine the effect of neuromuscular blockade (NMB) on working space in a porcine laparoscopy model.

BACKGROUND

Conflicting results on the effect of NMB on laparoscopic working space are found in literature. Almost all studies are limited by absence of objective assessment of working space or use surrogate outcomes.

METHODS

In a standardized porcine laparoscopy model, laparoscopic working-space dimensions with and without NMB were investigated in 16 animals using computed tomography at intra-abdominal pressures of 0, 5, 10, and 15 mmHg during multiple runs of abdominal insufflation.

RESULTS

No statistically significant effect of NMB on abdominal dimensions and laparoscopic working-space volume was found during CO2 pneumoperitoneum. In contrast, the effect of pre-stretching of the abdominal wall by a previous abdominal insufflation was found to be significant.

CONCLUSIONS

This experimental study confirms the results from several clinical studies that NMB does not influence laparoscopic working space. Studies dealing with working space during laparoscopy should take note of pre-stretching bias.

Use of multidetector computed tomography to guide management of pneumothorax

PURPOSE OF REVIEW

Pneumothorax, a potentially life-threatening condition, is present in about one-third of chest trauma patients. Traditionally, pneumothorax has been diagnosed and managed by use of chest radiography, which has been found inaccurate and inconsistent. With the ubiquitous application of multidetector computed tomography (MDCT) in emergency care, MDCT quantification of pneumothoraces becomes an emerging technique for accurate determination of the size of pneumothoraces. The use of MDCT quantification provides a promising means to improve pneumothorax management.

RECENT FINDINGS

Recent studies have demonstrated that MDCT is the gold standard for detecting pneumothorax and MDCT provides an effective imaging modality for the accurate measurement of the volume of pneumothoraces. The use of MDCT volumetric quantification of pneumothoraces has been evidenced in the improvement of performance in pneumothorax management for clinically stable chest trauma patients.

SUMMARY

The MDCT volumetric quantification of pneumothoraces is a new concept in the care of chest trauma patients and has the potential to improve pneumothorax management. Further clinical studies are needed to establish a MDCT-based clinical guideline for pneumothorax management.

Using thoracic ultrasonography to accurately assess pneumothorax progression during positive pressure ventilation: a comparison with CT scanning

BACKGROUND

Although thoracic ultrasonography accurately determines the size and extent of occult pneumothoraces (PTXs) in spontaneously breathing patients, there is uncertainty about patients receiving positive pressure ventilation. We compared the lung point (ie, the area where the collapsed lung still adheres to the inside of the chest wall) using the two modalities ultrasonography and CT scanning to determine whether ultrasonography can be used reliably to assess PTX progression in a positive-pressure-ventilated porcine model.

METHODS

Air was introduced in incremental steps into fi ve hemithoraces in three intubated porcine models. The lung point was identified on ultrasound imaging and referenced against the lateral limit of the intrapleural air space identified on the CT scans. The distance from the sternum to the lung point (S-LP) was measured on the CT scans and correlated to the insufflated air volume.

RESULTS

The mean total difference between the 131 ultrasound and CT scan lung points was 6.8 mm (SD, 7.1 mm; range, 0.0-29.3 mm). A mixed-model regression analysis showed a linear relationship between the S-LP distances and the PTX volume ( P , .001).

CONCLUSIONS

In an experimental porcine model, we found a linear relation between the PTX size and the lateral position of the lung point. The accuracy of thoracic ultrasonography for identifying the lung point (and, thus, the PTX extent) was comparable to that of CT imaging. These clinically relevant results suggest that ultrasonography may be safe and accurate in monitoring PTX progression during positive pressure ventilation.

Automatic segmentation and measurement of pleural effusions on CT

Pleural effusion is an important biomarker for the diagnosis of many diseases. We develop an automated method to evaluate pleural effusion on CT scans, the measurement of which is prohibitively time consuming when performed manually. The method is based on parietal and visceral pleura extraction, active contour models, region growing, Bezier surface fitting, and deformable surface modeling. Twelve CT scans with three manual segmentations were used to validate the automatic segmentation method. The method was then applied on 91 additional scans for visual assessment. The segmentation method yielded a correlation coefficient of 0.97 and a Dice coefficient of 0.72±0.13 when compared to a professional manual segmentation. The visual assessment estimated 83% cases with negligible or small segmentation errors, 14% with medium errors, and 3% with large errors.

Optimizing working-space in laparoscopy: measuring the effect of mechanical bowel preparation in a porcine model

BACKGROUND

Adequate working space is a prerequisite for safe and efficient minimal access surgery. No objective data exist in literature about the effect of mechanical bowel preparation (MBP) on working space in laparoscopic surgery. We objectively measured this effect with computed tomography in a porcine laparoscopy model.

METHODS

Using standardized anesthesia, twelve 20-kg pigs without MBP and eight 20-kg pigs with MBP were studied with computed tomography at intra-abdominal pressure (IAP) levels of 0, 5, 10, and 15 mmHg. Volumes and dimensions of the pneumoperitoneum were measured on reconstructed CT images and compared between the pigs with and those without MBP.

RESULTS

A reproducible and statistically significant increase of approximately 500 ml in pneumoperitoneum volume was found in the MBP group at all levels of IAP. This represents a 43 % relative increase at a pneumoperitoneum pressure of 5 mmHg, 21 % at IAP 10 mmHg, and 18 % at IAP 15 mmHg. Peak inspiratory pressure was lower at IAP 0 and 5 mmHg in the MBP group. Anteroposterior diameter in the group with MBP was lower at 0 mmHg, but abdominal dimensions were similar in both groups at all other IAPs. This shows that the gain in working space is due to a diminished volume of the intra-abdominal content and not to compression or displacement of the bowel.

CONCLUSIONS

MBP increases working space by reducing bowel content. Especially at low intra-abdominal working pressures, the increase in working space associated with MBP could represent an important benefit in challenging laparoscopic surgery.

Optimizing working space in porcine laparoscopy: CT measurement of the effects of intra-abdominal pressure

BACKGROUND

Several factors may affect volume and dimensions of the working space in laparoscopic surgery. The precise impact of these factors has not been well studied. In a porcine model, we used computed tomographic (CT) scanning for measuring working space volume and distances. In a first series of experiments, we studied the relationship between intra-abdominal pressure (IAP) and working space.

METHODS

Eleven 20 kg pigs were studied under standardized anesthesia and volume-controlled ventilation. Cardiorespiratory parameters were monitored continuously, and blood gas samples were taken at different IAP levels. Respiratory rate was increased when ETCO₂ exceeded 7 kPa. Breath-hold CT scans were made at IAP levels of 0, 5, 10, and 15 mmHg. Insufflator volumes were compared to CT-measured volumes. Maximum dimensions of pneumoperitoneum were measured on reconstructed CT images.

RESULTS

Respiratory rate had to be increased in three animals. Mild hypercapnia and acidosis occurred at 15 mmHg IAP. Peak inspiratory pressure rose significantly at 10 and 15 mmHg. CT-measured volume increased relatively by 93 % from 5 to 10 mmHg IAP and by 19 % from 10 to 15 mmHg IAP. Comparing CT volumes to insufflator volumes gave a bias of 76 mL. The limits of agreement were -0.31 to +0.47, a range of 790 mL. The internal anteroposterior diameter increased by 18 % by increasing IAP from 5 to 10 mmHg and by 5 % by increasing IAP from 10 to 15 mmHg. At 15 mmHg, the total relative increase of the pubis-diaphragm distance was only 6 %. Abdominal width did not increase.

CONCLUSIONS

CT allows for precise calculation of the actual CO₂ pneumoperitoneum volume, whereas the volume of CO₂ released by the insufflator does not. Increasing IAP up to 10 mmHg achieved most gain in volume and in internal anteroposterior diameter. At an IAP of 10 mmHg, higher peak inspiratory pressure was significantly elevated.

Automated quantification of pneumothorax in CT

An automated, computer-aided diagnosis (CAD) algorithm for the quantification of pneumothoraces from Multidetector Computed Tomography (MDCT) images has been developed. Algorithm performance was evaluated through comparison to manual segmentation by expert radiologists. A combination of two-dimensional and three-dimensional processing techniques was incorporated to reduce required processing time by two-thirds (as compared to similar techniques). Volumetric measurements on relative pneumothorax size were obtained and the overall performance of the automated method shows an average error of just below 1%.

MDCT quantification is the dominant parameter in decision-making regarding chest tube drainage for stable patients with traumatic pneumothorax

It is commonly believed that the size of a pneumothorax is an important determinant of treatment decision, in particular regarding whether chest tube drainage (CTD) is required. However, the volumetric quantification of pneumothoraces has not routinely been performed in clinics. In this paper, we introduced an automated computer-aided volumetry (CAV) scheme for quantification of volume of pneumothoraces in chest multi-detect CT (MDCT) images. Moreover, we investigated the impact of accurate volume of pneumothoraces in the improvement of the performance in decision-making regarding CTD in the management of traumatic pneumothoraces. For this purpose, an occurrence frequency map was calculated for quantitative analysis of the importance of each clinical parameter in the decision-making regarding CTD by a computer simulation of decision-making using a genetic algorithm (GA) and a support vector machine (SVM). A total of 14 clinical parameters, including volume of pneumothorax calculated by our CAV scheme, was collected as parameters available for decision-making. The results showed that volume was the dominant parameter in decision-making regarding CTD, with an occurrence frequency value of 1.00. The results also indicated that the inclusion of volume provided the best performance that was statistically significant compared to the other tests in which volume was excluded from the clinical parameters. This study provides the scientific evidence for the application of CAV scheme in MDCT volumetric quantification of pneumothoraces in the management of clinically stable chest trauma patients with traumatic pneumothorax.

Micropower impulse radar: a novel technology for rapid, real-time detection of pneumothorax

Pneumothorax detection in emergency situations must be rapid and at the point of care. Current standards for detection of a pneumothorax are supine chest X-rays, ultrasound, and CT scans. Unfortunately these tools and the personnel necessary for their facile utilization may not be readily available in acute circumstances, particularly those which occur in the pre-hospital setting. The decision to treat therefore, is often made without adequate information. In this report, we describe a novel hand-held device that utilizes Micropower Impulse Radar to reliably detect the presence of a pneumothorax. The technology employs ultra wide band pulses over a frequency range of 500 MHz to 6 GHz and a proprietary algorithm analyzes return echoes to determine if a pneumothorax is present with no user interpretation required. The device has been evaluated in both trauma and surgical environments with sensitivity of 93% and specificity of 85%. It is has the CE Mark and is available for sale in Europe. Post market studies are planned starting in May of 2011. Clinical studies to support the FDA submission will be completed in the first quarter of 2012.

MDCT for computerized volumetry of pneumothoraces in pediatric patients

RATIONALE AND OBJECTIVES

Our purpose in this study was to develop an automated computer-aided volumetry (CAV) scheme for quantifying pneumothorax in multidetector computed tomography (MDCT) images for pediatric patients and to investigate the imaging parameters that may affect its accuracy.

MATERIALS AND METHODS

Fifty-eight consecutive pediatric patients (mean age 12 ± 6 years) with pneumothorax who underwent MDCT for evaluation were collected retrospectively for this study. All cases were imaged by a 16- or 64-MDCT scanner with weight-based kilovoltage, low-dose tube current, 1.0-1.5 pitch, 0.6-5.0 mm slice thickness, and a B70f (sharp) or B31f (soft) reconstruction kernel. Sixty-three pneumothoraces ≥1 mL were visually identified in the left (n = 30) and right (n = 33) lungs. Each identified pneumothorax was contoured manually on an Amira workstation V4.1.1 (Mercury Computer Systems, Chelmsford, MA) by two radiologists in consensus. The computerized volumes of the pneumothoraces were determined by application of our CAV scheme. The accuracy of our automated CAV scheme was evaluated by comparison between computerized volumetry and manual volumetry, for the total volume of pneumothoraces in the left and right lungs.

RESULTS

The mean difference between the computerized volumetry and the manual volumetry for all 63 pneumothoraces ≥1 mL was 8.2%. For pneumothoraces ≥10 mL, ≥50 mL, and ≥200 mL, the mean differences were 7.7% (n = 57), 7.3% (n = 33), and 6.4% (n = 13), respectively. The correlation coefficient was 0.99 between the computerized volume and the manual volume of pneumothoraces. Bland-Altman analysis showed that computerized volumetry has a mean difference of -5.1% compared to manual volumetry. For all pneumothoraces ≥10 mL, the mean differences for slice thickness ≤1.25 mm, = 1.5 mm, and = 5.0 mm were 6.1% (n = 28), 3.5% (n = 10), and 12.2% (n = 19), respectively. For the two reconstruction kernels, B70f and B31f, the mean differences were 6.3% (n = 42, B70f) and 11.7% (n = 15, B31f), respectively.

CONCLUSION

Our automated CAV scheme provides an accurate measurement of pneumothorax volume in MDCT images of pediatric patients. For accurate volumetric quantification of pneumothorax in children in MDCT images by use of the automated CAV scheme, we recommended reconstruction parameters based on a slice thickness ≤1.5 mm and the reconstruction kernel B70f.