Detection of acute rib fractures on CT images with convolutional neural networks: effect of location and type of fracture and reader's experience

Detection, Classification, and Segmentation of Rib Fractures From CT Data Using Deep Learning Models: A Review of Literature and Pooled Analysis

PURPOSE

Trauma-induced rib fractures are common injuries. The gold standard for diagnosing rib fractures is computed tomography (CT), but the sensitivity in the acute setting is low, and interpreting CT slices is labor-intensive. This has led to the development of new diagnostic approaches leveraging deep learning (DL) models. This systematic review and pooled analysis aimed to compare the performance of DL models in the detection, segmentation, and classification of rib fractures based on CT scans.

MATERIALS AND METHODS

A literature search was performed using various databases for studies describing DL models detecting, segmenting, or classifying rib fractures from CT data. Reported performance metrics included sensitivity, false-positive rate, F1-score, precision, accuracy, and mean average precision. A meta-analysis was performed on the sensitivity scores to compare the DL models with clinicians.

RESULTS

Of the 323 identified records, 25 were included. Twenty-one studies reported on detection, four on segmentation, and 10 on classification. Twenty studies had adequate data for meta-analysis. The gold standard labels were provided by clinicians who were radiologists and orthopedic surgeons. For detecting rib fractures, DL models had a higher sensitivity (86.7%; 95% CI: 82.6%-90.2%) than clinicians (75.4%; 95% CI: 68.1%-82.1%). In classification, the sensitivity of DL models for displaced rib fractures (97.3%; 95% CI: 95.6%-98.5%) was significantly better than that of clinicians (88.2%; 95% CI: 84.8%-91.3%).

CONCLUSIONS

DL models for rib fracture detection and classification achieved promising results. With better sensitivities than clinicians for detecting and classifying displaced rib fractures, the future should focus on implementing DL models in daily clinics.

LEVEL OF EVIDENCE

Level III-systematic review and pooled analysis.

A deep learning-based approach to automated rib fracture detection and CWIS classification

PURPOSE

Trauma-induced rib fractures are a common injury. The number and characteristics of these fractures influence whether a patient is treated nonoperatively or surgically. Rib fractures are typically diagnosed using CT scans, yet 19.2-26.8% of fractures are still missed during assessment. Another challenge in managing rib fractures is the interobserver variability in their classification. Purpose of this study was to develop and assess an automated method that detects rib fractures in CT scans, and classifies them according to the Chest Wall Injury Society (CWIS) classification.

METHODS

198 CT scans were collected, of which 170 were used for training and internal validation, and 28 for external validation. Fractures and their classifications were manually annotated in each of the scans. A detection and classification network was trained for each of the three components of the CWIS classifications. In addition, a rib number labeling network was trained for obtaining the rib number of a fracture. Experiments were performed to assess the method performance.

RESULTS

On the internal test set, the method achieved a detection sensitivity of 80%, at a precision of 87%, and an F1-score of 83%, with a mean number of FPPS (false positives per scan) of 1.11. Classification sensitivity varied, with the lowest being 25% for complex fractures and the highest being 97% for posterior fractures. The correct rib number was assigned to 94% of the detected fractures. The custom-trained nnU-Net correctly labeled 95.5% of all ribs and 98.4% of fractured ribs in 30 patients. The detection and classification performance on the external validation dataset was slightly better, with a fracture detection sensitivity of 84%, precision of 85%, F1-score of 84%, FPPS of 0.96 and 95% of the fractures were assigned the correct rib number.

CONCLUSION

The method developed is able to accurately detect and classify rib fractures in CT scans, there is room for improvement in the (rare and) underrepresented classes in the training set.

Intelligent detection and grading diagnosis of fresh rib fractures based on deep learning

BACKGROUND

Accurate detection and grading of fresh rib fractures are crucial for patient management but remain challenging due to the complexity of rib structures on CT images.

METHODS

Chest CT images from 383 patients with rib fractures were retrospectively analyzed. The dataset was divided into a training set (n = 306) and an internal testing set (n = 77). An external testing set of 50 patients from the public RibFrac dataset was included. Fractures were classified into severe and non-severe categories. A modified YOLO-based deep learning model was developed for detection and grading. Performance was compared with thoracic surgeons using precision, recall, mAP50, and F1 score.

RESULTS

The deep learning model showed excellent performance in diagnosing fresh rib fractures. For all fractures types in internal test set, the precision, recall, mAP50, and F1 score were 0.963, 0.934, 0.972, and 0.948, respectively. The model outperformed thoracic surgeons of varying experience levels (all p < 0.01).

CONCLUSION

The proposed deep learning model can automatically detect and grade fresh rib fractures with accuracy comparable to that of physicians. This model helps improve diagnostic accuracy, reduce physician workload, save medical resources, and strengthen health care in resource-limited areas.

CLINICAL TRIAL NUMBER

Not applicable.

Applications of Artificial Intelligence in Acute Thoracic Imaging

The applications of artificial intelligence (AI) in radiology are rapidly advancing with AI algorithms being used in a wide range of disease pathologies and clinical settings. Acute thoracic pathologies including rib fractures, pneumothoraces, and acute PE are associated with significant morbidity and mortality and their identification is crucial for prompt treatment. AI models which increase diagnostic accuracy, improve radiologist efficiency and reduce time to diagnosis of acute abnormalities in the thorax have the potential to significantly improve patient outcomes. The purpose of this review is to summarize the current applications of AI in acute thoracic imaging, highlighting their strengths, limitations, and future research opportunities.

Detection of Costal Cartilage Fractures on CT Images With Computer-aided Detection System for Rib Fractures

BACKGROUND/AIM

Costal cartilage fractures are associated with poor prognosis in patients with blunt chest trauma. A Computer-Aided Detection (CAD) system for detecting rib fractures has been used in practice, but it is unclear whether this system recognizes costal cartilage fractures. This study investigated whether the CAD system for rib fracture can detect costal cartilage fractures.

PATIENTS AND METHODS

A total of 89 patients with costal cartilage fractures from participating hospitals over an 18-year period were included in the study. The presence of a costal cartilage fracture was determined by three radiologists. We reviewed fracture location, cartilage calcification, dislocation, and callus formation. The percentage of agreement between the radiologists and the CAD system (Rib fracture CAD, Fujifilm Medical Co., Ltd) was assessed.

RESULTS

We detected 246 costal cartilage fractures in 89 patients. The costal cartilage of rib 7 was injured most frequently. Costal cartilage fractures were categorized as either mid-chondral, costochondral, or chondro-sternal. The CAD system detected 33 lesions; 16 were consistent with the costal cartilage fractures determined by the radiologists (costochondral junction 13, mid-chondral 2, chondro-sternal 1).

CONCLUSION

The CAD system for rib fracture can detect costal cartilage fractures at the costochondral junction but is not sufficiently sensitive to detect costal cartilage fractures without calcification. The CAD system for rib fracture needs further development before it can be used to detect rib cartilage fractures.

Diagnostic evaluation of blunt chest trauma by imaging-based application of artificial intelligence

Artificial intelligence (AI) is becoming increasingly integral in clinical practice, such as during imaging tasks associated with the diagnosis and evaluation of blunt chest trauma (BCT). Due to significant advances in imaging-based deep learning, recent studies have demonstrated the efficacy of AI in the diagnosis of BCT, with a focus on rib fractures, pulmonary contusion, hemopneumothorax and others, demonstrating significant clinical progress. However, the complicated nature of BCT presents challenges in providing a comprehensive diagnosis and prognostic evaluation, and current deep learning research concentrates on specific clinical contexts, limiting its utility in addressing BCT intricacies. Here, we provide a review of the available evidence surrounding the potential utility of AI in BCT, and additionally identify the challenges impeding its development. This review offers insights on how to optimize the role of AI in the diagnostic evaluation of BCT, which can ultimately enhance patient care and outcomes in this critical clinical domain.

The impact of large language models on radiology: a guide for radiologists on the latest innovations in AI

The advent of Deep Learning (DL) has significantly propelled the field of diagnostic radiology forward by enhancing image analysis and interpretation. The introduction of the Transformer architecture, followed by the development of Large Language Models (LLMs), has further revolutionized this domain. LLMs now possess the potential to automate and refine the radiology workflow, extending from report generation to assistance in diagnostics and patient care. The integration of multimodal technology with LLMs could potentially leapfrog these applications to unprecedented levels.However, LLMs come with unresolved challenges such as information hallucinations and biases, which can affect clinical reliability. Despite these issues, the legislative and guideline frameworks have yet to catch up with technological advancements. Radiologists must acquire a thorough understanding of these technologies to leverage LLMs' potential to the fullest while maintaining medical safety and ethics. This review aims to aid in that endeavor.

Deep learning for acute rib fracture detection in CT data: a systematic review and meta-analysis

OBJECTIVES

To review studies on deep learning (DL) models for classification, detection, and segmentation of rib fractures in CT data, to determine their risk of bias (ROB), and to analyse the performance of acute rib fracture detection models.

METHODS

Research articles written in English were retrieved from PubMed, Embase, and Web of Science in April 2023. A study was only included if a DL model was used to classify, detect, or segment rib fractures, and only if the model was trained with CT data from humans. For the ROB assessment, the Quality Assessment of Diagnostic Accuracy Studies tool was used. The performance of acute rib fracture detection models was meta-analysed with forest plots.

RESULTS

A total of 27 studies were selected. About 75% of the studies have ROB by not reporting the patient selection criteria, including control patients or using 5-mm slice thickness CT scans. The sensitivity, precision, and F1-score of the subgroup of low ROB studies were 89.60% (95%CI, 86.31%-92.90%), 84.89% (95%CI, 81.59%-88.18%), and 86.66% (95%CI, 84.62%-88.71%), respectively. The ROB subgroup differences test for the F1-score led to a p-value below 0.1.

CONCLUSION

ROB in studies mostly stems from an inappropriate patient and data selection. The studies with low ROB have better F1-score in acute rib fracture detection using DL models.

ADVANCES IN KNOWLEDGE

This systematic review will be a reference to the taxonomy of the current status of rib fracture detection with DL models, and upcoming studies will benefit from our data extraction, our ROB assessment, and our meta-analysis.

Oblique Axis Rib Stretch and Curved Planar Reformats in Patients for Rib Fracture Detection and Characterization: Feasibility and Clinical Application

Objective

To assess the use of CT with oblique axis rib stretch (OARS) and curved planar reformats (CPRs) for rib fracture detection and characterization.

Methods

A total of 108 forensically diagnosed patients with rib fractures were evaluated retrospectively. OARS and CPRs were independently used during the diagnosis in two groups. In each group, the final diagnosis was made after a junior radiologist's initial diagnosis was reviewed by a senior radiologist. The images were evaluated for the presence and characterization of rib fractures.

Results

A total of 2,592 ribs were analyzed, and 326 fractured ribs and 345 fracture sites were diagnosed using reference standard. Two groups of radiologists identified 331 and 333 fracture sites using the OARS method, 291 and 288 fracture sites using the CPRs method, and 274 fracture sites in forensically diagnosed patients (CR: conventional reconstruction), respectively; and all missed diagnoses were nondisplaced rib fractures. The ROC Az value of OARS1,2 was 0.98, which is higher than CPRs1,2 0.91, and CR 0.90 (all p < 0.01). The Az value for detecting rib fractures using CPRs1,2 and CR has no statistical difference (p = 0.14 and 0.29). More misdiagnosed patients were found using CPRs1,2 (42 and 44 cases) than OARS1,2 (1 and 2 cases) and CR (2 cases). The displaced fracture detection ratio of all methods showed no difference.

Conclusions

Doctors using the OARS method could improve diagnostic performance for detecting rib fractures without the requirement of specialized software and workstation when compared with CPRs.

Development of an artificial intelligence-assisted computed tomography diagnosis technology for rib fracture and evaluation of its clinical usefulness

Artificial intelligence algorithms utilizing deep learning are helpful tools for diagnostic imaging. A deep learning-based automatic detection algorithm was developed for rib fractures on computed tomography (CT) images of high-energy trauma patients. In this study, the clinical effectiveness of this algorithm was evaluated. A total of 56 cases were retrospectively examined, including 46 rib fractures and 10 control cases from our hospital, between January and June 2019. Two radiologists annotated the fracture lesions (complete or incomplete) for each CT image, which is considered the “ground truth.” Thereafter, the algorithm’s diagnostic results for all cases were compared with the ground truth, and the sensitivity and number of false positive (FP) results per case were assessed. The radiologists identified 199 images with a fracture. The sensitivity of the algorithm was 89.8%, and the number of FPs per case was 2.5. After additional learning, the sensitivity increased to 93.5%, and the number of FPs was 1.9 per case. FP results were found in the trabecular bone with the appearance of fracture, vascular grooves, and artifacts. The sensitivity of the algorithm used in this study was sufficient to aid the rapid detection of rib fractures within the evaluated validation set of CT images.