3D CT-scan Images Research Articles

Lightweight Advanced Deep Neural Network (DNN) Model for Early-Stage Lung Cancer Detection.

Background: Lung cancer, also known as lung carcinoma, has a high mortality rate; however, an early prediction helps to reduce the risk. In the current literature, various approaches have been developed for the prediction of lung carcinoma (at an early stage), but these still have various issues, such as low accuracy, high noise, low contrast, poor recognition rates, and a high false-positive rate, etc. Thus, in this research effort, we have proposed an advanced algorithm and combined two different types of deep neural networks to make it easier to spot lung melanoma in the early phases. Methods: We have used WDSI (weakly supervised dense instance-level lung segmentation) for laborious pixel-level annotations. In addition, we suggested an SS-CL (deep continuous learning-based deep neural network) that can be applied to the labeled and unlabeled data to improve efficiency. This work intends to evaluate potential lightweight, low-memory deep neural net (DNN) designs for image processing. Results: Our experimental results show that, by combining WDSI and LSO segmentation, we can achieve super-sensitive, specific, and accurate early detection of lung cancer. For experiments, we used the lung nodule (LUNA16) dataset, which consists of the patients' 3D CT scan images. We confirmed that our proposed model is lightweight because it uses less memory. We have compared them with state-of-the-art models named PSNR and SSIM. The efficiency is 32.8% and 0.97, respectively. The proposed lightweight deep neural network (DNN) model archives a high accuracy of 98.2% and also removes noise more effectively. Conclusions: Our proposed approach has a lot of potential to help medical image analysis to help improve the accuracy of test results, and it may also prove helpful in saving patients' lives.

Prediction of the as Low as Diagnostically Acceptable CT Dose for Identification of the Inferior Alveolar Canal Using 3D Convolutional Neural Networks with Multi-Balancing Strategies

Ionizing radiation is necessary for diagnostic imaging and deciding the right radiation dose is extremely critical to obtain a decent quality image. However, increasing the dosage to improve the image quality has risks due to the potential harm from ionizing radiation. Thus, finding the optimal as low as diagnostically acceptable (ALADA) dosage is an open research problem that has yet to be tackled using artificial intelligence (AI) methods. This paper proposes a new multi-balancing 3D convolutional neural network methodology to build 3D multidetector computed tomography (MDCT) datasets and develop a 3D classifier model that can work properly with 3D CT scan images and balance itself over the heavy unbalanced multi-classes. The proposed models were exhaustively investigated through eighteen empirical experiments and three re-runs for clinical expert examination. As a result, it was possible to confirm that the proposed models improved the performance by an accuracy of 5% to 10% when compared to the baseline method. Furthermore, the resulting models were found to be consistent, and thus possibly applicable to different MDCT examinations and reconstruction techniques. The outcome of this paper can help radiologists to predict the suitability of CT dosages across different CT hardware devices and reconstruction algorithms. Moreover, the developed model is suitable for clinical application where the right dose needs to be predicted from numerous MDCT examinations using a certain MDCT device and reconstruction technique.

A 3D-printed surgical guide for ischemic scar targeting and ablation.

3D printing technology development in medical fields allows to create 3D models to assist preoperative planning and support surgical procedures. Cardiac ischemic scar is clinically associated with malignant arrhythmias. Catheter ablation is aimed at eliminating the arrhythmogenic tissue until the sinus rhythm is restored. The scope of this work is to describe the workflow for a 3D surgical guide able to define the ischemic scar and target catheter ablation. For the patient-specific 3D surgical guide and 3D heart phantom model realization, both CT scan and cardiac MRI images were processed; this was necessary to extract anatomical structures and pathological information, respectively. Medical images were uploaded and processed in 3D Slicer. For the surgical guide modeling, images from CT scan and MRI were loaded in Meshmixer and merged. For the heart phantom realization, only the CT segmentation was loaded in Meshmixer. The surgical guide was printed in MED625FLX with Polyjet technology. The heart phantom was printed in polylactide with FDM technology. 3D-printed surgical model was in agreement with prespecified imputed measurements. The phantom fitting test showed high accuracy of the 3D surgical tool compared with the patient-specific reproduced heart. Anatomical references in the surgical guide ensured good stability. Ablation catheter fitting test showed high suitability of the guide for different ablation tools. A 3D-printed guide for ventricular tachycardia ablation is feasible and accurate in terms of measurements, stability, and geometrical structure. Concerning clinical use, further clinical investigations are eagerly awaited.

Siamese-Based Deep Learning for Markerless Lung Tumor Tracking During Stereotactic Radiotherapy

<h3>Purpose/Objective(s)</h3> During stereotactic lung radiotherapy we want to monitor the target position to ensure it remains inside the planning target volume (PTV). This ensures accurate treatment delivery and may allow for margin reduction. Many LINAC systems have a kilo-voltage (kV) imaging system mounted orthogonally to the MV beam. This can acquire images during volumetric modulated arc therapy delivery. The 2D-kV images can be used to identify the tumor position, e.g., using template matching and triangulation (TMT). However, small lung targets are difficult to track, due to e.g., low tumor contrast, occlusion (e.g., by spine), or noisy images. We investigated deep learning (DL) as an alternative method for lung tumor tracking. <h3>Materials/Methods</h3> Clinical data from 6 patients (P1-P6) with small lung tumors (0.5-3 cm diameter) and a 3D-printed anthropomorphic phantom (P0), including a low-density lung tumor, was used. For each patient, data included 4D-CT scans, fluoroscopic kV images and breathing-related surface motion (Real-Time Position Management, RPM). For the phantom, a 3D-CT scan and kV images taken with a static and moving couch (regular and irregular motion) were available. Due to lack of ground truth tumor locations for the kV images, a patient-specific DL model is trained on Digitally Reconstructed Radiographs (DRRs), generated from the planning CT scans. To enhance the tumor's visual diversity in DRRs, we augmented the training data with locally deformed CT scans. A Siamese-based tracking by similarity method is hypothesized to be better suited for the tumor tracking task than other DL methods, due to its more localized search space and the use of previous frames (temporal information). <h3>Results</h3> Table 1 shows the superior-inferior (SI) direction results obtained by the Siamese model and TMT, on all kV images. For the patient data, the <b>Correlation coefficient</b> between estimated tumor trajectory and RPM was computed, due to lack of ground truth locations. For the phantom, the <b>Mean Absolute Distance</b> (MAD) between detected and ground truth center coordinates was computed. For all 6 patients and the phantom, the Siamese model outperformed TMT. <h3>Conclusion</h3> A novel approach for real-time markerless lung tumor tracking is presented, based on deep Siamese networks. The model compares favorably to the template matching/triangulation-based technique and merits further investigation. The current analysis presents only SI tracking. Reliable 3D tracking, with sub-mm accuracy, is the ultimate goal. Further work is needed to achieve this.

CORR Insights®: 3D-printed Handheld Models Do Not Improve Recognition of Specific Characteristics and Patterns of Three-part and Four-part Proximal Humerus Fractures.

CORR Insights®: 3D-printed Handheld Models Do Not Improve Recognition of Specific Characteristics and Patterns of Three-part and Four-part Proximal Humerus Fractures.

Micro-Structure Analysis of Quasi-Static Crushing and Low-Velocity Impact Behavior of Graded Composite Metallic Foam Filled Tube

Foam filled tubes (FFT) are novel structures with high energy absorption, enhanced strength to weight ratio, and tailoring capability. In the present paper, we have analyzed quasi-static uniaxial compression and low-velocity impact behavior of FFT with closed-cell metallic foam cores and functionally graded densities both experimentally and numerically. Alporas foams were manufactured using liquid state method with TiH2 blowing agent. We prepared Specimens with graded composition and densities by stacking of several layers of pure aluminum and A356 alloy Alporas foams with cubic geometry. We conducted several standard uniaxial compression experiments to determine the non-linear mechanical properties and hardening. Square aluminum tubes are manipulated to enhance the performance and tailoring specification of the structure. We generated microstructural models using a hybrid 3D Voronoi diagram and CT-scan images to predict mechanical behavior numerically. Computed tomography is used to determine the inner cells morphological characterization. Also, the modified Kelvin cell with a beam element in edge regions is manipulated to enhance accuracy. Comparing the quasi-static experiment and FEA results show good accordance, and hence, we achieved the calibrated model. Finally, we used the numerical model in FFT tailoring and mechanical properties design.

Accuracy of Radiographic Evaluation of Ulnar Styloid Base Fracture by Simple Radiogram

Purpose: There remains uncertain whether to fix an ulnar styloid fracture accompanied by distal radius fracture. Fixation should be required in cases of the fracture involving a fovea of ulnar head, an attachment site of deep portion of triangular fibrocartilage. We analyzed the patients with ulnar styloid fracture accompanied by distal radius fracture using simple radiograph and 3D computed tomography (3D CT) scan images, respectively, and compared accuracy of evaluation for the types of ulnar styloid fractures, which might designate the treatment methods. Materials and Methods: We retrospectively reviewed 168 patients who underwent surgery for distal radius fracture in our hospital from January 2005 to March 2015, and evaluated whether ulnar styloid base was involved using simple radiographs and 3D CT scan images. The accuracy of evaluation by simple radiographs was compared with the findings of 3D CT scan images. To avoid the effect of memory, evaluation using 3D CT scan images was performed in a randomized order 2 weeks after initial evaluation for simple radiograph. Results: On simple radiographs, 64 patients (38%) had ulnar styloid base fractures; however, 21 cases of these revealed nonbase fractures by 3D CT scan images. And 7 out of 104 cases, on simple radiographs, determined as nonbase fracture by simple radiographs, were diagnosed as base fractures by 3D CT scans. Based on the findings of 3D CT scan images, the accuracy of simple radiographic evaluation for types of ulnar styloid fractures was 83%, with 86% sensitivity and 82% specificity. The positive predictive value was 67% and the negative predictive value was 93%. Conclusion: Accuracy of radiographic evaluation of ulnar styloid base fracture accompanied by distal radius fracture, when compared with 3D CT, was 83%. Therefore, we recommend that 3D CT scan image–based evaluation should be performed in case of unclear involvement sites so as to determine whether to fix it or not.

Measurement and simplified numerical prediction of effective thermal conductivity of open-cell ceramic foams at high temperature

In the present article, the effective thermal conductivity (ETC) of 10ppi open-cell ceramic foams is measured at temperatures ranging from 22°C to 750°C using the transient plane source (TPS) technique. The detailed morphology of the foam structures is reconstructed from 3D CT-scan images and their structural analysis is carried out using a volume image processing tool. The experimental results are compared with the predictions of directional ETC, obtained using a simplified 1D homogeneous model that considers coupled conduction–radiation heat transfer. The parameters, required by the model, are evaluated based only upon the information that could be extracted from 3D CT-scan images under certain simplifying assumptions. It can be concluded that the predictions of ETC for the present ceramic foam samples clearly correlate with the geometric parameters obtained from their structural analysis. It is also observed that the estimated microscopic porosity of the foam, which is not captured by 3D CT-scan images, has considerable influence on the reconstructed foam structure and hence on the predicted ETC. The predictions of ETC in the measured direction, obtained after the proposed directional averaging, compare extremely well with TPS measurements for the complete range of investigated temperatures. This suggests that the measurements of ETC, obtained employing TPS technique, are the directionally averaged representations of ETC in three perpendicular directions. The present investigation also clearly demonstrates the potential of the simplified modeling approach for characterizing even the directional ETC of open-cell ceramic foams, based primarily on their structural information.

Can orbital angle morphology distinguish dogs from wolves?

For more than a century, the orbital angle has been studied by many authors to distinguish dog skulls from their progenitor, the wolf. In early studies, the angle was reported to be different between dogs (49°–55°) and wolves (39°–46°). This clear difference was, however, questioned in a more recent Scandinavian study that shows some overlap. It is clear that in all studies several methodological issues were unexplored or unclear and that group sizes and the variety of breeds and wolf subspecies were small. Archaeological dog skulls had also not been studied. Our goal was to test larger and more varied groups and add archaeological samples as they are an evolutionary stage between wolves and modern dogs. We also tested the influence of measuring methods, intra- and inter-reliability, angle symmetry, the influence of variations in skull position and the possibility of measuring and comparing this angle on 3D CT scan images. Our results indicate that there is about 50 % overlap between the angle range in wolves and modern dogs. However, skulls with a very narrow orbital angle were only found in wolves and those with a very wide angle only in dogs. Archaeological dogs have a mean angle very close to the one of the wolves. Symmetry is highest in wolves and lowest in archaeological dogs. The measuring method is very reliable, for both inter- and intra-reliability (0.99–0.97), and most skull position changes have no statistical influence on the angle measured. Three-dimensional CT scan images can be used to measure OA, but the angles differ from direct measuring and cannot be used for comparison. Evolutionary changes in dog skulls responsible for the wider OA compared to wolf skulls are mainly the lateralisation of the zygomatic process of the frontal bone. Our conclusion is that the orbital angle can be used as an additional morphological measuring method to discern wolves from recent and archaeological dogs. Angles above 60° are certainly from recent dogs. Angles under 35° are certainly of wolves.

Experimental and numerical determination of effective thermal conductivity of open cell FeCrAl-alloy metal foams

Heat transfer in metal foams is one of the fields of particular research interest due to the possibility of imparting tailored effective thermo-physical properties for various applications. Their Effective Thermal Conductivity (ETC) varies in a wide range and it is strongly influenced by the morphology of metal foam. This study presents the evaluation of ETC of open cell FeCrAl-alloy metal foams with different porosities and characteristic pore dimensions. The ETC was measured with the transient plane source method at room temperature. Measurements of the ETC for different metal foams were compared with each other and with literature data in order to investigate the influence of morphology of metal foam. Furthermore, measurements were also compared with numerical predictions, obtained using both detailed approach based on 3D thermal lattice-Boltzmann method as well as a simplified modelling approach, and reasonably good agreements were observed. The real foam geometry, obtained from 3D CT-scan images, was used for detailed and simplified numerical simulations.

Experimental validation of simplified conduction–radiation models for evaluation of Effective Thermal Conductivity of open-cell metal foams at high temperatures

In the present study, the Effective Thermal Conductivity (ETC) of open-cell metal foams at high temperatures is numerically predicted using two different simplified homogeneous models, where relevant effective properties of foams are estimated considering their true geometry obtained from three-dimensional (3D) Computed Tomography (CT)-scan data. The performance of these models is judged by comparing their predictions with experimental results of the ETC measured using the panel test technique in the temperature range of around 130°C–850°C. Two samples of open-cell FeCrAl foams with similar porosity but different characteristic pore dimensions are considered for this purpose. The present results clearly show that both simplified homogeneous models offer efficient alternatives to detailed models for the estimation of ETC at both ambient and higher temperatures. Although Model-2 uses the extremely simplified Rosseland approximation for radiation and hence is expected to be less accurate than Model-1, where the radiation heat transfer is solved with the one-dimensional Radiative Transfer Equation, it may be successfully applied provided the optical thickness of the porous foam is reasonably high. This investigation also shows that effective properties of porous foams, required by these models for both conduction and radiation (i.e. the ETC due to pure heat conduction and the extinction coefficient), can be accurately estimated using simple numerical tools based only upon the information extracted from 3D CT-scan images, which is a cheap alternative as compared to using experimental techniques. Since these simplified models already require the Effective Thermal Conductivity due to pure heat conduction as input, they are not considered self-sufficient as far as the overall modeling is concerned.

SENSITIVITY ANALYSIS OF EFFECTIVE THERMAL CONDUCTIVITY OF OPEN-CELL CERAMIC FOAMS USING A SIMPLIFIED MODEL BASED ON DETAILED STRUCTURE

The Effective Thermal Conductivity (ETC) of open-cell porous foams can be predicted from the detailed numerical simulation, considering the complex foam structure obtained from three-dimensional (3D) Computed Tomography (CT)-scan images. An alternative approach could be to consider simplified models for a quick and accurate estimation of the ETC. A model for ETC of open-cell porous foams, using such a simplified approach, has been proposed recently which relies upon a single numerical prediction of the dimensionless ETC under vacuum condition, evaluated using the detailed foam structure obtained from 3D CTscan information. This model is applied in the present study in order to analyze the influence of different parameters, namely the microscopic porosity within the bulk solid material and the direction of heat transfer, on the ETC of open-cell ceramic foams. The present investigation demonstrates that the considered simplified modeling approach offers reasonable accuracy with reduced computational effort for the sensitivity analysis of ETC to different parameters. INTRODUCTION Porous foams are commonly used for a wide range of applications in numerous technological fields due to their special structural and thermo-physical properties. In this respect, the evaluation of their Effective Thermal Conductivity (ETC) has become a major requirement for the accurate prediction of heat transfer behavior of many systems, such as heat exchangers, thermal insulators, combustion systems, evaporators, etc., using homogenization approach [1-3]. The complex morphology of foam structures plays an important role in the overall heat transfer characteristics and hence the ETC can vary substantially among different foams of same material and porosity. Moreover, owing to the very high porosity of foams, total heat transfer can take place by both conduction as well as thermal radiation, particularly when the operating temperature as well as the applied temperature difference are relatively high [4]. The present investigation, however, is restricted only to the conduction heat transfer at room temperature, where the contribution of thermal radiation is seemingly small. The ETC of porous foams can be numerically predicted from simulations of conduction heat transfer through them due to imposed temperature gradient by considering their detailed morphology. The structural information of foams, required by these simulations, nowadays can be obtained from the high resolution 3D CT-scan images. Such detailed approach is generally quite accurate, as well as time consuming. Therefore, in order to achieve a reasonable compromise between the accuracy and the computational effort, an alternative approach is to use simplified models for quick and accurate evaluation of the ETC, ideally presenting a wide range of applicability. A review on models for the ETC, presenting different levels of complexity, has been presented by Coquad et al. [5]. Recently, a model for evaluating the ETC of opencell porous foams, using a simplified approach with one adjustable parameter, has been proposed by Mendes et al. [6]. This, parameter appearing in an explicit expression for the ETC, can be uniquely determined from a single numerical prediction of the dimensionless ETC under vacuum condition (made dimensionless with respect to the thermal conductivity of solid phase), based on the detailed geometry of foam. Quite obviously, evaluation of the dimensionless ETC under vacuum condition requires significantly reduced computational time, which scales at least to the order of porosity of foams and owing to the definition, the result is independent of the thermal conductivity of solid phase. Most importantly, as has been demonstrated by Mendes et al. [6], the dimensionless ETC under vacuum condition implicitly contains the relevant quantitative morphological

Unilateral duplication of the acromioclavicular joint: case report and literature review

Clavicle duplication is a rare anatomical variation of the scapular belt: only seven cases have been reported in the literature to date, and only one took note of the existence of a duplication of the acromioclavicular joint. Two hypotheses have been proposed to interpret this variation: genetic factors, or trauma occurred in the growth period. Clavicle duplication should not be mistaken for a quite frequent coracoclavicular joint widely described. The authors report the case of a left acromioclavicular joint duplication in a 51-year-old male patient presenting with left shoulder pain. This case was the first of literature providing 3D CT-scan images.

Percutaneous cerclage wiring, does it disrupt femoral blood supply? A cadaveric injection study

BackgroundA percutaneous cerclage wiring technique has been developed to reduce iatrogenic soft tissue and vascular disruption associated with classic cerclage fixation. ObjectivesThe purpose of this study was to evaluate the extent of femoral vascular disruption resulting iatrogenically from the application of two percutaneous cerclage wire loops. MethodsPairs of cerlage wire loops were percutaneously inserted on 18 fresh cadaveric femurs. The position of the wire loops varied. The wire loops were either inserted 10 and 15cm, 10 and 20cm, or 15 and 20cm distal to the tip of the greater trochanter. Each study group had 6 cadavers. Contralateral femurs without cerclage wiring were used as controls. Liquid contrast–gelatin was injected into the common femoral artery. Using axial and 3D CT scan images the superficial femoral artery (SFA), deep femoral artery (DFA), perforating arteries and their anastomotic patterns as well as endosteal perfusion were identified and their patency was graded. ResultsPercutaneous cerclage wiring did not disrupt femoral endosteal blood supply and maintained the integrity of all of the superficial femoral arteries. Four specimens demonstrated maintenance of all 4 perforators, 11 showed disruption of 1 perforator, and 3 showed disruption of 2 perforators. One deep femoral artery was disrupted after its first perforator branched off; however, perfusion was maintained by fill from an alternative anastamosis. There was no significant difference between disruption of deep femoral arteries and perforating arteries (P=1.000), location of wiring (P=0.905) or spacing between wire loops (P=1.000). ConclusionPercutaneous cerclage wiring resulted in minimal disruption of the femoral blood supply. When partial disruption occurred the SFA, DFA, and their associated perforators compensated to maintain femoral perfusion through their anastomoses. The location of the cerclage wire and the distance between the wire loops in the proximal femur showed no significant difference in the rate of iatrogenic perforator injury.

Intensity-based registration of freehand 3D ultrasound and CT-scan images of the kidney

Objectives This paper presents a method to register a pre-operative computed-tomography (CT) volume to a sparse set of intra-operative ultra-sound (US) slices. In the context of percutaneous renal puncture, the aim is to transfer planning information to an intra-operative coordinate system. Materials and methods The spatial position of the US slices is measured by optically localizing a calibrated probe. Assuming the reproducibility of kidney motion during breathing, and no deformation of the organ, the method consists in optimizing a rigid 6 degree of freedom transform by evaluating at each step the similarity between the set of US images and the CT volume. The correlation between CT and US images being naturally rather poor, the images were preprocessed in order to increase their similarity. Among the similarity measures formerly studied in the context of medical image registration, correlation ratio turned out to be one of the most accurate and appropriate, particularly with the chosen non-derivative minimization scheme, namely Powell-Brent’s. The resulting matching transforms are compared to a standard rigid surface registration involving segmentation, regarding both accuracy and repeatability. Results The obtained results are presented and discussed.

Using mathematical morphology for the anatomical labeling of vertebrae from 3D CT-scan images

In this article we propose an original method for the anatomical labeling of vertebrae from 3D CT-scan images. The primary purpose of this work is to obtain a robust referential of the abdomen. This referential can be used to locate anatomical structures like organs or blood vessels. The main problematic concerns the separation of the vertebrae, which are structures that are very close from each other. In order to detect the intervertebral spaces, we use a morphological operator which detects the dark spaces corresponding to intervertebral discs in combination with an analysis of the shape of the vertebrae in the axial plane. To reconstruct the vertebrae we use the paradigm of mathematical morphology, which consists in finding markers inside the vertebrae and compute the watershed from markers. Then we label the vertebrae according to their anatomical names. To do this, we automatically detect T12 vertebrae. We have evaluated our algorithm on 26 images.