Transfer Learning-Based B-Line Assessment of Lung Ultrasound for Acute Heart Failure.

OBJECTIVE: B-lines assessed by lung ultrasound (LUS) outperform physical exam, chest radiograph, and biomarkers for the associated diagnosis of acute heart failure (AHF) in the emergent setting. The use of LUS is however limited to trained professionals and suffers from interpretation variability. The objective was to utilize transfer learning to create an AI-enabled software that can aid novice users to automate LUS B-line interpretation. METHODS: Data from an observational AHF LUS study provided standardized cine clips for AI model development and evaluation. A total of 49,952 LUS frames from 30 patients were hand scored and trained on a convolutional neural network (CNN) to interpret B-lines at the frame level. A random independent evaluation set of 476 LUS clips from 60 unique patients assessed model performance. The AI models scored the clips on both a binary and ordinal 0-4 multiclass assessment. RESULTS: A multiclassification AI algorithm had the best performance at the binary level when applied to the independent evaluation set, AUC of 0.967 (95% CI 0.965-0.970) for detecting pathologic conditions. When compared to expert blinded reviewer, the 0-4 multiclassification AI algorithm scale had a reported linear weighted kappa of 0.839 (95% CI 0.804-0.871). CONCLUSIONS: The multiclassification AI algorithm is a robust and well performing model at both binary and ordinal multiclass B-line evaluation. This algorithm has the potential to be integrated into clinical workflows to assist users with quantitative and objective B-line assessment for evaluation of AHF.

Using Deep Learning to Detect the Presence and Location of Hemoperitoneum on the Focused Assessment with Sonography in Trauma (FAST) Examination in Adults.

Abdominal ultrasonography has become an integral component of the evaluation of trauma patients. Internal hemorrhage can be rapidly diagnosed by finding free fluid with point-of-care ultrasound (POCUS) and expedite decisions to perform lifesaving interventions. However, the widespread clinical application of ultrasound is limited by the expertise required for image interpretation. This study aimed to develop a deep learning algorithm to identify the presence and location of hemoperitoneum on POCUS to assist novice clinicians in accurate interpretation of the Focused Assessment with Sonography in Trauma (FAST) exam. We analyzed right upper quadrant (RUQ) FAST exams obtained from 94 adult patients (44 confirmed hemoperitoneum) using the YoloV3 object detection algorithm. Exams were partitioned via fivefold stratified sampling for training, validation, and hold-out testing. We assessed each exam image-by-image using YoloV3 and determined hemoperitoneum presence for the exam using the detection with highest confidence score. We determined the detection threshold as the score that maximizes the geometric mean of sensitivity and specificity over the validation set. The algorithm had 95% sensitivity, 94% specificity, 95% accuracy, and 97% AUC over the test set, significantly outperforming three recent methods. The algorithm also exhibited strength in localization, while the detected box sizes varied with a 56% IOU averaged over positive cases. Image processing demonstrated only 57-ms latency, which is adequate for real-time use at the bedside. These results suggest that a deep learning algorithm can rapidly and accurately identify the presence and location of free fluid in the RUQ of the FAST exam in adult patients with hemoperitoneum.

Automated detection and localization of pericardial effusion from point-of-care cardiac ultrasound examination.

Focused Assessment with Sonography in Trauma (FAST) exam is the standard of care for pericardial and abdominal free fluid detection in emergency medicine. Despite its life saving potential, FAST is underutilized due to requiring clinicians with appropriate training and practice. To aid ultrasound interpretation, the role of artificial intelligence has been studied, while leaving room for improvement in localization information and computation time. The purpose of this study was to develop and test a deep learning approach to rapidly and accurately identify both the presence and location of pericardial effusion on point-of-care ultrasound (POCUS) exams. Each cardiac POCUS exam is analyzed image-by-image via the state-of-the-art YoloV3 algorithm and pericardial effusion presence is determined from the most confident detection. We evaluate our approach over a dataset of POCUS exams (cardiac component of FAST and ultrasound), comprising 37 cases with pericardial effusion and 39 negative controls. Our algorithm attains 92% specificity and 89% sensitivity in pericardial effusion identification, outperforming existing deep learning approaches, and localizes pericardial effusion by 51% Intersection Over Union with ground-truth annotations. Moreover, image processing demonstrates only 57 ms latency. Experimental results demonstrate the feasibility of rapid and accurate pericardial effusion detection from POCUS exams for physician overread.

A consensus list of ultrasound competencies for graduating emergency medicine residents.

Objectives: Emergency ultrasound (EUS) is a critical component of emergency medicine (EM) resident education. Currently, there is no consensus list of competencies for EUS training, and graduating residents have varying levels of skill and comfort. The objective of this study was to define a widely accepted comprehensive list of EUS competencies for graduating EM residents through a modified Delphi method. Methods: We developed a list of EUS applications through a comprehensive literature search, the American College of Emergency Physicians list of core EUS benchmarks, and the Council of Emergency Medicine Residency-Academy of Emergency Ultrasound consensus document. We assembled a multi-institutional expert panel including 15 faculty members from diverse practice environments and geographical regions. The panel voted on the list of competencies through two rounds of a modified Delphi process using a modified Likert scale (1 = not at all important, 5 = very important) to determine levels of agreement for each application-with revisions occurring between the two rounds. High agreement for consensus was set at >80%. Results: Fifteen of 15 panelists completed the first-round survey (100%) that included 359 topics related to EUS. After the first round, 195 applications achieved high agreement, four applications achieved medium agreement, and 164 applications achieved low agreement. After the discussion, we removed three questions and added 13 questions. Fifteen of 15 panelists completed the second round of the survey (100%) with 209 of the 369 applications achieving consensus. Conclusion: Our final list represents expert opinion on EUS competencies for graduating EM residents. We hope to use this consensus list to implement a more consistent EUS curriculum for graduating EM residents and to standardize EUS training across EM residency programs.

Transfer Learning for Automated COVID-19 B-Line Classification in Lung Ultrasound.

Lung ultrasound (LUS) as a diagnostic tool is gaining support for its role in the diagnosis and management of COVID-19 and a number of other lung pathologies. B-lines are a predominant feature in COVID-19, however LUS requires a skilled clinician to interpret findings. To facilitate the interpretation, our main objective was to develop automated methods to classify B-lines as pathologic vs. normal. We developed transfer learning models based on ResNet networks to classify B-lines as pathologic (at least 3 B-lines per lung field) vs. normal using COVID-19 LUS data. Assessment of B-line severity on a 0-4 multi-class scale was also explored. For binary B-line classification, at the frame-level, all ResNet models pretrained with ImageNet yielded higher performance than the baseline nonpretrained ResNet-18. Pretrained ResNet-18 has the best Equal Error Rate (EER) of 9.1% vs the baseline of 11.9%. At the clip-level, all pretrained network models resulted in better Cohen's kappa agreement (linear-weighted) and clip score accuracy, with the pretrained ResNet-18 having the best Cohen's kappa of 0.815 [95% CI: 0.804-0.826], and ResNet-101 the best clip scoring accuracy of 93.6%. Similar results were shown for multi-class scoring, where pretrained network models outperformed the baseline model. A class activation map is also presented to guide clinicians in interpreting LUS findings. Future work aims to further improve the multi-class assessment for severity of B-lines with a more diverse LUS dataset.

Point-of-care Lung Ultrasound Is More Sensitive than Chest Radiograph for Evaluation of COVID-19.

INTRODUCTION: Current recommendations for diagnostic imaging for moderately to severely ill patients with suspected coronavirus disease 2019 (COVID-19) include chest radiograph (CXR). Our primary objective was to determine whether lung ultrasound (LUS) B-lines, when excluding patients with alternative etiologies for B-lines, are more sensitive for the associated diagnosis of COVID-19 than CXR. METHODS: This was a retrospective cohort study of all patients who presented to a single, academic emergency department in the United States between March 20 and April 6, 2020, and received LUS, CXR, and viral testing for COVID-19 as part of their diagnostic evaluation. The primary objective was to estimate the test characteristics of both LUS B-lines and CXR for the associated diagnosis of COVID-19. Our secondary objective was to evaluate the proportion of patients with COVID-19 that have secondary LUS findings of pleural abnormalities and subpleural consolidations. RESULTS: We identified 43 patients who underwent both LUS and CXR and were tested for COVID-19. Of these, 27/43 (63%) tested positive. LUS was more sensitive (88.9%, 95% confidence interval (CI), 71.1-97.0) for the associated diagnosis of COVID-19 than CXR (51.9%, 95% CI, 34.0-69.3; p = 0.013). LUS and CXR specificity were 56.3% (95% CI, 33.2-76.9) and 75.0% (95% CI, 50.0-90.3), respectively (p = 0.453). Secondary LUS findings of patients with COVID-19 demonstrated 21/27 (77.8%) had pleural abnormalities and 10/27 (37%) had subpleural consolidations. CONCLUSION: Among patients who underwent LUS and CXR, LUS was found to have a higher sensitivity than CXR for the evaluation of COVID-19. This data could have important implications as an aid in the diagnostic evaluation of COVID-19, particularly where viral testing is not available or restricted. If generalizable, future directions would include defining how to incorporate LUS into clinical management and its role in screening lower-risk populations.

Central venous catheter placement after ultrasound guided peripheral IV placement for difficult vascular access patients.

OBJECTIVES: Ultrasound guided peripheral intravenous catheters (USPIV) are frequently utilized in the Emergency Department (ED) and lead to reduced central venous catheter (CVC) placements. USPIVs, however, are reported to have high failure rates. Our primary objective was to determine the proportion of patients that required CVC after USPIV. Our secondary objective was to determine if classic risk factors for difficult vascular access were predictive of future CVC placement. METHODS: We performed a retrospective review for patients treated at a large academic hospital. Patients were identified by electronic health record and were restricted to age older than 21 years, had received USPIV, and admittance. Exclusion criteria included an existing CVC. Descriptive statistics, t-tests, chi-square proportions, and logistic regression were performed to test associations. RESULTS: Of 363 eligible patients, 20 were excluded allowing for 343 for analysis. Of 343, 45 (13.1% 95% CI 9.9-17.1%) required CVC after USPIV. For secondary outcomes, no expected characteristics (diabetes, end-stage renal disease, IV drug abuse, peripheral vascular disease, or sickle cell disease) were predictive of CVC placement. The only predictive variables were admission to ICU/stepdown and length of stay. Each additional day of hospitalization had an OR 1.11 (95% CI 1.06-1.16%) of having a CVC placed. CONCLUSION: Of those admitted after USPIV placement, approximately 7 out of every 8 patients did not require a subsequent CVC. Of the nearly 1 in 8 patients that required a CVC, factors associated with CVC placement were admission to a higher level of care and length of stay.

Ultrasound vs. Computed Tomography for Severity of Hydronephrosis and Its Importance in Renal Colic.

INTRODUCTION: Supporting an "ultrasound-first" approach to evaluating renal colic in the emergency department (ED) remains important for improving patient care and decreasing healthcare costs. Our primary objective was to compare emergency physician (EP) ultrasound to computed tomography (CT) detection of hydronephrosis severity in patients with suspected renal colic. We calculated test characteristics of hydronephrosis on EP-performed ultrasound for detecting ureteral stones or ureteral stone size >5mm. We then analyzed the association of hydronephrosis on EP-performed ultrasound, stone size >5mm, and proximal stone location with 30-day events. METHODS: This was a prospective observational study of ED patients with suspected renal colic undergoing CT. Subjects had an EP-performed ultrasound evaluating for the severity of hydronephrosis. A chart review and follow-up phone call was performed. RESULTS: We enrolled 302 subjects who had an EP-performed ultrasound. CT and EP ultrasound results were comparable in detecting severity of hydronephrosis (x2=51.7, p<0.001). Hydronephrosis on EP-performed ultrasound was predictive of a ureteral stone on CT (PPV 88%; LR+ 2.91), but lack of hydronephrosis did not rule it out (NPV 65%). Lack of hydronephrosis on EP-performed ultrasound makes larger stone size >5mm less likely (NPV 89%; LR- 0.39). Larger stone size > 5mm was associated with 30-day events (OR 2.30, p=0.03). CONCLUSION: Using an ultrasound-first approach to detect hydronephrosis may help physicians identify patients with renal colic. The lack of hydronephrosis on ultrasound makes the presence of a larger ureteral stone less likely. Stone size >5mm may be a useful predictor of 30-day events.

Point-of-Care Ultrasound Use, Accuracy, and Impact on Clinical Decision Making in Rwanda Hospitals.

OBJECTIVES: Few studies of point-of-care ultrasound training and use in low resource settings have reported the impact of examinations on clinical management or the longer-term quality of trainee-performed studies. We characterized the long-term effect of a point-of-care ultrasound program on clinical decision making, and evaluated the quality of clinician-performed ultrasound studies. METHODS: We conducted point-of-care ultrasound training for physicians from Rwandan hospitals. Physicians then used point-of-care ultrasound and recorded their findings, interpretation, and effects on patient management. Data were collected for 6 months. Trainee studies were reviewed for image quality and accuracy. RESULTS: Fifteen participants documented 1158 ultrasounds; 590 studies (50.9%) had matched images and interpretations for review. Abdominal ultrasound for free fluid was the most frequently performed application. The mean image quality score was 2.36 (95% confidence interval, 2.28-2.44). Overall sensitivity and specificity for trainee-performed examinations was 94 and 98%. Point-of-care ultrasound use most commonly changed medications administered (42.4%) and disposition (30%). CONCLUSIONS: A point-of-care ultrasound training intervention in a low-resource setting resulted in high numbers of diagnostic-quality studies over long-term follow-up. Ultrasound use routinely changed clinical decision making.

Intensive point-of-care ultrasound training with long-term follow-up in a cohort of Rwandan physicians.

OBJECTIVE: We delivered a point-of-care ultrasound training programme in a resource-limited setting in Rwanda, and sought to determine participants' knowledge and skill retention. We also measured trainees' assessment of the usefulness of ultrasound in clinical practice. METHODS: This was a prospective cohort study of 17 Rwandan physicians participating in a point-of-care ultrasound training programme. The follow-up period was 1 year. Participants completed a 10-day ultrasound course, with follow-up training delivered over the subsequent 12 months. Trainee knowledge acquisition and skill retention were assessed via observed structured clinical examinations (OSCEs) administered at six points during the study, and an image-based assessment completed at three points. RESULTS: Trainees reported minimal structured ultrasound education and little confidence using point-of-care ultrasound before the training. Mean scores on the image-based assessment increased from 36.9% (95% CI 32-41.8%) before the initial 10-day training to 74.3% afterwards (95% CI 69.4-79.2; P < 0.001). The mean score on the initial OSCE after the introductory course was 81.7% (95% CI 78-85.4%). The mean OSCE performance at each subsequent evaluation was at least 75%, and the mean OSCE score at the 58-week follow up was 84.9% (95% CI 80.9-88.9%). CONCLUSIONS: Physicians providing acute care in a resource-limited setting demonstrated sustained improvement in their ultrasound knowledge and skill 1 year after completing a clinical ultrasound training programme. They also reported improvements in their ability to provide patient care and in job satisfaction.

Image Segmentation and Machine Learning for Detection of Abdominal Free Fluid in Focused Assessment With Sonography for Trauma Examinations: A Pilot Study.

The objective of this pilot study was to test the feasibility of automating the detection of abdominal free fluid in focused assessment with sonography for trauma (FAST) examinations. Perihepatic views from 10 FAST examinations with positive results and 10 FAST examinations with negative results were used. The sensitivity and specificity compared to manual classification by trained physicians was evaluated. The sensitivity and specificity (95% confidence interval) were 100% (69.2%-100%) and 90.0% (55.5%-99.8%), respectively. These findings suggest that computerized detection of free fluid on abdominal ultrasound images may be sensitive and specific enough to aid clinicians in their interpretation of a FAST examination.

Goal-directed Focused Ultrasound Milestones Revised: A Multiorganizational Consensus.

In 2012 the Accreditation Council for Graduate Medical Education and the American Board of Emergency Medicine released the emergency medicine milestones. The Patient Care 12 (PC12) subcompetency delineates staged and progressive accomplishment in emergency ultrasound. While valuable as an initial framework for ultrasound resident education, there are limitations to PC12. This consensus paper provides a revised description of criteria to define the subcompetency. A multiorganizational task force was formed between the American College of Emergency Physicians Ultrasound Section, the Council of Emergency Medicine Residency Directors, and the Academy of Emergency Ultrasound of the Society for Academic Emergency Medicine. Representatives from each organization created this consensus document and revision.

A practical guide to self-sustaining point-of-care ultrasound education programs in resource-limited settings.

The value of point-of-care ultrasound education in resource-limited settings is increasingly recognized, though little guidance exists on how to best construct a sustainable training program. Herein we offer a practical overview of core factors to consider when developing and implementing a point-of-care ultrasound education program in a resource-limited setting. Considerations include analysis of needs assessment findings, development of locally relevant curriculum, access to ultrasound machines and related technological and financial resources, quality assurance and follow-up plans, strategic partnerships, and outcomes measures. Well-planned education programs in these settings increase the potential for long-term influence on clinician skills and patient care.

Neuro-ophthalmology.

Neuro-ophthalmologic disorders arise from all areas of the neuro-ophthalmologic tract. They may be expressed simply as loss of vision or double vision, or as complex syndromes or systemic illnesses, depending on the location and type of lesion. Problems may occur anywhere along the visual pathway, including the brainstem, cavernous sinus, subarachnoid space, and orbital apex, and may affect adjacent structures also. A firm understanding of the neuroanatomy and neurophysiology of the eye is essential to correct diagnosis.