Assessment of automatic rib fracture detection on chest CT using a deep learning algorithm.

OBJECTIVES: To evaluate deep neural networks for automatic rib fracture detection on thoracic CT scans and to compare its performance with that of attending-level radiologists using a large amount of datasets from multiple medical institutions. METHODS: In this retrospective study, an internal dataset of 12,208 emergency room (ER) trauma patients and an external dataset of 1613 ER trauma patients taking chest CT scans were recruited. Two cascaded deep neural networks based on an extended U-Net architecture were developed to segment ribs and detect rib fractures respectively. Model performance was evaluated with a 95% confidence interval (CI) on both the internal and external dataset, and compared with attending-level radiologist readings using t test. RESULTS: On the internal dataset, the AUC of the model for detecting fractures at per-rib level was 0.970 (95% CI: 0.968, 0.972) with sensitivity of 93.3% (95% CI: 92.0%, 94.4%) at a specificity of 98.4% (95% CI: 98.3%, 98.5%). On the external dataset, the model obtained an AUC of 0.943 (95% CI: 0.941, 0.945) with sensitivity of 86.2% (95% CI: 85.0%, 87.3%) at a specificity of 98.8% (95% CI: 98.7%, 98.9%), compared to the sensitivity of 70.5% (95% CI: 69.3%, 71.8%) (p < .0001) and specificity of 98.8% (95% CI: 98.7%, 98.9%) (p = 0.175) by attending radiologists. CONCLUSIONS: The proposed DL model is a feasible approach to identify rib fractures on chest CT scans, at the very least, reaching a level on par with attending-level radiologists. KEY POINTS: • Deep learning-based algorithms automatically detected rib fractures with high sensitivity and reasonable specificity on chest CT scans. • The performance of deep learning-based algorithms reached comparable diagnostic measures with attending level radiologists for rib fracture detection on chest CT scans. • The deep learning models, similar to human readers, were susceptible to the inconspicuity and ambiguity of target lesions. More training data was required for subtle lesions to achieve comparable detection performance.

Development and assessment of deep learning system for the location and classification of rib fractures via computed tomography.

PURPOSE: The purpose of this study was to evaluate the performance of a deep learning system for the automatic diagnosis and classification of rib fractures. METHODS: This retrospective study analyzed computed tomography (CT) data of patients diagnosed with a rib fracture between 1 January 2019 and 23 July 2020 in two hospitals, including 591 patients from Suzhou TCM hospital and 75 patients from Jintan TCM hospital. A deep learning system (Dr.Wise@ChestFracture v1.0) based on a convolutional neural network framework was used as a diagnostic tool, and a human-model comparison experiment was designed to compare the diagnostic efficiencies of the deep learning system and radiologists. Furthermore, a secondary classification model was established to distinguish the different types of fracture. First, a classification model to differentiate between fresh and old fractures was developed. Second, a submodel to determine any misalignment in fresh fractures was established. RESULTS: For all fracture types, the detection efficiency (recall) of the system was statistically significantly better than that of radiologists with different levels of experience (all p < 0.0167 except for senior radiologists). The F1-score of the system for diagnosing rib fractures was similar to that of the radiologists. The system was much faster than the radiologists in assessing rib fractures (all p < 0.0167). The two classification models can distinguish between fresh and old fractures (accuracy = 87.63%) and determine whether there is any misalignment in fresh fractures (accuracy = 95.22%) or not. CONCLUSION: The use of a deep learning system can accurately, automatically, and rapidly diagnose and classify rib fractures, helping doctors improve the diagnostic efficiency and reducing their workload. The classification models can distinguish different types of rib fracture well.

A new quantitative assessment method for predicting pneumonia caused by chest wall injury.

BACKGROUND: The severity of rib fractures has been previously evaluated by combining categorical data, but these methods have only low predictive capability for respiratory complications and mortality. This study aimed to establish a more accurate method for predicting the development of pneumonia, a frequent complication in chest injuries, using anatomical relationships. METHODS: We analyzed three-dimensional reconstructed images of 644 consecutive trauma patients who underwent whole-body computed tomography (CT) in our institution within a 36-month study period from April 2017. The anatomical relationship between the right and left thoracic volumes of non-rib fracture patients was used to estimate thoracic volume changes on the injured side in unilateral rib fracture patients. The predictive capability of changes in thoracic volume for the development of pneumonia was evaluated according to the area under the receiver operating characteristic curve and compared with that of previous chest wall severity evaluation methods. RESULTS: Of the 644 patients, 133 and 478 patients had unilateral rib fractures and non-rib fractures, respectively. The amount of change in thoracic volume due to unilateral rib fractures was significantly greater in pneumonia patients (400 mL vs. 160 mL, p < 0.01). The area under the receiver operating characteristic curve for the development of pneumonia was 0.83, which tended to be higher than that of the previous severity scoring methods. CONCLUSION: The amount of change in chest volume, which can be estimated using CT images, has better predictive capability for pneumonia than previous severity assessment methods based on categorical data. The amount of change in chest volume measured using whole-body CT can be used to rapidly determine the optimal treatment for severe chest wall injuries. LEVEL OF EVIDENCE: Prognostic study, level IV.

Assessment of thoracic volume changes after the collapse of lateral rib fractures based on chest computed tomography data: computer simulation and a multiple variable linear regression analysis.

BACKGROUND: Chest blunt trauma (CBT) and the resultant rib fractures often lead to thoracic collapse. The purpose of this study was to explore the effect of displacement of the rib fracture and thoracic collapse on the thoracic volume by using normal chest CT data. METHODS: In this retrospective study, seven consecutive normal participants were selected from our hospital between June and July 2018. Normal thoracic models were reconstructed, followed by simulation of lateral fractures through the 4th to 9th ribs under three collapse modes with 1-5 cm of collapse. The thoracic collapse models (n = 630) were reconstructed using 3Dmax 2014. We calculated the thoracic volume and reduction percentage for each thoracic collapse model. Linear regression-based comparisons of thoracic volume reductions were performed. RESULTS: In all three collapse modes, the degree of the collapse was linearly correlated with the mean thoracic volume reduction. The reduction percentage in the posterior collapse mode was higher than that in the anterior collapse mode (P < 0.001). The largest volume reductions in the anterior, posterior, and simultaneous collapse models were in the 6th rib fracture model (P < 0.001), 8th rib fracture model (P < 0.001), and 7th rib fracture model (P < 0.001), respectively. CONCLUSIONS: The influences of rib fracture displacement and collapse on the thoracic volume in the 6th through 8th ribs are critical in lateral rib fractures. For patients with 6th to 8th rib fractures and posterior rib collapse, surgical intervention to restore thoracic volume may be more essential.

Comparison of the Effectiveness of Surgical Versus Nonsurgical Treatment for Multiple Rib Fractures Accompanied with Pulmonary Contusion.

OBJECTIVE: To compare the effectiveness of surgical versus nonsurgical treatment for multiple rib fractures accompanied with pulmonary contusion. METHODS: The clinical records of consecutive 167 patients with multiple rib fractures accompanied with pulmonary contusion, who were treated from June 2014 to June 2017, were retrospectively analyzed. Of them, 75 and 92 underwent surgery (surgery group) and non-surgical treatment (non-surgery group), respectively. Patient pain score, complications, length of hospital stay, cost of hospitalization, and post-treatment 3-month follow-up results were compared. RESULTS: The mean number of days and moderate pain in the surgery group was significantly lower than that of the non-surgery group (p <0.01). The incidence of post-treatment complications was significantly lower in the surgery group than in the non-surgery group. The length of hospital stay of the surgery group was also significantly shorter than that of the non-surgery group (p <0.01). The cost of hospitalization was significantly higher in the surgery group than in the non-surgery group (p <0.01). The chest computed tomography (CT) scan which was performed 3 months after the treatment revealed that the surgery group had a better recovery than the non-surgery group. Physical recovery of the surgery group was also significantly better than that of the non-surgery group. CONCLUSION: Surgery to treat multiple rib fractures (≥ 4 fractures) accompanied with pulmonary contusion is safe and effective.

New bone post-processing tools in forensic imaging: a multi-reader feasibility study to evaluate detection time and diagnostic accuracy in rib fracture assessment.

PURPOSE: The aim of this multi-reader feasibility study was to evaluate new post-processing CT imaging tools in rib fracture assessment of forensic cases by analyzing detection time and diagnostic accuracy. MATERIALS AND METHODS: Thirty autopsy cases (20 with and 10 without rib fractures in autopsy) were randomly selected and included in this study. All cases received a native whole body CT scan prior to the autopsy procedure, which included dissection and careful evaluation of each rib. In addition to standard transverse sections (modality A), CT images were subjected to a reconstruction algorithm to compute axial labelling of the ribs (modality B) as well as "unfolding" visualizations of the rib cage (modality C, "eagle tool"). Three radiologists with different clinical and forensic experience who were blinded to autopsy results evaluated all cases in a random manner of modality and case. RESULTS: Rib fracture assessment of each reader was evaluated compared to autopsy and a CT consensus read as radiologic reference. A detailed evaluation of relevant test parameters revealed a better accordance to the CT consensus read as to the autopsy. Modality C was the significantly quickest rib fracture detection modality despite slightly reduced statistic test parameters compared to modalities A and B. CONCLUSION: Modern CT post-processing software is able to shorten reading time and to increase sensitivity and specificity compared to standard autopsy alone. The eagle tool as an easy to use tool is suited for an initial rib fracture screening prior to autopsy and can therefore be beneficial for forensic pathologists.

An assessment of outcomes with intramedullary fixation of fractured ribs.

BACKGROUND: Surgical management of fractured ribs with internal fixation is an increasingly accepted therapy. Concurrently, specific rib fixation prostheses are being developed which should improve results and minimise hardware and rib/splint construct failures. The Synthes titanium intramedullary splint lends itself to difficult to access areas such as posterior rib fractures and fractures under the scapula. We analyse a case series of patients in whom this rib fixation prosthesis has been used. METHODS: Fifteen patients received 35 intramedullary splints. Follow up at 3 and 6 months was performed with three dimensional computed tomography scanning to assess for bone alignment, callus formation and healing, residual deformity, hardware failure or cut through. Computerized finite element analysis (FEA) was used to model forces acting on a posterior fracture with and without an intramedullary fixation splint in situ. RESULTS: Complete healing (bony union) was noted in only 3 (9 %) of the fractures fixed with splints by 3 months. Partial healing (cartilaginous union) was noted in 28 of the 33 fractures (85 %), and non healing was noted in only 2 (6 %). In both those two patients, failure at the rib / splint interface was noted after both patients reported sneezing. No hardware failures were noted. By 6 months the fractures which had shown partial healing, had all completely healed. There were no late failures (between 3 and 6 months) of either hardware or rib/splint interfaces. FEA modelling identified sites of increased stress in the rib at the rib / splint interface and in a modelled intramedullary splint where it spans the fracture. CONCLUSIONS: Further analysis of outcomes with intramedullary splints is warranted as well as further development of intramedullary rib fixation solutions.

[COMPARATIVE STUDY OF THE TACTICS OF SURGICAL FIXATION WITH THE CLOSED INJURY OF THE CHEST: THE VIEW OF THE RESUSCITATOR, ECONOMIC ASPECTS].

The closed injury of chest with the breaks of edges is the vital problem of traumatology, anesthesiology and resuscitation For the change to conservative treatment with the aid of mechanical ventilation of lungs today come the methods of surgical fixation with the closed injury of chest. The conducted investigation showed the clinical and economic expediency of introducing the method of active surgical tactics.

The diagnostic utility of rib series in assessing rib fractures.

Rib series rarely add information to the posteroanterior (PA) film for the diagnosis of rib fractures. In this investigation, we evaluated the utility of rib X-rays using turnaround time (TAT), radiation exposure, and cost-efficiency as the key parameters. This investigation was conducted from January 2008 to December 2012. We included patients who had rib series performed for suspected rib fractures. TAT for patients was calculated from the time exam was ordered by the emergency department (ED) physician/staff to time the report was finalized by the attending radiologist. Effective radiation dose for rib series was calculated as a summation of radiation dose from the standard rib series images for each patient. Cost-efficiency was determined based on the number of interventions that took place as a result of a complicated study. Our investigation consisted of 422 patients, 208 females aged (57 ± 20.8) and 214 males aged (48 ± 17.3). A total of 74(17.5 %) abnormal findings were noted, out of which only 1(0.23 %) underwent management change. The mean turnaround time for patients undergoing rib series had a value of 133.5 (±129.8) min as opposed to a single chest PA of 61.8(± 64) min. Average effective radiation dose for a rib series was 0.105 (±0.04) mSv, whereas average effective radiation dose of a single chest PA was 0.02 mSv. Dedicated rib series has a low-yield diagnostic value as it pertains to management change. The overall impact on patient care based on our findings is small when compared to the risks associated with prolonged TAT, repeated exposure to radiation, and extensive medical costs.

Clinical rib fractures: are follow-up chest X-rays a waste of resources?

Rib fractures (RFs) are estimated to be present in 10 per cent of all traumatic injuries. However, up to 50 per cent of all fractures go undetected on the screening chest X-ray (CXR). The purpose of this study was to identify the incidence of clinical (CRFs) and objective rib fractures (ORFs) as well as to examine the utility of the routine follow-up CXR with regard to patient recovery and healthcare cost. We identified patients sustaining RF in addition to other traumatic injuries with an Injury Severity Score (ISS) < or = 15 and RF as the primary pathology. Five hundred fifty-two patients sustained blunt thoracic trauma with resultant RF. Two hundred nine patients had RFs and an ISS < or = 15. The average ISS was 8. Follow-up films illustrated that 93 per cent of CRFs had resolution of any pathology, 4 per cent had persistent X-ray findings, and 4 per cent were lost to follow-up. Ultimately 93 per cent of patients with CRF were able to resume daily activities without disability and 3 per cent incurred lifestyle changes at home or work, which was significantly better than those with ORFs (P < 0.05). Follow-up films produced no change in clinical management and cost approximately $2000/year. The prognosis for CRFs is excellent if treatment consists of appropriate pain management and pulmonary rehabilitation. We do not advocate routine follow-up CXRs in addition to physical examination for the evaluation of CRFs unless clinical deterioration is evident.

The evaluation of diagnostic accordance between plain radiography and bone scintigraphy for the assessment of sternum and rib fractures in the early period of blunt trauma.

OBJECTIVE: To evaluate the diagnostic accordance between scintigraphy and radiography in the early period following blunt thoracic trauma. DESIGN: Double-blind, prospective clinical study. METHODS: Thirty-four adult patients with blunt thoracic trauma were examined by plain radiography and scintigraphy within 24 hours of trauma. RESULTS: Radiography displayed pathologies in 25 (73.5%), while scintigraphy suggested pathologies in 24 (70.5%) patients. Rib fractures were identified radiographically in 16 (47%) and scintigraphically in 17 (50%) patients. Lateral radiograph of the sternum showed fracture in 10 (52.6%), while scintigraphy suggested fracture in 14 (73.6%) patients with suspected sternal trauma. CONCLUSION: Scintigraphy gave better results in the evaluation of sternal fractures during the early period of thorax trauma. Conventional plain radiography is thought to be the initial imaging modality, because it saves time and shows hemothorax or pneumothorax besides the osseous abnormality.

Non-life-threatening blunt chest trauma: appropriate investigation and treatment.

Non-life-threatening blunt chest trauma is a commonly occurring phenomenon. No clear guidelines exist in the literature regarding the appropriate investigation and treatment, and most clinicians' practice patterns are based on anecdote, individual experience, and the theoretical risk of complications. A prospective study was undertaken of all patients with non-life-threatening blunt chest trauma presenting to a high-volume, community-based teaching hospital emergency department. No clinical features were found to have a high enough positive predictive value for rib fracture or any other injury to warrant their use as a screening tool for further investigations. No significant lung injury was detected in the study cohort of patients, on either an immediate or delayed basis. In the absence of compelling evidence of an underlying lung injury, the performance of chest or rib radiography is of no benefit, and considerable cost savings can potentially be realized.

[The echographic assessment of traumatic rib lesions]. / Valutazione ecografica delle lesioni traumatiche costali.

INTRODUCTION: Conventional frontal radiography is mandatory in trauma patients and usually sufficient to detect rib fractures, even though the alignment of bone fragments or projection difficulties may sometimes hinder their depiction. Therefore, there exists moderate disagreement between clinically suspected rib fractures and radiographic findings. MATERIALS AND METHODS: We report our personal experience with US in the detection of rib conditions (fractures and bruises), missed at preliminary radiography. Five symptomatic patients with suspected rib fractures underwent radiographic and US studies. All patients were examined with a real time unit (RT 2800, GE Medical Systems, Milwaukee, USA) with a 7.5 MHz linear probe. RESULTS AND CONCLUSION: US demonstrated rib fractures in 4 patients with negative radiographic findings and a parosteal hematoma in one patient. We conclude that US can detect the rib fractures missed at conventional radiography and show rib cartilage conditions, costochondral dislocations and parosteal hematomas. Therefore we suggest the routine use of this technique in all chest trauma patients, as a useful complement to frontal radiographs, for both management and forensic purposes.

Rib radiographs for trauma: useful or wasteful?

In trauma to the chest, the clinical impression and the physical findings of rib fractures are nonspecific. Fractures often are not seen on initial films. The principal diagnostic goal should be the detection of significant complications (pneumothorax, hemothorax, major vascular injury, or pulmonary contusion) requiring admission. The therapeutic effort should be to provide pain relief and prevent the delayed development of atelectasis or pneumonia in patients with painful chest wall injuries, whether or not a fracture is detected initially. An upright posteroanterior chest radiograph has the greatest yield in detecting fractures and complications resulting from them. Tomograms and expiratory, oblique, and "coned-down" views should not be done routinely. The use of these more specific examinations may be indicated, however, in such cases as trauma to ribs 1 to 3 or 9 to 12. Their selective use in isolated cases (trauma to ribs 1 to 3 or 9 to 12) and suspected child abuse may indicate the need for these more specific examinations. Because detection of pulmonary complications of chest trauma is most important, a delayed or repeat upright posteroanterior chest radiograph may be the most cost-effective second radiograph. Significant medical care cost savings may be appreciated by limiting the use of specific rib views to instances in which it might influence the patient's therapy.

Radiographic evaluation of rib fractures.

A prospective study of 100 radiographs performed for possible rib fracture due to blunt trauma was conducted to determine if some of the films routinely obtained should be eliminated for medical or economic reasons. Of the 100 patients evaluated for fracture, 29 fractures were detected radiographically. The physical findings of fracture are seen to be nonspecific, and the clinical impression of the likelihood of fracture is unreliable. Therapy is symptomatic, aimed at relief of pain, and except for the complications of fractures, therapy does not differ significantly in patients with and without fractures. The complications of fractures that may influence therapy were seen in 13 of the 29 patients with fractures. All 13 complications were detected by the posteroanterior chest film while only three were seen on the rib films. Radiographs obtained solely to detect fractures are not warranted, but a posteroanterior chest film is necessary to assess pleural or pulmonary complication of chest trauma.