Introduction: U.S. Army Special Forces Medics (18Ds) operate in austere environments where decisions regarding patient management may be limited by available resources. Portable ultrasound may allow for the detection of fractures in environments where other imaging modalities such as radiography are not readily available or practical. Objective: We used a simulation training model for the ultrasound diagnosis of long bone fractures to study the ability of 18Ds to detect the presence or absence of a fracture using a portable ultrasound. Methods: The fracture simulation model is composed of a bare turkey leg bone that is mechanically fractured and housed in a shallow plastic container within an opaque gelatin base solution. Five fracture patterns were created: transverse, segmental, oblique, comminuted, and no fracture. After a brief orientation session, twenty 18Ds evaluated the models in a blinded fashion with a SonoSite M-Turbo portable ultrasound device for the presence or absence of a fracture. Results: 18Ds demonstrated 100% sensitivity (95% CI: 94.2% to 100%) in fracture detection and an overall specificity of 90% (95% CI: 66.8-98.2%) due to two false positive assessments of the no fracture model. Conclusions: Using a portable ultrasound device, 18Ds were able to correctly detect the presence or absence of a simulated long bone fracture with a high degree of sensitivity and specificity. Future studies are needed to investigate the clinical impact of this diagnostic ability.