Improving rib fracture detection accuracy and reading efficiency with deep learning-based detection software: a clinical evaluation

Objectives: To investigate the impact of deep learning (DL) on radiologists’ detection accuracy and reading efficiency of rib fractures on CT. Methods: Blunt chest trauma patients (n = 198) undergoing thin-slice CT were enrolled. Images were read by two radiologists (R1, R2) in three sessions: S1, unassisted reading; S2, assisted by DL as the concurrent reader; S3, DL as the second reader. The fractures detected by the readers and total reading time were documented. The reference standard for rib fractures was established by an expert panel. The sensitivity and false-positives per scan were calculated and compared among S1, S2, and S3. Results: The reference standard identified 865 fractures on 713 ribs (102 patients) The sensitivity of S1, S2, and S3 was 82.8, 88.9, and 88.7% for R1, and 83.9, 88.7, and 88.8% for R2, respectively. The sensitivity of S2 and S3 was significantly higher compared to S1 for both readers (all p < 0.05). The sensitivity between S2 and S3 did not differ significantly (both p > 0.9). The false-positive per scan had no difference between sessions for R1 (p = 0.24) but was lower for S2 and S3 than S1 for R2 (both p < 0.05). Reading time decreased by 36% (R1) and 34% (R2) in S2 compared to S1. Conclusions: Using DL as a concurrent reader can improve the detection accuracy and reading efficiency for rib fracture. Advances in knowledge: DL can be integrated into the radiology workflow to improve the accuracy and reading efficiency of CT rib fracture detection.

Download Full-text

Rib fracture detection system based on deep learning

Scientific Reports ◽

10.1038/s41598-021-03002-7 ◽

2021 ◽

Vol 11 (1) ◽

Author(s):

Liding Yao ◽

Xiaojun Guan ◽

Xiaowei Song ◽

Yanbin Tan ◽

Chun Wang ◽

...

Keyword(s):

Deep Learning ◽

Learning Algorithm ◽

Detection System ◽

Rib Fracture ◽

Diagnostic Efficiency ◽

Fracture Detection ◽

Deep Learning Algorithm ◽

Experienced Radiologist ◽

Comparable Performance ◽

Detection And Diagnosis

AbstractRib fracture detection is time-consuming and demanding work for radiologists. This study aimed to introduce a novel rib fracture detection system based on deep learning which can help radiologists to diagnose rib fractures in chest computer tomography (CT) images conveniently and accurately. A total of 1707 patients were included in this study from a single center. We developed a novel rib fracture detection system on chest CT using a three-step algorithm. According to the examination time, 1507, 100 and 100 patients were allocated to the training set, the validation set and the testing set, respectively. Free Response ROC analysis was performed to evaluate the sensitivity and false positivity of the deep learning algorithm. Precision, recall, F1-score, negative predictive value (NPV) and detection and diagnosis were selected as evaluation metrics to compare the diagnostic efficiency of this system with radiologists. The radiologist-only study was used as a benchmark and the radiologist-model collaboration study was evaluated to assess the model’s clinical applicability. A total of 50,170,399 blocks (fracture blocks, 91,574; normal blocks, 50,078,825) were labelled for training. The F1-score of the Rib Fracture Detection System was 0.890 and the precision, recall and NPV values were 0.869, 0.913 and 0.969, respectively. By interacting with this detection system, the F1-score of the junior and the experienced radiologists had improved from 0.796 to 0.925 and 0.889 to 0.970, respectively; the recall scores had increased from 0.693 to 0.920 and 0.853 to 0.972, respectively. On average, the diagnosis time of radiologist assisted with this detection system was reduced by 65.3 s. The constructed Rib Fracture Detection System has a comparable performance with the experienced radiologist and is readily available to automatically detect rib fracture in the clinical setting with high efficacy, which could reduce diagnosis time and radiologists’ workload in the clinical practice.

Download Full-text

An Advanced Deep Learning Approach for Multi-Object Counting in Urban Vehicular Environments

Future Internet ◽

10.3390/fi13120306 ◽

2021 ◽

Vol 13 (12) ◽

pp. 306

Author(s):

Ahmed Dirir ◽

Henry Ignatious ◽

Hesham Elsayed ◽

Manzoor Khan ◽

Mohammed Adib ◽

...

Keyword(s):

Deep Learning ◽

Intelligent Transportation Systems ◽

Smart Cities ◽

Weather Conditions ◽

Research Area ◽

Transportation Systems ◽

Detection Accuracy ◽

Adverse Weather ◽

Object Counting ◽

The Impact

Object counting is an active research area that gained more attention in the past few years. In smart cities, vehicle counting plays a crucial role in urban planning and management of the Intelligent Transportation Systems (ITS). Several approaches have been proposed in the literature to address this problem. However, the resulting detection accuracy is still not adequate. This paper proposes an efficient approach that uses deep learning concepts and correlation filters for multi-object counting and tracking. The performance of the proposed system is evaluated using a dataset consisting of 16 videos with different features to examine the impact of object density, image quality, angle of view, and speed of motion towards system accuracy. Performance evaluation exhibits promising results in normal traffic scenarios and adverse weather conditions. Moreover, the proposed approach outperforms the performance of two recent approaches from the literature.

Download Full-text

Management of multiple rib fractures-results from a Major Trauma Centre with review of the existing literature

10.21203/rs.3.rs-118238/v2 ◽

2020 ◽

Author(s):

Debkumar Chowdhury ◽

P. Okoh ◽

H. Dambappa

Keyword(s):

Major Trauma ◽

Significant Proportion ◽

Fragility Fractures ◽

Trauma Centre ◽

Rib Fractures ◽

Rib Fracture ◽

Analgesic Requirement ◽

Multiple Rib Fractures ◽

Major Trauma Centre ◽

The Impact

Abstract Introduction Rib fractures are amongst the most common fractures following major trauma presenting to the Emergency Department. It accounts for more than 15% of ED presentations (1) on a global scale. As the population ages the incidence of rib fractures also rises often following falls from a relatively small height being part of fragility fractures. The impact of rib fractures is even more significant in the patient with underlying chronic respiratory conditions.Aim To assess our current management of rib fractures at our trauma centreMethod We collected our data from the TARN Registry primarily focussing on patients with multiple rib fractures. The main components were the analgesic requirement of our patients. We also studied the number of rib fracture stabilisation procedures and the average number of ribs fixed.Results The data was collected retrospectively over a period of 12 months. There were 313 patients identified as having chest wall injuries. From the data, 41.9% (131) of patients were over the age of 65 years. A significant proportion of our patients 34.5% (108) were noted to have multiple rib fractures (>3 Ribs). It was noted that 3% (9) of the 313 patients required operations. The average of the patients included in the study was noted to be 61 years with an age range of 17-92 years.Conclusion The mainstay management of rib fracture is provision of adequate analgesia and the prevention of respiratory complications that can all stem from poor ventilatory function amongst other patient factors and injury patterns. Through the decades, surgical stabilisation has gained pace and has found its niche in the management of rib fractures.

Download Full-text

The Effectiveness of Image Augmentation in Deep Learning Networks for Detecting COVID-19: A Geometric Transformation Perspective

Frontiers in Medicine ◽

10.3389/fmed.2021.629134 ◽

2021 ◽

Vol 8 ◽

Author(s):

Mohamed Elgendi ◽

Muhammad Umer Nasir ◽

Qunfeng Tang ◽

David Smith ◽

John-Paul Grenier ◽

...

Keyword(s):

Deep Learning ◽

Data Augmentation ◽

Learning Algorithms ◽

Training Dataset ◽

Geometric Transformation ◽

Detection Accuracy ◽

X Rays ◽

X Ray ◽

Testing Dataset ◽

The Impact

Chest X-ray imaging technology used for the early detection and screening of COVID-19 pneumonia is both accessible worldwide and affordable compared to other non-invasive technologies. Additionally, deep learning methods have recently shown remarkable results in detecting COVID-19 on chest X-rays, making it a promising screening technology for COVID-19. Deep learning relies on a large amount of data to avoid overfitting. While overfitting can result in perfect modeling on the original training dataset, on a new testing dataset it can fail to achieve high accuracy. In the image processing field, an image augmentation step (i.e., adding more training data) is often used to reduce overfitting on the training dataset, and improve prediction accuracy on the testing dataset. In this paper, we examined the impact of geometric augmentations as implemented in several recent publications for detecting COVID-19. We compared the performance of 17 deep learning algorithms with and without different geometric augmentations. We empirically examined the influence of augmentation with respect to detection accuracy, dataset diversity, augmentation methodology, and network size. Contrary to expectation, our results show that the removal of recently used geometrical augmentation steps actually improved the Matthews correlation coefficient (MCC) of 17 models. The MCC without augmentation (MCC = 0.51) outperformed four recent geometrical augmentations (MCC = 0.47 for Data Augmentation 1, MCC = 0.44 for Data Augmentation 2, MCC = 0.48 for Data Augmentation 3, and MCC = 0.49 for Data Augmentation 4). When we retrained a recently published deep learning without augmentation on the same dataset, the detection accuracy significantly increased, with a χMcNemar′s statistic2=163.2 and a p-value of 2.23 × 10−37. This is an interesting finding that may improve current deep learning algorithms using geometrical augmentations for detecting COVID-19. We also provide clinical perspectives on geometric augmentation to consider regarding the development of a robust COVID-19 X-ray-based detector.

Download Full-text

Optimal Accuracy Zone Identification in Object Detection Technique - A Learning Rate Methodology

International Journal of Engineering and Advanced Technology - Regular Issue ◽

10.35940/ijeat.a2258.109119 ◽

2019 ◽

Vol 9 (1) ◽

pp. 6470-6476

Keyword(s):

Deep Learning ◽

Object Detection ◽

Learning Rate ◽

Training Data ◽

Fine Tuning ◽

Batch Size ◽

Detection Accuracy ◽

Learning Models ◽

The Impact ◽

Selection Of

In the recent past, Deep Learning models [1] are predominantly being used in Object Detection algorithms due to their accurate Image Recognition capability. These models extract features from the input images and videos [2] for identification of objects present in them. Various applications of these models include Image Processing, Video analysis, Speech Recognition, Biomedical Image Analysis, Biometric Recognition, Iris Recognition, National Security applications, Cyber Security, Natural Language Processing [3], Weather Forecasting applications, Renewable Energy Generation Scheduling etc. These models utilize the concept of Convolutional Neural Network (CNN) [3], which constitutes several layers of artificial neurons. The accuracy of Deep Learning models [1] depends on various parameters such as ‘Learning-rate’, ‘Training batch size’, ‘Validation batch size’, ‘Activation Function’, ‘Drop-out rate’ etc. These parameters are known as Hyper-Parameters. Object detection accuracy depends on selection of Hyperparameters and these in-turn decides the optimum accuracy. Hence, finding the best values for these parameters is a challenging task. Fine-Tuning is a process used for selection of a suitable Hyper-Parameter value for improvement of object detection accuracy. Selection of an inappropriate Hyper-Parameter value, leads to Over-Fitting or Under-Fitting of data. Over-Fitting is a case, when training data is larger than the required, which results in learning noise and inaccurate object detection. Under-fitting is a case, when the model is unable to capture the trend of the data and which leads to more erroneous results in testing or training data. In this paper, a balance between Over-fitting and Under-fitting is achieved by varying the ‘Learning rate’ of various Deep Learning models. Four Deep Learning Models such as VGG16, VGG19, InceptionV3 and Xception are considered in this paper for analysis purpose. The best zone of Learning-rate for each model, in respect of maximum Object Detection accuracy, is analyzed. In this paper a dataset of 70 object classes is taken and the prediction accuracy is analyzed by changing the ‘Learning-rate’ and keeping the rest of the Hyper-Parameters constant. This paper mainly concentrates on the impact of ‘Learning-rate’ on accuracy and identifies an optimum accuracy zone in Object Detection

Download Full-text

Management of multiple rib fractures-results from a Major Trauma Centre with review of the existing literature

10.21203/rs.3.rs-118238/v1 ◽

2020 ◽

Author(s):

Debkumar Chowdhury ◽

P. Okoh ◽

H. Dambappa

Keyword(s):

Major Trauma ◽

Significant Proportion ◽

Fragility Fractures ◽

Trauma Centre ◽

Rib Fractures ◽

Rib Fracture ◽

Analgesic Requirement ◽

Multiple Rib Fractures ◽

Major Trauma Centre ◽

The Impact

Download Full-text

Multiple rib fractures: A novel and prognostic CT-based classification system

Trauma ◽

10.1177/1460408619895683 ◽

2020 ◽

Vol 22 (4) ◽

pp. 265-272

Author(s):

Rosalind B Simpson ◽

Jessica R Dorman ◽

William J Hunt ◽

John G Edwards

Keyword(s):

Clinical Outcomes ◽

Classification System ◽

Wide Spectrum ◽

Flail Chest ◽

Derived Categories ◽

Rib Fractures ◽

Rib Fracture ◽

Observer Agreement ◽

Multiple Rib Fractures ◽

The Impact

Background The accepted classification for multiple rib fractures is binary: flail chest or not. There is a wide spectrum of morphology with subsequent variation in the impact on chest wall mechanics and clinical outcomes. As the practice of surgical stabilisation of rib fractures evolves, there is a need for a better taxonomy. The aim of this study was to create a data-driven radiological classification system for multiple rib fractures, prognostic of both complications and surgical stabilisation of rib fracture. Methods The radiological pattern of injury was assessed for cases undergoing surgical stabilisation of rib fracture (n = 48) over a five-year period and a consecutive sample of non-operative controls (n = 48). Every rib fracture (n = 1032) was assessed on CT scans for location, displacement and comminution. An iterative classification system was developed and tested for inter-observer agreement and outcome prediction. Results The fractures occurred in a ‘series’ (≥3 consecutive ribs at a similar location) in 72% of cases: these were more likely to be displaced (p < 0.001). Variables included in the classification were the anatomical pattern (presence, length and overlap of series) and degree of displacement. The classification was prognostic for complications (p < 0.001), discriminated for fixation (C = 0.907) and had acceptable inter-observer agreement (k = 0.50). Conclusions The Sheffield Multiple Rib Fracture Classification derived categories of short/long series, and short/long flail chest, with sub-division according to the presence of displacement. It was prognostic for clinical outcomes and of surgical fixation. It may facilitate communication, comparison of outcomes and selection for management protocols.

Download Full-text

The impact of deep learning–based equipment usage detection on building energy demand estimation

Building Services Engineering Research and Technology ◽

10.1177/01436244211034737 ◽

2021 ◽

pp. 014362442110347

Author(s):

Shuangyu Wei ◽

Paige Wenbin Tien ◽

Yupeng Wu ◽

John Kaiser Calautit

Keyword(s):

Deep Learning ◽

Energy Consumption ◽

Real Time ◽

Energy Demand ◽

Internal Heat ◽

Building Energy ◽

Detection Accuracy ◽

Equipment Usage ◽

The Impact ◽

Detection And Recognition

As external temperatures and internal gains from equipment rise, office buildings’ cooling demand and issues are likely to increase. Solutions such as demand-driven controls can help minimise energy consumption and maintain thermal comfort in buildings by coordinating the real-time heating, ventilation and air-conditioning (HVAC) use to the requirements of the conditioned spaces. The present study introduces a real-time equipment usage detection and recognition approach for demand-driven controls using a deep learning method. A Faster R-CNN model was trained and deployed to a camera. The performance of this model was assessed through different evaluation metrics. Based on the initial field experiment results, a detection accuracy of 76.21% was achieved. To investigate the impact of the proposed approach on building heating and cooling energy demand, the case study building was modelled and simulated. The results showed that the deep learning–based method predicted up to 35.95% lower internal heat gains compared to static or ‘fixed’ schedules based on the set conditions. Practical Application: As the appliances and equipment in building spaces contribute to the internal heat gains, their usage can influence the building energy demand and indoor thermal environment. Linking equipment usage with occupants’ presence in space may not be fully accurate and may lead to the over- or under-estimation of heat emissions, especially when the space is unoccupied, and the equipment is powered ON or the opposite. This approach can be integrated with demand-driven controls for HVAC systems, which can minimise unnecessary building energy consumption while maintaining a comfortable indoor environment using computer vision and deep learning detection and recognition methods.

Download Full-text

Deep Learning-based Activity Detection for Grant-free Random Access

10.36227/techrxiv.16834552.v1 ◽

2021 ◽

Author(s):

Taufik Abrao ◽

João Henrique Inácio de Souza

Keyword(s):

Deep Learning ◽

Random Access ◽

Network Models ◽

Efficient Solutions ◽

Supervised Machine Learning ◽

Data Detection ◽

Detection Accuracy ◽

Activity Detection ◽

Neural Network Models ◽

The Impact

<div> <div> <div> <p>The cellular internet-of-things wireless network is a promising solution to provide massive connectivity for machine- type devices. However, designing grant-free random access (GF- RA) protocols to manage such connections is challenging, since they must operate in interference-aware scenarios with sporadic device activation patterns and a shortage of mutually orthogonal resources. Supervised machine learning models have provided efficient solutions for activity detection, non-coherent data detection, and non-orthogonal preamble design in scenarios with massive connectivity. Considering these promising results, in this paper, we develop two deep learning (DL) sparse support recovery algorithms to detect active devices in mMTC random access. The DL algorithms, developed to deploy GF-RA protocols, are based on the deep multilayer perceptron and the convolutional neural network models. Unlike previous works, we investigate the impact of the type of sequences for preamble design on the activity detection accuracy. Our results reveal that preambles based on the Zadoff-Chu sequences, which present good correlation properties, achieve better activity detection accuracy with the proposed algorithms than random sequences. Besides, we demonstrate that our DL algorithms achieve activity detection accuracy comparable to state-of-the-art techniques with extremely low computational complexity. </p> </div> </div> </div>

Download Full-text

The value of deep learning-based computer aided diagnostic system in improving diagnostic performance of rib fractures in acute blunt trauma

10.21203/rs.3.rs-413049/v1 ◽

2021 ◽

Author(s):

Hui Tan ◽

Hui Xu ◽

Nan Yu ◽

Yong Yu ◽

Haifeng Duan ◽

...

Keyword(s):

Positive Predictive Value ◽

Predictive Value ◽

Diagnostic Performance ◽

Reading Time ◽

Chest Trauma ◽

Diagnostic System ◽

Rib Fracture ◽

Diagnostic Sensitivity ◽

Reference Standard ◽

Diagnostic Confidence

Abstract Purpose To evaluate the value of a deep learning-based computer-aided diagnostic system (DL-CAD) in improving the diagnostic performance of acute rib fractures in patients with chest trauma. Methods CT images of 214 patients with acute blunt chest trauma were retrospectively analyzed by two interns and two attending radiologists independently firstly and then with the assistance of a DL-CAD one month later, in a blinded and randomized manner. The consensus diagnosis of fib fracture by another two senior thoracic radiologists was regarded as reference standard. The rib fracture diagnostic sensitivity, specificity, positive predictive value, diagnostic confidence and mean reading time with and without DL-CAD were calculated and compared. Results There were 680 rib fracture lesions confirmed as reference standard among all patients. The diagnostic sensitivity and positive predictive value of interns were significantly improved from (68.82%, 84.50%) to (91.76%, 93.17%) with the assistance of DL-CAD, respectively. Diagnostic sensitivity and positive predictive value of interns assisted by DL-CAD were comparative to those of attendings aided by DL-CAD (94.56%, 86.47%) or not aided (95.67%, 93.83%), respectively. In addition, when radiologists were assisted by DL-CAD, the mean reading time was significantly reduced and diagnostic confidence was significantly enhanced. Conclusions DL-CAD improves the diagnostic performance of acute rib fracture in chest trauma patients, which increases the diagnostic confidence, sensitivity and positive predictive value for radiologists. DC-CAD can advance the diagnostic consistency of radiologists with different experiences.

Download Full-text