Interpretation Of Ultrasound Images Research Articles

398 Emergency Medical Services and Bilateral Lung Ultrasound in Emergency (EMS BLUE) Study

Out-of-hospital ultrasound is not yet widely implemented. Most studies report on convenience samples and trauma patients. We assessed the feasibility of paramedic performed out-of-hospital lung ultrasound in medical patients with respiratory distress. Paramedics at two ambulance stations in the city of Pittsburgh, PA, USA underwent a 2-hour training session in out-of-hospital lung ultrasound using the SonoSite iViz, a handheld ultrasound device. Emergency medical services (EMS) physicians were instructed in the interpretation of lung ultrasound images. Paramedics enrolled patients presenting with signs and symptoms of respiratory distress over a 3-month period. The ultrasound exam included anterior and lateral views from both sides of the chest. Images were transmitted wirelessly using a mobile hotspot device and uploaded into an online image archiving system. Images were interpreted remotely by EMS physicians, and two expert sonographers provided an overread. We assessed agreement between EMS physicians and experts, as well as between chart review-derived ED diagnosis and expert interpretation. We defined 4 a priori hypotheses that would need to be met for the intervention to be considered “feasible.” 34 out of 78 eligible patients had an ultrasound exam completed. Image transmission was successful in 25 cases. The primary reason for not enrolling an otherwise eligible patient was equipment failure (25%), followed by patient acuity and patient refusal (18.2% each). 58.8% of completed scans were deemed uninterpretable upon expert review. Agreement between EMS physicians and experts was low. The predetermined thresholds for feasibility were not met. Paramedic-performed out-of-hospital lung ultrasound for patients with respiratory distress and remote interpretation by EMS physicians did not meet the predetermined thresholds to be considered “feasible” in a real-world environment.

Feasibility of Paramedic Performed Prehospital Lung Ultrasound in Medical Patients with Respiratory Distress

Objective: Prehospital ultrasound is not yet widely implemented. Most studies report on convenience samples and trauma patients, often by prehospital physicians or critical care clinicians. We assessed the feasibility of paramedic performed prehospital lung ultrasound in medical patients with respiratory distress. Methods: Paramedics at 2 ambulance stations in the city of Pittsburgh, Pennsylvania, USA underwent a 2-hour training session in prehospital lung ultrasound using the SonoSite iViz, a handheld ultrasound device. Emergency medical services (EMS) command center (EMS-CC) physicians were instructed in the interpretation of lung ultrasound images. Paramedics enrolled patients presenting with signs and symptoms of respiratory distress over a 3-month period. The ultrasound exam included anterior and lateral views from both sides of the chest. Images were transmitted wirelessly using a mobile hotspot device and uploaded into an online image archiving system. Images were interpreted remotely by the EMS-CC physicians, and 2 expert sonographers provided an overread. We assessed agreement between EMS-CC physicians and experts, as well as between chart-review derived ED diagnosis and both EMS-CC physician and expert interpretation. We defined four a priori hypotheses that would need to be met for the intervention to be considered “feasible.” Results: A total of 34 of 78 (43.6%) eligible patients had an ultrasound exam completed. Image transmission was successful in 25 (73.5%) of cases where ultrasound was performed. The primary reason for not enrolling an otherwise eligible patient was equipment failure (25.0%), followed by patient acuity and patient refusal (18.2% each). A total of 20 (58.8%) completed scans were deemed uninterpretable upon expert review. Agreement between EMS physicians and experts was poor. Agreement between EMS-CC physicians and ED diagnosis, as well as between experts and ED diagnosis, was fair. The predetermined thresholds for feasibility were not met. Conclusions: Paramedic performed prehospital lung ultrasound for patients with respiratory distress and remote interpretation by EMS physicians did not meet the predetermined thresholds to be considered “feasible” in a real-world environment with currently available technologies. This study identified important barriers to the implementation of prehospital lung ultrasound, which should be addressed in future studies.

Teaching bedside ultrasound to medical students.

Point-of-care ultrasound (POCUS) assists in the rapid diagnosis of conditions in the Emergency Department (ED). POCUS has been introduced to international medical curricula; however, there is no described implementation of clinically focused POCUS education in Australian medical schools. We wanted to investigate whether a formal curriculum can be effective and feasible in an Australian medical school. Pre-post intervention study of a focused curriculum based on the Extended Focused Assessment with Sonography in Trauma (E-FAST) examination, consisting of online and practical teaching, was implemented for Year-4 and -5 medical students. An online questionnaire was used to measure knowledge, image interpretation and confidence prior to the intervention. After the intervention and ED placement, the questionnaire was repeated and students were assessed performing the E-FAST examination on a healthy volunteer. Twenty-seven students participated in both the pre-intervention and post-intervention questionnaires. There was a significant improvement in confidence in performing the E-FAST after the intervention [p<0.001]. There was also a significant improvement in ultrasound knowledge and image interpretation skills. For the formative assessment, the mean score was 31.8 out of 33 and 22 of 27 students (82%) passed the assessment. There is no described implementation of clinically focused point-of-care ultrasound education in Australian medical schools CONCLUSIONS: We have demonstrated that a focused curriculum can improve POCUS knowledge and skills. The curriculum was feasible and well received. With global trends to include POCUS in medical education, Australian institutions should consider upskilling their medical graduates.

Preliminary Experience Using Eye-Tracking Technology to Differentiate Novice and Expert Image Interpretation for Ultrasound-Guided Regional Anesthesia.

Objective measures are needed to guide the novice's pathway to expertise. Within and outside medicine, eye tracking has been used for both training and assessment. We designed this study to test the hypothesis that eye tracking may differentiate novices from experts in static image interpretation for ultrasound (US)-guided regional anesthesia. We recruited novice anesthesiology residents and regional anesthesiology experts. Participants wore eye-tracking glasses, were shown 5 sonograms of US-guided regional anesthesia, and were asked a series of anatomy-based questions related to each image while their eye movements were recorded. The answer to each question was a location on the sonogram, defined as the area of interest (AOI). The primary outcome was the total gaze time in the AOI (seconds). Secondary outcomes were the total gaze time outside the AOI (seconds), total time to answer (seconds), and time to first fixation on the AOI (seconds). Five novices and 5 experts completed the study. Although the gaze time (mean ± SD) in the AOI was not different between groups (7 ± 4 seconds for novices and 7 ± 3 seconds for experts; P = .150), the gaze time outside the AOI was greater for novices (75 ± 18 versus 44 ± 4 seconds for experts; P = .005). The total time to answer and total time to first fixation in the AOI were both shorter for experts. Experts in US-guided regional anesthesia take less time to identify sonoanatomy and spend less unfocused time away from a target compared to novices. Eye tracking is a potentially useful tool to differentiate novices from experts in the domain of US image interpretation.

Imaging Performance of a Handheld Ultrasound System With Real-Time Computer-Aided Detection of Lumbar Spine Anatomy: A Feasibility Study.

The aim of this study was to evaluate the imaging performance of a handheld ultrasound system and the accuracy of an automated lumbar spine computer-aided detection (CAD) algorithm in the spines of human subjects. This study was approved by the institutional review board of the University of Virginia. The authors designed a handheld ultrasound system with enhanced bone image quality and fully automated CAD of lumbar spine anatomy. The imaging performance was evaluated by imaging the lumbar spines of 68 volunteers with body mass index between 18.5 and 48 kg/m. The accuracy, sensitivity, and specificity of the lumbar spine CAD algorithm were assessed by comparing the algorithm's results to ground-truth segmentations of neuraxial anatomy provided by radiologists. The lumbar spine CAD algorithm detected the epidural space with a sensitivity of 94.2% (95% confidence interval [CI], 85.1%-98.1%) and a specificity of 85.5% (95% CI, 81.7%-88.6%) and measured its depth with an error of approximately ±0.5 cm compared with measurements obtained manually from the 2-dimensional ultrasound images. The spine midline was detected with a sensitivity of 93.9% (95% CI, 85.8%-97.7%) and specificity of 91.3% (95% CI, 83.6%-96.9%), and its lateral position within the ultrasound image was measured with an error of approximately ±0.3 cm. The bone enhancement imaging mode produced images with 5.1- to 10-fold enhanced bone contrast when compared with a comparable handheld ultrasound imaging system. The results of this study demonstrate the feasibility of CAD for assisting with real-time interpretation of ultrasound images of the lumbar spine at the bedside.

The feasibility and efficacy of implementing a focused cardiac ultrasound course into a medical school curriculum

BackgroundTeaching cardiac ultrasound to medical students in a brief course is a challenge. We aimed to evaluate the feasibility of teaching large groups of medical students the acquisition and interpretation of cardiac ultrasound images using a pocket ultrasound device (PUD) in a short, specially designed course.MethodsThirty-one medical students in their first clinical year participated in the study. All were novices in the use of cardiac ultrasound. The training consisted of 4 hours of frontal lectures and 4 hours of hands-on training. Students were encouraged to use PUD for individual practice. Finally, the students’ proficiency in the acquisition of ultrasound images and their ability to recognize normal and pathological states were evaluated.ResultsSixteen of 27 (59%) students were able to demonstrate all main ultrasound views (parasternal, apical, and subcostal views) in a six-minute test. The most obtainable view was the parasternal long-axis view (89%) and the least obtainable was the subcostal view (58%). Ninety-seven percent of students correctly differentiated normal from severely reduced left ventricular function, 100% correctly differentiated a normal right ventricle from a severely hypokinetic one, 100% correctly differentiated a normal mitral valve from a rheumatic one, and 88% correctly differentiated a normal aortic valve from a calcified one, while 95% of them correctly identified the presence of pericardial effusion.ConclusionsTraining of medical students in cardiac ultrasound during the first clinical year using a short, focused course is feasible and enables students with modest ability to acquire the main transthoracic ultrasound views and gain proficiency in the diagnosis of a limited number of cardiac pathologies.

Plantar fascia segmentation and thickness estimation in ultrasound images

Ultrasound (US) imaging offers significant potential in diagnosis of plantar fascia (PF) injury and monitoring treatment. In particular US imaging has been shown to be reliable in foot and ankle assessment and offers a real-time effective imaging technique that is able to reliably confirm structural changes, such as thickening, and identify changes in the internal echo structure associated with diseased or damaged tissue. Despite the advantages of US imaging, images are difficult to interpret during medical assessment. This is partly due to the size and position of the PF in relation to the adjacent tissues. It is therefore a requirement to devise a system that allows better and easier interpretation of PF ultrasound images during diagnosis. This study proposes an automatic segmentation approach which for the first time extracts ultrasound data to estimate size across three sections of the PF (rearfoot, midfoot and forefoot). This segmentation method uses artificial neural network module (ANN) in order to classify small overlapping patches as belonging or not-belonging to the region of interest (ROI) of the PF tissue. Features ranking and selection techniques were performed as a post-processing step for features extraction to reduce the dimension and number of the extracted features. The trained ANN classifies the image overlapping patches into PF and non-PF tissue, and then it is used to segment the desired PF region. The PF thickness was calculated using two different methods: distance transformation and area-length calculation algorithms. This new approach is capable of accurately segmenting the PF region, differentiating it from surrounding tissues and estimating its thickness.

Fast feature-preserving speckle reduction for ultrasound images via phase congruency

Ultrasonography is widely used in clinical diagnosis and therapeutic procedures, but speckle noise often obscures important features and complicates interpretation and analysis of ultrasound images. In this regard, speckle reduction is a crucial prerequisite of many computer aided ultrasound diagnosis and treatment systems However, removing speckle noise while simultaneously preserving features in ultrasound images is a challenging task. We propose a novel optimization framework for speckle reduction by leveraging the concept of phase congruency and incorporating a feature asymmetry metric into the regularization term of the objective function to effectively distinguish the features and speckle noise. The feature asymmetry metric can productively separate features from speckle noise by analyzing the local frequency information. Compared with traditional methods employing intensity gradients as regularization terms, our framework is invariant to the intensity amplitude of features so that low contrast features are almost equally protected as high contrast features. In addition, rather than adopting the gradient descent, we propose a novel solver by decomposing the original non-convex optimization into solving several linear systems, leading to an efficient solution of the optimization. Owing to different penalties on speckle noise and features, our method can efficiently remove speckle noise and preserve features at the same time. Experiments on simulated and real ultrasound images demonstrate our method can better maintain features with speckle removal than state-of-the-art methods, especially for the low contrast features.

Basics of ultrasonographic examination in sheep

Diagnostic ultrasound has become an essential tool in veterinary medicine and is well established in a variety of species. Since its introduction into sheep clinical practice, it has been mainly used in pregnancy diagnosis and identification of litter size, as well as in experimental settings in reproductive medicine. Furthermore, sheep frequently serve as an experimental model in several disciplines and sub-disciplines of human medicine. However, despite of its great potential, ultrasonography does not yet play a major role in general clinical practice. The review focuses on ultrasonography in sheep and provides information in basics of ultrasound imaging. These include physical principles, instrumentation, handling and preparation of animals, imaging modes (amplitude-depth mode, brightness mode, time-motion mode, 3- and 4-dimensional ultrasound, Doppler ultrasonography), and interpretation of ultrasound images and artefacts. Examples of clinical and experimental applications are included with regard to the different imaging modes, to demonstrate how ultrasonography can be used in sheep clinical practice and research. Additionally, animal welfare as well as safety and hygiene aspects are considered.

Teaching Old Dogs New Tricks: A Course for Faculty Learners on Fundamentals of Critical Care Ultrasonography (FoCUS)

SESSION TITLE: Education SESSION TYPE: Original Investigation Slide PRESENTED ON: Tuesday, October 25, 2016 at 02:45 PM - 04:15 PM PURPOSE: Point-of-care ultrasonography has rapidly become a necessary skillset for clinicians caring for critically ill patients. Fellows, residents, and medical students are increasingly being taught how to perform and understand bedside ultrasound (US). Faculty sufficiently competent in critical care ultrasonography to supervise and teach these learners, however, are lacking. Moreover, many faculty feel left behind by the emergence of this modality. We developed a 1-day course specifically tailored to critical care faculty without extensive ultrasound experience in order to develop baseline competence and foster confidence in fundamental critical care ultrasound examinations. METHODS: All faculty learners completed pre- and post-course surveys recording learner characteristics, confidence on a 5-point Likert scale for performing specific US exams and identifying relevant pathology, effectiveness of specific components of the course, and desires for future US training. All faculty learners also completed pre- and post-tests regarding US knowledge and image interpretation. Prerequisite reading materials were provided prior to the course. The course was a combination of didactic lectures (Introduction to US, US for Vascular Access, Lung/Pleural US, Abdominal US, focused TTE, and DVT study) with 5 hours of hands-on small group sessions with expert faculty in our simulation center using standardized patients and task trainers. RESULTS: Eleven critical care faculty learners completed the course. The majority (54%) had ≥15 years of Attending experience, with 64% having had no prior US training. Mean pre-test score was 58±17% and mean post-test score was 73±14% for the group, p=0.006. Faculty learners felt more confident in using ultrasound in performance of procedures and diagnostic examinations after the course (catheter placement, thoracentesis, paracentesis, DVT study, TTE, FAST exam; all p<0.05), and felt more confident in identifying pathology on US after the course (IVC respiratory variation, severe valvular regurgitation, pericardial effusion, hypo/hyper-dynamic left ventricle, loculated pleural effusion, pneumothorax, hydronephrosis, and ascites; all p<0.05). Faculty learners were very interested in future US-based faculty development, especially US for Acute Respiratory Failure and US for Acute Shock. CONCLUSIONS: A 1-day course for critical care faculty with limited ultrasound experience was well received and resulted in tangible improvements in knowledge and image interpretation while increasing confidence performing basic critical care ultrasound examinations/procedures and identifying relevant pathology on ultrasound imaging. CLINICAL IMPLICATIONS: Critical care faculty with limited prior ultrasound experience can learn and improve their ultrasound knowledge, image acquisition, and image interpretation via a short course targeting this learner group. DISCLOSURE: The following authors have nothing to disclose: Deepak Pradhan, Bishoy Zakhary, Vikramjit Mukherjee, Harald Sauthoff No Product/Research Disclosure Information

Integrating Ultrasound Teaching into Preclinical Problem-based Learning.

Objectives:The aim is to provide students in the preclinical with ultrasound image interpretation skills. Research question: Are students in smaller groups with access to a combination of lectures and hands-on patient contact most likely to have better ultrasound image interpretation skills, than students in larger groups with only interactive didactic lectures?Methodology:First-year students at the preclinical Program of the College of Medicine, participated in two 2-h introductory interactive ultrasound sessions. The study comprised two cohorts: 2012/2013 students, who were offered large group teaching (LGT) sessions (control group), and 2013/2014 students, who received the intervention in small group learning problem-based learning (PBL) sessions (experimental group). The overall learning objectives were identical for both groups. The success of the module was evaluated using pre- and post-tests as well as students’ feedback.Results:The students in the experimental group showed significantly higher scores in interpretations of images than those in the control group. The experimental group showed achievement of learning outcomes along with higher levels of satisfaction with the module compared to the latter.Conclusion:Posttest knowledge of the basics of ultrasound improved significantly over the pretest in the experimental group. In addition, students’ overall satisfaction of the ultrasound module was shown to be higher for the PBL compared to the LGT groups. Small groups in an interactive and PBL setting along with opportunities for hands-on practice and simultaneous visualization of findings on a high definition screen should enhance preclinical student learning of the basics of ultrasound. Despite the potential of ultrasound as a clinical, teaching and learning tool for students in the preclinical years, standardized recommendations have yet to be created regarding its integration into the curricula within academic institutions and clinical medicine. The interactive and PBL is here to stay at the college of medicine. Further research would be carried out to see if this trend persists in the upcoming vertical system-based curriculum of the college of medicine.

AIUM Practice Parameter for the Performance of Selected Ultrasound‐Guided Procedures

AIUM Practice Parameter for the Performance of Selected Ultrasound‐Guided Procedures

AIUM Practice Parameter for the Performance of Selected Ultrasound-Guided Procedures

AIUM Practice Parameter for the Performance of Selected Ultrasound-Guided Procedures

Hemodynamic Bedside Ultrasound Image Quality and Interpretation After Implementation of a Training Curriculum for Pediatric Critical Care Medicine Providers*

Bedside ultrasound for hemodynamic evaluation in critically ill children is increasingly recognized as an important skill for pediatric critical care medicine providers. Our institution implemented a training curriculum leading to institutional credentialing for pediatric critical care providers in nonprocedural bedside ultrasound core applications. We hypothesized that hemodynamic studies performed or supervised by credentialed providers (credentialed providers group) have better image quality and greater accuracy in interpretation than studies performed by non-credentialed providers without supervision (non-credentialed providers group). Retrospective descriptive study. Single-center tertiary non-cardiac 55-bed PICU in a children's hospital. Patients from October 2013 to January 2015, with hemodynamic bedside ultrasound performed and interpreted by pediatric critical care providers exposed to bedside ultrasound training. A cardiologist blinded to performer scored hemodynamic bedside ultrasound image quality for five core cardiac views (excellent = 3, good = 2, fair = 1, unacceptable = 0; median = quality score) and interpretation within 5 hemodynamic domains (agreement = 3, minor disagreement = 2, major disagreement = 1; median = interpretation score), as well as a global assessment of interpretation. Eighty-one studies (45 in the credentialed providers group and 36 in the non-credentialed providers group) were evaluated. There was no statistically significant difference in quality score between groups (median: 1.4 [interquartile range: 0.8-1.8] vs median: 1.2 [interquartile range: 0.75-1.6]; p = 0.14]. Studies in the credentialed providers group had higher interpretation score than those in the non-credentialed providers group (median: 3 [interquartile range: 2.5-3) vs median: 2.67 [interquartile range: 2.25-3]; p = 0.04). Major disagreement between critical care provider and cardiology review occurred in 25 of 283 hemodynamic domains assessed (8.8%), with no statistically significant difference between credentialed providers and non-credentialed providers groups (6.1% vs 11.9%; p = 0.12). Hemodynamic bedside ultrasound performed or supervised by credentialed pediatric critical care providers had more accurate interpretation than studies performed by unsupervised non-credentialed providers. A rigorous pediatric critical care medicine bedside ultrasound credentialing program can train intensivists to attain adequate images and interpret those images appropriately.

Quels signes échographiques doit-on rechercher (2D/3D) pour affirmer la bonne localisation des implants tubaires ? À propos d’une étude rétrospective de 92 cas

ObjectiveThe aim of this study is to propose an analysis of the route and the curvature of the Essure® system in 3D ultrasound to determine their correct location so as not to miss a tubal perforation. MethodsThis is a retrospective single center study during 2 years analyzing 92 3D ultrasound performed by a single sonographer. Implant placement was performed by different operators. One prescribed 3D ultrasound control in case of difficulties with the installation; other indications where systematics. ResultsThe Essure®’s position is right in 79% of cases. Twenty-eight implants appear incorrectly positioned on ultrasound 3D positioning. Abnormalities found are those described by the classification proposed in the literature. In one case, the curvature of the implant, not taking into account this classification, allowed to evoke a tubal perforation. Among the 28 cases of non-visualized implants in place, 3 cases of wrong position of the implant were confirmed by additional examinations (laparoscopy or HSG). For 14 cases, the ASP or HSG disproved the wrong position of the implants. In other cases, we did not have other complementary examinations. ConclusionUltrasonography 3D seems to be the method of choice to control implants for simple implementation and good reproducibility. However, the interpretation of 3D ultrasound images is sometimes difficult. The study of the curvature of the implant should be systematically analyzed not to miss a tubal perforation.

Basic interpretation of intracoronary ultrasound and optical coherence tomography images: examples

Basic interpretation of intracoronary ultrasound and optical coherence tomography images: examples

Registration of Pre- and Postresection Ultrasound Volumes With Noncorresponding Regions in Neurosurgery.

Brain tissue deforms significantly after opening the dura and during tumor resection, invalidating preoperative imaging data. Ultrasound is a popular imaging modality for providing the neurosurgeon with real-time updated images of brain tissue. Interpretation of postresection ultrasound images is difficult due to large brain shift and tissue resection. Furthermore, several factors degrade the quality of postresection ultrasound images such as the strong reflection of waves at the interface of saline water and brain tissue in resection cavities, air bubbles, and the application of blood-clotting agents around the edges of resection. Image registration allows the comparison of postresection ultrasound images with higher quality preresection images, assists in interpretation of postresection images and may help identify residual tumor, and, as such, is of significant clinical importance. In this paper, we propose a nonrigid symmetric registration (NSR) framework for accurate alignment of pre- and postresection volumetric ultrasound images in near real time. We first formulate registration as minimization of a regularized cost function and analytically derive its derivative to efficiently optimize the cost function. An outlier detection algorithm is proposed and utilized in this framework to identify noncorresponding regions (outliers) and therefore improve the robustness and accuracy of registration. We use an efficient second-order minimization method for fast and robust optimization. Furthermore, we exploit a symmetric and inverse-consistent method to generate realistic deformation fields. The results show that NSR significantly improves the quality of the alignment between pre- and postresection ultrasound images.

No Longer a Novelty&amp;[hellip]. Formalizing Ultrasound Based Activities in Gross Anatomy through Objective Structural Practical Examination

Use of ultrasound as a diagnostic tool is fast becoming an expectation in clinical practice. With formal outcomes now being included as competencies in many residency programs, earlier exposure to use of ultrasound is being achieved through informal activities embedded within medical school curricula. Using ultrasound to learn anatomy is now standard practice for many courses in the United States. However, very few if any offer formal opportunities to assess student skill and anatomical knowledge at the same time. As part of the Mini Ultrasound Skills Institute (MUSI) in Anatomy, a component of the first year anatomy course, students were expected to complete a team‐based, hands‐on ultrasound skills assessment. Throughout the course, students received both didactic briefing sessions and hands‐on training on ultrasound basics, image interpretation, and scanning techniques. The ultrasound content was presented in parallel to anatomy and radiology curriculum. In the past, ultrasound knowledge was assessed via static images; however, due to the contextual nature of ultrasound interpretation, a live, hands‐on method was devised to assess student learning. At regular intervals during the course, students completed an ultrasound skills assessment. In their dissection teams of four, students completed four tasks: setup the machine, obtained, oriented, and labeled a desired image using a live model. Each student was given a unique role within this exercise. Students were graded on a rubric scale for each of the four components and given an overall team score. At the conclusion of the course, students completed a graded ultrasound objective structural practical examination (OSPE). Rotating through four stations, students completed the four tasks on four different structures. Students rotated through roles within their team as they moved between stations. OSPE grade was incorporated into their final laboratory practical grade. Preliminary data from formative OSPE assessments ‐‐‐‐showed an average score of 72% (±21%) with a minimum of 50%, maximum of 100%. Within the skill specific breakdown, students were very comfortable at labeling obtained images (100%) and had the most challenging time with either selecting the correct probe or setting up the scanner 40% (±51%). Overall, team‐based OSPE is an effective method to promote and evaluate ultrasound skills learned in an anatomy course.

Segmentation of the spinous process and its acoustic shadow in vertebral ultrasound images

Spinal ultrasound imaging is emerging as a low-cost, radiation-free alternative to conventional X-ray imaging for the clinical follow-up of patients with scoliosis. Currently, deformity measurement relies almost entirely on manual identification of key vertebral landmarks. However, the interpretation of vertebral ultrasound images is challenging, primarily because acoustic waves are entirely reflected by bone. To alleviate this problem, we propose an algorithm to segment these images into three regions: the spinous process, its acoustic shadow and other tissues. This method consists, first, in the extraction of several image features and the selection of the most relevant ones for the discrimination of the three regions. Then, using this set of features and linear discriminant analysis, each pixel of the image is classified as belonging to one of the three regions. Finally, the image is segmented by regularizing the pixel-wise classification results to account for some geometrical properties of vertebrae. The feature set was first validated by analyzing the classification results across a learning database. The database contained 107 vertebral ultrasound images acquired with convex and linear probes. Classification rates of 84%, 92% and 91% were achieved for the spinous process, the acoustic shadow and other tissues, respectively. Dice similarity coefficients of 0.72 and 0.88 were obtained respectively for the spinous process and acoustic shadow, confirming that the proposed method accurately segments the spinous process and its acoustic shadow in vertebral ultrasound images. Furthermore, the centroid of the automatically segmented spinous process was located at an average distance of 0.38mm from that of the manually labeled spinous process, which is on the order of image resolution. This suggests that the proposed method is a promising tool for the measurement of the Spinous Process Angle and, more generally, for assisting ultrasound-based assessment of scoliosis progression.

A Fusion-Based Approach for Breast Ultrasound Image Classification Using Multiple-ROI Texture and Morphological Analyses.

Ultrasound imaging is commonly used for breast cancer diagnosis, but accurate interpretation of breast ultrasound (BUS) images is often challenging and operator-dependent. Computer-aided diagnosis (CAD) systems can be employed to provide the radiologists with a second opinion to improve the diagnosis accuracy. In this study, a new CAD system is developed to enable accurate BUS image classification. In particular, an improved texture analysis is introduced, in which the tumor is divided into a set of nonoverlapping regions of interest (ROIs). Each ROI is analyzed using gray-level cooccurrence matrix features and a support vector machine classifier to estimate its tumor class indicator. The tumor class indicators of all ROIs are combined using a voting mechanism to estimate the tumor class. In addition, morphological analysis is employed to classify the tumor. A probabilistic approach is used to fuse the classification results of the multiple-ROI texture analysis and morphological analysis. The proposed approach is applied to classify 110 BUS images that include 64 benign and 46 malignant tumors. The accuracy, specificity, and sensitivity obtained using the proposed approach are 98.2%, 98.4%, and 97.8%, respectively. These results demonstrate that the proposed approach can effectively be used to differentiate benign and malignant tumors.