3-dimensional reconstruction and mixed reality in thoracic surgery: a narrative review and user guide

Introduction

The thoracic surgery field is rapidly transforming with a variety of new technology, including 3-dimensional (3D) reconstruction, augmented reality (AR), and virtual reality (VR) (1). There are multiple programs currently approved for use in thoracic surgery with many more in development, but overall adoption into practice has been slow. This is likely due to a number of factors, not the least of which is the unfamiliarity with how to access and use them. We will cover what each of these platforms entail, how they can be used to benefit patients, how physicians can access them, what products are approved for use, and current surgeon perceptions and limitations.

3D reconstruction can refer to the creation of models in both the physical and virtual world. 3D reconstruction in the physical world typically involves 3D-printed models or implants. These are usually created from computed tomography (CT) scans, but can also come from other imaging modalities, such as magnetic resonance imaging (MRI) or external laser scans (2). 3D-printed models may be used by surgeons for either surgical planning or better visualization of anatomy prior to determining suitability for surgery. 3D-printed implants have been described largely in the context of chest wall pathology, such as rib replacement, chest wall reconstruction or prostheses for sternal deformities like pectus excavatum (2,3).

In the virtual world, 3D reconstruction can take multiple forms. Many hospitals already provide visual 3D reconstructions of CT scans, for example, in the case of rib and skull fractures, but this technology has become more advanced over time (4). Prior 3D reconstructions were generally created by hospital radiologists and radiology technicians and could be viewed directly on the hospital’s own imaging software. Surgeons could rotate the images to see different angles, but they were still presented on a 2D screen. Newer technology can create similar reconstructions manually or with artificial intelligence that may be viewed on AR or VR headsets allowing for a more realistic, immersive 3D experience (5). VR involves the projection of these images in a completely simulated environment whereas AR projects the generated images over the real world (6). In some situations, there may be a combination of the two, termed “mixed reality” or “extended reality” (6).

Although adoption of these platforms and techniques into practice may seem daunting, the underlying process can be easily broken down and a number of Food and Drug Administration (FDA) approved products exist. In this context, we aim to form a review of AR, VR, and 3D models to guide thoracic surgeons in the adoption of this technology with the ultimate goal of improving surgical safety and patient care. We present this article in accordance with the Narrative Review reporting checklist (available at https://jtd.amegroups.com/article/view/10.21037/jtd-2025-1779/rc).

Methods

PubMed was searched using MeSH terms “thoracic surgery” combined with “augmented reality”, “virtual reality”, and “3D reconstruction”. We limited the search to the last 15 years (January 1, 2010–June 30, 2025) yielding a total of 287 publications. We identified the highest impact articles involving each of these and included them in our review. Particularly, we identified articles that covered how these platforms are developed, what products are available, and how they may be used within thoracic surgery, including lung, chest wall, mediastinal, and esophageal pathology. A list of hand-selected studies are shown in Table S1. We supplemented our search with a list of 510k-approved VR and AR technologies found on the FDA website. Specifically, we identified all approved products that may apply to thoracic surgery. An outline of our search strategy can be seen in Table 1, and a PRISMA-style flow diagram can be seen in Figure S1.

Creating a model

The general schematic for generating a 3D model involves three steps: (I) image acquisition; (II) image segmentation and post-processing; and (III) segmented image model application. In the first step, image acquisition is performed via multiple conventional modalities with consideration for 3D models requiring volumetric imaging data. Such modalities may include CT, volumetric 3D ultrasounds, and MRI. Of these, CT has been the primary imaging modality of choice for 3D modelling due to its high spatial resolution (submillimeter slices) and use in diagnosis of most cardiothoracic pathology. It is less subject to artifact found in other modalities like shadowing in ultrasound and does not have the metal implant limitations of MRI. Image acquisition typically results in a dataset that can be stored into formats, such as Digital Imaging and Communications in Medicine (DICOM), that can then be used for segmentation and post-processing (7).

After image acquisition, the imaging data is imported into software programs that allow for processing. Multiple software development environments are readily available and range from testing programs published on open-source coding forums to sophisticated commercial software products with proprietary algorithms (8). Some of the more prominent examples of such image processing software include 3D Slicer, MATLAB (MathWorks), MIMICS (Materialise NV) and Synapse (Fuji) which have widespread applications in thoracic surgery (9-12). These environments allow the user to import volumetric data files such as DICOM and create a 3D volumetric output to visualize the image. Depending on the objectives of the 3D application, various methods for segmentation can be applied to obtain the desired output. On a high level, image segmentation involves importing a volumetric image dataset, performing various software techniques that allow for feature detection, and classifying these features into distinct categories for further post-processing (13). In the context of 3D reconstruction of medical images, the objective of image segmentation is typically to delineate different anatomical structures or tissue types, such as muscle, bone, vessels, organs, and tumor tissue. This is achieved through specific characteristics of the image, such as structural edges, color variations, and texture analysis (14). Once the characteristic features of the target anatomy are discovered, the structure can then be classified or “segmented” from the entire image dataset for further analysis. The segmented dataset can often be visualized to validate the model through masks, a visual overlay onto a CT or native image (15). Additional modifications to the segmentation methods are then made as needed until the desired segmentation output is obtained. This output can then be further processed to align with 3D reconstruction goals such as color adjustment or texture modification. Once the model has been validated, the application phase of the segmentation model follows (16).

Segmentation models can be utilized in 3D printing, AR or VR. The output from the segmentation model can be exported into different file types [i.e., computer-aided design (CAD) or stereolithography (STL)]. The final dataset contained within these files can be imported into various systems including AR/VR headsets or 3D printers, which use various methods such as STL, fused deposition, and selective laser melting (13,17). AR and VR applications typically require a head mounted display, such as the commercially available Oculus (Meta, Menlo Park, CA, USA) to provide a platform for projecting the image dataset onto either the real-world through a camera (AR) or into a virtual environment generated through software (VR). An overview of this whole process can be seen in Figure 1.

Figure 1 Flowchart demonstrating creation of 3D models from DICOM file (left) to segmentation and creation of STL file (middle) to the final model via 3D printing, virtual reality and augmented reality (right). 3D, three-dimensional; DICOM, Digital Imaging and Communications in Medicine; STL, stereolithography.

The financial implication of such technology is not inconsequential. For 3D-printed models, the overall expenses largely involve the upfront cost of a 3D printer as well as recurring expenses for machine maintenance and resin. Fortunately, the cost of the 3D printers and resin has decreased over time. The cost to create a small 3D model with two segmented layers is roughly the same as that of a single stapler fire used intra-operatively (18). The 3D-printed model itself can cost anywhere from $1.50 to $203 (19). However, larger models with multiple layers of segmentation or movable pieces with magnets may cost more. On the other hand, VR and AR models require the upfront cost of the headset device, and recurring expenses vary depending on the vendor/product used. If a user is creating their own VR/AR models, then this cost is minimal. However, some commercial platforms are subscription-based while other platforms may have per case fees to create the models, which can add up over time.

As with other technology, companies that develop these programs for commercial use may spend millions of dollars on their development, which is eventually passed down to the consumer. One study estimated the use of VR in training a surgical novice to proficiency could cost up to $100,000 (20). Another study estimated VR simulator capital expenditures for a single ophthalmology module can total approximately $250,000 (21). Creation of a single VR 3D model for surgical planning on one patient can cost roughly $5,000 (22).

Time to commercial production of different platforms can take multiple years and represents a significant investment of resources, capital and education (23). The timeline is affected by components such as software development, pilot testing, and arguably most by regulatory controls and market approvals (24). The economic benefit of these technologies may be derived from studies that demonstrate shorter operating times. The actual cost savings depend on the cost-charge ratio models used as well as associated labor costs with operating room staff. However, decreased overall operating time likely has a financial breakeven point (25).

Uses in thoracic surgery

AR and VR platforms have been developed for many different surgical areas (26,27); however, they are especially relevant in thoracic surgery due to the complexity of the space, numerous potential anatomic variants, and constant physiologic motion of major organs like the heart and lungs (28). Therefore, 3D reconstruction models are paramount for identifying these anomalies pre-operatively, proper surgical planning, and precise lesion localization intra-operatively (29,30). This can be applied to nearly any intra-thoracic pathology.

Lung pathology

AR and VR platforms have been developed and tested for multiple purposes regarding lung pathology. One example is for the training of residents and fellows to operate in this high-risk space. Multiple VR simulations of video-assisted thoracoscopic surgery (VATS) lobectomy have been created over the last decade, some with the ability to provide participants with a score based on estimated blood loss and operative length (31-33). One experimental study found VR VATS simulation could successfully quantify significant differences in operating metrics based on level of experience (i.e., novice vs. experienced) in operating time (41.1 vs. 24.3 seconds, P<0.001), blood loss (692.3 vs. 315.0 mL, P=0.006), and total instrument path (78.7 vs. 59.1 meters, P=0.02) (32). Widespread implementation of these platforms could allow trainees to have more realistic practice prior to operating on real patients, thus making the overall training environment safer.

3D reconstruction of lungs and lung tumors may also be useful in diagnosis and surgical planning. One retrospective pilot study used automated artificial intelligence and manual techniques for creation of virtual lung reconstruction models from patients’ non-contrast CT scans to detect anatomical variants, which were then confirmed intra-operatively (34). Pulmonary vessel and airway variants were detected with similar accuracy in both sets of reconstruction, revealing that these techniques may be useful for identifying anatomic variants pre-operatively. Similar VR platforms have been developed for segmentectomy planning. One VR program, PulmoVR (Rotterdam, The Netherlands) used an artificial intelligence-based algorithm to create 3D virtual models from patients’ CT scans (35). The model automatically identified different segments of the lungs, allowing a surgeon to more accurately identify which segment a tumor lies within. In the study, the planned segment for resection changed after viewing the PulmoVR model in 4 out of 10 cases, and all tumors were successfully removed after VR planning (35). Multiple groups have created similar VR models to plan segmentectomies (36,37). One of these studies utilized the technology on 50 cases and found that 52% of surgical plans were adjusted after viewing the VR, ultimately leading to a complete resection in 98% of cases (36). Lung segments can be difficult to delineate on traditional CT scans, so there is a lot of potential for future use of 3D reconstruction models in segmentectomy planning.

Similar technology can be applied to other types of lung resections, such as lobectomy. These may combine VR, AR and 3D printing to allow for optimal pre-operative manipulation of the model, ultimately making the operation quicker and safer (38). One study compared surgeons who looked at 3D models on a screen vs. those using AR and 3D printing in pre-operative planning (39). They found that those who used AR and 3D printing together had shorter operative times (65.00–87.25 vs. 78.00–107.25 minutes; P<0.001), less estimated blood loss (22.00–46.25 vs. 33.75–60.00 mL; P=0.006), and shorter length of hospital stay (3–4 vs. 4–5 days; P=0.001) (39). Another group used representative videos of lobectomies and segmentectomies to generate virtual images that would create a “resection process map” and help simulate the procedure for surgeons pre-operatively (40). These studies all serve as proof of concept and show how pre-operative 3D reconstruction may be superior to the typical review of CT scans alone or 3D reconstructions viewed on a 2-dimensional (2D) screen, as was historically done.

Other models have taken this a step further and overlaid 3D simulations onto intra-operative surgical images via AR (40). AR models can be manipulated in the operating room to identify lung anatomy, simulate lung deflation and help guide optimal port placement (41). This may ultimately increase surgeon ergonomics, economy of motion and productivity in the case. Additionally, multiple studies have used AR or mixed reality to identify nonpalpable nodules smaller than 2 cm intra-operatively (42,43). One case series used cone beam CT to create a 3D reconstruction model, which was then combined with fluoroscopic images to create augmented fluoroscopy for identification of non-palpable lesions intra-operatively with high accuracy typically within 1 cm (range, 0–20 mm) (42). Another group created a 3D-printed model and mixed reality for combined use intra-operatively to locate nonpalpable tumors (43). In this study, there was some deviation between actual location of the nodule and the mixed reality location, but they were generally accurate to within ~1 cm and all subsequent resection margins were negative. Therefore, further work is certainly needed to improve accuracy of some models, especially as it relates to localization after lung deflation, but results thus far are promising. Current localization techniques, such as bronchoscopic microcoil placement or dye injection, could potentially be replaced by AR techniques, thus reducing unnecessary procedures and potential for their associated complications (44,45).

Chest wall pathology

3D reconstruction techniques are particularly useful for both benign and malignant chest wall pathology. Benign conditions can include deformities of the chest wall, such as pectus excavatum, or benign tumors and structural lesions, such as desmoid tumors or fibrous dysplasia of the ribs (46,47). Malignant tumors, such as chest wall sarcoma, may require a wide local excision, leading to a large defect that requires reconstruction (48). Numerous studies and companies have looked at the role of 3D reconstruction in surgical planning and creation of prosthetics for implantation intra-operatively, of which, we will provide a few examples.

One group created a 3D model of a chest wall desmoid tumor to better understand its invasion into surrounding structures (49). This allowed a multidisciplinary team to view the model together, determine resectability, and plan the operation. Similar 3D-printed models can be seen in Figure 2. Another group used a VR software that allowed cutting, drawing, and measuring of patients’ chest wall tumors within a virtual environment to simulate surgery (50). These programs demonstrate successful cases of pre-operative 3D reconstruction use for surgical planning; however, it’s important to understand how these compare to standard surgical planning with 2D CT scans. One retrospective cohort study compared operative planning with VR vs. CT scans in 28 patients with non-small cell lung cancer invading into the chest wall. They found that chest wall resection and reconstruction planning was significantly more accurate when using the VR platform compared to CT scans alone (28.6% vs. 18.3%; P=0.02) (51). Another retrospective cohort study compared surgical planning with 3D reconstruction and 3D-printed models to surgeries planned with CT scans only in 34 cases, and found that the 3D reconstruction group had shorter operative times, less estimated blood loss, and fewer changes to the incision approach (P<0.05) (52).

Figure 2 3D printed models from multiple cases of chest wall tumors. (A) Relationship between intrathoracic tumor (green) and internal mammary artery and sternum. (B) Relationship between extrathoracic tumor (dark blue) and pectoralis muscle (clear). (C) Intrathoracic view of tumor from (B) to assess relationship with ribs and internal mammary artery. 3D, three-dimensional.

Notably, 3D-printed models can also be useful for planning of rib-plating. One retrospective cohort study compared 3D printing rib fractures for planning before surgery to CT review only and found that the 3D printing group had decreased rates of poor fracture fixation, lower off-plate rate, and lower dislocation rates (53). These studies clearly represent an advantage to the use of 3D reconstruction models, whether virtual or printed. In addition to surgical planning, some VR models can be overlaid onto the patient in the operating room via AR to assist with precise marking of tumors and margins, as shown in Figure 3. These models can help bridge surgical planning and execution of the actual procedure in the operating room.

Figure 3 Example of virtual reality and augmented reality use on a patient with a chest wall chondrosarcoma (green mass). (A) Virtual reality view of thoracic cavity with incorporated computed tomography images. (B) Augmented reality view intra-operatively during patient marking. This image is published with the patient’s and surgeon’s consent. All images were created using ImmersiveAR^TM.

These models can also be turned into 3D-printed prosthetics that may be used intra-operatively. For example, the literature contains many reports of 3D-printed titanium rib- and sternal-prostheses for a variety of pathology, including desmoid fibromatosis, chondrosarcoma, lung cancer invading the chest wall, and pathologic fractures (3,54-57). Other studies have shown the long-term benefits of these prosthetics with minimal complications, durable fixation, and favorable histocompatibility (58,59). Outside of chest wall defects and rib fractures, 3D-printed prostheses can also be useful in the treatment of deformities, such as pectus excavatum. Multiple studies have used different algorithms and artificial intelligence models to create optimal Nuss bars (2,60). These bars can either be 3D-printed in the exact specified form or bent to a pre-printed mold in the operating room; both methods have led to smooth and fast surgical procedures without post-operative complications (61,62).

Overall, the chest wall can have numerous different deformities and diseases requiring surgery. Measurements for reconstruction can vary significantly by patient, so 3D models are vital for resection and reconstruction planning. Additionally, pre-printed prostheses for intra-operative use can lead to safe, quick, and effective surgery in these cases.

Mediastinal pathology

Reconstruction models are also useful for surgical planning, education and intra-operative assistance around mediastinal structures, such as the esophagus, tracheobronchial tree, and thymus. This tight space has countless structures that can often be difficult to delineate on standard imaging, such as CT and MRI. Therefore, being able to remove structures in a virtual model or physically manipulate a 3D-printed model can help multidisciplinary teams determine resectability, endoscopic or bronchoscopic needs, and ultimately treatment plans.

There are multiple models that have been developed for learning mediastinal anatomy, both normal and abnormal. Unique programs have been made from real human scans to segment mediastinal anatomy, which can then be broken down into individual structures (63). Creating models from real patients’ scans allows for a more realistic review and study of the anatomy, particularly when viewed in a 3D format, such as VR or 3D-printed models. Examples of some of these 3D-printed models can be seen in Figure 4. One randomized controlled trial gave a group of students “3D”-reconstructed CT images displayed on a 2D screen and another group of students VR reconstructions of these same scans (64). The VR group achieved a better understanding of the CT images and had a greater interest in mediastinal anatomy and surgery (64). This can ultimately translate into real practice for trainees. One group created 3D reconstruction images that could simulate insertion of a thoracoscope and removal of individual structures to gain a better understanding of the patient’s anatomy prior to surgery (65). Manipulation of a 3D model in a risk-free environment may become vital to performing safe surgery in the future.

Figure 4 3D-printed models of mediastinal tumors and associated anatomy. (A) Middle mediastinal tumor (green) abutting esophagus, tracheobronchial tree and great vessels. (B) Mediastinal tumor (green) sandwiched between spine, tracheobronchial tree, pulmonary arteries, and azygos vein. 3D, three-dimensional.

In particular, multiple studies have been published on the use of 3D simulations in esophageal surgery, which is often one of the most complex operations given the adjacent vascular and airway structures. 3D models of the bronchial arteries in relation to the esophagus can be particularly helpful in the resection of esophageal cancer given significant variation in this anatomy by patient (66,67). Reconstructions of CT images may also allow for improved diagnostic capabilities in esophageal cancer. One retrospective cohort study found that measurements of tumors and lymph nodes correlated well with those taken during endoscopy, so eventually it’s possible that reconstructions could replace these more invasive methods of diagnosis and surveillance (68). These techniques for diagnosis and surgical planning are particularly useful in complex or re-operative patients (69).

Similar benefits have been seen for anterior mediastinal pathology. For example, one group used a 3D mediastinal model pre-operatively to view the relationship between bronchi and vessels to a thymic tumor (70). Other groups have used 3D models for virtual bronchoscopy and airway management planning in patients with large anterior mediastinal tumors that deform the airways (71,72). Therefore, they may not only guide thoracoscopic surgery but also associated bronchoscopic procedures and intra-operative airway management.

The tracheobronchial tree is a particularly complex structure that is difficult to fully visualize both pre-operatively and intra-operatively. Smaller airways are especially difficult to visualize and determine precise areas of stenosis, so there are multiple reports detailing the use of 3D-printed models in bronchoscopy training and localization of bronchial stenosis for stent placement (10,73,74). These may also be useful from a thoracoscopic approach when there is variant anatomy. For example, one report used a 3D-printed bronchial tree to visualize an aberrant S3 bronchus arising from the middle lobe bronchus and determine that a segmentectomy would be required in addition to initially planned middle lobectomy (75). Understanding a patient’s unique anatomic variants before beginning a procedure is a major benefit of having a 3D model that may demonstrate them better than a standard CT scan. Even with larger airways, such as the trachea, 3D-printed models can be useful for procedural training purposes, planning or creation of bioprosthetics for tracheal reconstruction (76,77). Examples of 3D-printed tracheobronchial anatomy can be seen in Figure 5. Overall, there are countless uses for 3D models in the mediastinum from education and surgical planning to intra-operative use or prosthetic implantation.

Figure 5 3D-printed models of tracheal tumors or tumors abutting the trachea and surrounding tracheobronchial anatomy. (A) Tumor (yellow) abutting trachea. (B) Left: adenoid cystic carcinoma (yellow) of trachea; right: trachea after resection; model used to determine necessary length of tracheal resection for surgical planning. (C) Same tumor (green) from (A) with vasculature added to model. (D) Tumor (green) abutting esophagus and trachea and surrounding vasculature. 3D, three-dimensional.

FDA approval process and existing products

Although we have demonstrated numerous models found in the literature and their potential uses, it is also important to understand the current state of commercially available models and what it takes to get to this point. There are three major pathways to United States FDA approval. The first, and more stringent, is premarket approval (PMA), which requires a comprehensive review of a device’s safety and effectiveness. This is typically meant for high-risk devices that may be life sustaining or pose potential risk, injury or illness. The second is the de novo pathway, which is meant for low-to-moderate risk devices without a similar prior device on the market. The third route is the Premarket Notification 510(k) approval process, which is intended for new devices that have a similar device previously on the market, termed a predicate (24). Most of these AR, VR and 3D reconstruction models will fall under the 510(k)-approval process as they are typically not high-risk devices and there are already many currently approved that may satisfy the substantial equivalence requirement to serve as a predicate. Substantial equivalence requires that relative to the predicate, the new device (I) has the same intended use; (II) has same or different technological characteristics and (III) is equally safe and effective (24). 510(k) approval is typically required when a device is introduced into marketing for the first time or there is a change or modification to a legally marketed device. This process only applies once the device is ready to be commercially distributed or marketed, and does not apply to those undergoing development, clinical evaluation or testing (24). However, devices during the clinical trials phase are subject to the Investigational Device Exemption (IDE) regulation (24).

The 510(k)-submission process starts with filling out the required application forms, assignment of a 510(k) number, and forwarding to the appropriate Office of Health Technology. Initial application review is undertaken and completed within 15 days of submission. The application review process involves (I) acceptance review; (II) substantive review; and (III) 510(k) decision (24). Acceptance review essentially refers to a standard checklist that the device must pass, determining whether it is eligible for 510(k), if clinical trials have been completed, and if the submission contains all required elements. After this, the submission moves to the substantive review phase, which involves a comprehensive review of the submission including communication between the lead reviewer and submitter to answer questions and provide additional information if requested. The cost of the 510(k) application is typically around $25,000 (24). The FDA provides a list of specific devices exempt from the Premarket Notification Submission process, which can be found on the FDAs website under the “Medical Device Exemptions 501(k) and Good Manufacturing Practice (GMP) requirements” page (78).

As previously discussed, there are many uses for 3D reconstruction models and devices in thoracic surgery, but most in the literature are not FDA approved and widely available. However, there are several existing models that do have FDA approval and are commercially available. Some, such as Ceevra, can be implemented into a hospital system and be easily used for pre-operative surgical planning (79). Others, like ImmersiveAR (ImmersiveTouch, Chicago, IL, USA) can be used intra-operatively for lesion localization and patient marking. One company called Osteobionix (Las Palmas de Gran Canaria, Spain) creates custom 3D-printed titanium implants for use in chest wall reconstruction. Overall, these models vary in price, acquisition time, required imaging, and required hardware for use. Examples of existing products can be seen in Table 2.

Current surgeon perceptions, limitations and hurdles

Despite many successful case reports and randomized controlled trials demonstrating the superiority of 3D reconstructive techniques available in the literature, universal uptake of this technology has been slow. Surveys of surgeons and doctors from different specialties throughout the world have revealed a few important points. For one, current perceptions of this technology vary by how familiar a person is with it and whether or not they’ve used it before (80). The majority of those that have used extended reality technology find it easy to use and feel it improves their surgical skills, efficiency and safety (80,81). However, many users also feel they are limited by unknown costs, time for acquisition, use, and learning on these platforms (80-82). In addition, certain groups of surgeons are more likely to desire the integration of 3D reconstruction models than others. In one survey, junior surgeons were more interested in integrating this technology into practice compared to senior surgeons (83). There is some heterogeneity in how surgeons believe this technology should be used as well. One cross-sectional, multispecialty study showed that 65% of respondents believed there were good clinical applications, but only 41% believed they would use it for clinical decision making and 35% believed it was superior to human decision making (84). Essentially, this tells us that the majority of surgeons are interested in using the technology, but don’t feel it will be entirely necessary or provide greater insights above that of their own clinical knowledge.

There are other limitations that will need to be addressed in the near future, such as costs associated with different platforms. Technology often gets cheaper as it becomes more established and widespread, but many of these models currently cost thousands of dollars per patient. This begs the question of who will cover the cost? Hospitals may be expected to front the bill, which is eventually passed on to the patient or their insurance provider. However, it’s likely that these institutions will push back on a service that is not deemed absolutely necessary. Therefore, significant evidence will be required attesting to the benefits of these reconstruction models on patient care. For example, every minute in the operating room can cost hundreds of dollars (25). Theoretically, if a 3D reconstruction program can cut down on operative time by a significant amount, this may offset the cost of acquiring the model. Additionally, if these models can help reduce expensive complications, this too may help offset the initial cost of using them. Another concern is disparities between patient populations. It’s likely that those with a higher socioeconomic status will be more likely to afford these supplementary tools, potentially increasing disparities in financial and structural barriers that are already present (85). Overall, getting insurance companies and hospitals to cover these costs will require extensive evidence, beyond what is currently available, to demonstrate increased safety and efficiency with use of these models. Another important point is that these regulatory processes, payment structures, and overall hospital costs may vary significantly by country.

Furthermore, at the current time, there are some technical barriers with current models. AR has been used intra-operatively to help with lesion localization and avoidance of critical structures, as previously discussed. However, integration of these VR and AR platforms is variable. Multiple groups have created video outputs of VR or AR models and fed them into a robotic platform using TilePro (86). However, this process is essentially a picture-in-picture feature and does not represent true integration, which we would envision as a similar process to FireFly imaging where it could be “toggled-on” during the operation and allow the entire viewfinder to show an AR overlay that adapts to motion and deformity in real time. One group was able to implement AR technology that adapts to the anatomy during robotic lower lobectomy by using artificial intelligence algorithms to automate segmentation and apply these images intra-operatively (87). However, this technology is not widely available. Therefore, although more integrated forms of AR and VR technology are being developed, it is limited at the current time. Additionally, VATS and open procedures may have difficulty integrating this technology unless lightweight cheaper headsets or projection systems are created that would allow everyone in the room to see the model at the same time (88). These auxiliary considerations will be important in the future development and integration of these systems.

Also important to note are potential legal implications of using this technology. It’s debatable whether or not we can entrust commercial suppliers with confidential patient information (89). If an error in the software leads to a surgical error or there is an exposure of patient information or breach of Health Insurance Portability and Accountability Act (HIPAA), who will be considered responsible? Integration of this new technology into regular practice will require careful consideration of existing platforms, hospital security, and information technology systems.

User guide

The successful implementation of available 3D reconstruction technology with either 3D models, VR, or AR can often be simpler than many expect. With the present review, we have laid out the spectrum of uses within thoracic surgery as well as many of the existing products that are on the market and their required approval. However, it is likely that many more products will come to market in the near future, which is why it is important for all surgeons to understand how to choose the best model for their purpose. Here, we outline a proposed “User Guide” in Figure 6 to help provide an initial framework in the adoption of such models.

Figure 6 Decision aid algorithm for choosing the appropriate 3D reconstruction technology. 3D, three-dimensional.

Additionally, it is important to understand what infrastructure is required. If the user is intending to use 3D-printed models for surgical education or surgical planning, this will require the proper equipment. Many hospitals already have 3D printers, but some may need to acquire them and the associated resins. Once materials have been obtained, radiologists and either information technology (IT) personnel or biomedical engineers can help with image segmentation, creation of the STL file and printing of that file on the institution’s 3D printer. Image segmentation, processing, and printing can take multiple days, so these models should be requested weeks prior to surgery to ensure completion with enough time to ensure their quality and study them appropriately.

Use of AR and VR models can depend on the user’s technological experience. As we’ve demonstrated in this review, many institutions have created their own pipelines and virtual models, but this generally requires experience in computer science or biomedical engineering. For many surgeons, this is not feasible, so use of pre-existing, publicly available platforms may be their best option. This typically involves contacting the respective company, sending a DICOM file, creating the 3D model and then using it on its respective platform. For products like Ceevra, this involves the company working with the institution’s IT department to install the program on end-user computers, connect the software to hospital picture archiving and communication systems (PACS), and upload of the patient’s CT scan to the company’s cloud for model creation. Models can then be easily viewed on surgeons’ phones, tablets, or computers. Companies that use headsets for VR or AR viewing often send company representatives who provide viewing devices compatible with their software. Creation of the model uses the same general process previously described. However, if a surgeon wants to have more flexible use of the model without a representative present, then it’s possible the hospital may need to invest in a VR headset for use, which can add to the expected costs. We provide a pre-implementation checklist in Figure 7 that provides an outline of resources needed for each type of model.

Figure 7 Pre-implementation checklist for future 3D reconstruction. AR, augmented reality; 3D, three-dimensional; STL, stereolithography; VR, virtual reality.

Conclusions

3D reconstruction platforms are an important development with many uses within thoracic surgery, including surgical planning, intra-operative guidance, prosthetic creation and implantation, and surgical training and education. Further studies into their efficacy, use, and safety will be vital in the coming years, particularly when compared to traditional imaging and localization techniques, such as wire-guided localization, dye injection, and radioactive seed placement. This review should serve as a guide to thoracic surgeons planning to incorporate this technology into their everyday practice. It is important for surgeons to understand the uses and limitations associated with 3D reconstruction models prior to implementing them to ensure the safety and best outcomes for patients.

Acknowledgments

None

Footnote

Reporting Checklist: The authors have completed the Narrative Review reporting checklist. Available at https://jtd.amegroups.com/article/view/10.21037/jtd-2025-1779/rc

Peer Review File: Available at https://jtd.amegroups.com/article/view/10.21037/jtd-2025-1779/prf

Funding: None.

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://jtd.amegroups.com/article/view/10.21037/jtd-2025-1779/coif). Z.M.A. serves as an unpaid editorial board member of Journal of Thoracic Disease from December 2023 to November 2025. The other authors have no conflicts of interest to declare.

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. Publication of the patient’s images was obtained with the patient’s and surgeon’s consent.

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. Chen-Yoshikawa TF, Fukui T, Nakamura S, et al. Current trends in thoracic surgery. Nagoya J Med Sci 2020;82:161-74. [Crossref] [PubMed]
2. Rodrigues PL, Rodrigues NF, Pinho AC, et al. Automatic modeling of pectus excavatum corrective prosthesis using artificial neural networks. Med Eng Phys 2014;36:1338-45. [Crossref] [PubMed]
3. Yang C, Chen L, Xie X, et al. Three-dimensional (3D)-printed custom-made titanium ribs for chest wall reconstruction post-desmoid fibromatosis resection. Comput Assist Surg (Abingdon) 2025;30:2456303. [Crossref] [PubMed]
4. Chen-Yoshikawa TF. Evolution of Three-Dimensional Computed Tomography Imaging in Thoracic Surgery. Cancers (Basel) 2024;16:2161. [Crossref] [PubMed]
5. Pohlman A, Abdelsattar ZM. The Use of Artificial Intelligence and Machine Learning in Thoracic Surgery. Thorac Surg Clin 2025;35:417-32. [Crossref] [PubMed]
6. Hayes MD, Amanda. What is augmented reality? IBM.com. 2024. [Accessed July 23, 2025]. Available online: https://www.ibm.com/think/topics/augmented-reality#:~:text=Augmented%20reality%20(AR)%20refers%20to,in%20commerce%2C%20manufacturing%20and%20entertainment
7. Sadeghi AH, Bakhuis W, Van Schaagen F, et al. Immersive 3D virtual reality imaging in planning minimally invasive and complex adult cardiac surgery. Eur Heart J Digit Health 2020;1:62-70. [Crossref] [PubMed]
8. Silva JNA, Southworth M, Raptis C, et al. Emerging Applications of Virtual Reality in Cardiovascular Medicine. JACC Basic Transl Sci 2018;3:420-30. [Crossref] [PubMed]
9. Cheng GZ, San Jose Estepar R, Folch E, et al. Three-dimensional Printing and 3D Slicer: Powerful Tools in Understanding and Treating Structural Lung Disease. Chest 2016;149:1136-42. [Crossref] [PubMed]
10. Tam MD, Laycock SD, Jayne D, et al. 3-D printouts of the tracheobronchial tree generated from CT images as an aid to management in a case of tracheobronchial chondromalacia caused by relapsing polychondritis. J Radiol Case Rep 2013;7:34-43. [Crossref] [PubMed]
11. Cheng GZ, Folch E, Brik R, et al. Three-dimensional modeled T-tube design and insertion in a patient with tracheal dehiscence. Chest 2015;148:e106-8. [Crossref] [PubMed]
12. Morrison RJ, Hollister SJ, Niedner MF, et al. Mitigation of tracheobronchomalacia with 3D-printed personalized medical devices in pediatric patients. Sci Transl Med 2015;7:285ra64. [Crossref] [PubMed]
13. Vukicevic M, Mosadegh B, Min JK, et al. Cardiac 3D Printing and its Future Directions. JACC Cardiovasc Imaging 2017;10:171-84. [Crossref] [PubMed]
14. Feature Extraction in Image Processing: Techniques and Applications. Geeks for Geeks. 2025. [Accessed July 2025]. Available online: https://www.geeksforgeeks.org/computer-vision/feature-extraction-in-image-processing-techniques-and-applications/
15. Liu C, Cheng Y, Tamura S. Masked image modeling-based boundary reconstruction for 3D medical image segmentation. Comput Biol Med 2023;166:107526. [Crossref] [PubMed]
16. LaLonde R, Xu Z, Irmakci I, et al. Capsules for biomedical image segmentation. Med Image Anal 2021;68:101889. [Crossref] [PubMed]
17. Kato K, Ishiguchi T, Maruyama K, et al. Accuracy of plastic replica of aortic aneurysm using 3D-CT data for transluminal stent-grafting: experimental and clinical evaluation. J Comput Assist Tomogr 2001;25:300-4. [Crossref] [PubMed]
18. Smelt JLC, Suri T, Valencia O, et al. Operative Planning in Thoracic Surgery: A Pilot Study Comparing Imaging Techniques and Three-Dimensional Printing. Ann Thorac Surg 2019;107:401-6. [Crossref] [PubMed]
19. Chen JV, Dang ABC, Dang A. Comparing cost and print time estimates for six commercially-available 3D printers obtained through slicing software for clinically relevant anatomical models. 3D Print Med 2021;7:1.
20. Parham G, Bing EG, Cuevas A, et al. Creating a low-cost virtual reality surgical simulation to increase surgical oncology capacity and capability. Ecancermedicalscience 2019;13:910. [Crossref] [PubMed]
21. Ng DS, Yip BHK, Young AL, et al. Cost-effectiveness of virtual reality and wet laboratory cataract surgery simulation. Medicine (Baltimore) 2023;102:e35067. [Crossref] [PubMed]
22. Okachi S, Sugimoto M, Ina T, et al. Clinical Applicability of Three-Dimensional Holographic Virtual Bronchoscopy with Mixed Reality. Ann Am Thorac Soc 2025;22:609-11. [Crossref] [PubMed]
23. Colt HG, Crawford SW, Galbraith O 3rd. Virtual reality bronchoscopy simulation: a revolution in procedural training. Chest 2001;120:1333-9. [Crossref] [PubMed]
24. FDA Regulatory Controls. U.S. Food and Drug Administration, FDA.gov. 2018. [Accessed 7/23/2025]. Available online: https://www.fda.gov/medical-devices/overview-device-regulation/regulatory-controls
25. Macario A. What does one minute of operating room time cost? J Clin Anesth 2010;22:233-6. [Crossref] [PubMed]
26. Hersh A, Mahapatra S, Weber-Levine C, et al. Augmented Reality in Spine Surgery: A Narrative Review. HSS J 2021;17:351-8. [Crossref] [PubMed]
27. Brollo PP, Bresadola V. Enhancing visualization and guidance in general surgery: a comprehensive and narrative review of the current cutting-edge technologies and future perspectives. J Gastrointest Surg 2024;28:179-85. [Crossref] [PubMed]
28. Akiba T. Utility of three-dimensional computed tomography in general thoracic surgery. Gen Thorac Cardiovasc Surg 2013;61:676-84. [Crossref] [PubMed]
29. Doornbos MJ, Peek JJ, Maat APWM, et al. Augmented Reality Implementation in Minimally Invasive Surgery for Future Application in Pulmonary Surgery: A Systematic Review. Surg Innov 2024;31:646-58. [Crossref] [PubMed]
30. Arjomandi Rad A, Vardanyan R, Thavarajasingam SG, et al. Extended, virtual and augmented reality in thoracic surgery: a systematic review. Interact Cardiovasc Thorac Surg 2022;34:201-11. [Crossref] [PubMed]
31. Jensen K, Bjerrum F, Hansen HJ, et al. A new possibility in thoracoscopic virtual reality simulation training: development and testing of a novel virtual reality simulator for video-assisted thoracoscopic surgery lobectomy. Interact Cardiovasc Thorac Surg 2015;21:420-6. [Crossref] [PubMed]
32. Haidari TA, Bjerrum F, Hansen HJ, et al. Simulation-based VATS resection of the five lung lobes: a technical skills test. Surg Endosc 2022;36:1234-42. [Crossref] [PubMed]
33. Solomon B, Bizekis C, Dellis SL, et al. Simulating video-assisted thoracoscopic lobectomy: a virtual reality cognitive task simulation. J Thorac Cardiovasc Surg 2011;141:249-55. [Crossref] [PubMed]
34. Chen X, Wang Z, Qi Q, et al. A fully automated noncontrast CT 3-D reconstruction algorithm enabled accurate anatomical demonstration for lung segmentectomy. Thorac Cancer 2022;13:795-803. [Crossref] [PubMed]
35. Sadeghi AH, Maat APWM, Taverne YJHJ, et al. Virtual reality and artificial intelligence for 3-dimensional planning of lung segmentectomies. JTCVS Tech 2021;7:309-21. [Crossref] [PubMed]
36. Bakhuis W, Sadeghi AH, Moes I, et al. Essential Surgical Plan Modifications After Virtual Reality Planning in 50 Consecutive Segmentectomies. Ann Thorac Surg 2023;115:1247-55. [Crossref] [PubMed]
37. Bakhuis W, Max SA, Nader M, et al. Video-assisted thoracic surgery S7 segmentectomy: use of virtual reality surgical planning and simulated reality intraoperative modelling. Multimed Man Cardiothorac Surg 2023; [Crossref]
38. Frajhof L, Borges J, Hoffmann E, et al. Virtual reality, mixed reality and augmented reality in surgical planning for video or robotically assisted thoracoscopic anatomic resections for treatment of lung cancer. J Vis Surg 2018;4:143.
39. Li C, Zheng B, Yu Q, et al. Augmented Reality and 3-Dimensional Printing Technologies for Guiding Complex Thoracoscopic Surgery. Ann Thorac Surg 2021;112:1624-31. [Crossref] [PubMed]
40. Tokuno J, Chen-Yoshikawa TF, Nakao M, et al. Creation of a video library for education and virtual simulation of anatomical lung resection. Interact Cardiovasc Thorac Surg 2022;34:808-13. [Crossref] [PubMed]
41. Perkins SL, Krajancich B, Yang CJ, et al. A Patient-Specific Mixed-Reality Visualization Tool for Thoracic Surgical Planning. Ann Thorac Surg 2020;110:290-5. [Crossref] [PubMed]
42. Rouzé S, de Latour B, Flécher E, et al. Small pulmonary nodule localization with cone beam computed tomography during video-assisted thoracic surgery: a feasibility study. Interact Cardiovasc Thorac Surg 2016;22:705-11. [Crossref] [PubMed]
43. Fu R, Zhang C, Zhang T, et al. A three-dimensional printing navigational template combined with mixed reality technique for localizing pulmonary nodules. Interact Cardiovasc Thorac Surg 2021;32:552-9. [Crossref] [PubMed]
44. Chang SH, Geraci TC, Johnson KR, et al. Narrative review: preoperative localization techniques for small lung nodules. Curr Chall Thorac Surg 2022;4:36.
45. Zhang H, Zhang C, Li L, et al. Small pulmonary nodule localization techniques in the era of lung cancer screening: a narrative review. Int J Surg 2025;111:2624-32. [Crossref] [PubMed]
46. Colella S, Brandimarte A, Marra R, et al. Chest wall reconstruction in benign and malignant tumors with non-rigid materials: An overview. Front Surg 2022;9:976463. [Crossref] [PubMed]
47. Gonfiotti A, Salvicchi A, Voltolini L. Chest-Wall Tumors and Surgical Techniques: State-of-the-Art and Our Institutional Experience. J Clin Med 2022;11:5516. [Crossref] [PubMed]
48. Ozaniak A, Galova D, Benesova I, et al. Treatment approaches and outcomes of major chest wall resections and reconstructions in patients with soft tissue and bone sarcomas: a retrospective observational study. J Thorac Dis 2024;16:6863-78. [Crossref] [PubMed]
49. Marinozzi F, Carleo F, Novelli S, et al. 3D Reconstruction Model of an Extra-Abdominal Desmoid Tumor: A Case Study. Front Bioeng Biotechnol 2020;8:518. [Crossref] [PubMed]
50. Feodorovici P, Schnorr P, Bedetti B, et al. Collaborative Virtual Reality Real-Time 3D Image Editing for Chest Wall Resections and Reconstruction Planning. Innovations (Phila) 2023;18:525-30. [Crossref] [PubMed]
51. Thumerel M, Belaroussi Y, Prisciandaro E, et al. Immersive Three-dimensional Computed Tomography to Plan Chest Wall Resection for Lung Cancer. Ann Thorac Surg 2022;114:2379-82. [Crossref] [PubMed]
52. Tan D, Yao J, Hua X, et al. Application of 3D modeling and printing technology in accurate resection of complicated thoracic tumors. Ann Transl Med 2020;8:1342. [Crossref] [PubMed]
53. Zhou X, Zhang D, Xie Z, et al. Application of preoperative 3D printing in the internal fixation of posterior rib fractures with embracing device: a cohort study. BMC Surg 2023;23:237. [Crossref] [PubMed]
54. Goldsmith I, Evans PL, Goodrum H, et al. Chest wall reconstruction with an anatomically designed 3-D printed titanium ribs and hemi-sternum implant. 3D Print Med 2020;6:26.
55. Goldsmith I, Dovgalski L, Evans PL. 3D Printing Technology for Chest Wall Reconstruction With a Sternum-Ribs-Cartilage Titanium Implant: From Ideation to Creation. Innovations (Phila) 2023;18:67-72. [Crossref] [PubMed]
56. Kamel MK, Cheng A, Vaughan B, et al. Sternal Reconstruction Using Customized 3D-Printed Titanium Implants. Ann Thorac Surg 2020;109:e411-4. [Crossref] [PubMed]
57. Pontiki AA, Lampridis S, De Angelis S, et al. Creation of personalised rib prostheses using a statistical shape model and 3D printing: Case report. Front Surg 2022;9:936638. [Crossref] [PubMed]
58. Yoon DW, Kim TH, Cha MJ, et al. Three-dimensional printed pure-titanium implantation for chest wall reconstruction involving the sternum and ribs: a novel approach. Interdiscip Cardiovasc Thorac Surg 2024;38:ivae037. [Crossref] [PubMed]
59. Wang W, Yang S, Han M, et al. Three-dimensional printed titanium chest wall reconstruction for tumor removal in the sternal region. J Cardiothorac Surg 2024;19:579. [Crossref] [PubMed]
60. Papp JG, Kiss Á, Balogh K, et al. New Computerized Planning Algorithm and Clinical Testing of Optimized Nuss Bar Design for Patients with Pectus Excavatum. Med Sci Monit 2024;30:e943705. [Crossref] [PubMed]
61. Rocco R, Morris J, Wentworth A, et al. Customized Three-Dimensional Printed Nuss Bar Molds for Repair of Pectus Excavatum. Ann Thorac Surg Short Rep 2023;1:713-7. [Crossref] [PubMed]
62. Gaspar Pérez M, Núñez García B, Álvarez García N, et al. Initial experience with 3D printing in the use of customized Nuss bars in pectus excavatum surgery. Cir Pediatr 2021;34:186-90.
63. Wu Y, Luo N, Tan L, et al. Three-dimensional reconstruction of thoracic structures: based on Chinese Visible Human. Comput Math Methods Med 2013;2013:795650. [Crossref] [PubMed]
64. Imai T, Tanaka Y, Hatanaka Y, et al. Incorporation of virtual reality in the clinical training of medical students studying esophageal and mediastinal anatomy and surgery. Surg Today 2022;52:1212-7. [Crossref] [PubMed]
65. Morita Y, Takase K, Yamada T, et al. Virtual CT thoracoscopy: preoperative simulation for thoracoscopic surgery of esophageal cancer. Abdom Imaging 2007;32:679-87. [Crossref] [PubMed]
66. Morita Y, Takase K, Ichikawa H, et al. Bronchial artery anatomy: preoperative 3D simulation with multidetector CT. Radiology 2010;255:934-43. [Crossref] [PubMed]
67. Maeda T, Fujiwara H, Konishi H, et al. Preoperative 3D-CT evaluation of the bronchial arteries in transmediastinal radical esophagectomy for esophageal cancer. Esophagus 2022;19:77-84. [Crossref] [PubMed]
68. Cai H, Wang R, Li Y, et al. Role of 3D reconstruction in the evaluation of patients with lower segment oesophageal cancer. J Thorac Dis 2018;10:3940-7. [Crossref] [PubMed]
69. Dickinson KJ, Matsumoto J, Cassivi SD, et al. Individualizing Management of Complex Esophageal Pathology Using Three-Dimensional Printed Models. Ann Thorac Surg 2015;100:692-7. [Crossref] [PubMed]
70. Akiba T, Nakada T, Inagaki T. A three-dimensional mediastinal model created with rapid prototyping in a patient with ectopic thymoma. Ann Thorac Cardiovasc Surg 2015;21:87-9. [Crossref] [PubMed]
71. Arora D, Tewari P, Shamshery C, et al. 3D Virtual Bronchoscopy as an Aid to Airway Management in a Patient with Anterior Mediastinal Mass. Ann Card Anaesth 2024;27:165-8. [Crossref] [PubMed]
72. Shaylor R, Golden E, Verenkin V, et al. Virtual reality and 3D printing in clinical anesthesia: a case series of two years' experience in a single tertiary medical centre. Can J Anaesth 2023;70:1433-40. [Crossref] [PubMed]
73. Bustamante S, Bose S, Bishop P, et al. Novel application of rapid prototyping for simulation of bronchoscopic anatomy. J Cardiothorac Vasc Anesth 2014;28:1122-5. [Crossref] [PubMed]
74. Miyazaki T, Yamasaki N, Tsuchiya T, et al. Airway stent insertion simulated with a three-dimensional printed airway model. Ann Thorac Surg 2015;99:e21-3. [Crossref] [PubMed]
75. Akiba T, Inagaki T, Nakada T. Three-dimensional printing model of anomalous bronchi before surgery. Ann Thorac Cardiovasc Surg 2014;20:659-62. [Crossref] [PubMed]
76. Doucet G, Ryan S, Bartellas M, et al. Modelling and Manufacturing of a 3D Printed Trachea for Cricothyroidotomy Simulation. Cureus 2017;9:e1575. [Crossref] [PubMed]
77. Park JH, Ahn M, Park SH, et al. 3D bioprinting of a trachea-mimetic cellular construct of a clinically relevant size. Biomaterials 2021;279:121246. [Crossref] [PubMed]
78. Administration USFaD. Medical Device Exemptions 510(k) and GMP Requirements. 2025. [Accessed 07/30/2025]. Available online: https://www.accessdata.fda.gov/scripts/cdrh/cfdocs/cfpcd/315.cfm
79. Ceevra. Ceevra for Thoracic Surgery. 2025. [Accessed May 29 2025]. Available online: https://ceevra.com/thoracic-surgery
80. Gupta N, Barrington NM, Panico N, et al. Assessing views and attitudes toward the use of extended reality and its implications in neurosurgical education: a survey of neurosurgical trainees. Neurosurg Focus 2024;56:E18. [Crossref] [PubMed]
81. Gupta N, Walker J, Turnow M, et al. Use of Mixed Reality Technologies by Orthopedic Surgery Residents: A Cross-Sectional Study of Trainee Perceptions. Journal of Orthopaedic Experience & Innovation 2024;5.
82. Amparore D, Pecoraro A, Checcucci E, et al. 3D imaging technologies in minimally invasive kidney and prostate cancer surgery: which is the urologists' perception? Minerva Urol Nephrol 2022;74:178-85. [Crossref] [PubMed]
83. Aggarwal R, Balasundaram I, Darzi A. Training opportunities and the role of virtual reality simulation in acquisition of basic laparoscopic skills. J Surg Res 2008;145:80-6. [Crossref] [PubMed]
84. Alhumaidi WA, Alqurashi NN, Alnumani RD, et al. Perceptions of Doctors in Saudi Arabia Toward Virtual Reality and Augmented Reality Applications in Healthcare. Cureus 2023;15:e42648. [Crossref] [PubMed]
85. Dinh A, Tseng E, Yin AL, et al. Perceptions About Augmented Reality in Remote Medical Care: Interview Study of Emergency Telemedicine Providers. JMIR Form Res 2023;7:e45211. [Crossref] [PubMed]
86. Le Moal J, Peillon C, Dacher JN, et al. Three-dimensional computed tomography reconstruction for operative planning in robotic segmentectomy: a pilot study. J Thorac Dis 2018;10:196-201. [Crossref] [PubMed]
87. Sadeghi AH, Mank Q, Tuzcu AS, et al. Artificial intelligence-assisted augmented reality robotic lung surgery: Navigating the future of thoracic surgery. JTCVS Tech 2024;26:121-5. [Crossref] [PubMed]
88. Yoon JW, Chen RE, Han PK, et al. Technical feasibility and safety of an intraoperative head-up display device during spine instrumentation. Int J Med Robot 2017;
89. Khor WS, Baker B, Amin K, et al. Augmented and virtual reality in surgery-the digital surgical environment: applications, limitations and legal pitfalls. Ann Transl Med 2016;4:454. [Crossref] [PubMed]