Detection Of Fractures Research Articles

Toward the Application of Dual-Energy Computed Tomography with Virtual Non-Hydroxyapatite Color-Coded Maps to Identify Traumatic Fractures in Daily Emergency Settings

To evaluate the advantages of dual-energy computed tomography (DECT) virtual non-hydroxyapatite color mapping (VNHAP) in combination with standard bone CT (BCT) in the identification of subtle or occult traumatic fractures referred to emergency and acceptance departments (DEAs). Forty patients (22 men; mean age 83 ± 23.7 y) with suspected traumatic fractures referred to our emergency department and examined with a fast kilovoltage-switching single-source spectral CT scan between January and October 2023 were retrospectively reviewed. The BCT and VNHAP images were blindly evaluated by two radiologists with >10 years and <2 years of experience in musculoskeletal imaging. Both techniques were evaluated in terms of sensitivity (SE), specificity (SP), positive and negative predictive values (PPVs and NPVs) and accuracy for fracture detection, as confirmed at a 3-month clinical–instrumental follow-up. Inter-observer agreement and examination times were also analyzed. Fractures were confirmed in 18/40 cases. The highest values of diagnostic performance for VNHAP images were obtained in terms of SP (90.9% and 95%) and PPV (87.5% and 92.8%) and for the less experienced operator. No statistically significant differences were observed between the diagnostic accuracy of the two readers in the evaluation of VNHAP images. Inter-observer agreement was moderate (κ = 0.536) for BCT and substantial (κ = 0.680) for VNHAP. Comparing the two operators, a significantly longer examination time for BCT and no significant difference for VNHAP were registered. Our preliminary experience may encourage the employment of VNHAP maps in combination with BCT images in emergency settings. Their use could be time-saving and valuable in terms of diagnostic performance, especially for less experienced operators.

Deep Learning Approach for Arm Fracture Detection Based on an Improved YOLOv8 Algorithm

Artificial intelligence (AI)-assisted computer vision is an evolving field in medical imaging. However, accuracy and precision suffer when using the existing AI models for small, easy-to-miss objects such as bone fractures, which affects the models’ applicability and effectiveness in a clinical setting. The proposed integration of the Hybrid-Attention (HA) mechanism into the YOLOv8 architecture offers a robust solution to improve accuracy, reliability, and speed in medical imaging applications. Experimental results demonstrate that our HA-modified YOLOv8 models achieve a 20% higher Mean Average Precision (mAP 50) and improved processing speed in arm fracture detection.

Comparative Analysis of Neural Network Architectures for Automated Fracture Detection in Hand X-ray Images

The application of several neural network architectures—including Fully Connected Networks (FCN), Convolutional Neural Networks (CNN), pretrained ResNet, Vision Transformer (ViT-B-16)—for the classification of hand X-ray images into "Fractured" and "Not Fractured"—categories is investigated in this work. The main goals are to evaluate these models' fracture detection ability and determine which architectural design fits this work. Because transfer learning let the model use past information from big-scale picture datasets, the pretrained ResNet model emerged as the most effective with high accuracy, stability, and resilience. The bespoke CNN also performed well, displaying excellent feature extraction powers especially for medical imaging. But the non-pretrained ResNet model overfitted, meaning deeper networks find it difficult to generalize without pretraining. Though innovative, the Vision Transformer performed poorly since it depends on a lot of training data and finds difficult learning of intricate spatial properties from little datasets. Although acting as a baseline, the FCN's simple architecture and incapacity to detect spatial hierarchies in images meant it could not match the efficacy of CNN models. Emphasizing the important function of transfer learning in clinical applications, the results show that pretrained CNN architectures, especially ResNet, offer the most consistent and accurate method for automatic fracture diagnosis in medical pictures.

Artificial intelligence in traumatology.

The aim of this study was to create artificial intelligence (AI) software with the purpose of providing a second opinion to physicians to support distal radius fracture (DRF) detection, and to compare the accuracy of fracture detection of physicians with and without software support. The dataset consisted of 26,121 anonymized anterior-posterior (AP) and lateral standard view radiographs of the wrist, with and without DRF. The convolutional neural network (CNN) model was trained to detect the presence of a DRF by comparing the radiographs containing a fracture to the inconspicuous ones. A total of 11 physicians (six surgeons in training and five hand surgeons) assessed 200 pairs of randomly selected digital radiographs of the wrist (AP and lateral) for the presence of a DRF. The same images were first evaluated without, and then with, the support of the CNN model, and the diagnostic accuracy of the two methods was compared. At the time of the study, the CNN model showed an area under the receiver operating curve of 0.97. AI assistance improved the physician's sensitivity (correct fracture detection) from 80% to 87%, and the specificity (correct fracture exclusion) from 91% to 95%. The overall error rate (combined false positive and false negative) was reduced from 14% without AI to 9% with AI. The use of a CNN model as a second opinion can improve the diagnostic accuracy of DRF detection in the study setting.

Modal and harmonic analysis of femur bone using different boundary conditions by Finite Element Analysis.

The longest and largest bone in the human body is the femur. Pelvic bone sustains the weight of the body to which the femur bone is connected. Many researches have been carried out to identify the behaviour of the femur bone. The study aimed to explore the natural frequencies and mode shapes of vibrating devices to gain insights into the dynamic behavior of the femur bone during various physical activities. Examining the impact of patient-specific bone shape and density on bone natural frequencies is crucial. The primary goals of femur bone analysis involve employing computer simulations for fracture detection and employing Finite Element (FE) models to determine natural frequencies and vibration modes. To obtain the natural frequency of the femur bone, different boundary conditions such as free-fixed and fixed-fixed are applied. Avoiding the coincidence of the natural frequency with external excitation frequencies is crucial to prevent femur bone fractures. Also, for different magnitude of loads, femur bone is involved in harmonic analysis is to identify the amplitude and stress against acceleration.

Artificial intelligence in commercial fracture detection products: a systematic review and meta-analysis of diagnostic test accuracy

Conventional radiography (CR) is primarily utilized for fracture diagnosis. Artificial intelligence (AI) for CR is a rapidly growing field aimed at enhancing efficiency and increasing diagnostic accuracy. However, the diagnostic performance of commercially available AI fracture detection solutions (CAAI-FDS) for CR in various anatomical regions, their synergy with human assessment, as well as the influence of industry funding on reported accuracy are unknown. Peer-reviewed diagnostic test accuracy (DTA) studies were identified through a systematic review on Pubmed and Embase. Diagnostic performance measures were extracted especially for different subgroups such as product, type of rater (stand-alone AI, human unaided, human aided), funding, and anatomical region. Pooled measures were obtained with a bivariate random effects model. The impact of rater was evaluated with comparative meta-analysis. Seventeen DTA studies of seven CAAI-FDS analyzing 38,978 x-rays with 8,150 fractures were included. Stand-alone AI studies (n = 15) evaluated five CAAI-FDS; four with good sensitivities (> 90%) and moderate specificities (80–90%) and one with very poor sensitivity (< 60%) and excellent specificity (> 95%). Pooled sensitivities were good to excellent, and specificities were moderate to good in all anatomical regions (n = 7) apart from ribs (n = 4; poor sensitivity / moderate specificity) and spine (n = 4; excellent sensitivity / poor specificity). Funded studies (n = 4) had higher sensitivity (+ 5%) and lower specificity (-4%) than non-funded studies (n = 11). Sensitivity did not differ significantly between stand-alone AI and human AI aided ratings (p = 0.316) but specificity was significantly higher the latter group (p < 0.001). Sensitivity was significant lower in human unaided compared to human AI aided respectively stand-alone AI ratings (both p ≤ 0.001); specificity was higher in human unaided ratings compared to stand-alone AI (p < 0.001) and showed no significant differences AI aided ratings (p = 0.316). The study demonstrates good diagnostic accuracy across most CAAI-FDS and anatomical regions, with the highest performance achieved when used in conjunction with human assessment. Diagnostic accuracy appears lower for spine and rib fractures. The impact of industry funding on reported performance is small.

External validation of an artificial intelligence multi-label deep learning model capable of ankle fracture classification

BackgroundAdvances in medical imaging have made it possible to classify ankle fractures using Artificial Intelligence (AI). Recent studies have demonstrated good internal validity for machine learning algorithms using the AO/OTA 2018 classification. This study aimed to externally validate one such model for ankle fracture classification and ways to improve external validity.MethodsIn this retrospective observation study, we trained a deep-learning neural network (7,500 ankle studies) to classify traumatic malleolar fractures according to the AO/OTA classification. Our internal validation dataset (IVD) contained 409 studies collected from Danderyd Hospital in Stockholm, Sweden, between 2002 and 2016. The external validation dataset (EVD) contained 399 studies collected from Flinders Medical Centre, Adelaide, Australia, between 2016 and 2020. Our primary outcome measures were the area under the receiver operating characteristic (AUC) and the area under the precision-recall curve (AUPR) for fracture classification of AO/OTA malleolar (44) fractures. Secondary outcomes were performance on other fractures visible on ankle radiographs and inter-observer reliability of reviewers.ResultsCompared to the weighted mean AUC (wAUC) 0.86 (95%CI 0.82–0.89) for fracture detection in the EVD, the network attained wAUC 0.95 (95%CI 0.94–0.97) for the IVD. The area under the precision-recall curve (AUPR) was 0.93 vs. 0.96. The wAUC for individual outcomes (type 44A-C, group 44A1-C3, and subgroup 44A1.1-C3.3) was 0.82 for the EVD and 0.93 for the IVD. The weighted mean AUPR (wAUPR) was 0.59 vs 0.63. Throughout, the performance was superior to that of a random classifier for the EVD.ConclusionAlthough the two datasets had considerable differences, the model transferred well to the EVD and the alternative clinical scenario it represents. The direct clinical implications of this study are that algorithms developed elsewhere need local validation and that discrepancies can be rectified using targeted training. In a wider sense, we believe this opens up possibilities for building advanced treatment recommendations based on exact fracture types that are more objective than current clinical decisions, often influenced by who is present during rounds.

Vertical root fracture detection with cone-beam computed tomography in Biodentine™ filled teeth

PurposeThis study aimed to evaluate the accuracy of detecting vertical root fractures in Biodentine™-filled teeth using the Promax 3Dmax cone-beam computed tomography (CBCT) unit compared to periapical radiographs. It tested hypotheses regarding CBCT’s diagnostic superiority in non-root-filled and Biodentine™-root-filled maxillary central incisors and assessed the impact of smaller field of view and lower intensity settings on detection accuracy.Materials and methodsExtracted maxillary incisors were divided into groups based on fracture status and root filling material, then placed in a Thiel-embalmed skull to simulate clinical conditions. The teeth were imaged using periapical radiographs and the CBCT unit under different settings. Fracture thickness was measured with microcomputed tomography for accuracy benchmarking. Multiple observers assessed the images, and statistical analyses were conducted to evaluate diagnostic performance.ResultsIntra-rater reliabilities of consensus scores ranged from good to very good. Specificities were generally higher than sensitivities across all imaging modalities, but sensitivities remained constantly low. None of the Area Under the Curve scores exceeded 0.6, indicating poor overall accuracy for all imaging modalities. Paired comparisons of the area differences under Receiver Operator Characteristic curves revealed no significant differences between the CBCT and periapical radiograph techniques for detecting vertical root fractures in either Biodentine™-filled or non-root-filled teeth.ConclusionsThere was no significant accuracy improvement of the current CBCT device (Promax 3Dmax, Planmeca, Finland) over periapical radiographs in detecting small vertical root fractures in both non-root-filled and Biodentine™-root-filled maxillary central incisors. A smaller field of view with lower intensity did not enhance detection accuracy. These results highlight the challenges in accurately detecting small VRFs, emphasizing the need for further research and technological advancements in this domain.

Artificial intelligent-driven decision-making for automating root fracture detection in periapical radiographs

BackgroundThe detection and early diagnosis of root fractures can be challenging; this difficulty applies particularly to newly qualified dentists. Aside from clinical examination, diagnosis often requires radiographic assessment. Nonetheless, human fallibility can introduce errors due to a lack of experience.AimThe proposed system aimed to assist in detecting root fractures through the integration of artificial intelligence techniques into the diagnosis process as a step for automating dental diagnosis and decision-making processes.Materials and methodA total of 400 radiographic images of fractured and unfractured teeth were obtained for the present research. Data handling techniques were implemented to balance the distribution of the samples. The AI-based system used the voting technique for five different pretrained models namely, VGG16, VGG19, ResNet50. DenseNet121, and DenseNet169 to perform the analysis. The parameters used for the analysis of the models are loss and accuracy curves.ResultsVGG16 exhibited notable success with low training and validation losses (0.09% and 0.18%, respectively), high specificity, sensitivity, and positive predictive value (PPV). VGG19 showed potential overfitting concerns, while ResNet50 displayed progress in minimizing loss but exhibited bias toward unfractured cases. DenseNet121 effectively addressed overfitting and noise issues, achieving balanced metrics and impressive PPVs for both fractured and unfractured cases (0.933 and 0.898 respectively). With increased depth, DenseNet169 demonstrated enhanced generalization capability.ConclusionThe proposed AI- based system demonstrated high precision and sensitivity for detecting root fractures in endodontically treated teeth by utilizing the voting method.

IA-HLD: an improved AlexNet for hairline fracture detection in orthopedic images

iA-HLD: an improved AlexNet for hairline fracture detection in orthopedic images

Artificial Intelligence Detection of Cervical Spine Fractures Using Convolutional Neural Network Models.

To develop and evaluate a technique using convolutional neural networks (CNNs) for the computer-assisted diagnosis of cervical spine fractures from radiographic x-ray images. By leveraging deep learning techniques, the study might potentially lead to improved patient outcomes and clinical decision-making. This study obtained 500 lateral radiographic cervical spine x-ray images from standard open-source dataset repositories to develop a classification model using CNNs. All the images contained diagnostic information, including normal cervical radiographic images (n=250) and fracture images of the cervical spine fracture (n=250). The model would classify whether the patient had a cervical spine fracture or not. Seventy percent of the images were training data sets used for model training, and 30% were for testing. Konstanz Information Miner (KNIME)'s graphic user interface-based programming enabled class label annotation, data preprocessing, CNNs model training, and performance evaluation. The performance evaluation of a model for detecting cervical spine fractures presents compelling results across various metrics. This model exhibits high sensitivity (recall) values of 0.886 for fractures and 0.957 for normal cases, indicating its proficiency in identifying true positives. Precision values of 0.954 for fractures and 0.893 for normal cases highlight the model's ability to minimize false positives. With specificity values of 0.957 for fractures and 0.886 for normal cases, the model effectively identifies true negatives. The overall accuracy of 92.14% highlights its reliability in correctly classifying cases by the area under the receiver operating characteristic curve. We successfully used deep learning models for computer-assisted diagnosis of cervical spine fractures from radiographic x-ray images. This approach can assist the radiologist in screening, detecting, and diagnosing cervical spine fractures.

Commentary on "Artificial Intelligence Detection of Cervical Spine Fractures Using Convolutional Neural Network Models".

Commentary on "Artificial Intelligence Detection of Cervical Spine Fractures Using Convolutional Neural Network Models".

Texture Aware Deep Features for Precise Wrist Fracture Detection

Texture Aware Deep Features for Precise Wrist Fracture Detection

Digital methods for diagnosing dental occlusion disorders

Relevance. Occlusal disorders are not only predictors of oral diseases but are also associated with systemic conditions such as endocrine, cardiovascular, and cognitive disorders, highlighting their broader significance in medicine. The digitization of diagnostic processes and the integration of computer technologies into traditional methods offer promising advancements in assessing occlusal status. This makes the analysis of digital methods for diagnosing occlusal relationship disorders particularly relevant.Purpose. To evaluate digital methods and devices used for diagnosing occlusal relationship disorders.Material and methods. A systematic review was conducted, analyzing 40 scientific publications from 2016 to 2024 that focus on digital methods for diagnosing occlusal imbalance. The study assessed electronic devices that measure occlusal force by converting mechanical force into electrical signals using various sensors. These digital methods are employed to evaluate the functional state of the masticatory system, including the detection of temporomandibular joint disorders, mandibular fractures, and occlusal deformities, as well as to monitor occlusal force dynamics during treatment. Recent advancements in portable electronic sensors suggest potential applications for their integration with control chips into dental prostheses or occlusal stabilization splints, allowing real-time monitoring of tooth compression forces, potentially enhanced by artificial intelligence.Results. The development of new diagnostic methods and devices equipped with highly sensitive sensors for measuring masticatory pressure opens new avenues for identifying clinical parameters that were previously difficult to analyze. Recent advances in hardware technology, particularly those utilizing neural network analysis, have the potential to expedite accurate diagnosis and predict the risk of complications.Conclusion. Both in Russia and internationally, significant progress is being made in the development of digital methods for diagnosing occlusal imbalance. These methods are poised to become a priority in the future of dental practice. However, the adoption of digital occlusal methods into routine dental practice is currently hindered by challenges related to the availability of digital equipment in clinics and the insufficient digital skills of many dentists. Addressing these challenges is crucial for improving the quality of dental care for patients with occlusal disorders.

Assessing the Performance of Models from the 2022 RSNA Cervical Spine Fracture Detection Competition at a Level I Trauma Center

Assessing the Performance of Models from the 2022 RSNA Cervical Spine Fracture Detection Competition at a Level I Trauma Center

Retraction Note: Diagnostic power of ChatGPT 4 in distal radius fracture detection through wrist radiographs.

Retraction Note: Diagnostic power of ChatGPT 4 in distal radius fracture detection through wrist radiographs.

Comparison of Spectral CT and MRI in Pelvic Ring Fragility Fractures: A Prospective Diagnostic Accuracy Study.

Background/Objectives: Fragility fractures of the pelvis (FFP) are characterized by inadequate trauma to a structurally compromised bone, primarily in osteoporosis. Conventional CT studies can be inadequate in identifying FFPs. An MRI of the pelvis is considered the gold standard in diagnosing FFPs. Spectral CT or Dual-Energy CT may have comparable diagnostic accuracy. It provides additional insights into associated bone marrow edema. The aim of this prospective monocentric study is to evaluate the diagnostic accuracy of Spectral CT compared to the gold standard MRI in diagnosing FFP. Methods: Over a 2-year period, patients presenting in the emergency department with clinical suspicion of an FFP were consecutively included. They underwent Spectral CT (GE Revolution 16 cm GSI) upon admission, followed by an MRI. The gold standard for diagnosing FFP is pelvic MRI, showing sensitivity and specificity ranging from 97% to 100%. The acquired images were evaluated and classified using the osteoporotic fractures of the pelvis (OFP) classification. Results: Compared to the reference test, which was the MRI pelvis, the sensitivity of the CT pelvis was determined to be 86.8 (95% confidence interval (CI) 71.9-95.6%) with a specificity of 84.6% (95% CI: 54.6-98.1%, p = 0.453). Spectral CT could identify an additional FFP correctly, exhibiting a sensitivity of 89.5% (95% CI: 75.2-97.1%, p = 0.688), while maintaining the same specificity as the conventional CT. The inter-rater reliability assessment for Spectral CT, conducted by four independent raters, resulted in a Fleiss' Kappa value of 0.516 (95% CI: 0.450-0.582, p < 0.001). Conclusion: The sensitivity of Spectral CT in the detection of pelvic ring fragility fractures shows a slightly lower sensitivity compared to MRI. There were no statistically significant differences observed when compared to conventional CT or MRI. In conclusion, Spectral CT may be beneficial in distinguishing FFP, particularly in cases where a definitive diagnosis is uncertain. Level of Evidence: II.

HVPS-DFN-DL: Intelligent capture and characterization of geological fracture outcrops based on a hybrid vision-photogrammetric system and discrete fracture network

The main objective of this article is to provide a framework for intelligent capture-acquisition analysis of geometric information from geological outcrops. By combining deep learning methods with photogrammetric data from unmanned aerial vehicles (UAVs), FPV drones, and terrestrial cameras acquired by a hybrid vision-photogrammetric system (HVPS), intelligent fracture detection and geometric information segmentation of multiscale field geological outcrops were achieved. The extraction results were subsequently used to generate a three-dimensional discrete fracture network (DFN) of real rock masses for studying the influence of the spatial connectivity of discontinuity structural planes on the mechanical and hydrodynamic characteristics of rock masses. By testing data collected in situ from a variety of field rock masses in several regions of China, this framework was shown to be a very efficient method for geostatistical work, exhibiting very low measurement errors. Furthermore, this framework is extremely safe for geologists and applicable to a wide range of site geological environments. It is also suitable for field geological surveys, geometry acquisition of outcropping lithologies, obtaining tunnel face and surrounding fissure statistics, and geological stability assessment of unstable rock masses. This framework can also provide a method for unmanned topographic-geological exploration. Furthermore, the fracture network realism and the data acquisition efficiency have been greatly improved, and the difficulty of developing field measurements and validating the DFN model has been overcome.

Microwave Technique for Linear Skull Fracture Detection-Simulation and Experimental Study Using Realistic Human Head Models.

Microwave (MW) sensing is regarded as a promising technique for various medical monitoring and diagnostic applications due to its numerous advantages and the potential to be developed into a portable device for use outside hospital settings. The detection of skull fractures and the monitoring of their healing process would greatly benefit from a rapidly and frequently usable application that can be employed outside the hospital. This paper presents a simulation- and experiment-based study on skull fracture detection with the MW technique using realistic models for the first time. It also presents assessments on the most promising frequency ranges for skull fracture detection within the Industrial, Scientific and Medical (ISM) and ultrawideband (UWB) ranges. Evaluations are carried out with electromagnetic simulations using different head tissue layer models corresponding to different locations in the human head, as well as an anatomically realistic human head simulation model. The measurements are conducted with a real human skull combined with tissue phantoms developed in our laboratory. The comprehensive evaluations show that fractures cause clear differences in antenna and channel parameters (S11 and S21). The difference in S11 is 0.1-20 dB and in S21 is 0.1-30 dB, depending on the fracture width and location. Skull fractures with a less than 1 mm width can be detected with microwaves at different fracture locations. The detectability is frequency dependent. Power flow representations illustrate how fractures impact on the signal propagation at different frequencies. MW-based detection of skull fractures provides the possibility to (1) detect fractures using a safe and low-cost portable device, (2) monitor the healing-process of fractures, and (3) bring essential information for emerging portable MW-based diagnostic applications that can detect, e.g., strokes.

Research on ultrasonic guided wave-based high-speed turnout switch rail base flaw detection

Turnout switch rail fracture detection is currently a serious issue in the field of railway transportation. Guided wave detection, a non-destructive testing method, is a good way of studying this issue and looking for a suitable solution. For this paper, a guided wave mode and an excitation position were selected based on the phase velocity dispersion curve and the wave structure. After this, a model was devised for the turnout switch area using the finite element (FE) method. This model considered the straight switch rail, curved stock rail, bolt hole, spacer block, and sub-rail foundation, and verifies the validity of the simulation model through the experiment. The propagation characteristics of guided waves in the switch rail were then simulated in different fracturing states by means of a 30-kHz excitation applied vertically to the rail base. The results showed that a single mode of the guided wave could be generated by this excitation method, showing that it could be used as an effective means for fracture detection. The growth rate of the root-mean-square (RMS) value of the time-domain acceleration signal could then be analysed to identify the state of fracture in the straight switch rail. This discrimination method is suitable for finding fractures parallel to the rail cross-section with a width of 5 mm or more at the bottom of the straight switch rail.