High sensitivity methods for automated rib fracture detection in pediatric radiographs.

Rib fractures are highly predictive of non-accidental trauma in children under 3 years old. Rib fracture detection in pediatric radiographs is challenging because fractures can be obliquely oriented to the imaging detector, obfuscated by other structures, incomplete, and non-displaced. Prior studies have shown up to two-thirds of rib fractures may be missed during initial interpretation. In this paper, we implemented methods for improving the sensitivity (i.e. recall) performance for detecting and localizing rib fractures in pediatric chest radiographs to help augment performance of radiology interpretation. These methods adapted two convolutional neural network (CNN) architectures, RetinaNet and YOLOv5, and our previously proposed decision scheme, "avalanche decision", that dynamically reduces the acceptance threshold for proposed regions in each image. Additionally, we present contributions of using multiple image pre-processing and model ensembling techniques. Using a custom dataset of 1109 pediatric chest radiographs manually labeled by seven pediatric radiologists, we performed 10-fold cross-validation and reported detection performance using several metrics, including F2 score which summarizes precision and recall for high-sensitivity tasks. Our best performing model used three ensembled YOLOv5 models with varied input processing and an avalanche decision scheme, achieving an F2 score of 0.725 ± 0.012. Expert inter-reader performance yielded an F2 score of 0.732. Results demonstrate that our combination of sensitivity-driving methods provides object detector performance approaching the capabilities of expert human readers, suggesting that these methods may provide a viable approach to identify all rib fractures.

Women in Radiology: A perspective from Spain.

INTRODUCTION: There are gender inequalities in all fields, including radiology. Although the situation is improving, the presence of radiologists in leadership positions continues to be a minority. The objective of this article is to analyse the situation of women in the spanish radiology, comparing it with Europe and the United States. MATERIALS AND METHODS: We selected the years 2000-2022 as reference period to make a comparison with feminization data throughout history. In addition, relevant specific data from the just begun 2023 were also included. The variables in which we investigated feminization were the following: medical students, medical graduates, radiology residents and specialists, section chiefs, department chairs, radiology residency programme directors, radiology university professors, presidents of the main radiological entities and societies in Spain, Europe and the United States, recipients of the main awards given by these radiological societies and chief editors of their journals. In order to perform this analysis we conducted an in-depth bibliographic research, we contacted the radiological societies of Spain, Europe and the USA and we carried out a survey in the main Spanish radiology departments. RESULTS: The female presence in radiology decreases as we rise to leadership positions, a situation that is patent in Spain, Europe and the US, comparison that will be analysed in depth throughout the article. In Spanish hospitals in 2021 there were 58.1% female radiology residents, 55% female radiologists, 42.9% female section chiefs and 24.4% female department chairs. In SERAM's history there have been 10% female presidents, 22% female gold medallists and 5% female editors-in-chief. If we analyse data from 2000 to 2023, female presidents reach 32% and female gold medallists 31%. CONCLUSIONS: Although gender inequality is declining, in radiology women continue to be underrepresented in leadership positions. Work must be done in order to build a diverse and inclusive profession that reflects demographic reality.

Is the use of gonad protection protectors necessary during infants chest radiography?

INTRODUCTION AND OBJECTIVES: To compare gonad doses with and without a gonad protector and to optimize the use of gonadal protectors in infants thorax radiography. MATERIALS AND METHODS: Two pediatric anthropomorphic phantoms are used: an X-ray system for KXO-50SS/DRX-3724HD, and a digital radiography system for CALNEO Smart C12, with and without a gonad protector during infants thorax radiography. A real time skin dosimeter is placed on the X-ray system, and a real time skin dosimeter is inserted on the front side of the mammary gland, the front and back sides of the true pelvis level, and on the ovaries and testes. The X-ray system is irradiated 15 times using phantoms with and without a gonad protector. The measured entrance patient doses values of for the real time skin dosimeter are compared for each phantom, with and without the gonad protector. RESULTS: The medium of measured entrance patient doses values for front side dose of the true pelvis level with and without the protector are 10.00 and 5.00 µGy at newborn, and 10.00 and 0.00µGy at one year, respectively. The medium of measured entrance patient doses values for the back side dose of the true pelvis level with and without the protector are 0.00 and 0.00 µGy at both newborn one year, respectively. The measured entrance patient doses cannot be detected in the ovaries and testes with or without the protector. No significant differences are observed in the measured entrance patient doses values for the front and back side doses of the pelvis, ovaries, and testes at newborn and one year, with and without the protector (p>0.05). CONCLUSIONS: No significant difference was observed in gonad dose measurements with and without the gonad protector during infants chest radiography. We believe that gonadal protector wearing is not necessary.

Teaching radiology in integrated educational programs.

Radiology is a medical discipline, an area of transversal knowledge integrated into any clinical situation. The optimal training of learning knowledge, skills and aptitudes in Radiology in the Degree in Medicine requires the integration of any imaging modality in the different areas of knowledge; from the basic subjects to any clinical subject of the Degree. This article describes the integration of Radiology teaching into the curriculum throughout the Medicine Degree at the University of Girona (UdG), describing the different radiology teaching activities that are taught. The specific activities of the subject "Radiology" are detailed; through workshops, seminars, practices, interactive computer game; and describing the characteristics of the main teaching methodological activity of the UdG, Problem-Based Learning.

Challenges in radiological evaluation of brain metastases, beyond progression.

MRI is the cornerstone in the evaluation of brain metastases. The clinical challenges lie in discriminating metastases from mimickers such as infections or primary tumors and in evaluating the response to treatment. The latter sometimes leads to growth, which must be framed as pseudo-progression or radionecrosis, both inflammatory phenomena attributable to treatment, or be considered as recurrence. To meet these needs, imaging techniques are the subject of constant research. However, an exponential growth after radiotherapy must be interpreted with caution, even in the presence of results suspicious of tumor progression by advanced techniques, because it may be due to inflammatory changes. The aim of this paper is to familiarize the reader with inflammatory phenomena of brain metastases treated with radiotherapy and to describe two related radiological signs: "the inflammatory cloud" and "incomplete ring enhancement", in order to adopt a conservative management with close follow-up.

The teaching of Radiology in other university studies.

After the implementation of the European Space for Higher Education, the contents of the Radiology and Physical Medicine Area that were taught in the Medicine Degree have also been incorporated into the new degrees of Dentistry, Nursing, Physiotherapy, Podiatry, and, to a lesser extent, Pharmacy, Occupational Therapy, Logopedia, and Biomedical Engineering As a whole, the basic concepts of radiology and radiological protection are taught in Murcia in 5 different degrees with a total of 52.5 ECTS credits, participating in the training of 1219 students each academic year. This incorporation in the new degrees has tripled the number of subjects in which undergraduate teaching is taught, and doubled both the number of ECTS credits and the number of undergraduate students to whom it directs its training work. Thus, given the possible creation of new university degrees in the near future (Diagnostic Imaging and Radiotherapy Technicians), it would be necessary to involve a greater number of accredited professionals, from different specialties, and to optimize teaching resources (bibliography, material teacher, clinical cases, etc.,) for its usefulness in the different subjects that share similar contents.

Morphological and radiographic studies on the Manus region in the Arabian one-humped camel (Camelus dromedaries).

The study aims to analyse the normal anatomical and radiographical features of the Manus of the southern Aswanian-adapted Arabian one-humped camel, providing crucial data for diagnosing and treating various ailments. Our study was applied to 10 cadaver forelimbs of adult male one-humped camels (4-5 years old) for an explanation of the gross anatomy of the bones of the Manus region from under the carpal bones by using traditional techniques, including the gross anatomical, radiographic and x-ray (at the dorsopalmar and lateral planes) of the preparation of Manus bones. Our results showed that the large fused (third and fourth) metacarpal bones, in which the fusion extended along the entire length of the bone except at the distal end, diverged to form separate articulations with cross-ponding digits. As described in all ruminant species, especially the camel, there were two digits, and each digit consisted of three phalanges and two proximal sesamoid bones. Our radiographic x-ray data revealed that the complete radiopaque septum that completely divided the medullary cavity into two separate parts was clear from the dorsopalmar view, while the lateral view showed the proximal sesamoid bones that were placed over each other and located palmar to the head of the large metacarpal bone. In conclusion, our study reveals the adaptations of the Arabian one-humped camel to Egyptian conditions, aiding in the early diagnosis of lameness and digit problems and enabling veterinarians and camel owners to better address these issues, thereby improving the overall health and well-being of these animals.

Potential of GPT-4 for Detecting Errors in Radiology Reports: Implications for Reporting Accuracy.

Background Errors in radiology reports may occur because of resident-to-attending discrepancies, speech recognition inaccuracies, and large workload. Large language models, such as GPT-4 (ChatGPT; OpenAI), may assist in generating reports. Purpose To assess effectiveness of GPT-4 in identifying common errors in radiology reports, focusing on performance, time, and cost-efficiency. Materials and Methods In this retrospective study, 200 radiology reports (radiography and cross-sectional imaging [CT and MRI]) were compiled between June 2023 and December 2023 at one institution. There were 150 errors from five common error categories (omission, insertion, spelling, side confusion, and other) intentionally inserted into 100 of the reports and used as the reference standard. Six radiologists (two senior radiologists, two attending physicians, and two residents) and GPT-4 were tasked with detecting these errors. Overall error detection performance, error detection in the five error categories, and reading time were assessed using Wald χ2 tests and paired-sample t tests. Results GPT-4 (detection rate, 82.7%;124 of 150; 95% CI: 75.8, 87.9) matched the average detection performance of radiologists independent of their experience (senior radiologists, 89.3% [134 of 150; 95% CI: 83.4, 93.3]; attending physicians, 80.0% [120 of 150; 95% CI: 72.9, 85.6]; residents, 80.0% [120 of 150; 95% CI: 72.9, 85.6]; P value range, .522-.99). One senior radiologist outperformed GPT-4 (detection rate, 94.7%; 142 of 150; 95% CI: 89.8, 97.3; P = .006). GPT-4 required less processing time per radiology report than the fastest human reader in the study (mean reading time, 3.5 seconds ± 0.5 [SD] vs 25.1 seconds ± 20.1, respectively; P < .001; Cohen d = -1.08). The use of GPT-4 resulted in lower mean correction cost per report than the most cost-efficient radiologist ($0.03 ± 0.01 vs $0.42 ± 0.41; P < .001; Cohen d = -1.12). Conclusion The radiology report error detection rate of GPT-4 was comparable with that of radiologists, potentially reducing work hours and cost. © RSNA, 2024 See also the editorial by Forman in this issue.

[Application of Deep Learning to Diagnose and Classify Adolescent Idiopathic Scoliosis].

A deep learning-based model for automatic diagnosis and classification of adolescent idiopathic scoliosis has been constructed. This model mainly included key points detection and Cobb angle measurement. 748 full-length standing spinal X-ray images were retrospectively collected, of which 602 images were used to train and validate the model, and 146 images were used to test the model performance. The results showed that the model had good diagnostic and classification performance, with an accuracy of 94.5%. Compared with experts' measurement, 94.9% of its Cobb angle measurement results were within the clinically acceptable range. The average absolute difference was 2.1°, and the consistency was also excellent (r2≥0.9552, P<0.001). In the future, this model could be applied clinically to improve doctors' diagnostic efficiency.

[Deep Learning-Based Key Frame Recognition Algorithm for Adrenal Vascular in X-Ray Imaging].

Adrenal vein sampling is required for the staging diagnosis of primary aldosteronism, and the frames in which the adrenal veins are presented are called key frames. Currently, the selection of key frames relies on the doctor's visual judgement which is time-consuming and laborious. This study proposes a key frame recognition algorithm based on deep learning. Firstly, wavelet denoising and multi-scale vessel-enhanced filtering are used to preserve the morphological features of the adrenal veins. Furthermore, by incorporating the self-attention mechanism, an improved recognition model called ResNet50-SA is obtained. Compared with commonly used transfer learning, the new model achieves 97.11% in accuracy, precision, recall, F1, and AUC, which is superior to other models and can help clinicians quickly identify key frames in adrenal veins.

New insights into pterosaur cranial anatomy: X-ray imaging reveals palatal structure and evolutionary trends.

Among the least studied portion of the pterosaur skeleton is the palate, which tends to be poorly preserved and commonly only visible from one side (the ventral portion). Even in well-preserved specimens, the bones tend to be fused, with the limits of individual palatal elements obscured. To shed new light on this region, we employed advanced X-ray imaging techniques on the non-pterodactyloid Kunpengopterus (Wukongopteridae), and the pterodactyloids Dsungaripterus (Dsungaripteridae), Hongshanopterus (Istiodactylidae), and Hamipterus (Hamipteridae). Our analyses revealed the presence of sutures between palatal bones in Dsungaripterus and Kunpengopterus, which resulted in different interpretations of the relation between palatine, ectopterygoid, and pterygoid, leading to a new identification of the palatal openings. Furthermore, our study shows six main observations such as the variation of the angle between the palatine rami and the variation in the relative sizes of the palatal openings. We also point out that the presence of a maxillopalatine fenestra (previously identified as postpalatine fenestra), is unique within Diapsida. Although much more work needs to be done, we showed that advanced X-ray imaging techniques open a window for understanding pterosaur cranial anatomy and provide a new perspective for investigating the evolutionary history of these flying reptiles.

RT-SRTS: Angle-agnostic real-time simultaneous 3D reconstruction and tumor segmentation from single X-ray projection.

Radiotherapy is one of the primary treatment methods for tumors, but the organ movement caused by respiration limits its accuracy. Recently, 3D imaging from a single X-ray projection has received extensive attention as a promising approach to address this issue. However, current methods can only reconstruct 3D images without directly locating the tumor and are only validated for fixed-angle imaging, which fails to fully meet the requirements of motion control in radiotherapy. In this study, a novel imaging method RT-SRTS is proposed which integrates 3D imaging and tumor segmentation into one network based on multi-task learning (MTL) and achieves real-time simultaneous 3D reconstruction and tumor segmentation from a single X-ray projection at any angle. Furthermore, the attention enhanced calibrator (AEC) and uncertain-region elaboration (URE) modules have been proposed to aid feature extraction and improve segmentation accuracy. The proposed method was evaluated on fifteen patient cases and compared with three state-of-the-art methods. It not only delivers superior 3D reconstruction but also demonstrates commendable tumor segmentation results. Simultaneous reconstruction and segmentation can be completed in approximately 70 ms, significantly faster than the required time threshold for real-time tumor tracking. The efficacies of both AEC and URE have also been validated in ablation studies. The code of work is available at https://github.com/ZywooSimple/RT-SRTS.

[Diagnosis and classification of axial spondyloarthritis (axSpA) - the current status]. / Diagnostik und Klassifikation der axialen Spondyloarthritis (axSpA) ­ der aktuelle Stand.

Axial spondyloarthritis (axSpA) is an inflammatory rheumatic disease typically characterized by inflammatory back pain (IBP). The term axSpA has largely replaced the long-used term ankylosing spondylitis (AS). IBP is caused by inflammation in the axial skeleton, with the sacroiliac joints (SIJ) being particularly frequently affected initially. The spine is usually added in later stages, which is then increasingly characterized structurally by the formation of new bone. The overall concept of spondyloarthritis includes other disease manifestations such as uveitis, psoriasis and colitis and comorbidities such as cardiovascular disease and osteoporosis.The ASAS classification criteria for axSpA, in place since 2009, have replaced the 1984 modified New York criteria. In the former, in addition to conventional X-rays, changes in the SIJ detected by magnetic resonance imaging (MRI) and also the detection of HLA B27 have, for the first time, played a role. It is important to note that these are not diagnostic criteria, as they do not exist. This paper outlines 10 points that should be considered when making a diagnosis.

Ankle and hindfoot motion of healthy adults during running revealed by dynamic biplane radiography: Side-to-side symmetry, sex-specific differences, and comparison with walking.

This study aimed to characterize ankle and hindfoot kinematics of healthy men and women during overground running using biplane radiography, and to compare these data to those previously obtained in the same cohort during overground walking. Participants ran across an elevated platform at a self-selected pace while synchronized biplane radiographs of their ankle and hindfoot were acquired. Motion of the tibia, talus, and calcaneus was tracked using a validated volumetric model-based tracking process. Tibiotalar and subtalar 6DOF kinematics were obtained. Absolute side-to-side differences in ROM and kinematics waveforms were calculated. Side-to-side and sex-specific differences were evaluated at 10 % increments of stance phase with mixed model analysis. Pearson correlation coefficients were used to assess the relationship between stance-phase running and walking kinematics. 20 participants comprised the study cohort (10 men, mean age 30.8 ± 6.3 years, mean BMI 24.1 ± 3.1). Average absolute side-to-side differences in running kinematics waveforms were 5.6°/2.0 mm or less at the tibiotalar joint and 5.2°/3.2 mm or less at the subtalar joint. No differences in running kinematics waveforms between sides or between men and women were detected. Correlations were stronger at the tibiotalar joint (42/66 [64 %] of correlations were p < 0.05), than at the tibiotalar joint (38/66 [58 %] of correlations were p < 0.05). These results provide a normative reference for evaluating native ankle and hindfoot kinematics which may be informative in surgical or rehabilitation contexts. Sex-specific differences in ankle kinematics during overground running are likely not clinically or etiologically significant. Associations seen between walking and running kinematics suggest one could be used to predict the other.

Ensemble detection of hand joint ankylosis and subluxation in radiographic images using deep neural networks.

The modified total Sharp score (mTSS) is often used as an evaluation index for joint destruction caused by rheumatoid arthritis. In this study, special findings (ankylosis, subluxation, and dislocation) are detected to estimate the efficacy of mTSS by using deep neural networks (DNNs). The proposed method detects and classifies finger joint regions using an ensemble mechanism. This integrates multiple DNN detection models, specifically single shot multibox detectors, using different training data for each special finding. For the learning phase, we prepared a total of 260 hand X-ray images, in which proximal interphalangeal (PIP) and metacarpophalangeal (MP) joints were annotated with mTSS by skilled rheumatologists and radiologists. We evaluated our model using five-fold cross-validation. The proposed model produced a higher detection accuracy, recall, precision, specificity, F-value, and intersection over union than individual detection models for both ankylosis and subluxation detection, with a detection rate above 99.8% for the MP and PIP joint regions. Our future research will aim at the development of an automatic diagnosis system that uses the proposed mTSS model to estimate the erosion and joint space narrowing score.

Non-traumatic complete cervical spine dislocation with severe fixed kyphosis: successful multidisciplinary approach to a challenging case.

BACKGROUND: To our knowledge, there is no previous report in the literature of non-traumatic neglected complete cervical spine dislocation characterized by anterior spondyloptosis of C4, extreme head drop, and irreducible cervicothoracic kyphosis. CASE PRESENTATION: We report the case of a 33-year-old Caucasian man with a 17-year history of severe immune polymyositis and regular physiotherapy who presented with severe non-reducible kyphosis of the cervicothoracic junction and progressive tetraparesia for several weeks after a physiotherapy session. Radiographs, computed tomography, and magnetic resonance imaging revealed a complete dislocation at the C4-C5 level, with C4 spondyloptosis, kyphotic angulation, spinal cord compression, and severe myelopathy. Due to recent worsening of neurological symptoms, an invasive treatment strategy was indicated. The patient's neurological status and spinal deformity greatly complicated the anesthetic and surgical management, which was planned after extensive multidisciplinary discussion and relied on close collaboration between the orthopedic surgeon and the anesthetist. Regarding anesthesia, difficult airway access was expected due to severe cervical angulation, limited mouth opening, and thyromental distance, with high risk of difficult ventilation and intubation. Patient management was further complicated by a theoretical risk of neurogenic shock, motor and sensory deterioration, instability due to position changes during surgery, and postoperative respiratory failure. Regarding surgery, a multistage approach was carefully planned. After a failed attempt at closed reduction, a three-stage surgical procedure was performed to reduce displacement and stabilize the spine, resulting in correct spinal realignment and fixation. Progressive complete neurological recovery was observed. CONCLUSION: This case illustrates the successful management of a critical situation based on a multidisciplinary collaboration involving radiologists, anesthesiologists, and spine surgeons.

Volvulus of the gastrointestinal tract.

Volvulus describes the twisting of the intestine or colon around its mesentery. Intestinal obstruction and/or ischaemia are the most common complications of volvulus. Within the gastrointestinal tract, there is a preponderance towards colonic volvulus. The sigmoid is the most commonly affected segment, followed by the caecum, small intestine and stomach. Distinguishing between the differing anatomical locations of gastrointestinal volvulus can be challenging, but is important for the management and prognosis. This article focuses on the main anatomical sites of gastrointestinal volvulus encountered in clinical practice. The aetiology, presentation, radiological features and management options for each are discussed to highlight the key differences.