Cost-Effective Training Models in Point-of-Care Ultrasound for Medical Students in Emergency Medicine: An Evaluation of Current Resources

Background Ultrasound is becoming more widely utilized in clinical practice; however, its effectiveness is limited by the operator’s skills. Simulation models are attractive options for developing skills because they allow inexperienced users to practice without the risk of endangering patients. Objective The purpose of this study was to identify commercially available and homemade ultrasound models to describe them in terms of materials, cost, and whether they are high- or low-fidelity for medical student education. Methods This is an investigational study on cost-effective ultrasound training methods for medical students. Our study was performed using search engines in Google, Google Scholar, and PubMed to search for models for the following five modalities: foreign body identification, intravenous (IV) injection training, abdominal ultrasound, ocular ultrasound, and ultrasound-guided lumbar puncture training. Results Most homemade models for foreign body identification, IV injection training, and ocular ultrasound could be created for less than $20. IV injection training models were the cheapest commercially available models. There are multiple commercially available options for abdominal ultrasound models, but no options were found for homemade construction. The construction cost for lumbar puncture models was larger due to the need to purchase an anatomically accurate set of lumbar vertebrae. Conclusions This study provides initial guidance and suggestions for ultrasound training models that are currently available. Ultrasound models that can be cheaply made or purchased increase accessibility for medical students to gain early exposure in a cost-effective and safe manner.


Introduction
Point-of-care ultrasonography (POCUS) is an ultrasound that is performed at the bedside to address a specific clinical concern. It is commonly used in the emergency medicine setting because it decreases the amount of time to make a diagnosis as it can be performed and interpreted rapidly. POCUS also reduces patient risk when used to guide procedures [1][2][3].
Considering that a certain level of competence must be obtained before utilizing POCUS within a clinical setting, ultrasound training is more commonly being included in medical school curricula [4]. Medical students who receive this training and exposure to ultrasound techniques are better suited to make accurate diagnoses using this tool. They are more likely to obtain accurate images and interpret them [5].
Ultrasound simulators are tools that students can use to practice both ultrasound-guided procedures and diagnostic ultrasonography. A variety of commercially available simulators can be constructed, with the main differences being similarity to human tissue and cost [6]. Simulators can be classified as either high- 1 2 2 fidelity, typically commercially available, or low-fidelity, typically homemade. Commercially available simulators tend to be much more expensive, ranging from hundreds to tens of thousands of US dollars depending on the target modality. The advantage of using high-fidelity simulators is that they more accurately replicate the anatomy and sonographic appearance of the target tissue [7]. The primary benefit of using homemade low-fidelity simulators is that they are much more inexpensive than high-fidelity simulators, but they are less realistic [7,8].
The focus of this article was to identify commercially available and homemade ultrasound simulators for the following modalities: foreign body identification, ultrasound-guided lumbar puncture, intravenous (IV) injection training, abdominal ultrasound, and ocular ultrasound. These modalities were chosen because they are either commonly performed procedures or represent pathologies that are widely observed in the emergency department. They are also essential to be applied in medical school curricula [9].
This article was previously presented as a meeting abstract at the 2021 American College of Osteopathic Emergency Physicians (ACOEP) Annual Scientific Assembly on October 12, 2021.

Data collection
This was an investigational study done by conducting searches on Google, Google Scholar, and PubMed for ultrasound-compatible models for the following modalities: foreign body identification, IV injection training, abdominal ultrasound, ocular ultrasound, and ultrasound-guided lumbar puncture. Search terms of simple keywords included "__ Ultrasound Training Model OR __ Ultrasound Training Simulator," and a term describing each modality was used. Inserting "Foreign Body," "IV Injection," "Abdominal," "Ocular," and "Lumbar Puncture" as the specific terms for each modality produced the most meaningful search results. Additional searches included secondary references from bibliographies. Data collection was done during June 2021.

Inclusion criteria
The examinations for websites listing commercially available models included the first five pages of search results within Google in order to find the most products currently marketed and widely available for purchase. Examinations for homemade models included results from Google Scholar and PubMed. We set out to find five models for each modality. When more than five models were found, the commercially available models or the homemade models with the lowest cost of materials were included.

Exclusion criteria
Homemade models that required supplies and ingredients not readily available in the United States were excluded. Homemade models that required access to a 3D printer to be constructed were not included in this article due to the additional cost and skills needed to operate a 3D printer. We excluded commercially available virtual reality simulators that did not involve the use of a tangible model as there was a significant price difference between the two types of models, and excluding virtual reality models allowed for fair comparison due to additional material costs of the tangible models. Models that did not list ultrasound compatibility in the product description were also excluded.

Data analysis
Within each modality listed above, the models were described and compared based on materials, cost, availability (homemade versus commercial), and reference articles for homemade model instructions or manufacturing companies for commercial models.

Foreign body identification
Foreign bodies that become retained within the soft tissue may lead to complications such as localized infection and inflammation. POCUS can be used for both diagnosis and removal of the retained foreign body [10]. Three commercially available foreign body ultrasound models were found that ranged in price from $599 to $3,049. Homemade model construction involves suspending a target (i.e., wood, needles, and cotton) in a bulking agent. The bulking agent is the component that makes up the surrounding material of the model. Ideally, it should be similar in sonographic appearance to human tissue [8,11]. Five articles detailing instructions on foreign body identification model creation were categorized based on the bulking agent used, as this was the most costly aspect of construction for this modality ( Table 1). The model created by Wichtel et al. [12] was the most expensive, with an average cost of CAD $31.50 (about USD $25.15) per model. They used silicone foreign bodies as the embedded target submerged in ballistic gel. Other models used meat-based bulking agents such as chicken breast and SPAM (a brand of precooked canned meat) [13,14]. An advantage of using the SPAM model is the ease of construction and the little preparation time needed. The foreign body can be inserted directly, and then, the top can be smoothed over with your fingers, so there are no markings to hint to the student where the foreign body is located [13]. The advantage of using chicken breast is the wide availability and presence of anatomic structures such as blood vessels and nerves. The disadvantage of poultry models is their perishability and the risk of infection with handling raw meats. Qurash et al. [15] describe a model made of gelatin, which is very inexpensive, widely available, and less prone to infection than meat-based models.  The construction of homemade IV injection training models uses methods that are similar to foreign body identification models. They can be made from the same bulking materials and use similar methods for construction. Still, the main difference is that for IV models, a tubing filled with fluid is typically placed in the bulking material to simulate vessels instead of a foreign body. Most materials used to create simulated vessels are relatively inexpensive. The main difference in price for homemade model construction is due to the use of a more expensive bulking material such as a ballistic gel. Rippey et al. [16] and Johnson [17] describe using modeling balloons as vessels, which are very inexpensive (about $10 for 100 packs) and are widely available. Nolting et al. [13] describe a method that uses a straw to create a tunnel in a block of SPAM (the bulking material), which is then sealed with a latex glove and filled with ultrasound jelly to simulate a vessel. Other simulated vessels can be created from Penrose drain and latex tubing [18,19].

Abdominal ultrasound
Multiple commercially available models for simulation of emergent POCUS procedures such as FAST scans were found. These models were high-fidelity and all much more expensive than the commercially available models for other modalities within this article, with a price range from $3,999 to $28,000. The Blue Phantom™ FAST Exam Real-Time Ultrasound Training Model (CAE Healthcare, Sarasota, FL, USA) was the most expensive model ($28,000) and provide features such as the ability to also practice transthoracic echocardiography (TTE) and pericardiocentesis procedure training. The SonoSkin® Ultrasound Diagnostic Wearable for FAST and eFAST training (Simulab Corporation, Seattle, WA, USA) is the cheapest model ($3,999), includes a wearable trainer placed on a standardized device patient or a mannequin, and includes the software that can display different pathological conditions. No articles describing the construction of homemade abdominal ultrasound models were found.

Ocular ultrasound
No commercially available ocular ultrasound models were found. Four articles discussing the formulation of a homemade ocular ultrasound model were evaluated. Murphy et al. [20] and Hajat et al. [21] used ping pong balls to form the globe, Jafri et al. [22] used a vending machine capsule as the mold for the globe, and Cuévas Gonzales [23] used an aluminum rod to shape the mold of the globe. Jafri et al. [22] described the use of ECG leads, glue, and foreign bodies to simulate different pathologic eye conditions. Hajat et al. [21] used additional material such as a pediatric microcuffed endotracheal tube, which allowed for the simulated pathologic changing of optic nerve sheath diameter to represent increased intracranial pressure and increased fidelity of the model.

Lumbar puncture training
The commercially available models ranged in price from $1,520 to $3,999. Out of the five commercially available models found and listed ( Table 2), Simulab Lumbar Puncture/Epidural and Blue Phantom LP and Spinal Epidural had been validated in terms of realism and compatibilities in a study by Vaughan et al. [25]. Two sources were found that described how to make homemade lumbar spine models to practice ultrasoundguided lumbar punctures. Gelatin or ballistic gel were used as the main bulking ingredients, and the cost of construction ranged from ~$30 to $130. All homemade models required the purchase of a lumbar spine model that was the primary determinant of the model expense and varied in price with the cheapest model mentioned by Morrow

Discussion
Multiple recipes for homemade models are available to construct foreign body identification models and IV injection training models. This is likely due to the simplicity of construction. Methods that involved using meat models such as chicken breast and SPAM forego steps to prepare the bulking agent, and the foreign body or simulated vessels can be placed in the desired position. Meat models are similar in echogenicity to human tissue, but the downside of using meat models is that they are perishable and limited in reusability [6]. Most materials used as foreign bodies or simulated vessels are relatively inexpensive, and the main difference in price for homemade model construction is due to the use of more expensive bulking materials such as a ballistic gel. The price for constructing ultrasound-compatible lumbar puncture training models is higher than the cost of construction for other modalities due to the need to purchase an anatomically accurate set of lumbar vertebrae. This extra cost, however, is minimal compared to commercially available lumbar puncture models.
High-fidelity models should be used for practicing FAST ultrasounds with simulated pathological conditions due to the complication of creating homemade models that accurately mimic the anatomy in this area. Alternatively, live human volunteers can be used to practice image optimization of structures that need to be visualized in a FAST scan, but students will rarely see pathological findings [4].
This study represents a first attempt to compile a suggested list of both commercially available and homemade ultrasound training models for a range of common modalities that are specific to the field of emergency medicine. This information would be valuable for both medical educators and learners as they try to explore the potential options for the most effective ultrasound training. We believe that this study promotes further investigation into the availability of newer and more affordable ultrasound models in medical training.

Limitations
A limitation of this article is that new homemade and commercial models could be available by the time this article is published. The prices of the listed models could also be different since we last searched. Information about commercially available products retrieved from Google was collected from the descriptions listed by the companies selling the products and was not peer-reviewed. There is also a lack of studies directly comparing differences in cost and materials among these five modalities. There is a need for further prospective comparison studies to determine differences in types of models compared to cost. Despite these limitations, our article provides some useful resources and ideas for ultrasound students and educators.

Conclusions
This study aimed to provide initial guidance on ultrasound training models that are currently available and categorize them based on cost and materials through the search parameters described above. This study cannot make conclusions about learner utilization for each model or the effectiveness in terms of ultrasound education. Medical students and other learners interested in practicing ultrasound-guided techniques will ideally find this article helpful in identifying models to use based on cost and availability.

Additional Information Disclosures
Human subjects: All authors have confirmed that this study did not involve human participants or tissue. Animal subjects: All authors have confirmed that this study did not involve animal subjects or tissue.

Conflicts of interest:
In compliance with the ICMJE uniform disclosure form, all authors declare the following: Payment/services info: All authors have declared that no financial support was received from any organization for the submitted work. Financial relationships: All authors have declared that they have no financial relationships at present or within the previous three years with any organizations that might have an interest in the submitted work. Other relationships: All authors have declared that there are no other relationships or activities that could appear to have influenced the submitted work.