Search

Recognizing classes of objects from their shape is an unsolved problem in machine vision that entails the ability of a computer system to represent and generalize complex geometrical information on the basis of a finite amount of prior data. A practical approach to this problem is particularly difficult to implement, not only because the shape variability of relevant object classes is generally large, but also because standard sensing devices used to capture the real world only provide a partial view of a scene, so there is partial information pertaining to the objects of interest. In this work, we develop an algorithmic framework for recognizing classes of deformable shapes from range data. The basic idea of our component-based approach is to generalize existing surface representations that have proven effective in recognizing specific 3D objects to the problem of object classes using our newly introduced symbolic-signature representation that is robust to deformations, as opposed to a numeric representation that is often tied to a specific shape. Based on this approach, we present a system that is capable of recognizing and classifying a variety of object shape classes from range data. We demonstrate our system in a series of large-scale experiments that were motivated by specific applications in scene analysis and medical diagnosis. Index Terms--Three-dimensional object recognition and classification, deformable shapes, range data, numeric and symbolic signatures, Mercer kernel, scene analysis, craniosynostosis, craniofacial malformations.

The design of hospital environments is receiving increased attention as an important contributor to patient satisfaction and experience, which have a direct impact on reimbursement. Well-designed health care environments can decrease stress, improve concentration, and contribute to improved patient outcomes and enhanced staff morale. Most radiologists and business directors lack formal training in design and may feel they have little to contribute to design planning, yet creating an optimal environment for patients requires a strong understanding of local demographics and both patient and staff needs, which is a core responsibility of radiology leadership. This article presents practical guidelines for selecting a design partner for an imaging construction project, developing a design theme and design sensibilities, and engaging a multidisciplinary radiology team in working with a designer; the goal is to enable radiology leadership to collaborate with designers to cocreate health care environments that aspire to be integral components of patient-centered care and experience. [ABSTRACT FROM AUTHOR]

The rapid pace of technologic advancement and increasing expectations for patient- and family-friendly environments make it common for radiology leaders to be involved in imaging remodel and construction projects. Most radiologists and business directors lack formal training in architectural and construction processes but are expected to play significant and often leading roles in all phases of an imaging construction project. Avoidable mistakes can result in significant increased costs and scheduling delays; knowledgeable participation and communication can result in a final product that enhances staff workflow and morale and improves patient care and experience. This article presents practical guidelines for preparing for and leading a new imaging architectural and construction project. We share principles derived from the radiology and nonradiology literature and our own experience over the past decade completely remodeling a large pediatric radiology department and building a full-service outpatient imaging center. [ABSTRACT FROM AUTHOR]

The article discusses Imaging 3.0 that aims to improve patient and family centered care, highlighting its use in the radiology department. The conventional radiology environment design is explored which often does not consider the potential impact on families. Also tackled are the drivers of reimbursement in health care and the importance of training employees.

In Duchenne muscular dystrophy (DMD), a genetic disruption of dystrophin protein expression results in repeated muscle injury and chronic inflammation. Magnetic resonance imaging shows promise as a surrogate outcome measure in both DMD and rehabilitation medicine that is capable of predicting clinical benefit years in advance of functional outcome measures. The mdx mouse reproduces the dystrophin deficiency that causes DMD and is routinely used for preclinical drug testing. There is a need to develop sensitive, non-invasive outcome measures in the mdx model that can be readily translatable to human clinical trials. Here we report the use of magnetic resonance imaging and spectroscopy techniques for the non-invasive monitoring of muscle damage in mdx mice. Using these techniques, we studied dystrophic mdx muscle in mice from 6 to 12 weeks of age, examining both the peak disease phase and natural recovery phase of the mdx disease course. T2 and fat-suppressed imaging revealed significant levels of tissue with elevated signal intensity in mdx hindlimb muscles at all ages; spectroscopy revealed a significant deficiency of energy metabolites in 6-week-old mdx mice. As the mdx mice progressed from the peak disease stage to the recovery stage of disease, each of these phenotypes was either eliminated or reduced, and the cross-sectional area of the mdx muscle was significantly increased when compared to that of wild-type mice. Histology indicates that hyper-intense MRI foci correspond to areas of dystrophic lesions containing inflammation as well as regenerating, degenerating and hypertrophied myofibers. Statistical sample size calculations provide several robust measures with the ability to detect intervention effects using small numbers of animals. These data establish a framework for further imaging or preclinical studies, and they support the development of MRI as a sensitive, non-invasive outcome measure for muscular dystrophy. [ABSTRACT FROM AUTHOR]

Background: A multidisciplinary meeting was held from March 4 to 6, 2010, in Atlanta, Georgia, entitled "Craniosynostosis: Developing Parameters for Diagnosis, Treatment, and Management." The goal of this meeting was to create parameters of care for individuals with craniosynostosis. Methods: Fifty-two conference attendees represented a broad range of expertise, including anesthesiology, craniofacial surgery, dentistry, genetics, hand surgery, neurosurgery, nursing, ophthalmology, oral and maxillofacial surgery, orthodontics, otolaryngology, pediatrics, psychology, public health, radiology, and speech-language pathology. These attendees also represented 16 professional societies and peer-reviewed journals. The current state of knowledge related to each discipline was reviewed. Based on areas of expertise, four breakout groups were created to reach a consensus and draft specialty- specific parameters of care based on the literature or, in the absence of literature, broad clinical experience. In an iterative manner, the specialty-specific draft recommendations were presented to all conference attendees. Participants discussed the recommendations in multidisciplinary groups to facilitate exchange and consensus across disciplines. After the conference, a pediatric intensivist and social worker reviewed the recommendations. Results: Consensus was reached among the 52 conference attendees and two post hoc reviewers. Longitudinal parameters of care were developed for the diagnosis, treatment, and management of craniosynostosis in each of the 18 specialty areas of care from prenatal evaluation to adulthood. Conclusions: To our knowledge, this is the first multidisciplinary effort to develop parameters of care for craniosynostosis. These parameters were designed to help facilitate the development of educational programs for the patient, families, and health-care professionals; stimulate the creation of a national database and registry to promote research, especially in the area of outcome studies; improve credentialing of interdisciplinary craniofacial clinical teams; and improve the availability of health insurance coverage for all individuals with craniosynostosis. [ABSTRACT FROM AUTHOR]

Objective: To describe a novel set of scaphocephaly severity indices (SSls) for predicting and quantifying head- and skull-shape deformity in children diagnosed with isolated sagittal synostosis (ISS) and compare their sensitivity and specificity with those of the traditional cranial index (Cl). Methods: Computed tomography head scans were obtained from 60 patients diagnosed with ISS and 41 age-matched control patients. Volumetric reformations of the skull and overlying skin were used to trace two-dimensional planes defined in terms of skull-base plane and internal or surface landmarks. For each patient, novel SSIs were computed as the ratio of head width and length as measured on each of these planes. A traditional Cl was also calculated and a receiver operating characteristic curve analysis was applied to compare the sensitivity and specificity of the proposed indices with those of Cl. Results: Although the Cl is a sensitive measure of scaphocephaly, it is not specific and therefore not a suitable predictor of ISS in many practical applications. The SSI-A provides a specificity of 95% at a sensitivity level of 98%, in contrast with the 68% of Cl. On average, the sensitivity and specificity of all proposed indices are superior to those of Cl. Conclusions: Measurements of cranial width and length derived from planes that are defined in terms of internal or surface landmarks and skull-base plane produce SSIs that outperform traditional CI measurements. [ABSTRACT FROM AUTHOR]

Characterizes the epidemiology and imaging findings of several pediatric gastrointestinal emergencies. Use of ultrasound and computed tomography in diagnosing possible cases of malrotation and midgut volvulus; Findings of conventional radiographs in intussusception; Importance of ultrasound in diagnosing hypertrophic pyloric stenosis.

Background: In evaluating the effectiveness of ultrasound as a screening tool for craniosynostosis it was discovered that sonologists and sonographers needed more experience scanning and visualizing cranial sutures on ultrasound. Objective: To create an ultrasound simulator to train radiologists and technologists to locate and recognize patent and fused cranial sutures in children. Materials and methods: The hypoechoic appearance of patent sutures was simulated by cutting lines into life-sized plastic doll heads and filling them with a commercial hypoechogenic material. Fused hyperechoic sutures were simulated by not cutting into the hard plastic region of a suture. The simulator’s teaching value was evaluated on three radiology residents and three fellows. Subjects performed pre-training scans on unknown simulators, received feedback and an opportunity to scan a training simulator, and then performed post-training scans on random unknown simulators. Accuracy was recorded as percentage of correctly demonstrated sutures. Results: The suture simulator reproduces the sonographic appearance of patent and fused cranial sutures. Accuracy of acquisition, interpretation, and overall diagnosis increased from 64 to 91%, 79 to 91%, 61 to 97%, respectively, between pre and post training scans. Conclusion: An ultrasound simulator can reproduce the appearance of patent and fused cranial sutures in children and can be used to train radiologists and technologists in the performance of a screening protocol. [ABSTRACT FROM AUTHOR]

Objective: In this proof-of-concept study, we aimed to develop deep-learning-based classifiers to identify rib fractures on frontal chest radiographs in children under 2 years of age., Methods: This retrospective study included 1311 frontal chest radiographs (radiographs with rib fractures, n = 653) from 1231 unique patients (median age: 4 m). Patients with more than one radiograph were included only in the training set. A binary classification was performed to identify the presence or absence of rib fractures using transfer learning and Resnet-50 and DenseNet-121 architectures. The area under the receiver operating characteristic curve (AUC-ROC) was reported. Gradient-weighted class activation mapping was used to highlight the region most relevant to the deep learning models' predictions., Results: On the validation set, the ResNet-50 and DenseNet-121 models obtained an AUC-ROC of 0.89 and 0.88, respectively. On the test set, the ResNet-50 model demonstrated an AUC-ROC of 0.84 with a sensitivity of 81% and specificity of 70%. The DenseNet-50 model obtained an AUC of 0.82 with 72% sensitivity and 79% specificity., Conclusion: In this proof-of-concept study, a deep learning-based approach enabled the automatic detection of rib fractures in chest radiographs of young children with performances comparable to pediatric radiologists. Further evaluation of this approach on large multi-institutional data sets is needed to assess the generalizability of our results., Advances in Knowledge: In this proof-of-concept study, a deep learning-based approach performed well in identifying chest radiographs with rib fractures. These findings provide further impetus to develop deep learning algorithms for identifying rib fractures in children, especially those with suspected physical abuse or non-accidental trauma.

Background Radiographic measurement of leg length discrepancy (LLD) is time consuming yet cognitively simple for pediatric radiologists. Purpose To compare deep learning (DL) measurements of LLD in pediatric patients to measurements performed by radiologists. Materials and Methods For this HIPAA-compliant retrospective study, radiographs obtained to evaluate LLD in children between January and August 2018 were identified. LLD was automatically measured by means of image segmentation followed by leg length calculation. On training data, a DL model was trained to segment femurs and tibias on radiographs. The validation set was used to select the optimized model. On testing data, leg lengths were calculated from segmentation masks and compared with measurements from the radiology report. Statistical analysis was performed by using a paired Wilcoxon signed-rank test to compare DL calculations and radiology reports. In addition, the measurement time was manually assessed by a pediatric radiologist and automatically assessed by the DL model on a randomly chosen group of 26 cases; the values were compared with the paired Wilcoxon signed-rank test. Results Radiographs obtained to evaluate LLD in 179 children (mean age ± standard deviation, 12 years ± 3; age range, 5-19 years; 89 boys and 90 girls) were evaluated. Radiographs were randomly divided into training, validation, and testing sets and consisted of studies from 70, 32, and 77 patients, respectively. In the training and validation sets, the DL model showed a high spatial overlap between manual and automatic segmentation masks of pediatric legs (Dice similarity coefficient, 0.94). For the testing set, the correlation between radiology reports and DL-calculated lengths of separated femurs and tibias ( r = 0.99; mean absolute error [MAE], 0.45 cm), full pediatric leg lengths ( r = 0.99; MAE, 0.45 cm), and full LLD ( r = 0.92; MAE, 0.51 cm) was high ( P < .001 for all correlations). Calculation time for the DL method per radiograph was faster than the mean time for radiologist manual calculation (1 second vs 96 seconds ± 7, respectively; P < .001). Conclusion A deep learning algorithm measured pediatric leg lengths with high spatial overlap compared with manual measurement at a rate 96 times faster than that of subspecialty-trained pediatric radiologists. © RSNA, 2020 See also the editorial by van Rijn and De Luca in this issue.

Theranostic nanoparticles offer the potential for mixing and matching disparate diagnostic and therapeutic functionalities within a single nanoparticle for the personalized treatment of diseases. In this article, we present composite iron oxide-gadolinium-containing Prussian blue nanoparticles (Fe 3 O 4 @GdPB) as a novel theranostic agent for T 1 -weighted magnetic resonance imaging (MRI) and photothermal therapy (PTT) of tumors. These particles combine the well-described properties and safety profiles of the constituent Fe 3 O 4 nanoparticles and gadolinium-containing Prussian blue nanoparticles. The Fe 3 O 4 @GdPB nanoparticles function both as effective MRI contrast agents and PTT agents as determined by characterizing studies performed in vitro and retain their properties in the presence of cells. Importantly, the Fe 3 O 4 @GdPB nanoparticles function as effective MRI contrast agents in vivo by increasing signal:noise ratios in T 1 -weighted scans of tumors and as effective PTT agents in vivo by decreasing tumor growth rates and increasing survival in an animal model of neuroblastoma. These findings demonstrate the potential of the Fe 3 O 4 @GdPB nanoparticles to function as effective theranostic agents., Competing Interests: Disclosure The authors report no conflicts of interest in this work.