Abdominal Ultrasound Imaging Research Articles

SEGMENTATION OF KIDNEY STONE REGION IN ULTRA SOUND IMAGEBY USING REGION PARTITION AND MOUNTING SEGMENTATION ALGORITHM (RPM)

One of the safest techniques for disease diagnosis which can be used in any part of the body is ultrasound imaging. The cost when compared with MRI, PET etc are higher than using ultra sound images is the one of the major reason. Further, it is an efficient technique for initial diagnosis and it is free from any radiation exposure. This paper concentrates on segmentation of kidney from abdominal ultrasound images. There are many common ailments affecting kidney. Hence conducting study on this segmented image becomes easy with an efficient segmentation technique. In this paper Various algorithms to pull out kidney regions from abdominal ultrasound images which are discussed by many researchers are also investigated. Due to the complicated internal organs of the abdominal region, extraction of only the kidney region is very challenging and is the major drawback of ultrasound imaging. a new technique where the collected abdominal ultrasound image is cleaned, to remove unwanted noise produced due to various interferences has been processed by this paper, the kidney region is segmented after applying the filtering technique. the subjected to Region indicator contour segmentation method to extract the renal calculi which is the region of interest in this study is this extracted kidney image. with a reasonable number of dataset and applied the statistical performance test to check for the accuracy , the method is experimented .

Giant Abdominal Lipoma in Adult

Lipomas arising from the omentum are extremely rare in adults. Omental lipomas are typically asymptomatic, but very large ones may cause nonspecific abdominal symptoms and discomfort. Rarely they can cause omental torsion and present with an acute abdomen. We report a 41-year-old female patient with a giant lipoma (40 × 26 × 8 cm and 11,520 g) who presented with mild abdominal discomfort. Workup included abdominal ultrasound (USG) and magnetic resonance imaging (MRI). Surgical resection was performed without complication. No recurrence was observed during 4-year follow-up.

Deep learning-based automated quantification of the hepatorenal index for evaluation of fatty liver by ultrasonography.

PurposeThe aim of this study was to develop and validate a fully-automatic quantification of the hepatorenal index (HRI) calculated by a deep convolutional neural network (DCNN) comparable to the interpretations of radiologists experienced in ultrasound (US) imaging.MethodsIn this retrospective analysis, DCNN-based organ segmentation with Gaussian mixture modeling for automated quantification of the HRI was developed using abdominal US images from a previous study. For validation, 294 patients who underwent abdominal US examination before living-donor liver transplantation were selected. Interobserver agreement for the measured brightness of the liver and kidney and the calculated HRI were analyzed between two board-certified radiologists and DCNN using intraclass correlation coefficients (ICCs).ResultsMost patients had normal (n=95) or mild (n=198) fatty liver. The ICCs of hepatic and renal brightness measurements and the calculated HRI between the two radiologists were 0.892 (95% confidence interval [CI], 0.866 to 0.913), 0.898 (95% CI, 0.873 to 0.918), and 0.681 (95% CI, 0.615 to 0.738) for the first session and 0.920 (95% CI, 0.901 to 0.936), 0.874 (95% CI, 0.844 to 0.898), and 0.579 (95% CI, 0.497 to 0.650) for the second session, respectively; the results ranged from moderate to excellent agreement. Using the same task, the ICCs of the hepatic and renal measurements and the calculated HRI between the average values of the two radiologists and DCNN were 0.919 (95% CI, 0.899 to 0.935), 0.916 (95% CI, 0.895 to 0.932), and 0.734 (95% CI, 0.676 to 0.782), respectively, showing high to excellent agreement.ConclusionAutomated quantification of HRI using DCNN can yield HRI measurements similar to those obtained by experienced radiologists in patients with normal or mild fatty liver.

Imaging prediction with ultrasound and MRI of long-term medical outcome in native liver survivor patients with biliary atresia after kasai portoenterostomy: a pilot study.

To comparatively assess the role of abdominal ultrasound (US) and magnetic resonance imaging (MRI) in predicting long-term medical outcome in native liver survivor patients with biliary atresia (BA) after Kasai portoenterostomy (KP). Twenty-four retrospectively enrolled patients were divided in two groups according to clinical and laboratory data at initial evaluation after KP (median follow-up = 9.7years; range = 5-25years) as with ideal (Group 1; n = 15) or non-ideal (Group 2; n = 9) medical outcome. All patients were re-evaluated for a period of additional 4years using clinical and laboratory indices. US and MRI studies were qualitatively analyzed assessing imaging signs suggestive of chronic liver disease (CLD). At re-evaluation, 6 patients (40%) of Group 1 changed their medical outcome in non-ideal (Group 1A); the other 9 patients (60%) remained stable (Group 1B); the mean time to change the medical outcome in non-ideal status at re-evaluation was 43.5 ± 2.3months. The area under the ROC curve was 0.84 and 0.87 for US and MRI scores to predict long-term medical outcome with the best cut-off value score > 4 for both modalities (p = 0.89). In Group 2, 6 (67%) patients showed a clinical progression (Group 2A) with a mean time of 39.8 ± 3.8months; in the other 3 (33%) patients, no clinical progression was observed (Group 2B). In BA patients with ideal medical outcome after KP, US and MRI may both predict long-term outcome. US, non-invasive and widely available technique, should be preferred.

Liver-ultrasound-guided lung tumour tracking for scanned proton therapy: a feasibility study

Pencil beam scanned (PBS) proton therapy of lung tumours is hampered by respiratory motion and the motion-induced density changes along the beam path. In this simulation study, we aim to investigate the effectiveness of proton beam tracking for lung tumours both under ideal conditions and in conjunction with a respiratory motion model guided by real-time ultrasound imaging of the liver. Multiple-breathing-cycle 4DMRIs of the thorax and abdominal 2D ultrasound images were acquired simultaneously for five volunteers. Deformation vector fields extracted from the 4DMRI, referred to as ground truth motion, were used to generate 4DCT(MRI) data sets of two lung cancer patients, resulting in 10 data sets with variable motion patterns. Given the 4DCT(MRI) and the corresponding ultrasound images as surrogate data, a patient-specific motion model was built. The model consists of an autoregressive model and Gaussian process regression for the temporal and spatial prediction, respectively. Two-field PBS plans were optimised on the reference CTs, and 4D dose calculations (4DDC) were used to simulate dose delivery for (a) unmitigated motion, (b) ideal 2D and 3D tracking (both beam adaption and 4DDC based on ground truth motion), and (c) realistic 2D and 3D tracking (beam adaption based on motion predictions, 4DDC on ground truth motion). Model-guided tracking retrieved clinically acceptable target dose homogeneity, as seen in a substantial reduction of the D5%–D95% compared to the non-mitigated simulation. Tracking in 2D and 3D resulted in a similar improvement of the dose homogeneity, as did ideal and realistic tracking simulations. In some cases, however, the tracked deliveries resulted in a shift towards higher or lower dose levels, leading to unacceptable target over- or under-coverage. The presented motion modelling framework was shown to be an accurate motion prediction tool for the use in proton beam tracking. Tracking alone, however, may not always effectively mitigate motion effects, making it necessary to combine it with other techniques such as rescanning.

Hepatic angiosarcoma in a 6-month-old infant

We report a case of 6-month-old infant presented with abdominal distension in whom an abdominal ultrasound and hepatic magnetic resonance imaging (MRI) revealed a large hepatic mass of malignant appearance.

Feature Extraction from Uterus Signal and Image for Diagnosing Pregnancy Complications

Preeclampsia is a disorder that arises during pregnancy which results in maternal death during delivery, fetal death in the womb and growth retention in infants based on the degree of severity and duration of onset. The markers for identification are found to be Strength of uterine muscles, Decreased velocity & volume of Uterine Artery, Poor placentation, Deficient in Remodeling of Spiral artery. The proposed work emphasis on extraction of features of EHG signal and Ultrasound Image of normal and suspected preeclampsia patients at second trimester to identify the condition at the earliest. The ability of the Uterine Muscle can be identified by recording of electrical activity of the uterus, by a device called Electrohysterograph that uses surface bipolar electrodes placed at various points in the lower abdominal of pregnant women that can indicate the possibility of Preterm labor. The anatomy of Uterus, an indicator of weaker cervix and anatomy of kidney can be analyzed by abdominal Ultrasound Imaging. Based on the combined analysis of images and signal along with the support of biochemical tests it is possible to identify the disease at the start of second trimester.

Liver-ultrasound based motion modelling to estimate 4D dose distributions for lung tumours in scanned proton therapy

Motion mitigation strategies are crucial for scanned particle therapy of mobile tumours in order to prevent geometrical target miss and interplay effects. We developed a patient-specific respiratory motion model based on simultaneously acquired time-resolved volumetric MRI and 2D abdominal ultrasound images. We present its effects on 4D pencil beam scanned treatment planning and simulated dose distributions. Given an ultrasound image of the liver and the diaphragm, principal component analysis and Gaussian process regression were applied to infer dense motion information of the lungs. 4D dose calculations for scanned proton therapy were performed using the estimated and the corresponding ground truth respiratory motion; the differences were compared by dose difference volume metrics. We performed this simulation study on 10 combined CT and 4DMRI data sets where the motion characteristics were extracted from 5 healthy volunteers and fused with the anatomical CT data of two lung cancer patients. Median geometrical estimation errors below 2 mm for all data sets and maximum dose differences of = 43.2% and = 16.3% were found. Moreover, it was shown that abdominal ultrasound imaging allows to monitor organ drift. This study demonstrated the feasibility of the proposed ultrasound-based motion modelling approach for its application in scanned proton therapy of lung tumours.

Utility of quantitative ultrasound in community screening for hepatic steatosis

Nonalcoholic fatty liver disease (NAFLD) is the most common chronic liver disease. Quantitative ultrasound facilitates clinical grading of hepatic steatosis (the early stage of NAFLD). However, the utility of quantitative ultrasound as a first-line method for community screening of hepatic steatosis remains unclear. Therefore, this study aimed to investigate the utility of quantitative ultrasound to screen for hepatic steatosis and for metabolic evaluation at the community level. In total, 278 participants enrolled from a community satisfied the study criteria. Each subject underwent anthropometric and biochemical examinations, and abdominal ultrasound imaging was performed to measure the controlled attenuation (CAP), integrated backscatter (IB), and information Shannon entropy (ISE). The assessment outcomes were compared with the fatty liver index (FLI), hepatic steatosis index (HSI), metabolic syndrome (MetS), and insulin resistance to evaluate the screening performance through the area under the receiver operating characteristic curve (AUROC) and Delong’s test. Ultrasound ISE, CAP, and IB were effective in screening hepatic steatosis, MetS, and insulin resistance. In screening for hepatic steatosis, the AUROCs of ISE, CAP, and IB were 0.85, 0.83, and 0.80 (the cutoff FLI = 60), respectively, and 0.84, 0.75, 0.77 (the cutoff HSI = 36), respectively, and those for the evaluation of MetS and insulin resistance were 0.79, 0.75, 0.79, respectively, and 0.83, 0.76, 0.78, respectively. Delong’s test revealed that ISE outperformed CAP and IB for the detection of hepatic steatosis and insulin resistance (P < .05). Based on the present results, ultrasound ISE is a potential imaging biomarker during first-line community screening of hepatic steatosis and insulin resistance.

Bile thickening syndrome in newborns and young children: risk factors, therapy, prognosis. А clinical case study

Bile thickening syndrome or biliary sludge causing obstruction of the common bile duct and impaired passage of bile into the duodenum ranks second among “surgical jaundice” in newborns and young children. The syndrome should be diagnosed early to conduct timely conservative treatment to avoid surgical intervention and complications such as acute pancreatitis, acute cholecystitis, cholelithiasis. Risk factors for bile thickening syndrome include prematurity, extremely low birth weight, long term parenteral nutrition, administration of diuretics, third generation cephalosporins. If a threat of biliary sludge arises, repeat abdominal ultrasound imaging and identification of the following signs of this syndrome are indicated: dilatation of CBD more than 3 mm, the presence of heterogeneous sediment in the gallbladder and common bile duct. The article presents a clinical discussion of a bile thickening syndrome case in a an extremely low birth weight girl born at 29 weeks’ gestational age. The condition of the newborn at birth was very severe. The severity of the condition was caused by the type 1 respiratory distress syndrome, the girl was immediately intubated and mechanically ventilated. At the age of 1 month 10 days, the girl was transferred from the intensive care unit to the neonatal unit. The ultrasonic imaging showed an oval-shaped gallbladder of 26 x 6 mm in size, 0.5 ml in volume, instable angulation of the gallbladder body and neck, the walls of up to 2 mm in thickness, laminated, hyperechogenic, hyperechoic contents in the lumen (ointment-like bile), which allowed to draw a conclusion about the presence of bile thickening syndrome. Ursodeoxycholic acid at a dose of 20 mg/kg/day was prescribed immediately after ultrasonic imaging. 10 days later the dose of the drug was increased to 30 mg/kg/day. Two weeks later, the child was discharged home in satisfactory condition without signs of bile thickening syndrome. Knowledge of the risk factors for biliary sludge makes it possible to assign timely instrumental examination, establish early diagnosis and prescribe conservative therapy, and exclude surgical methods of treatment.

Crisis adrenérgica como forma de debut de un neuroblastoma

La forma clínica de presentación más común del neuroblastoma es el de una masa abdominal, pero puede presentarse con sintomatología menos habitual, como es la crisis adrenérgica por liberación de catecolaminas.Objetivo: Describir una forma de presentación inusual de neuroblastoma y el amplio diagnóstico diferencial que existe en un lactante con síntomas adrenérgicos.Caso Clínico: Lactante femenina de 7 semanas de vida, consultó por historia de tres semanas de sudoración e irritabilidad a lo que se asoció fiebre de 24 h de evolución y dificultad respiratoria. Al ingreso presentaba mal estado general, irritabilidad, sudoración, enrojecimiento facial, taquipnea y palidez cutánea, taquicardia sinusal extrema e hipertensión arterial (HTA), interpretadas como sintomatología adrenérgica. Se completó el estudio con una ecografía abdominal y resonancia magnética que mostraron una gran masa retroperitoneal compatible con neuroblastoma. Las catecolaminas en sangre y en orina mostraron altos niveles de dopamina, adrenalina y noradrenalina, probablemente de origen tumoral. Se inició tratamiento antihipertensivo con fármacos alfa bloqueantes con buen control de la tensión arterial. Se resecó quirúrgicamente el tumor sin incidencias y con una adecuada recuperación posterior. La paciente presentó evolución favorable a tres años de seguimiento.Conclusiones: en un lactante con sintomatología adrenérgica como irritabilidad, enrojecimiento, sudoración asociada a HTA, se debe descartar patología cardiaca, metabólica (hipoglucemia), intoxicaciones y/o patología suprarrenal. Dentro de esta última, el neuroblastoma es la primera posibilidad diagnóstica, por ser uno de los principales tumores en la infancia y aunque esta presentación no es habitual puede producir estos síntomas.

Neoplasia sólida pseudopapilar del páncreas. Reporte de un caso

Background. Solid pseudopapillary neoplasm of the pancreas is an uncommon exocrine pancreatic tumor of epithelial origin, which represents only 1% to 2% of all pancreatic tumors and mainly affects young women. Objective. To describe clinical and imaging features of these rare entity in a symptomatic young woman. Case report. 41-year-old female patient, without morbid history, consulted for intermittent epigastralgia of one-month evolution. Abdominal ultrasound and magnetic resonance imaging demonstrated the presence of a solid-cystic mass of neoplastic appearance in the body-tail of the pancreas. The case was presented to the institutional hepato-biliary committee where surgical resection was decided. Discussion. Although solid pseudopapillary neoplasm of the pancreas are very uncommon pancreatic tumors, it is necessary to keep them within the differential diagnosis on the one hand due to their low but existing malignant potential and on the other hand due to their good prognosis with surgical treatment.

Increased body mass index is associated with decreased imaging quality of point-of-care abdominal aortic ultrasonography.

Given that abdominal aorta is a retroperitoneal structure, increased body mass index (BMI) may have an adverse effect upon the quality of aortic ultrasonographic imaging. To assess the hypothesis that increased BMI is associated with worsening point-of-care abdominal aortic ultrasonographic image quality. This is a retrospective single-center study of point-of-care abdominal aortic ultrasound examinations performed in an academic emergency department (ED) with fellowship-trained emergency ultrasonography faculty performing quality assurance review. Mean ± SD BMI was 27.4 ± 6.2, among the 221 included records. The overall quality rating decreased as BMI increased (correlation coefficient - 0.24; P < .001) and this persisted after adjustment for age and sex (P < .001). Although BMI was higher on average in the records that were of insufficient quality for clinical decisions when compared with those of sufficient quality (mean BMI 28.7 vs 27.0), this did not reach statistical significance in a univariable setting (P = .11) or after adjusting for age and sex (P = .14). This study data shows a decrease in point-of-care abdominal aorta ultrasound imaging quality as BMI increases, though this difference did not result in a statistically significant impairment in achieving the minimum quality for clinical decisions. This finding may help ameliorate some clinician concerns about ultrasonography for patients with high BMI.

Quality of abdominal ultrasound image acquisition by novice practitioners following a minimal training session on healthy volunteers.

Point of care ultrasound (POCUS) is an essential tool for physicians to guide treatment decisions in both hospital and prehospital settings. Despite the potential patient care and system utilization benefits of prehospital ultrasound, the financial burden of a "hands-on" training program for large numbers of paramedics remains a barrier to implementation. In this study, we conducted a prospective, observational, double-blinded study comparing paramedics to emergency physicians in their ability to generate usable abdominal ultrasound images after a 1-hour didactic training session. Canadian aeromedical critical care paramedics were compared against emergency medicine physicians in their ability to generate adequate abdominal ultrasound images on five healthy volunteers. Quality of each scan was evaluated by a trained expert in POCUS who was blinded to the identity of the participant using a 5-point Likert scale and using the standardized QUICk Focused Assessment with Sonography in Trauma (FAST) assessment tool. Fourteen Critical care paramedics and four emergency department (ED) physicians were voluntarily recruited. Of paramedics, 57% had never used ultrasound before, 36% has used ultrasound without formal training, and 7% had previous training. Physicians had a higher proportion of usable scans compared with paramedics (100% v. 61.4%, Δ38.6%; 95% confidence interval, 19.3-50.28). Paramedics were not able to produce images of interpretable quality at the same frequency when compared with emergency medicine physicians. However, a 61.4% usable image rate for paramedics following a short 1-hour didactic training session is promising for future studies, which could incorporate a short hands-on tutorial while remaining cost-effective.

Abdominal ultrasound image acquisition and interpretation by novice practitioners after minimal training on a simulated patient model.

The Focused Assessment with Sonography in Trauma (FAST) exam is a rapid ultrasound test to identify evidence of hemorrhage within the abdomen. Few studies examine the accuracy of paramedic performed FAST examinations. The duration of an ultrasound training program remains controversial. This study's purpose was to assess the accuracy of paramedic FAST exam interpretation following a one hour didactic training session. The interpretation of paramedic performed FAST exams was compared to the interpretation of physician performed FAST examinations on a mannequin model containing 300ml of free fluid following a one hour didactic training course. Results were compared using the Chi-square test. Differences in accuracy rate were deemed significant if p < 0.05. Fourteen critical care flight paramedics and four emergency physicians were voluntarily recruited. The critical care paramedics were mostly ultrasound-naive whereas the emergency physicians all had ultrasound training. The correct interpretation of FAST scans was comparable between the two groups with accuracy of 85.6% and 87.5% (∆1.79 95%CI -33.85 to 21.82, p = 0.90) for paramedics and emergency physicians respectively. This study determined that critical care paramedics were able to use ultrasound to detect free fluid on a simulated mannequin model and interpret the FAST exam with a similar accuracy as experienced emergency physicians following a one hour training course. This suggests the potential use of prehospital ultrasound to aid in the triage and transport decisions of trauma patients while limiting the financial and logistical burden of ultrasound training.

Liver Care and Surveillance: The Global Impact of the COVID-19 Pandemic.

As social distancing and strict stay‐at‐home orders have been instituted to slow the spread of coronavirus disease 2019 (COVID‐19), unintentional outcomes among those with chronic diseases including screening for the lethal hepatocellular carcinoma (HCC) may be occurring. We aimed to describe the changes in liver care before and after COVID‐19 restricted access to health care. We obtained data on the number of liver clinic visits, abdominal ultrasound, computed tomography, and magnetic resonance imaging using electronic query or clinic registry at three medical centers in the United States, Japan, and Singapore for the following periods: February 1 to March 14, 2018, 2019, and 2020; and March 15 to May 1, 2018, 2019, and 2020. We performed trend analysis using logistic regression. In total, 14,403 visits were made to the liver clinics at the three centers: 5,900 in 2018, 5,270 in 2019, and 3,233 in 2020. Overall, there were no significant changes in the distribution of males and females between February 1 and May 1 from 2018 to 2020, but there was a lower proportion of seniors ages 65 years and older (P < 0.001). There were significant decreasing trends in the total number of liver clinic visits overall (p‐trend = 0.038) and in the subanalysis for chronic hepatitis B, C, and other liver diseases. HCC/cirrhosis visits also dropped from 883 to 538 (39.07% decrease) overall and 665 to 355 (46.62% decrease) for the US site. In addition, there was a significant decreasing trend in the number of abdominal ultrasounds (P‐trend = 0.004) and computed tomography/magnetic resonance imaging (P‐trend = 0.007) performed overall. Conclusion: Liver clinic visits, hepatoma surveillance, and diagnostic abdominal imaging fell dramatically as social distancing measures were instituted. Care providers must find ways to recall patients for important care monitoring, including HCC surveillance.

Tumor Vascular Networks Depicted in Contrast-Enhanced Ultrasound Images as a Predictor for Transarterial Chemoembolization Treatment Response

Hepatocellular carcinoma (HCC) is prevalent worldwide. Among the various therapeutic options, transarterial chemoembolization (TACE) can be applied to the tumor vascular network by restricting the nutrients and oxygen supply to the tumor. Unique morphologic properties of this network may provide information predictive of future therapeutic responses, which would be significant for decision making during treatment planning. The extraction of morphologic features from the tumor vascular network depicted in abdominal contrast-enhanced ultrasound (CEUS) images faces several challenges, such as organ motion, limited resolution caused by clutter signal and segmentation of the vascular structures at multiple scales. In this study, we present an image processing and analysis approach for the prediction of HCC response to TACE treatment using clinical CEUS images and known pathologic responses. This method focuses on addressing the challenges of CEUS by incorporating a two-stage motion correction strategy, clutter signal removal, vessel enhancement at multiple scales and machine learning for predictive modeling. The morphologic features, namely, number of vessels (NV), number of bifurcations (NB), vessel to tissue ratio (VR), mean vessel length, tortuosity and diameter, from tumor architecture were quantified from CEUS images of 36 HCC patients before TACE treatment. Our analysis revealed that NV, NB and VR are the dominant features for the prediction of long-term TACE response. The model had an accuracy of 86% with a sensitivity and specificity of 89% and 82%, respectively. Reliable prediction of the TACE therapy response using CEUS-derived image features may help to provide personalized therapy planning, which will ultimately improve patient outcomes.

Neck Circumference is an Effective Supplement for Nonalcoholic Fatty Liver Disease Screening in a Community-Based Population.

Background Accumulating evidence has shown that neck circumference (NC) is associated with obesity-related metabolic abnormalities. Nonalcoholic fatty liver disease (NAFLD) is regarded as a liver manifestation of metabolic syndrome. This study aimed to investigate the relationship between NC and liver fat content (LFC) and NAFLD. Methods A total of 1698 subjects (577 men and 1121 women) from the Shanghai community were enrolled. All the subjects underwent NC measurement and biochemical measurements. LFC was calculated using the parameters from abdominal ultrasound images. Elevated NC was defined as NC ≥38.5 cm in men and NC ≥34.5 cm in women. Results Subjects with NAFLD based on the LFC measurement had higher values of NC, liver enzyme profiles, homoeostasis model assessment-insulin resistance index, and LFC than those without NAFLD (all P < 0.05), irrespective of sex. NC showed an upward trend with the increase of LFC in both men and women (both P < 0.05). An elevated NC could identify 55.22% of men and 50.29% of women with NAFLD based on quantitative ultrasonography. The positive correlation between NC and LFC remained significant even after adjustment for central obesity (both P < 0.05). After adjusting for confounding factors, the risk of NAFLD in subjects with an elevated NC was 1.52-fold higher in men (P=0.036) and 2.31-fold higher in women (P < 0.001). Conclusions There was a significant and positive correlation between NC and LFC. The risk of NAFLD increased significantly in subjects with an elevated NC.

MP12: Abdominal ultrasound image acquisition and interpretation by novice practitioners after minimal training on a simulated patient model

Introduction: The FAST exam – Focused Assessment with Sonography in Trauma - is a rapid test using ultrasound to identify sonographic evidence of hemorrhage within the abdomen. In the prehospital setting, the information from a FAST examination can help triage patients, direct patients to the most appropriate facilities, assist with management strategies and potentially expedite time to definitive intervention. Few studies examine the accuracy of paramedic-only-performed FAST examinations. However, despite the potential benefits to the Canadian prehospital system, a potential barrier to implementation is the tremendous financial and operational burden if paramedics require prolonged ultrasound training courses. In this study, we conducted a double-blinded observational study comparing the accuracy of paramedic-performed FAST versus physician-performed tests on a sonographic Phantom, after a one-hour didactic training session. Methods: The interpretation of paramedic performed FAST exams was compared to the interpretation of physician performed FAST examinations on a mannequin model. The mannequin utilized in this study was a realistic model of a human torso where fluid could be injected into the abdomen to create a realistic ultrasound image of abdominal free fluid. Participants were required to scan the mannequin twice, once with 300 mL of fluid instilled and once with the abdomen free of fluid. Participants were blinded to the status of hemoperitoneum. The primary outcome of the study was accuracy rate of FAST examination by paramedics compared to emergency room physicians. Results were compared using the Chi-square test. Differences in accuracy rate were deemed significant if p &lt; 0.05. Total scan time was reported using means, standard deviations and 95% CIs and was compared between groups using standard t-test. Results: Fourteen critical care flight paramedics and four emergency physicians were voluntarily recruited. The critical care paramedics were ultrasound-naive whereas the emergency physicians had ultrasound training. The correct interpretation of FAST scans was comparable between the two groups 85.6% and 87.5% (Δ1.79 95%CI -33.85 to 21.82, p = 0.90) for paramedics and emergency physicians respectively. Total scan time differed between groups but did not reach statistical significance. Paramedics took longer to complete the FAST examination with a mean (SD) time to complete the two scans of 10.35 (3.43) minutes compared to 7.34 (2.74) minutes for physicians, (Δ3.01 minutes 95%CI -0.97 to 7.00, p = 0.13). Conclusion: This study determined that critical care paramedics were able use ultrasound to detect free fluid on a simulated mannequin model and interpret the FAST exam with a similar accuracy as experienced emergency physicians following a one hour training course. This suggests the potential use of ultrasound in prehospital programs to determine the most appropriate transport destination and aid in the triage of trauma patients while limiting the financial and logistical burden of ultrasound training.

Automated detection of kidney abnormalities using multi-feature fusion convolutional neural networks

Kidney abnormalities have a high incidence in people of all ages. The requisite manual examinations to detect these abnormalities are costly and time consuming. The rapid and accurate detection of kidney abnormalities has emerged as a focus of computer-aided medical research. Traditional methods work on images of kidneys and rely on hand-crafted features to identify abnormal symptoms as independent classes that lack generalization to different kidney diseases. In this study, an automated architecture to detect various kidney abnormalities is proposed that works on abdominal ultrasound images using convolutional neural networks. Our model consists of three components: the first selects appropriate ultrasound images of kidneys, the second is a detection model used to locate the area occupied by the kidney in the given image, and the third component using the multi-feature fusion neural network (Mf-Net) discriminates normal and abnormal kidneys, by determining whether there are abnormal symptoms in images. The detection model is combined with a weighted ensemble method to improve performance. A multi-feature fusion layer is also designed in the Mf-Net to extract distinctive features for multiple views of images. The three components work together to automatically recognize abnormalities associated with kidneys. A large dataset containing 3,722 abdominal ultrasound images with classification and localization annotations is established to train and evaluate the model. Experimental results show that the proposed ensemble detection model performs best with an average TPF of 98.0%, and the Mf-Net achieves an average classification accuracy of 94.67%. The obtained high classification and detection accuracy demonstrate the effectiveness of the proposed method for recognizing kidney abnormalities.