3D Printing in Radiology Education


Introduction

Diagnostic radiology programs currently encompass image-based diagnosis and image-guided therapeutic techniques using multiple available imaging modalities. Three-dimensional (3D) image post-processing of radiologic images routinely uses high-resolution computed tomography (CT) and magnetic resonance imaging (MRI) datasets for diagnostic evaluation and treatment planning. Dedicated training in 3D modeling may be incorporated into some radiology training programs, although it is not required, and formal training programs are limited. In radiology, a comprehensive medical 3D printing training program should prepare radiologists to be knowledgeable and proficient in creating 3D printed medical models from radiological imaging data. This chapter will give background on 3D modeling for medical education and will provide an overview of the fundamentals required for a radiological program that includes medical 3D printing.

Historical Perspective on 3D Modeling for Medical Education

3D modeling in medical education has been used for generations. Together with two-dimensional (2D) drawings, scaled and realistic models have been used to record the discovery of anatomists, record unique and innovative patients and pathologies, and more recently, to widely disseminate both normal and abnormal teaching examples to students of anatomy, physiology, and medicine. 3D models are a staple of medical education; and the use of 3D models in medical schools is being expanded as the use of cadaveric dissection decreases. Surgical training has benefitted greatly from the availability of 3D models to plainly visualize pathology and simulate surgical approaches. The addition of readily available, patient-specific 3D printed models is a further step in the progression toward more personalized medical and surgical care.

One of the earliest existing anatomical models is an Early Classic Mayan head, dated to 300–600 AD. Half of this sculpture shows the head in life and the other half shows the underlying bony skull. In 1027 in China, an imperial physician, Wang Wei-Yi had two life-size bronze statues made for teaching surface anatomy for locating the acupuncture points. Sometime between 400–600 BC, an Indian sage named Sushruta was recorded to have used patient simulators for practice of surgical skills and suggested that such simulation-based education leads to competence and confidence. More recently, the 17th century father and son of the family Grégoire of Paris created obstetrical manikins for teaching midwives. Also, in 17th century, Gaetano Giulio Zummo (1656–1701) a Sicilian abbot created 3D models from wax and recommended them for anatomy training.

Over time, the materials used in creating realistic anatomic models have changed from wax to plaster and plastination. Preparing plastination models is a time-consuming and expensive process, which requires expertise for model preparation. These models are realistic but degrade readily and are easily damaged. More modern static models use plastic materials which rely on premade molds, limiting the flexibility of their application to individual patient care. As a tool for demonstration, however, virtual models and 3D printed models form the historical backbone and future for both hands-on learning and simulation.

3D Printing in Anatomy Education

Anatomy education, a traditional and key element of medical training, has evolved over the past few decades. The gold standard in anatomy education is cadaveric dissection, which is for many students their first encounter with a nonliving body. The process of dissection was traditionally viewed as a right-of-passage for students. Though this passage came with emotional and ethical conflict, dissection helped young physicians form a strong emotional connection to the form of the human body, which they sought to heal. Advances in technology have offered students even greater access to virtual dissection, which also closely mimics the way most nonsurgical physicians view clinical examples of pathology, that is, through imaging.

Anatomical dissection promotes deep anatomical understanding, and because each cadaver represents a unique individual, the process of dissection underscores the breadth of possible anatomical variations. As a 3D hands-on experience, cadaveric dissection offers tactile feedback and enhances manual skill sets. , Moreover, as students work in teams, cadaveric dissection promotes problem-based, team learning.

The total number of hours spent in anatomy teaching labs has decreased over the past 20 years for several reasons: the financial burden of having a fully equipped anatomy laboratory, limited cadaver availability, and the increased availability of e-learning platforms. While e-learning platforms have not fully replaced cadaveric dissection, they have greatly changed the ways students traditionally accessed anatomical information. In addition, these computer-based models are popular with students; however, studies have shown that students who rely solely on computer-based models perform worse compared to students who use traditional resources in learning anatomy.

3D printing as a novel method opens up opportunities to create anatomical models for medical training on an individual scale. Like all printed models, patient-specific 3D printed models allow students at all training levels to review both normal and abnormal anatomical structures outside the cadaveric laboratory. A wide variety of materials can be used to create 3D printed models, which can help accentuate anatomic details. These models are reproducible, safe to handle, and can represent variety of normal and pathologic anatomy.

When combined with cadaveric dissection (as many medical schools now perform CT scans on cadavers prior to dissection), 3D printing expands the possibilities for anatomy students. The creation of anatomical models by students which replicate the body's form promotes engagement with cadaveric specimens themselves. Prelearning through 3D modeling stimulates anatomical review, forcing students to understand anatomical relationships on a more direct level and facilitates kinesthetic learning by engaging the tactile senses.

Using 3D printing for anatomy training has its limitations. Compared to true anatomical dissection, fine details such as small nerve branches or microstructures, which can be explored in the cadaveric subjects using expert techniques, can be difficult if not impossible to replicate with 3D printing techniques. Whole organ printing with detachable parts requires a tradeoff between precision-printing, and the form and function needed to facilitate active engagement with the printed model.

A further limitation of using 3D printed models as a cadaveric replacement is the model printing time, which may limit the routine use of 3D printing in an ongoing course of study. Industrial 3D printers are better suited to producing multicolored models suitable for visualizing finer structures; however, local efforts to print with such fine detail make routine and on-demand 3D printed models expensive for most training purposes. Additionally, accurate size representation is an important element of student learning, which must be balanced by the time and material cost required for printing; the use of scaled 3D models is discouraged as it may potentially lead to an incomplete understanding of true organ size additional spatial relationships to nearby anatomical structures. The utility of 3D printing for medical education is a growing field of study. One recent systematic review validated the utility of 3D printed models for teaching medical students; and it was postulated that these models positively impacted medical students, especially because of their limited knowledge of anatomy.

3D Printed Models as a Tool in Clinical Radiology Training

Radiology practice, at its core, uses technology to visualize internal structures, assess anatomic relationships, and to infer pathology. These same tools are now used to quantify tissue structure and assess disease progression on a microstructural level. Key to success in radiology training programs is understanding anatomical relationships of increasingly greater complexity than those required of anatomy students, both in normal and in abnormal patients, as well as mastering anatomical description. 3D models can be used in education to visualize and conceptualize complex anatomical structures and are a useful tool for facilitating learning in a range of normal and abnormal patient-focused settings. Models can even include fine detailed structures such as ophthalmology anatomy and can be created based on cadaver prosections. Furthermore, since a catalog of models may not be available in many clinical learning environments, 3D printing allows a resident or student to select a specific area of interest that may be difficult to evaluate, and facilitates using individually printed models to teach these relationships to others.

Normal and Complex Anatomical Relationships

Due to the inherent complexity of normal anatomic structures and the fact that the human body is not made up of straight lines, smooth edges, and 2D interfaces, normal anatomical relationships are often difficult to comprehend. 3D models can be used to visualize and conceptualize complex anatomical structures, and have been shown to be effective tools in medical training. Studies have shown the utility of 3D printed models for teaching complex surface anatomy and as an alternative to traditional didactic instruction.

Beyond identifying key structures on imaging, students often struggle to recognize the relationship between adjacent structures, for example, the ductal anatomy of the pancreas and common duct within the pancreas head. Surface anatomy and its relationship to underlying structures can be difficult to estimate using standard cross-sectional imaging. 3D models, in contrast, more easily demonstrate complex interfaces and allow students to better understand these spatial relationships. 3D printed models have similarly been used to teach complex segmental anatomy of organs such as lungs, liver, and prostate, or branching anatomy of the coronary arteries and circle of Willis.

Another example of using 3D printing to visualize complex anatomical relationships relates to vascular structures in the setting of both common and less common anatomical variants. For example, the left renal vein typically crosses anterior to the aorta when communicating to the IVC. However, important vascular variants including retroaortic and circumaortic renal veins are critical to recognize. The relationship between the aorta, IVC, and renal veins is difficult to conceptualize and students who encounter this variant anatomy benefit from advanced 3D visualization. Similarly, the number and length of the renal veins and arteries is an important consideration that drives presurgical imaging prior to renal transplant surgery and can be difficult to accurately demonstrate to surgeons using 2D sectional anatomy alone ( Fig. 10.1 ) .

Fig. 10.1, (A) Patient with retroaortic left renal vein feeding the lower pole ( yellow arrow ) in addition to main left renal vein. (B) Patient with duplicate bilateral renal arteries feeding the upper and lower poles.

Abnormal Pathologies

There are many common injuries and pathologies which recur in a clinical setting for which trainees rely on representative examples to make diagnoses and highlight contrasts. This occurs most commonly on call, when residents practice with greater independence.

Understanding and referencing classification schemes for abnormal pathologies is a challenging task among radiology trainees for which they typically rely on external reference comparisons including anatomical models, textbooks, and case review examples when making a diagnosis. The opportunity to use patient-specific examples of complex anatomical structures can provide an added benefit in academic hospitals and training environments. Specialty reading rooms such as for musculoskeletal and neurological imaging are well suited to identify and to archive printed examples of these complex cases. Clinical conferences held together with surgeons serve to amplify the benefits to trainees in both diagnostic and procedural subspecialties when 3D models are available as a visual reference during case discussion.

For example, the classification of hip acetabulum fracture types or Le Fort midface fracture classifications can be aided by using printed models as a reference, given the complex anatomy and 3D geometry of these structures. A study on radiology residents showed that residents who received 3D printed models during a didactic lecture regarding acetabular fractures had better learning outcomes compared to control group which only received the didactic presentation. Printed clinical examples move this teaching tool into the clinical learning environment.

Realistic Phantoms for Hardware and Software Evaluation

In radiology departments, phantoms with known material properties and geometries are utilized to properly calibrate imaging equipment and optimize image protocols. Commonly available phantoms are simple geometric phantoms or anthropomorphic phantoms which usually represent typical or average adult or pediatric patients. 3D printing allows for the creation of more realistic models based on patient-specific imaging data, thereby providing more accurate and reliable models for quality assurance (QA) and research investigation. In one example, patient-specific 3D printed phantoms of peripheral and central pulmonary embolism were used to optimize a CT pulmonary artery protocol. Researchers used varying kVp and pitch values and assessed their impact on radiation dose and image quality using 3D printed models of peripheral and central pulmonary embolism, achieving 80% dose reduction. In addition to learning about radiation dosing including “image gently” and “as low as reasonably achievable,” in a simulated environment without exposing a patient to radiation, using 3D printed phantoms gives an opportunity for trainees to experiment with the physics concepts in radiology. More details regarding 3D printed imaging phantoms can be found in Chapter 14 .

You're Reading a Preview

Become a Clinical Tree membership for Full access and enjoy Unlimited articles

Become membership

If you are a member. Log in here