PBL is a student-centered teaching model that starts with complex and real “problems”, and through group cooperation and independent exploration, students learn the scientific knowledge hidden behind the problems and develop problem-solving skills and autonomous learning abilities [5][6]. In medical education, PBL can effectively promote the connection between basic theory and clinical practice, and cultivate students’ clinical reasoning, critical thinking, and lifelong learning habits. When applied to emergency CT teaching, it can integrate scattered knowledge points (such as cranial anatomy, CT parameters, stroke pathology) into specific clinical scenarios (such as “an elderly patient with sudden hemiplegia”), making the learning goals more targeted and intrinsically motivating.

Virtual simulation technology uses computers to generate highly realistic simulation environments, and students can interact with them through interactive devices [7][8]. In medical education, it can create a safe environment, allowing students to repeatedly practice dealing with various critical situations in a risk-free virtual environment, even including the consequences of operational errors, without incurring any clinical risks. It can also replicate complex processes: high-fidelity simulation covers the entire process from patient arrival, emergency triage, communication with clinical doctors, equipment preparation, patient transportation, scanning, to image transmission. It enables visualization and repeatability of the process: abstract pathophysiological changes (such as the dynamic process of aneurysm rupture and bleeding) and the internal working principles of equipment (such as the impact of different reconstruction algorithms on image quality) can be visually presented. Additionally, classic cases are available for students to repeatedly study, analyze, and be assessed.

The virtual simulation case library provides an “immersive problem context carrier” for PBL, transforming a textually described case into a “virtual patient” and “virtual work scene” that is audible, visible, and interactive. Meanwhile, PBL infuses “inquiry-based learning spirit” into virtual simulation, preventing the technology from becoming a simple operation simulator and guiding students to delve from “how to do it” to “why to do it” and “how to do it better”. This integration creates a “learning based on virtual context problems”, achieving the unity of “knowledge, action, and thought”. The “dual-cycle PBL teaching model” referred to in this study means the close coupling of the autonomous inquiry cycle of PBL (problem analysis-knowledge construction-solution-reflection) and the practical verification cycle of virtual simulation (situation interaction-operation decision-result generation-feedback), forming a closed-loop learning process of “inquiry-practice-reflection-re-inquiry”.

The construction of the case library is the cornerstone of the implementation of the teaching model and should follow the following principles: clinical authenticity—cases are derived from real emergency clinical work, with de-identification and educational design; the evolution of the condition, signs, and laboratory test results all conform to medical logic. Competency-oriented—each case focuses on cultivating a set of core competencies (such as the decision-making of scanning protocols for the acute chest pain triad). Progressive difficulty—the difficulty of the cases is set from simple to complex, from a single system (such as plain non-enhanced head CT) to the full process of multi-system trauma (whole-body CT), from clear indications to ambiguous ones. Multi-path interaction—key decision points are designed, and students’ different choices will lead to different process branches and outcomes (such as choosing the wrong scanning range leading to missed diagnosis, or improper communication causing complaints), and immediate feedback is provided.

The core data of the case library comes from the real and typical cases of the emergency CT room in the Department of Radiology of the First Affiliated Hospital of Wannan Medical College in the past three years. The data transformation process includes: 1) Ethical review and de-identification: With the approval of the hospital’s ethics committee, all personal information, such as names, ID numbers, and hospitalization numbers of all cases, is completely deleted or anonymized. 2) Case selection and “true value” determination: Two senior radiologists or above, based on the final clinical diagnosis, surgical pathology, or imaging follow-up results, jointly select cases that meet the teaching objectives and determine the key imaging signs (“true values”) and standardized scanning and post-processing protocols. 3) Teaching design: Integrate the complete clinical data (medical history, physical signs, laboratory tests) and the corresponding DICOM format CT images. The expert group designs the clinical manifestations, decision-making branch points, and expected learning objectives of the cases based on teaching principles. 4) Technical transformation: The de-identified image data, structured case scripts, and “true value” annotation information are handed over to the technical development team, which imports them into the virtual simulation engine to build an interactive three-dimensional virtual patient model and dynamic image generation logic.

This case library adopts a modular design, systematically constructing five teaching modules based on common core diseases in emergency and key skill dimensions. Each module selects typical clinical cases and clearly defines the focus of core competency cultivation, aiming to achieve stepwise training from single knowledge to comprehensive clinical competence through a virtual simulation environment (Table 1).

Table 1. Teaching modules for the construction of the emergency CT case library.

A complete virtual simulation case for emergency CT mainly consists of five core elements: three-dimensional virtual environment, virtual patient model, virtual equipment simulation, dynamic image generation engine, and intelligent evaluation and feedback system. These elements jointly build a highly realistic, safe, controllable, and repeatable training digital environment (Table 2).

Table 2. System composition of the emergency CT virtual simulation case.

The rule base of the intelligent assessment system is jointly developed by clinical and educational experts, covering multiple dimensions such as process, technology, safety, and communication. Representative scoring items include: 1) Reasonableness of scanning parameters: For instance, for a routine adult brain scan, whether the tube voltage is within the range of 120 - 140 kV and whether the noise index exceeds the preset threshold. 2) Efficiency of scanning time: For stroke green channel cases, whether the total time from the patient’s “arrival” at the CT room to the completion of the scan is within the target time. 3) Safety checks and contraindication identification: The system records whether the student virtually inquires about the patient’s allergy history, renal function (such as creatinine value), and pregnancy status before the examination. Points are deducted for missing key items. Feedback methods include real-time prompts (such as popping up a prompt box when the set parameters significantly deviate from the norm) and post-task comprehensive reports (generating a detailed score sheet after the case ends, listing each item’s score, error points, and improvement suggestions). The assessment of critical value identification is judged by whether the student clicks and marks the key lesion (such as a large area of cerebral hemorrhage) within the specified time after browsing the images, and whether the system’s built-in communication tool simulates the sending of a report.

4.1.1. Case Selection and Problem Design

Teachers select matching cases from the case database based on the teaching objectives of the syllabus. Carefully design driving questions, such as: “As the on-duty technician, upon receiving a check request from a patient with sudden severe chest pain and breathing difficulties, how would you plan the subsequent work?”

4.1.2. Preparation of Learning Resources

Provide an index of background materials related to the case (such as clinical guidelines, sections of equipment manuals, and classic literature), but do not directly provide the answers.

4.1.3. Student Grouping and Role Assignment

Divide students into groups of 4 to 6, and within each group, students can take turns playing the roles of the main operator, assistant operator, communicator with the emergency doctor, and patient’s family member, etc., to promote multi-angle thinking.

4.2.1. First Stage: Context Introduction and Problem Analysis (Online/Offline)

Each group enters the virtual case scenario through the terminal and gets the initial information of the “virtual patient” (age, gender, chief complaint, vital signs). Under the guidance of the PBL instructor, they conduct a brainstorming session around the driving question and list the problems that need to be solved: What disease might the patient have? What CT examination is most needed? What risks (allergy, renal function) need to be evaluated before the scan? What if the patient is agitated? What are the expected key imaging manifestations?

4.2.2. Phase Two: Autonomous Exploration and Knowledge Construction

Group members divide tasks and collaborate, using the provided resource package and independently searching for information to learn about the pathophysiology of related diseases, CT examination protocols, imaging manifestations, and precautions.

4.2.3. Phase Three: Virtual Simulation Practice and Decision-making

The group operates the virtual simulation system together, completing the entire process from patient reception to examination completion. Communication and assessment: Engage in dialogue with virtual doctors, nurses, or family members to obtain more detailed medical history; conduct safety assessments on virtual patients. Plan formulation and operation: Set the scanning range, parameters, contrast agent injection plan, etc., on the virtual CT console; the system will provide real-time feedback on whether the parameters are reasonable. Image acquisition and initial judgment: After the scan is completed, the system generates images; the group needs to browse the images together, identify key signs, determine if the examination was successful, and whether additional scans are needed. Handling emergencies: The system presets emergency situations (such as patients vomiting during the scan or having an allergic reaction to the contrast agent), testing students’ emergency response capabilities.

Each group organizes the practice process, focusing on analyzing the basis for decision-making, gains and losses from operations, encountered difficulties and solutions, and prepares for presentation.

4.4.1. Multi-Dimensional Evaluation

Systematic Automatic Evaluation: Based on operation logs and image results, generate quantitative evaluation reports (such as operation standardization scores, image quality scores, and time efficiency scores). Peer Evaluation and Self-evaluation within Groups: Evaluate teamwork and problem-solving processes. Teacher Evaluation: Qualitatively assess students’ clinical thinking and knowledge integration abilities based on group presentation performances and reflection reports.

4.4.2. In-Depth Reflection and Expansion

Teachers organize summary discussions to reveal the core knowledge chain behind the cases, compare different groups’ solutions, and elevate the learning theme. Guide students to write reflection logs, connecting virtual experiences with future real clinical work.