Digital Radiographic Images Research Articles

Digital radiography image quality evaluation using various phantoms and software.

To investigate the effect of the exposure parameters on image quality (IQ) metrics of phantom images, obtained automatically using software or from visual evaluation. Three commercial phantoms and a homemade phantom constructed according to the instructions given in the IAEA Human Health Series No. 39 publication were used, along with the respective software that estimate automatically various IQ metrics. Images with various exposure parameters were acquired in a digital radiography (DR) unit. For the commercial phantoms, visual evaluations were also performed. The IQ scores obtained were analyzed to investigate the effects of increasing incident air kerma (IAK), tube potential (kVp), additional filtration, and acquisition protocol on IQ. The effects of the exposure parameters on the IQ metrics, determined with the commercial and the IAEA phantoms, were not the same. For example, clear trends of improvement of IQ scores with increased IAK and reduction of most IQ scores with increased kVp were observed mostly with the IAEA phantom, but not with the commercial phantoms (for both automatic and visual scoring methods). For all phantoms, the maximum variations in IQ scores observed for repeated identical exposures were almost always below 10% with automatic evaluation whereas, for visual evaluation, reached 17%. Failure to detect some expected trends with the complex commercial phantoms may be attributed to the fact that IQ in DR is more strongly affected by the post-processing procedures, which may mask the effect of other parameters on IQ, something that was not observed with the simple IAEA phantom.

PO21: Commissioning of an HDR Universal Endometrial Applicator Set

<h3>Purpose</h3> To commission for HDR brachytherapy a Varian CT/MR compatible universal endometrial applicator set used to treat inoperable endometrial cancer (i.e., vagina, cervix, paracervical area, endometrium, uterus, and ovaries), and to develop all necessary documents and procedures for clinical implementation. The kit contains a straight central tandem, Left/Right tandems (20 deg. angle, 60, 70, and 80 mm length), clamping unit, and a vaginal cylinder with lever. <h3>Materials and Methods</h3> The tests performed for commissioning and clinical implementation are described in the following. <i>Functionality tests</i>: All kit components were checked for structural integrity. The applicator was assembled using the information received from the manufacturer. A GYN source guide tube (SGT) was connected to each tandem and the total length was checked using the gauge wire. GYN dummy wires were placed inside tandems and digital radiographic images (kVp 85, mA 80, ms 200, small focal spot, no buildup) were acquired to check the pathway and assess for any obstruction. <i>Autoradiographs</i>: A piece of XR-SP2 GafChromic film was placed on the conventional simulator couch, then the seven tandems were placed over the film, with a 2-cm solid water slab used as buildup. Tandems were connected via GYN SGTs to the HDR unit and a test plan was delivered with 9 dwell positions/channel, 0.8 sec/position, starting at 129.5 cm distally, with a 1-cm step size. GYN dummy wires were then placed inside tandems and the film was exposed twice with 135 kV, 80 mA, 400 mAs. Alternation of source positions and dummy dots displayed on the film was assessed. <i>Clinical implementation tests</i>: The applicator was assembled and placed in a small water-filled container and CT-scanned with GYN dummy wires inserted in each tandem. Sagittal/Coronal T2 and Axial MDIXON MR images were acquired for the applicator (without dummy wires) using a 1.5 T scanner. The CT and MR images were imported to Eclipse and rigidly registered, matching the tandems and cylinder. The device can be clinically used with two or three tandems. Test plans were generated for all clinical scenarios. Radiation Oncologists provided information regarding planning goals and fractionation regimen (number of fractions/number of implants, Rx dose). The physics team generated all necessary documents and procedures for clinical implementation. <h3>Results</h3> All set components were found to be functional with no obstruction. Their length was within 1 mm from nominal. Autoradiographs showed that source positions were within 1 mm from designed locations. To simulate a clinical scenario, a CTV was contoured to cover "cervical area" (above cylinder), entire "uterus" (up to the tip of the tandems), and a portion of the "upper vagina" (a portion of the cylinder). The tandems were manually digitized, paying attention to their curvature on sagittal and coronal images. The applicator was also loaded from the library of applicators to assess if that approach is more efficient than manual digitization. We found that Left and Right tandems loaded from the library are reversed (Varian was notified). An HDR plan was generated, assuming 500 cGy/fraction. The three tandems were loaded with sources inside the CTV and the plan was 3D optimized to cover the CTV with D90>90%. The process was repeated for combinations of two tandems. Treatment and Plan checklists, a BED Cumulative form, and a Planning Note document were developed, to be evaluated and refined after the first clinical case. A live demo with a manufacturer representative was organized to discuss the applicator assembly/disassembly in the OR setting. Sterilization procedures were developed based on the manufacturer's instructions. An in-service was organized for therapists/nurses, and pre-treatment huddles will be incorporated into the process for the first several cases. <h3>Conclusions</h3> The applicator set was validated and procedural documents are ready for clinical use. Coronal MRI series proved to be essential for the CT-MR registration process and tandem digitization. Via dose shaping, the dose distribution can be fine-tuned and adjusted to cover a longer portion of the upper vagina if needed. Also, a differential prescription may be used, as needed, for the uterus and upper vagina. The ability to rotate and shift the tandems allows adaptation to patient anatomy. An annual QA procedure was developed, including the tests performed for commissioning. A 3D printed holder with a 2-cm buildup was designed to ensure consistency in the autoradiographs.

A Comparative Evaluation of Digital Radiography and Ultrasound Imaging to Detect Periapical Lesions in the Oral Cavity.

Purpose This study evaluates the efficacy of digital radiography and ultrasound (USG) for the distinction between periapical cysts and granulomas, determines the nature and extent of the periapical lesion, visualizes the lumen of the lesion, assesses its size, content, and vascularity. Material and Methods Thirty patients, ages 18 to 40, with well-defined periapical radiolucencies on maxillary or mandibular teeth, indicated for the extraction or periapical surgery, underwent digital radiography examination using the paralleling technique, followed by USG examination. A sonologist evaluated the lesions' size, echogenicity, and vascular content. The diagnosis was compared to histopathological examinations of tissues obtained through extraction or periapical surgery. Results The diagnostic value of USG compared to the histopathological diagnosis of the periapical cyst was greater than that of the radiographic diagnosis, with an ultrasonographic diagnostic sensitivity (SN) value of 60% and a radiographic diagnostic SN value of 40%, respectively. The diagnostic value of USG imaging against the histopathological diagnosis of periapical granuloma was slightly lower than that of digital radiography, with an SN value of 72.2% for USG and 83.33% for digital radiography. However, the specificity (SP) value and precision of USG imaging were superior to those of digital radiographic diagnosis. USG imaging and radiographic diagnosis had 58.33% and 50% SP values, respectively. In cases of periapical abscess, the diagnostic values of USG against histopathological diagnosis were lower than those of radiographic diagnosis, which had an SN value of 100%. Conclusion USG with color doppler is a more effective tool than digital radiography for diagnosing periapical lesions. The echo structure of the lesions and the presence of vascularity on USG with color doppler correlated with histopathology better than the radiological diagnosis.

3D positioning of defects for gas turbine blades based on digital radiographic projective imaging

During the manufacturing and service process of gas turbine blades, various types of defects may be formed and bring huge threat to the safe operation of gas turbine. To confirm whether the gas turbine blades with defects could be in safe service for some period of time, the residual strength and fatigue life of gas turbine blade should be analyzed and predicted based on the 3D spatial position of defect, which significantly determines the degree of effect on the strength of gas turbine blade. Considering the characteristics of shape (free-from surface) and material (high density superalloy), the digital radiographic imaging techniques could be used in the non-destructive testing of gas turbine blades. However, the projective nature of digital radiography (DR) technique results in the unavailability of position in the transmitting direction of X-ray beam and brings difficulty to the 3D characterization of defect. Aiming at this critical issue, based on the exploration of the changing rule for the projection of a point feature relative to the irradiating angle, a new method is presented to obtain the 3D spatial position of defect. Finally, the accuracy of the proposed method is validated using a metrological computed tomography (CT) system. It can be learned that the proposed DR based method could obtain the 3D spatial position of defect with a relative error less than 2%. The 3D spatial position of defect could be used in the following residual strength analysis and fatigue life prediction of gas turbine blade.

Retracted: Chest digital dynamic radiography to detect changes in human pulmonary perfusion in response to alveolar hypoxia.

Hypoxic pulmonary vasoconstriction optimises oxygenation in the lung by matching the local-blood perfusion to local-ventilation ratio upon exposure to alveolar hypoxia. It plays an important role in various pulmonary diseases, but few imaging evaluations of this phenomenon in humans. This study aimed to determine whether chest digital dynamic radiography could detect hypoxic pulmonary vasoconstriction as changes in pulmonary blood flow in healthy individuals. Five Asian men underwent chest digital dynamic radiography before and after 60 sec breath-holding at the maximal inspiratory level in upright and supine positions. Alveolar partial pressure of oxygen and atmospheric pressure were calculated using the blood gas test and digital dynamic radiography imaging, respectively. To evaluate the blood flow, the correlation rate of temporal change in each pixel value between the lung fields and left cardiac ventricles was analysed. Sixty seconds of breath-holding caused a mean reduction of 26.7 ± 6.4 mmHg in alveolar partial pressure of oxygen. The mean correlation rate of blood flow in the whole lung was significantly lower after than before breath-holding (before, upright 51.5%, supine 52.2%; after, upright 45.5%, supine 46.1%; both P < 0.05). The correlation rate significantly differed before and after breath-holding in the lower lung fields (upright, 11.8% difference; supine, 10.7% difference; both P < 0.05). The mean radiation exposure of each scan was 0.98 ± 0.09 mGy. No complications occurred. Chest digital dynamic radiography could detect the rapid decrease in pulmonary perfusion in response to alveolar hypoxia. It may suggest hypoxic pulmonary vasoconstriction in healthy individuals.

Identifikasi Citra Paru-Paru pada Pasien COVID-19 dengan Teknik Edge Detection

X-ray chest radiographs produce digital radiographic images of the chest area such as the lungs, heart and ribs. This image can visualize the lung condition of patients with COVID-19. The Edge Detection technique can see the edges of objects in the lungs of COVID-19 patients more clearly. The lungs of patients with COVID-19 are damaged due to the GGO (Ground Glass Opacity) COVID-19, namely there is white fog, the lungs look blurry at the edges or areas affected by the disease and also affect the area of ​​​​the area. This technique can make it easier for health workers to see X-ray results on objects in the lungs of COVID-19 patients and assist doctors in handling COVID-19 patients. With advances in the computer field in the application of image processing techniques, this Edge Detection Technique uses Matlab software to obtain edge and area images of clean lungs in COVDI-19 patients. The data used in this study were 30 samples of lung images of COVID patients and 10 images of healthy patients' lungs as a comparison sourced from Embung Fatimah Hospital Batam, which were processed in the grayscale stage, then the image was preprocessed with Intensity Adjustment then segmented with Active Contour and then using Edge Detection Technique. The results of this study were as many as 22 lung images produced an accuracy value of 73%. The image of the test results with the Edge Detection Technique to identify the edges of the object's lungs of COVID-19 patients which are quite clear by producing white pixels that are so visible. The perimeter of the right lung of COVID-19 patients ranged from 110,897 - 261,254 mm and Area 267,719 - 940,668 mm2, Perimeter of the left lung ranged from 114,613 - 262,943 and Area 170.616 - 856,993 mm2, while healthy patients as comparison had the right lung Perimeter range of 187,598 -270,624 and Area 514,947 - 1025.44 mm2, Perimeter of the left lung 182,226 - 287,358 and Area 480,592 - 901,418 mm2 means that COVID-19 virus infection reduces lung area, the range of COVID-19 lungs is lower than healthy lungs.

Long cone versus short cone for digital intraoral radiographic imaging

<h3>Background</h3> Intraoral radiographs are acquired by one of two methods: the long cone paralleling technique that uses a 30 cm distance from the focal spot to the end of the beam indicating device (BID) versus the short cone technique of 20 cm. The long cone paralleling technique (LCPT) reduces x-ray beam divergence, scatter radiation, penumbra, and magnification as well as entrance skin exposure. LCPT has been recommended for use with digital intraoral radiographic imaging. However, most intraoral generators are supplied with only a short cone BID. <h3>Objective</h3> This research project evaluated image quality of digital radiographs acquired with both long cone and short cone techniques. <h3>Study Design</h3> A series of incrementally increasing exposures of a digital intraoral quality assurance phantom was acquired with both long and short cone methods over a range of exposures from 10 ms to 800 ms to encompass all exposures in the clinical range. Image quality was evaluated using a scientifically validated method that has been published in other research studies on digital intraoral image quality. The image quality parameters assessed were dynamic range, contrast perceptibility, and spatial resolution as specified in ANSI /ADA Standard No. 1094 Quality Assurance for Digital Intra-oral Radiographic Systems. Images were visually evaluated and then objectively evaluated using the NIH ImageJ software for validation. <h3>Results</h3> The results support the use of the LCPT for the acquisition of digital intraoral radiographic images. <h3>Conclusion</h3> The additional purchase of the BID extension is substantiated by the results. <b>Statement of Ethical Review</b> Ethical Review or exemption was not warranted for this study

EVALUATION OF THE DENTAL AGE ESTIMATION ON DIGITAL PANORAMIC RADIOGRAPHIES

<h3>Objectives</h3> To investigate the relationship between the chronology of dental age and the chronological age of children from panoramic radiographs. <h3>Study Design</h3> This was a cross-sectional and observational study. The study consisted of 20 digital panoramic radiographic images dated from the 2018, obeying the following criteria: patient age and panoramic radiographs with good image quality. We then obtained access to patients' files to obtain their chronological age. The data were divided by sex and analyzed statistically through the variance of 2 factors (chronological and dental age). <h3>Results</h3> When analyzing female patients, there was an advance in dental age up to 2 years from the chronological age. In male patients, there was a greater similarity between ages. During the analysis, similarity was found in the parity of the arcs, showing bilateral symmetry. <h3>Conclusions</h3> The result showed a high correlation between dental age and chronological age in male patients. This condition is useful to monitor dental-facial health in sync with postnatal growth and development. No significant differences were observed between arches during the eruption of permanent teeth.

Applying ADA/ANSI Standard 1094: Impact of image adjustments and display device on subjective assessments

<h3>Objective</h3> To examine operator and device factors that impact subjective assessment of digital radiographic image quality. <h3>Study design</h3> The DDQA phantom was used to generate images to assess spatial resolution, contrast perceptibility, and latitude. The phantom was imaged using an XDR size 2 sensor with a Gendex Expert DC intraoral x-ray source (65 kVp, 7.5mA) at 8 exposure times. Five observers analyzed the images in the native XDR software. Observers were allowed to adjust brightness and contrast to facilitate visualization. Images were analyzed in two environments: a diagnostic radiology reading room with lowered ambient lighting and a Dell Ultrasharp monitor, and an intraoral radiographic room with bright ambient light (450 lux) and an HP EliteOne Computer. The optimum exposure, defined as the lowest exposure that provided the highest feature visualization, was recorded. <h3>Results</h3> In the radiology reading environment, observers were able to discern the full latitude of grayscales at all exposure times. The mean optimal exposure was 0.25s for discerning contrast wells and 0.32s for discerning spatial resolution. Scoring the same images in the intraoral radiography room environment yielded identical results. All observers were able to discern the full image latitude, and the identified optimal exposures were unchanged (0.25s for contrast wells and 0.32s for spatial resolution). <h3>Conclusions</h3> We demonstrate that establishing optimal exposure using the DDQA phantom is reliable, with minimal individual variation. Image adjustments cannot compensate for decreased information. Interestingly, room lighting conditions and display device did not change observer assessments of image quality and optimal exposure times when the observers were allowed to manually adjust brightness and contrast of the presented image. This further validates image quality-guided approaches to establish optimal exposure times in dental clinic settings. In ongoing studies, we will study different display models and ambient lighting settings that are more typical of chair-side settings. <b>Statement of Ethical Review</b> Ethical Review or exemption was not warranted for this study

Machine Learning in Recognition of Basic Pulmonary Pathologies

Nowadays, during the diagnosis process, the doctor is able to obtain access to much information describing the patient’s condition using appropriate tools. However, there are always two sides to the coin. The doctor has certain limitations regarding the amount of data they can process at once. Information technology comes to the rescue, which with the help of computers is able to quickly and effectively separate important information from redundant information and support the doctor in making a diagnosis. In this work, a decision-making system was created to diagnose common lung pathologies in digital radiography images. Here, we consider four basic pulmonary diseases: pneumothorax, pneumonia, pulmonary consolidation, and lung lesions. Our objective is to develop a new automatic detection method of lung pathologies on chest X-ray radiographs using python programming language and its libraries. The approach uses solutions in the field of artificial intelligence, such as deep learning, convolutional neural network and segmentation to make a diagnosis that aims to help the radiologist at work. In the first sections, this work describes the fundamentals of the present form of diagnosis, a proposal to improve this process, the method of operation of the algorithms used, data acquisition, segmentation and processing methods. Then, the results of the operation of four different models and their implementation in a practical window program were presented. The best model, which detects pulmonary consolidation, achieves accuracy higher than 91%, which is a satisfactory result because they are not intended to replace radiologists but to improve their work. In the future, this type of program can be further developed by adding models that recognize other conditions.

Diffraction Enhanced Imaging Analysis with Pseudo-Voigt Fit Function.

Diffraction enhanced imaging (DEI) is an advanced digital radiographic imaging technique employing the refraction of X-rays to contrast internal interfaces. This study aims to qualitatively and quantitatively evaluate images acquired using this technique and to assess how different fitting functions to the typical rocking curves (RCs) influence the quality of the images. RCs are obtained for every image pixel. This allows the separate determination of the absorption and the refraction properties of the material in a position-sensitive manner. Comparison of various types of fitting functions reveals that the Pseudo-Voigt (PsdV) function is best suited to fit typical RCs. A robust algorithm was developed in the Python programming language, which reliably extracts the physically meaningful information from each pixel of the image. We demonstrate the potential of the algorithm with two specimens: a silicone gel specimen that has well-defined interfaces, and an additively manufactured polycarbonate specimen.

Afterglow-Suppressed Lu2O3:Eu3+ Nanoscintillators for High-Resolution and Dynamic Digital Radiographic Imaging.

Lu2(1-x)Eu2xO3 nanoscintillators (x = 0.005, 0.01, 0.03, 0.05, 0.07, and 0.10) with red emission were synthesized by a coprecipitation method. It is found that their photo- and radioluminescence intensities increase with increasing Eu3+ concentration until x = 0.05. According to their concentration-dependent luminescence intensity ratios (I610(C2)/I582(S6)), the existing energy transfer from Eu3+(S6) (occupying S6 sites) to Eu3+(C2) (occupying C2 sites) can be confirmed. Based on the spectral data and density functional theory (DFT) calculations, the origin of Lu2O3:Eu3+ persistent luminescence at low concentration might be related to the tunneling processes between Eu3+ (occupying C2 and S6 sites) and oxygen interstitials (Oi×). After dispersing afterglow-suppressed Lu2O3:Eu3+ nanoscintillators into polymethyl methacrylate (PMMA) polymer-acetone solution, flexible PMMA-Lu2O3:Eu3+ composite films with high thermal stability and radiation resistance were fabricated by a doctor blade method. As the flexible composite film was used as an imaging plate, static X-ray images with high spatial resolution (5.5 lp/mm) under an extremely low dose of ∼1.1 μGyair can be acquired. When a watch with a moving second hand was used as an object, the dynamic X-ray imaging can be realized under a dose rate of 55 μGyair·s-1. Our results demonstrate that Lu2O3:Eu3+ nanoscintillators can be regarded as candidate materials for dynamic digital radiographic imaging.

DR-only Carbon-ion radiotherapy treatment planning via deep learning

PurposeTo evaluate the feasibility of patient-specific digital radiography (DR)-only treatment planning for carbon ion radiotherapy in anthropomorphic thorax-and-abdomen phantom and head-and-neck patients. MethodsThe study was conducted on the anthropomorphic phantom and head-and-neck patients. We collected computed tomography (CT) and DR images of the phantom and cone beam CT (CBCT) and DR images of the patients, respectively. Two different deep neural networks were established to correlate the relationships between DR and digitally reconstructed radiograph (DRR) images, as well as DRR and CT images. The similarity between CT and predicted CT images was evaluated by computing the mean absolute error (MAE), root mean square error (RMSE), peak signal-to-noise ratio (PSNR) and structural similarity (SSIM), respectively. Dose calculations on the predicted CT images were compared against the true CT-based dose distributions for carbon-ion radiotherapy treatment planning with intensity-modulated pencil-beam spot scanning. Relative dose differences in the target volumes and organ-at-risks were computed and three-dimensional gamma analyses (3 mm, 3%) were performed. ResultsThe average MAE, RMSE, PSNR and SSIM of the framework were 0.007, 0.144, 37.496 and 0.973, respectively. The average relative dose differences between the predicted CT- and CT-based dose distributions at the same carbon-ion irradiation settings for the phantom and the patients were <2% and ≤4%, respectively. The average gamma pass-rates were >98% for the predicted CT-based versus CT-based carbon ion plans of the phantom and the patients. ConclusionWe have demonstrated the feasibility of a patient-specific DR-only treatment planning workflow for heavy ion radiotherapy by using deep learning approach.

Evaluation of resolution characteristics of digital intraoral radiographic images using a task transfer function.

This study aimed to quantitatively examine the effect of digital image processing of digital intraoral radiographic images on the resolution characteristics of the output image using a task transfer function (TTF). A photostimulable phosphor system with three types of image processing filters, including periodontal, endodontic, and dentine-to-enamel junction filters, was used. Each filter can be used in conjunction with the sharpness filter (+ S). Images were obtained from the original phantom, which combined aluminum disk and plate. The TTF, which indicates the resolution characteristics, was calculated. A one-dimensional profile curve was also measured, and the fluctuation in the pixel value was evaluated in detail. The results were compared to investigate the effects of digital image processing on digital intraoral radiographic images. The TTF values were specific to each filter. The change in the TTF strongly reflected the characteristics of the one-dimensional profile curve. The TTF was compared with a one-dimensional profile curve and was able to quantitatively express the resolution characteristics of all directions in the image. We attempted to evaluate the resolution characteristics of digital intraoral radiographic images with image processing filters using the TTF. The effect of each image processing filter and the + S filter on the resolution can be simply expressed using the TTF. Our results show that the TTF is useful for characterizing the resolution characteristics of image processing filters for image quality.

A high efficiency deep learning method for the x-ray image defect detection of casting parts

In the manufacturing industry, digital radiography (DR) images are often used to detect internal defects in casting parts. With the development of computer technology, increasingly more researchers use computer algorithms instead of manual inspection. However, traditional computer vision methods are generally not efficient and robust. In this study, we propose a DR image defect detection methodology based on deep learning technology. In order to train and evaluate the deep learning model, we create a casting defect DR image dataset, which includes 18 311 DR images labelled for two types of objects—defects and inclusions. In the methodology, an object detection method baseline named YOLOv3_EfficientNet, which replaces the backbone of YOLOv3_darknet53 with EfficientNet, is used. This operation leads to a significant improvement in the mean average precision value on YOLOv3 and greatly reduces the inference time and storage space. Then, a data enhancement method based on DR image features is used, which can increase the diversity of the clarity and the shapes of defects randomly. To further facilitate the deployment of models on embedded devices with an acceptable accuracy loss range, a depth separable convolution operation is adopted. Regarding the bounding box regression, we perform some relevant research in the training and inference stages of the model, and the accuracy of the model was improved in both stages of them according to the experiments. The experiments proved that every type we adopted could benefit the model’s performance.

PRINCIPLES &amp;FUNCTIONS OF CONVENTIONALS &amp; DIGITAL X-RAYS IMAGING SYSTEMS IN DIAGNOSIS

Digital radiography (DR) is an advanced form of x-ray inspection which produces a digital radiographic image instantly on a computer. This technique uses x-ray sensitive plates to capture data during object examination, which is immediately transferred to a computer without the use of an intermediate cassette. The incident x-ray radiation is converted into an equivalent electric charge and then to a digital image through a detector sensor. Compared to other imaging devices, flat panel detectors, also known as digital detector arrays (DDAs) provide high quality digital images. They can have better signal-to-noise ratio and improved dynamic range, which, in turn, provides high sensitivity for radiographic applications. Flat panel detectors work on two different approaches, namely, indirect conversion and direct conversion. Indirect conversion flat panel detectors have a scintillator layer which converts x-ray photons to photons of visible light and utilise a photo diode matrix of amorphous silicon to subsequently convert the light photons into an electrical charge. This charge is proportional to the number and energy of x-ray photons interacting with the detector pixel and therefore the amount and density of material that has absorbed the x-rays. Direct conversion flat panel detectors use a photo conductor like amorphous selenium (a-Se) or Cadmium telluride (Cd-Te) on a multi-micro electrode plate, providing the greatest sharpness and resolution. The information on both types of detectors is read by thin film transistors. In the direct conversion process, when x-ray photons impact over the photo conductor, like amorphous Selenium, they are directly converted to electronic signals which are amplified and digitised. As there is no scintillator, lateral spread of light photons is absent here, ensuring a sharper image. This differentiates it from indirect construction.

The Comparison of Convolutional Neural Networks and the Manual Measurement of Cobb Angle in Adolescent Idiopathic Scoliosis.

Comparative study. To compare manual and deep learning-based automated measurement of Cobb angle in adolescent idiopathic scoliosis. We proposed a fully automated framework to measure the Cobb angle of AIS patients. Whole-spine images of 500 AIS individuals were collected. 200 digital radiographic (DR) images were labeled manually as training set, and the remaining 300 images were used to validate by mean absolute error (MAE), Pearson or spearman correlation coefficients, and intra/interclass correlation coefficients (ICCs). The relationship between accuracy of vertebral boundary identification and the subjective image quality score was evaluated. The PT, MT, and TL/L Cobb angles were measured by the automated framework within 300 milliseconds. Remarkable 2.92° MAE, .967 ICC, and high correlation coefficient (r = .972) were obtained for the major curve. The MAEs of PT, MT, and TL/L were 3.04°, 2.72°, and 2.53°, respectively. The ICCs of these 3 curves were .936, .977, and .964, respectively. 88.7% (266/300) of cases had a difference range of ±5°, with 84.3% (253/300) for PT, 89.7% (269/300) for MT, and 93.0% (279/300) for TL/L. The decreased bone/soft tissue contrast (2.94 vs 3.26; P=.039) and bone sharpness (2.97 vs 3.35; P=.029) were identified in the images with MAE exceeding 5°. The fully automated framework not only identifies the vertebral boundaries, vertebral sequences, the upper/lower end vertebras and apical vertebra, but also calculates the Cobb angle of PT, MT, and TL/L curves sequentially. The framework would shed new light on the assessment of AIS curvature.

Progressive Changes in Lumbopelvic Alignment during the Three Month-Postpartum Recovery Period.

Pregnancy-related lumbopelvic pain is a common musculoskeletal problem, and postural changes are believed to be involved in these disorders. However, the lumbopelvic alignment changes in postpartum women remain unclear. This study aimed to determine whether there are changes in lumbopelvic alignment following vaginal or cesarean delivery and when these alignment changes occur after delivery. Thirty postpartum females (PP group) and 20 nulliparous female controls (CTL group) underwent anteroposterior, lateral pelvic, and lower-back X-ray in a static upright position. Digital radiographic images were analyzed and three radiographic variables, the pelvic incidence, pubic symphysis width, and sacral slope, were measured. The pubic symphysis width of the PP group was significantly larger immediately and one month after childbirth (PP group: 6.0 ± 1.1 mm (immediately), 5.0 ± 1.2 mm (one month); CTL group: 3.4 ± 0.4 mm; F = 31.79, p < 0.001). The sacrum slope in the PP group was significantly larger than in the CTL group 1 month after childbirth (PP group: 39.9 ± 6.6°; CTL group: 32.8 ± 5.1°; F = 2.59, p = 0.05). A two-way analysis of variance indicated no statistically significant main effects or interaction effects between the delivery modes on the pubic symphysis width or the sacrum slope. This study suggested that the course of lumbopelvic alignment progressed towards recovery for at least one month, and that these changes were independent of the delivery method.

Improved revealing of hidden structures and defects for historic art sculptures using poisson image editing

Radiography is a non-destructive tool and offers the acquisition of detailed information on the internal features of sculptures as a cultural heritage. However, radiographs contain different levels of blurriness mainly caused by the detection of scattered X-rays. Reduction of image blurriness provides improved contrast in targeted areas which enhances the extraction of information from the selected regions and features of the radiographs. In this study, we applied a set of convolution methods to a group of radiographic images of historic sculptures. Radiographs of the objects were provided with the associated documentation from the collection of the Radiographic Inspection Laboratory of the Universitat Politècnica de València. The selection of the particular objects was based on the difference in the materials used in their construction i.e. the objects were made of wood, paper, or wax. The Poisson Image Editing (PIE) based on L 2 -norm was applied for image enhancement of digital radiography images. The results showed that the PIE method was effective in selective region enhancement of the radiographic image contrast enabling better visualization of the objects’ internal structures. The application of the implemented algorithm enabled the conservators and radiographers involved in the study to improve the visualization of the sculptures' internal features and defects enhance the defects’ evaluation.

Mandibular Bone Changes in Children and Adolescents with Type 1 Diabetes Mellitus in Different Metabolic Control States

Objective: The aim of this study is to evaluate the cortical and trabecular mandibular bone morphology of children and adolescents with type 1 diabetes mellitus (DM) and control group utilizing fractal dimension analysis (FDA) and different panoramic radiomorphometric indices through digital panoramic radiographic images (DPRIs). Methods: The study included 57 patients for the type 1 DM group (25 male and 32 female with a mean age of 11.5±2.4 years) and 57 patients for the control group (28 male and 29 female with a mean age of 10.5±2.1 years). The type 1 DM group was divided into the well-controlled, moderately-controlled, and poorly-controlled subgroups based on HbA1c. Mandibular cortical width (MCW) (according to Lengerton et al.) and panoramic mandibular index (PMI) (according to Benson et al.) were measured, mandibular cortical index (MCI) (according to Klemetti et al) and simple visual estimation (SVE) (according to Lee et al.) were evaluated, and FDA was conducted according to White and Rudolph, resulting in three areas of interest (IAs) being obtained in all of the DPRIs. Results: There was no significant difference between type 1 DM group and control according to the mean MCW, mean PMI measurements, MCI and SVE. The mean FD values were not significantly different between type 1 DM group and the control and between type 1 DM subgroups and control. Conclusion: This study revealed no cortical and trabecular bone changes in mandibula in children and adolescents with type 1 DM compared to the control group. In addition, metabolic control states of DM did not affect the bone structure.