The use of ultrasound by physicians not specialized in diagnostic radiology, as an adjunct to physical examination, has been termed "point-of-care ultrasound". In the last decade, this technique has been implemented in various medical fields, particularly family medicine and in emergency and urgent care. A descriptive, cross-sectional observational study was conducted, involving a survey of physicians (mostly pediatricians) who had attended a training course in pediatric point-of-care ultrasound organized by different scientific societies. The aim was to obtain feedback from the pediatricians on the use, usefulness, barriers and training in ultrasound. A total of 126 physicians were interviewed, most of whom declared having access to ultrasound. However, less than half of them made use of such devices, due to considerable difficulties, including lack of time and training. A total of 96.8% of those surveyed considered that ultrasound should be part of the training protocol of physicians specializing in pediatrics. The vast majority believed ultrasound training to be very important but considered their level of experience with the technique to be basic. The implementation of point-of-care ultrasound in pediatrics is still limited in Spain, though the professionals agree on highlighting its relevance in daily practice. Among the most prominent benefits are resource optimization and shortening of the time to diagnosis. The participants agree that competence in the technique should be incorporated in a structured manner to the training of physicians specializing in pediatrics.
Currently, the role of lung ultrasound (LUS) in the diagnosis and treatment of patients with septic shock is widely recognized. Various LUS protocols and scoring criteria have been proposed, yet a unified LUS protocol for assessing lung water in these patients remains lacking. Forty‑six septic shock patients underwent LUS with three scanning schemes (4‑region, 8‑region, BLUE) alongside pulse indicated continuous cardiac output (PiCCO) monitoring. Spearman correlation analysis was used to compare the correlation between the three ultrasound scoring schemes, PiCCO and other clinical laboratory indicators. At the same time, the value of three LUS protocols for assessing pulmonary water in septic shock patients was evaluated using receiver operating characteristic (ROC) analysis. Spearman's correlation analysis showed a significant correlation between extravascular lung water index (EVLWI) and 4-region, 8-region, or the Bedside Lung Ultrasound Examination (BLUE) protocol. More importantly, the BLUE protocol showed a stronger correlation with EVLWI than the 4-region protocol and the 8-region protocol (r=0.634, 0.458, 0.546, p<0.001). The ROC curve analysis showed that ultrasound protocols could predict the early occurrence of pulmonary edema in septic shock patients (EVLWI>7 ml/kg), diagnosis of pulmonary edema (EVLWI>10 ml/kg), and evaluation of pulmonary fluid severity (EVLWI≥15 ml/kg). Our study found that septic shock patients were prone to pulmonary edema during fluid resuscitation, and ultrasound scoring could help us quantify pulmonary edema. Among the 4-region, 8-region and BLUE protocols, the BLUE protocol shows relatively better overall performance in predicting lung water elevation and evaluating edema severity.
Lung ultrasound (LUS) is nowadays an important tool to evaluate the state of lung surface. However, it is strongly operator-dependent, leading to reduced reproducibility of LUS analysis. Even though LUS acquisition protocols can improve LUS reproducibility and help standardizing LUS exams, human operators can guarantee only a limited precision in intercepting the optimal imaging plane. Hence, in this study, we assess the possibility to automatically intercept the optimal imaging plane in LUS examinations, i.e., the imaging plane perpendicular to the pleural plane (PP), by extracting three features, then utilized to guide a UR5e robotic arm handling an ultrasound probe. The main focus of this study consists on evaluating the potential of these three features in estimating the PP position with respect to the probe. To do so, we designed a simplified but highly controllable environment, where PP was mimicked with a steel plate (to simulate a highly reflective acoustic interface), while intercostal tissues were mimicked with a 2-cm-thick beef meat. The environment was imaged with a linear probe connected to an ULA-OP platform, which was held by an UR5e, programmed to explore 8 different paths of acquisitions with a rotational angle (RA) ranging from -20º to 20º (1º step size). This resulted in 328 positions that could be explored; each position with RA=0º corresponds to the optimal imaging plane. Radiofrequency data were acquired and post-processed to form normalized log-scale B-Mode images. A rectangular region of interest, defined to include PP, was considered to compute mean intensity at each depth of the region of interest, along lateral dimension. Mean intensity as a function of depth was then utilized to extract three different features, then fed to genetic algorithms to solve optimization problems to guide UR5e towards the optimal imaging plane. Genetic algorithms converged towards an average error < 1º after exploring only 18 positions, showing strong potential in automatic probe placement for LUS.
To evaluate the indirect head of rectus femoris tendon (IHRFT) enthesis as a stable adjunct anatomical landmark for standardizing coronal plane acquisition in ultrasound detection of developmental dysplasia of the hip (DDH) and characterize its diagnostic utility. In this prospective study, 200 infants (400 hips, median age 89 days) underwent hip ultrasound. Spatial relationships between the Graf standard coronal plane and IHRFT enthesis were quantified. A modified coronal plane was acquired using the posterior IHRFT margin as an adjunct landmark with flexed hips. Alpha angles from both planes were compared. IHRFT enthesis thickness was measured in an axial oblique plane; its diagnostic accuracy for DDH (Graf type ≥IIa) was assessed. The posterior IHRFT margin consistently aligned with the Graf standard plane (mean offset: -1.0 ± 1.1 mm). Alpha angles derived using the IHRFT as an adjunct landmark (62.2° ± 3.9°) showed no significant difference from Graf plane angles (62.9° ± 4.4°, p = 0.08), with good inter-operator reliability (ICC = 0.77). IHRFT enthesis thickness was significantly greater in dysplastic hips (n = 36) vs. normal hips (n = 364) (4.1 ± 1.1 mm vs. 2.4 ± 0.3 mm, p < 0.0001). Enthesis thickness ≥3.2 mm predicted DDH with high accuracy (AUC = 0.91, sensitivity 91%, specificity 88%). Incorporating the posterior margin of the IHRFT enthesis as an adjunct landmark for defining the standard coronal plane enhances inter-operator consistency in alpha angle measurements, reducing dependence on strict neutral positioning. IHRFT enthesis thickening (≥3.2 mm) on axial oblique view is a novel sonographic index for diagnosis of DDH.
To explore a fusion model designed for the quality evaluation of ultrasound images utilized in fetal crown-rump length (CRL) measurement, and to use SHapley Additive exPlanations (SHAP) method to elucidate the model's decision-making processes. We retrospectively collected 1149 images of midsagittal planes of the entire fetus during early pregnancy from two hospitals. Two senior radiologists categorized the images into standard and non-standard planes. Seven image segmentation models were trained to select the best model for automatically segmenting the region of interest. The radiomics features and deep transfer learning (DTL) features were extracted and selected to establish radiomics models and DTL models. We also constructed fusion models to enhance the classification performance and the optimal one underwent comparison with radiologists. The SHAP method was employed to interpret and visualize the model. The DeepLabV3 ResNet101 segmentation model demonstrated the highest performance (DSC: 97.15%). The early fusion model exhibited superior classification performance in validation set (AUC: 0.947, 95% CI: 0.924-0.970, accuracy: 88.4%, sensitivity: 83.0%, specificity: 92.7%, PPV: 90.1%, NPV: 87.3%, precision: 90.1%). The model demonstrated performance commensurate with that of senior radiologists while surpassing junior radiologists. Notably, when leveraging the model's support, there was a substantial improvement in their overall performance. The early fusion model demonstrated satisfactory performance in the intelligent quality evaluation of ultrasound images for CRL measurement. It has the potential to enhance the professional skills of junior radiologists.
Early detection of developmental dysplasia of the hip (DDH) is essential to prevent late presentation and long-term morbidity. Universal neonatal hip ultrasound screening using the Graf method improves early diagnosis. However, evidence-based training requirements for paediatricians remain unclear. This study aimed to determine the duration of training and the number of examinations required for paediatricians to achieve diagnostic competency in neonatal hip ultrasound. An analytical cross-sectional observational study was conducted. Fourteen paediatricians completed a standardized 40-h theoretical and practical training course and performed neonatal hip ultrasound examinations using the Graf method. A total of 467 newborns were screened between 4-6 and 12-16 weeks of age over a 10-month period. Images were compared with a blinded external expert reference standard. Diagnostic agreement was assessed using weighted κ statistics and learning curves and competency thresholds were evaluated using cumulative sum (CUSUM) analysis. A total of 3752 ultrasound images were obtained, of which 3182 (84.8%) met Graf standardization criteria. CUSUM analysis showed that diagnostic competency thresholds were reached after approximately 50 examinations, with performance consolidating between 50 and 100 examinations. Paediatricians completing around four months of supervised practice and at least 180 examinations achieved substantial interobserver agreement comparable to expert reviewers (weighted κ, 0.60-0.75). Diagnostic discrepancies occurred mainly during early training and involved borderline hip classifications (Graf IIa-IIc). Paediatricians can achieve reliable diagnostic competency in neonatal hip ultrasound using the Graf method following structured training and adequate supervised clinical experience. • Universal ultrasound screening using the Graf method improves early detection of developmental dysplasia of the hip (DDH). • Training requirements for paediatricians performing neonatal hip ultrasound remain poorly defined. • This study defines learning curve thresholds for paediatricians performing neonatal hip ultrasound using the Graf method. • Competency was achieved after 50 examinations, with consolidated performance after 4 months of practice and180 examinations.
Selective ultrasound screening for developmental dysplasia of the hip (DDH) was implemented in the North Denmark Region, Denmark in 2010 but has yet to be evaluated. This study aimed to evaluate selective hip screening in the North Denmark Region from 2010 to 2021, focusing on referral rates, the incidence of DDH and referral causes. We retrospectively identified 3,812 infants (2,091 females) who were referred during a 12-year period for hip ultrasound at Aalborg University Hospital in Denmark. DDH was defined as an ultrasound classification of Graf type IIb or worse. Referral rates and the incidence of DDH and their development over the study period were examined. During the study period, there was a steadily increasing referral rate, with the annual referral rate doubling from 200 (3.3%) in 2010 to 402 (6.8%) in 2021, whereas the incidence rate increased only minimally from 0.15% to 0.22%. The most common referral causes were breech position (41.4%) and positive clinical exam (27.8%). During the study period, the annual referral rate nearly doubled without a corresponding increase in DDH incidence, suggesting that the screening programme has become less efficient over time. This might explain the relatively low incidence of DDH in the North Denmark Region compared with national and international studies. The authors received no financial support for the research, authorship and/or publication of this article. This study was registered in the North Denmark Regional internal Project Register as IDF2022-187.
Developmental dysplasia of the hip (DDH) is the most common congenital joint disease in infant. The B-mode ultrasound (BUS) based computer-aided diagnosis (CAD) can help sonologists improve diagnostic accuracy for DDH. The routine CAD models mainly developed based on convolutional neural network or Transformer, which cannot fully learn the inherent structural and texture properties in the hip BUS images. The newly proposed Mamba model has shown its superior performance for learning feature representation with linear computation complexity. However, the scanning mechanism still effect the performance of state space model in Mamba to learns sequence information. To this end, a novel Hybrid Symmetry Mamba Network (HSMN) is proposed to improve the diagnostic performance of CAD model for DDH. The HSMN conducts both the symmetry convolution operation and symmetry scanning in Mamba to more effectively extract inherent information to represent the symmetrical structure in hip BUS images for DDH. The experimental results indicate that the proposed HSMN achieves a diagnostic accuracy of 89.28±2.32%, and outperforms all the compared algorithms, suggesting its effectiveness.Clinical RelevanceThis CAD model has the potential to be applied in clinical practice for help sonologist improve diagnostic accuracy of DDH.
Children with chronic diseases are at a significant risk of radiation exposure. This cohort study evaluates the effectiveness and reliability of ultrasonography (US) for detecting femoral head decentration in children with cerebral palsy (CP) and developmental dysplasia of the hip (DDH), comparing it with traditional radiographic techniques to reduce radiation exposure. A total of 169 patients were enrolled in the study. Both hips were evaluated in 158 patients, resulting in a total of 327 hips. Patients underwent clinical and radiological assessments, including standardized US. Parameters measured included the ventral and lateral bony and cartilaginous ultrasonographic migration indices (bUMI and cUMI), which were compared with standardized radiographic indices (Reimers index (RI) and extrusion index (EI)). The lateral bUMI (17.4%) was significantly lower than the lateral cUMI (25.9%). RI values were lower than EI values (16.8% vs. 27.7%). No significant differences were observed between the bUMI and RI, or between the cUMI and EI, indicating the reliability of US. All lateral parameters correlated well with the lateral centre-edge angle (LCE). Positive correlations were found between the lateral cUMI and the radiological indices, with high inter- and intra-rater reliability (ICC). Significant differences in lateral and ventral UMIs were noted when comparing DDH and CP patients. US is a reliable alternative to radiography for hip surveillance in detecting hip decentration in children with CP and DDH. It reduces radiation exposure while maintaining diagnostic accuracy. The findings support the adoption of US in clinical practice to improve early diagnosis and intervention.
Timely diagnosis of acute ischemic stroke can aid optimal treatment. Optic nerve sheath diameter (ONSD) can determine increased intracranial pressure (ICP) in such cases. The purpose of this study is to determine the value of ONSD in estimating the severity of acute ischemic stroke. Patients with acute ischemic stroke who were referred to a stroke center were studied. The ONSD of both the right and left sides was measured by ultrasound on the day of admission. Ischemic stroke severity was determined based on the NIHSS. A strong correlation was found between increased right and left ONSDs and severity of ischemic stroke determined by the initial NIHSS score. Based on ROC curve (receiver operating characteristic curve) analysis, both cut points of 5.65 mm for right ONSD (with 100% sensitivity of and 86% specificity) and 5.75 mm for left ONSD (with a 100% Sensitivity and 88% specificity) were able to predict severe stroke. The value of the right ONSD (Area Under the Curve = 0.959) and the left ONSD (Area Under the Curve = 0.942) indicated a strong predictive value. Ultrasound as a feasible and non-invasive modality might play a role in determining the severity of an acute ischemic stroke, and could be considered a promising first-line decision making tool.
Atrophy of the gluteus medius and gluteus minimus has been reported in patients with developmental dysplasia of the hip (DDH). We aimed to investigate whether atrophy of the gluteus medius and gluteus minimus is associated with DDH. Twenty white New Zealand rabbits were divided into 2 groups in our experiment. In group 1, one limb of each rabbit was randomly selected for tenotomy of the gluteus medius and gluteus minimus, with the other side serving as a sham control. In group 2, one limb of each rabbit was randomly selected for intramuscular injection of botulinum toxin to induce atrophy of the gluteus medius and gluteus minimus, with the other side serving as a sham control by injection of equal volumes of normal saline. Acetabulum and muscles were evaluated by ultrasound, X-ray imaging, shear wave elastography, and histology with hematoxylin-eosin (H&E) staining. Atrophy of the gluteus medius and gluteus minimus was observed on the experimental side in both groups. Early dysplastic changes in the hips on the tenotomy side was observed on the radiographs with an increased acetabular index (40.0 ± 6.7° vs 24.1 ± 1.9°, P < 0.001) compared with the control side in group 1. No dislocation of the hips was observed. No dysplasia of the hips was observed on the botulinum-toxin side in group 2 with similar acetabular index on both sides (25.0 ± 3.7° vs 24.1 ± 1.9°, P = 0.13). Tenotomy, not atrophy, of the gluteus medius and gluteus minimus is associated with early changes of DDH.
Identifying patients at the highest risk of developmental dysplasia of the hip (DDH) is critical to the success of an efficient selective screening program. Defining and evaluating risk factors is, therefore, vital. This study was designed to analyze the relationship between classic risk factors and rates of DDH in a contemporary United States-based population. All patients presenting before 12 months of age to a single tertiary care pediatric hospital's orthopaedic department for DDH screening, with explicit medical record documentation of DDH risk factors from 2018 to 2022 were included. Classic risk factors were defined as breech presentation, female sex, firstborn status, and family history of hip dysplasia. Ultrasounds and radiographs obtained at initial presentation were reviewed to determine Graf and IHDI classifications. Reasons for referral for DDH screening were obtained from the chart review. One thousand two hundred twenty-six patients were eligible for the study. The probability of having DDH increased as the number of risk factors present for DDH increased ( P <0.001). Female sex and firstborn status were found to be independent risk factors for DDH. Females were 2.5 times more likely to have DDH than males ( P <0.001). Firstborn children were 1.3 times more likely to have DDH than subsequent birth-order children ( P =0.048). There was no difference in the proportion of DDH in breech presenting patients compared with non-breech ( P =0.511) or in subjects with a family history of DDH ( P =0.16). The primary risk factor for DDH in this contemporary cohort was female sex. Female patients without any other risk factors for DDH had a 25.2% rate of DDH, significantly higher than the estimated population incidence. Fifteen percent of these patients had no notable physical exam findings consistent with DDH. Given the high rate of DDH in female patients in the absence of notable physical exam findings, universal ultrasound screening for females at the age of 6 weeks should be considered. Level IV.
To investigate the diagnostic value of three-dimensional superb micro-vascular imaging (3D-SMI) combined with quantitative analysis of Area and VI in differentiating benign and malignant renal tumors. A total of 256 renal lesions from 254 patients who underwent gray-scale ultrasound (Gray US), two-dimensional superb micro-vascular imaging (2D-SMI), and 3D-SMI examinations at Tianjin Medical University Cancer Institute and Hospital between January 2022 and June 2024 were retrospectively analyzed. The imaging features on Gray US, 2D-SMI and 3D-SMI were recorded. Based on 3D-SMI, Vascular Architecture were classified into five types: Type I (avascular), Type II (spotty flow), Type III (sparse flow), Type IV (encircling), and Type V (rich flow). The plane with the most abundant blood flow was selected, and the Area and VI were calculated using Image Pro Plus (IPP) software. Histopathology from surgery or biopsy served as the reference standard. The differences in Vascular Architecture, Area, and VI between benign and malignant renal tumors were compared, and their diagnostic performance was evaluated. Among the 256 lesions, 70 were benign and 186 were malignant. The interobserver agreement for Vascular Architecture classification was good (Kappa = 0.803), and the consistency for Area and VI was high (ICC = 0.835 and 0.864, respectively). Benign tumors Vascular Architecture were mainly type II or III, with mean Area and VI values of 945.87 ± 568.26 (range: 68-3125) and 5.93 ± 4.95 (range: 0.23-24.73), respectively. Malignant tumors were predominantly type IV or V, with mean Area and VI values of 3694.53 ± 2612.38 (range: 93-9965) and 18.21 ± 10.83 (range: 0.69-48.13), respectively. Significant differences were observed in Vascular Architecture, Area, and VI between benign and malignant lesions (all P < 0.001). The area under the ROC curve (AUC) values for 3D-SMI Vascular Architecture, Area, VI, 2D-SMI, and Gray US were 0.813, 0.807, 0.859, 0.750, and 0.718, respectively. VI demonstrated the highest diagnostic performance, with a cutoff value of 8.19 (sensitivity: 82.26%; specificity: 85.51%). Among benign subtypes, there were no significant differences in Vascular Architecture or Area (P > 0.05), while the VI of oncocytoma was significantly higher than epithelioid angiomyolipoma (EMAL), metanephric adenomas (MA), and angiomyolipoma (AML)(P < 0.01). Among malignant subtypes, clear cell renal cell carcinoma (ccRCC) showed distinct Vascular Architecture compared with papillary renal cell carcinoma(pRCC), chromophobe renal cell carcinoma(chRCC), and Xp11.2 translocation/TFE3 fusion-associated renal cell carcinoma(tRCC) (P < 0.01). The Area and VI of ccRCC were significantly higher than those of pRCC and chRCC (P < 0.05), but not significantly different from tRCC (P > 0.05). 3D-SMI provides three-dimensional visualization of Vascular Architecture. Quantitative analysis of the most vascularized plane using Area and VI differentiation between benign and malignant renal tumors, with VI demonstrating the best diagnostic efficacy. This technique offers a non-invasive diagnostic approach for renal tumors.
Metabolic-Associated Steatotic Liver Disease (MASLD) is recognized as one of the most common chronic liver diseases in children globally, rising in tandem with the childhood obesity pandemic. Although high-income countries focus on advanced phenotyping, Sub-Saharan Africa (SSA) faces a distinct "two-speed" epidemic characterized by rapid urbanization and a unique "double burden" of malnutrition and obesity. This review examines the global and regional epidemiology of paediatric MASLD, contrasting established Western practices with the unique genetic, environmental, and diagnostic challenges of SSA. A scoping review was conducted following the PRISMA-ScR guidelines and using the JBI methodological framework. PubMed/MEDLINE, Embase, and African Journals Online (AJOL) were searched for literature published between 2010 and 2025 focusing on epidemiology, risk factors, and diagnostic performance in children aged 0-19 years. Evidence was synthesized to compare global prevalence patterns with emerging African data and to evaluate the validity of conventional screening approaches in resource-limited settings. A total of 68 studies were included. Global evidence estimates paediatric MASLD prevalence between 7.6% and 14% in the general population and as high as 41% among children with obesity. In SSA, data remain sparse but alarming, with pooled prevalence among overweight children reaching 31.1%, a figure derived mostly from studies in the NAFLD-era utilizing ultrasound or ALT proxies, which may not align perfectly with newer MASLD criteria. The region exhibits a distinct "African Paradox" with a lower frequency of the PNPLA3 genetic risk variant (13.7%), which contributes to lower hepatic steatosis on imaging despite pronounced insulin resistance. As a result, reliance on alanine aminotransferase (ALT) and ultrasonography for screening risks under-detection, obscuring the metabolically high-risk yet hepatically lean phenotype common in SSA. Furthermore, environmental drivers such as high-fructose diets and endocrine-disrupting chemicals may be overriding genetic protection. Paediatric MASLD in SSA reflects a multifactorial pathology likely driven by environmental stressors, epigenetic "thrifty phenotype" programming, and rapid nutritional transition rather than simple caloric excess. Western-calibrated diagnostic algorithms are poorly suited to the African metabolic phenotype. To mitigate a future surge in advanced liver disease, public health strategy must prioritize low-cost innovations, including validating scalable markers such as Waist-to-Height Ratio (WHtR) and integrating task-shifting approaches within existing HIV and diabetes care platforms.
Ultrasound Localization Microscopy (ULM) is a milestone in the medical vascular imaging context, enabling the precise characterization of micro-vascular structures using ultrasound imaging. By accurately localizing contrast microbubbles (MBs) flowing in the circulatory system, ULM generates micro-resolved vascular images, overcoming the ultrasonic diffraction limit. However, as ULM relies on precise localization and tracking of individual MBs, high MB concentrations yield to increased localization errors and, ultimately, ULM failure. This constraint limits ULM to low MB concentrations, resulting in long acquisition times that pose challenges in clinical settings.    Methods: Here, we show the feasibility of uncoupling a bi-disperse MB population, composed of two monodisperse MB populations. The uncoupling is performed through a signal processing pipeline that exploits the strong nonlinear response of MBs having resonance frequency tuned with the transmission frequency. After uncoupling, ULM density and velocity flow maps are generated.   Results: Density and velocity maps are generated after uncoupling, when injecting the bi-disperse population individually and simultaneously in a vascular 3D-printed phantom. Furthermore, density maps generated after uncoupling are compared with the one obtained using standard ULM. Results demonstrate the capability of the proposed uncoupling pipeline to separate the bi-disperse population.     Conclusion: This work presents a signal processing pipeline to uncouple a bi-disperse MB population, formed by two monodisperse MB populations. Results are validated in a 3D-printed phantom and demonstrate the feasibility of the uncoupling which, in turn, would enable higher concentrations and reduce acquisition times for micro-vascular imaging.
Different pathogens cause pneumonia with overlapping symptoms, labs, and imaging, complicating early diagnosis. The modified lung ultrasound score (MLUS) shows value in lung disease evaluation, but its ability to differentiate pathogens is unclear. This study assessed MLUS for early identification of mycoplasma, viral, and bacterial pneumonia in children. We enrolled 186 children with suspected pneumonia (Jan-Dec 2023). Clinical data, labs, lung ultrasound findings (A-lines, B-lines, solid lesions, pleural effusion), and pathogens were recorded. MLUS was assigned. Based on pathogens, cases were grouped into mycoplasma (n=74), viral (n=63), and bacterial (n=49) pneumonia for comparison. 1. The median age of the mycoplasma pneumonia group was 6 (4.58,8) years, greater than that of the bacterial pneumonia group (3 [1.1,4.83] years) and the viral pneumonia group (3.41 [1.16,4.75] years) (P＜ 0.05). The mycoplasma pneumonia group had more febrile symptoms, extensive solid lung lesions, and fewer asthmatic symptoms than the other two pathogen groups (P < 0.05). 2. The median MLUS for mycoplasma pneumonia group was 15 (10,22), significantly higher than the median score of 9 (5,16) in the bacterial pneumonia group and 8 (5,15) in the viral pneumonia group (P < 0.05). 3. Coughing up sputum and small lung solid changes were significantly more common in the mycoplasma pneumonia group than in the bacterial pneumonia group (P < 0.05). A high modified lung ultrasound score, extensive solid lung lesions, older age, fever, and fewer shortness-of-breath symptoms strongly suggest mycoplasma pneumoniae infection. These findings provide critical evidence for early, targeted clinical management.
Ultrasound is the standard imaging test for infant developmental dysplasia of the hip (DDH) but is highly operator-dependent, leading to variable image quality and classification. Artificial intelligence (AI)-assisted ultrasound may standardize acquisition and interpretation and support DDH screening beyond specialist centers. To evaluate the diagnostic accuracy and feasibility of AI-assisted ultrasound for infant DDH. We performed a systematic review and diagnostic test-accuracy meta-analysis of studies enrolling infants (≤12 months) undergoing hip ultrasound, in which the index test was AI applied to two-dimensional (2D) or three-dimensional (3D) ultrasound and the reference standard was expert Graf-based interpretation or follow-up consensus. Risk of bias was assessed with QUADAS-2 (diagnostic accuracy bias tool). Sensitivity and specificity were pooled with a bivariate random-effects model. Twenty-nine studies were eligible; nine provided 2×2 data (6,351 hips) for pooling. Pooled sensitivity was 0.92 (95% CI 0.86-0.95) and specificity 0.96 (95% CI 0.91-0.98). Risk of bias was frequently high or unclear for patient selection and the index test. Feasibility signals included short operator training times (approx. 1-2 h) and scan acquisition time reductions (approx. 20-50%), while economic reporting was limited. AI-assisted ultrasound demonstrates high diagnostic accuracy for infant DDH and may help standardize hip imaging and facilitate safe use by nonexpert operators, but larger multicenter studies with external validation and robust economic evaluation are needed.
To identify the core components of obstetric point-of-care ultrasound (POCUS) training programs while simultaneously evaluating the effectiveness of these programs using the Analyze, Design, Develop, Implement, and Evaluate (ADDIE) model. This systematic review and meta-analysis followed a PROSPERO-registered protocol (CRD42024566260) and adhered to PRISMA2020, Cochrane Handbook, and JBI Manual guidelines. Comprehensive searches from database inception to September 22, 2024, covered international and Chinese databases to identify studies evaluating obstetric POCUS training. Two independent reviewers screened studies, assessed methodological quality with JBI tools, and extracted data on study, participant, intervention, and outcome characteristics. Training content was mapped to the ADDIE instructional design model via thematic and framework analyses. Meta-analyses of comparable quantitative outcomes used random-effects models. Integrating quantitative and qualitative findings, this review systematically evaluated the effectiveness and implementation of obstetric POCUS training programs. Systematic synthesis showed that obstetric POCUS training significantly improved healthcare providers' competencies, including knowledge, skills, sustained use, and clinical decision-making. Training also increased antenatal care attendance and identification of high-risk pregnancies, while reducing referrals and optimizing resource use. However, limitations were noted in needs assessment, implementation flexibility, and outcome evaluation. Using thematic and framework analyses combined with the ADDIE model, we systematically organized training phases and key components to provide a scientific basis for program improvement and optimization. Obstetric POCUS training effectively enhances clinical competencies and improves maternal and neonatal health outcomes. Applying the ADDIE model offers a replicable, practical, and sustainable approach for developing standardized training programs. Future obstetric POCUS training should leverage the ADDIE model and adapt to local contexts to improve maternal and neonatal health globally.
The rapid proliferation of artificial intelligence (AI) in medical education has outpaced the development of quality assurance methods for AI-generated content. This study provides the first systematic evaluation of AI-generated instructional materials for lung ultrasound (LUS) training. The ATLAS study employed a cross-sectional, multi-rater evaluation design comparing six instruction sources (five AI systems and human-generated content) across ten LUS content sessions. Expert evaluators (n=39) assessed materials using five standardized domains: Medical Accuracy, Evidence Completeness, Clarity, Practical Utility, and Pedagogical Quality. Statistical analysis included Kruskal-Wallis tests and pairwise comparisons with Bonferroni correction. Significant differences existed between instruction sources (H = 92.582, p < 0.001). Manus AI achieved the highest overall rating (4.55±0.83) and significantly outperformed human instructions in Medical Accuracy (p = 0.0002) and Evidence Completeness (p < 0.001). Gemini AI (3.94±0.97) performed statistically equivalent to human instructions (4.23±1.00). ChatGPT (2.62±1.35) and Meta (1.53±1.02) performed significantly worse than human instructions (p < 0.001). Clarity emerged as the most discriminating criterion with the widest performance range (1.53-4.54). Certain AI systems can generate high-quality LUS instructional materials that match or exceed human-generated content. However, significant quality variations across AI systems emphasize the critical importance of systematic evaluation before implementation. These findings support cautious but optimistic integration of high-performing AI systems into medical education with appropriate quality assurance measures.
Pneumonia is the leading cause of death from infectious diseases worldwide. Lung ultrasound (LUS) is highly accurate for chest infections diagnosis, yet its correlation with causative pathogens remains unclear. Respiratory cultures, combined with molecular techniques represent the gold standard, achieving etiological diagnosis in 90-95% of cases. We compared LUS findings with bronchoalveolar lavage (BAL) sample analyses; to our knowledge, no prior studies have investigated this in the emergency department (ED). Bronchoalveolar lavage (BAL)-LUS is a prospective observational non-profit study conducted in the ED, aiming to assess whether there is a correlation between the LUS sonographic appearance, assessed blindly across 12 lung fields, and the etiopathogenetic agent of pneumonia (bacterial and viral) detected with molecular syndromic panels (MSPs) and respiratory cultures obtained with BAL. 64 patients were enrolled (mean age 73.3 ± 14.6) with 11 diagnosed as viral pneumonia and 53 as bacterial pneumonia. Bacterial pneumonias were more commonly associated with consolidation (2.9 ± 2.2 vs. 1.5 ± 0.9, p < 0.01) and a higher incidence of pleural effusion (0.9 ± 1.3 vs. 0.3 ± 0.6, p < 0.01). Viral pneumonias were more often associated with interstitial syndrome (4.9 ± 3.3 vs. 0.5 ± 1.3, p < 0.01) and small subpleural consolidations (0.9 ± 1.8 vs. 0.2 ± 0.6, p = 0.01). The mean LUS score was significantly higher in bacterial than in viral pneumonia with a AUC of 0.81 (95% CI 0.68-0.93). Viral pneumonia is usually associated with interstitial syndrome and small subpleural consolidations; on the other hand, bacterial pneumonia is usually associated with consolidation, and pleural effusion.