A quality assessment tool for focused abdominal sonography for trauma examinations using artificial intelligence.

BACKGROUND: Current tools to review focused abdominal sonography for trauma (FAST) images for quality have poorly defined grading criteria or are developed to grade the skills of the sonographer and not the examination. The purpose of this study is to establish a grading system with substantial agreement among coders, thereby enabling the development of an automated assessment tool for FAST examinations using artificial intelligence (AI). METHODS: Five coders labeled a set of FAST clips. Each coder was responsible for a different subset of clips (10% of the clips were labeled in triplicate to evaluate intercoder reliability). The clips were labeled with a quality score from 1 (lowest quality) to 5 (highest quality). Clips of 3 or greater were considered passing. An AI training model was developed to score the quality of the FAST examination. The clips were split into a training set, a validation set, and a test set. The predicted scores were rounded to the nearest quality level to distinguish passing from failing clips. RESULTS: A total of 1,514 qualified clips (1,399 passing and 115 failing clips) were evaluated in the final data set. This final data set had a 94% agreement between pairs of coders on the pass/fail prediction, and the set had a Krippendorff α of 66%. The decision threshold can be tuned to achieve the desired tradeoff between precision and sensitivity. Without using the AI model, a reviewer would, on average, examine roughly 25 clips for every 1 failing clip identified. In contrast, using our model with a decision threshold of 0.015, a reviewer would examine roughly five clips for every one failing clip - a fivefold reduction in clips reviewed while still correctly identifying 85% of passing clips. CONCLUSION: Integration of AI holds significant promise in improving the accurate evaluation of FAST images while simultaneously alleviating the workload burden on expert physicians. LEVEL OF EVIDENCE: Diagnostic Test/Criteria; Level II.

Deep Learning for FAST Quality Assessment.

OBJECTIVES: To determine the feasibility of using a deep learning (DL) algorithm to assess the quality of focused assessment with sonography in trauma (FAST) exams. METHODS: Our dataset consists of 441 FAST exams, classified as good-quality or poor-quality, with 3161 videos. We first used convolutional neural networks (CNNs), pretrained on the Imagenet dataset and fine-tuned on the FAST dataset. Second, we trained a CNN autoencoder to compress FAST images, with a 20-1 compression ratio. The compressed codes were input to a two-layer classifier network. To train the networks, each video was labeled with the quality of the exam, and the frames were labeled with the quality of the video. For inference, a video was classified as poor-quality if half the frames were classified as poor-quality by the network, and an exam was classified as poor-quality if half the videos were classified as poor-quality. RESULTS: The results with the encoder-classifier networks were much better than the transfer learning results with CNNs. This was primarily because the Imagenet dataset is not a good match for the ultrasound quality assessment problem. The DL models produced video sensitivities and specificities of 99% and 98% on held-out test sets. CONCLUSIONS: Using an autoencoder to compress FAST images is a very effective way to obtain features that can be used to predict exam quality. These features are more suitable than those obtained from CNNs pretrained on Imagenet.

A Multifaceted Resident-Led Physical Exam Course.

This article was migrated. The article was marked as recommended. BACKGROUND: Physicians frequently report poor confidence applying the physical exam for medical decision making. We developed a novel, multifaceted, resident-led curriculum to teach medical students the physical exam for clinical practice. METHODS: We created a two-week elective comprised of didactics, journal discussions, bedside ultrasound, and physical exam rounds for fourth year medical students at the University of South Carolina School of Medicine - Greenville. JAMA: The Rational Clinical Exam and Evidence-Based Physical Diagnoses, by Steven McGee, MD, were used to develop content. The curriculum focused on cardiac, pulmonary, abdominal, endocrine, and neurologic exams. Faculty and residents facilitated all portions of the course. Chi-squared testing was used to calculate confidence intervals on pre- and post-course assessments. RESULTS: Twenty-two fourth year medical students enrolled in the elective over the course of three years. Seventeen faculty, three chief residents, and 13 residents provided instruction. Residents provided roughly half of the total instruction hours. Students demonstrated statistically significant improvement on multiple choice pre-course and post-course assessments (56.8% vs 77.1%, p < 0.001). 95.5% of students reported feeling "confident" in their physical exam skills after the course. CONCLUSION: After participating in the course, students demonstrated improved skill and comfort using the physical exam for clinical decision making.

Incidental Diagnosis of Adult Beta-Thalassemia With Point-of-Care Ultrasound in the Emergency Department: A Case Report.

Beta-thalassemia is an autosomal recessive hemoglobinopathy that can result in microcytic hypochromic anemia, splenomegaly, hypercoagulability, and long-term sequelae. Beta-thalassemia intermedia, specifically, is diagnosed based on the moderate severity of illness, which does not carry the early symptomatic urgency of beta-thalassemia major, although patients of both often become chronically or intermittently transfusion-dependent. A presenting symptom may be splenomegaly, which is most efficiently detected with a combination of physical examination and point-of-care ultrasound (POCUS). We present the case of a 25-year-old male patient with no significant past medical history who presented to the emergency department with abdominal discomfort for one week. The history of present illness, review of systems, and physical exam were nonrevealing. An ultrasound was performed to rule out renal colic; however, he was incidentally found to have an enlarged and infarcted spleen. This unexpected discovery warranted a laboratory workup, which indicated beta-thalassemia intermedia. His diagnosis was confirmed with serum protein electrophoresis and he was thereafter followed by hematology. Beta-thalassemia intermedia can present suddenly in adulthood, despite a benign past medical history. Splenomegaly may be a presenting symptom and can be effectively detected with a physical exam plus POCUS. Failure to detect these subtleties can lead to potentially life-threatening conditions such as profound anemia, thromboembolic accidents, pulmonary hypertension, and pathological fractures. This case demonstrates the importance of utilizing POCUS in combination with a physical examination to attain a comprehensive perspective of anatomy, even in those patients fast-tracked in the emergency department.

Limb threatening thigh hematoma diagnosis accelerated by emergency physician bedside ultrasound.

INTRODUCTION: Compartment syndrome is a serious condition that requires prompt diagnosis, specialty consultation, and definitive management to prevent significant morbidity. Traditionally, compartment syndrome is identified by physical exam findings including the presence of pain, pallor, paresthesia, pulselessness, and paralysis involving the affected limb. Identifying the presence of compartment syndrome prior to the onset of signs that portend a poor outcome (i.e. pallor, pulselessness, and paralysis) can be challenging since many other less serious traumatic conditions can lead to paresthesia and pain in a limb. Bedside ultrasound is increasingly being utilized by emergency providers to expedite identification of various emergent diagnoses and guide care for patients who present to emergency departments. Bedside ultrasound allows emergency providers to visualize pathologic processes occurring that may be difficult to identify through traditional physical exam findings. This case report highlights the use of bedside ultrasound to promptly identify the presence of a traumatic thigh hematoma, which led to expedited advanced imaging and specialty consultation for compartment syndrome prior to the onset of physical exam findings consistent with compartment syndrome. CONCLUSION: The identification of compartment syndrome in the early stages is challenging given the overlap of signs and symptoms with other less emergent conditions. Early diagnosis of compartment syndrome is important to decrease morbidity, which can result from a delayed diagnosis of compartment syndrome. To our knowledge, this is the first case report to describe the use of bedside ultrasound to aid in the diagnosis of compartment syndrome and accelerate the care for a patient who presented with a traumatic thigh hematoma, which rapidly progressed to compartment syndrome and required emergent operative intervention.

Point-of-Care Ultrasound in Established Settings.

The original and most widely accepted applications for point-of-care ultrasound (POCUS) are in the settings of trauma, shock, and bedside procedures. Trauma was the original setting for the introduction of POCUS and has been standardized under the four-plus view examination called the Focused Assessment with Sonography in Trauma (FAST). This examination was found to be especially practice changing for achieving rapid diagnoses in critically ill patients who are too unstable for the delays and transportation inherent in more advanced imaging with computed tomography. This application was broadened from the critically ill trauma patient to any critically ill patient, particularly the patient in undifferentiated shock. Although the Focused Assessment with Sonography in Trauma examination originally focused on sources of hemorrhage causing hypovolemic shock, POCUS also can quickly differentiate cardiogenic, obstructive, and distributive shock and help identify the more specific etiology such as massive pulmonary emboli, pericardial tamponade, and pneumothoraces. By expediting diagnosis, POCUS facilitates faster definitive treatment of life-threatening conditions. In pursuing treatment, US continues to serve a role in the form of visually guiding many procedures that were previously done blindly. US guidance of procedures has improved the safety of central line insertion, thoracentesis, and paracentesis, and has an emerging role in lumbar puncture. Experience in bedside US is becoming a vital tool in the clinician's bedside assessment and management, filling a void between the stethoscope and the more advanced studies and interventions available through radiology. Understanding the strengths and limitations of US enables clinicians to identify the appropriate situations in which they can apply this tool confidently.

Versatile, Reusable, and Inexpensive Ultrasound Phantom Procedural Trainers.

We have constructed simple and inexpensive models for ultrasound-guided procedural training using synthetic ballistic gelatin. These models are durable, leak resistant, and able to be shaped to fit a variety of simulation scenarios to teach procedures. They provide realistic tactile and sonographic training for our learners in a safe, idealized setting.

Cost-effective, Reusable, Leak-resistant Ultrasound-guided Vascular Access Trainer.

BACKGROUND: Ultrasound guidance for insertion of a peripheral venous catheter is becoming common practice in many emergency departments in the difficult-to-access patient, and simulation has become an important tool for health care practitioners to learn this technique. Commercial trainers are expensive, and low-cost alternatives described to date provide a sub-optimal training experience. We introduce ballistics gel as a new material for the creation of simulating phantoms. MATERIALS AND METHODS: Directions describe construction of a simulating phantom composed of 10% ballistic gelatin and commonly available latex tubing. The model's success as used by one residency training program and medical school is described. RESULTS: Cost per phantom was $22.83, with less than an hour preparation time per phantom. We found these phantoms to offer a comparable user experience to commercially available products and better than other homemade products. DISCUSSION: Ballistics gel is a novel material for production of simulation phantoms that provides a low-cost, realistic simulation experience. The clear gel material works well for novice learners, and opacifying agents can be added to increase difficulty for more advanced learners. The material offers flexibility in design to make models for a broad range of skill instruction. CONCLUSION: A relatively quick and easy process using ballistics gel allowed us to create a simulation experience similar to commercially available trainers at a fraction of the cost.

Identification and functional characterization of a stable, centrally active derivative of the neurotensin (8-13) fragment as a potential first-in-class analgesic.

The neurotensin hexapapetide fragment NT(8-13) is a potent analgesic when administered directly to the central nervous system but does not cross the blood-brain barrier. A total of 43 novel derivatives of NT(8-13) were evaluated, with one, ABS212 (1), being most active in four rat models of pain when administered peripherally. Compound 1 binds to human neurotensin receptors 1 and 2 with IC(50) of 10.6 and 54.2 nM, respectively, and tolerance to the compound in a rat pain model did not develop after 12 days of daily administration. When it was administered peripherally, serum levels and neurotensin receptor binding potency of 1 peaked within 5 min and returned to baseline within 90-120 min; however, analgesic activity remained near maximum for >240 min. This could be due to its metabolism into an active fragment; however, all 4- and 5-mer hydrolysis products were inactive. This pharmacokinetic/pharmacodynamic dichotomy is discussed. Compound 1 is a candidate for development as a first-in-class analgesic.