The application of robotic and artificial intelligence technologies in spinal surgery: a review focused on prospects in remote areas of China

Abstract

Spinal surgery in remote regions of China, including high-altitude plateau areas, faces unique challenges due to limited medical resources, harsh environmental conditions, and shortages of specialized expertise. Robotic spinal surgery and artificial intelligence (AI) technologies offer promising solutions to enhance diagnostic accuracy and support surgical decision-making in these underserved settings. Robotic-assisted systems enhance surgical precision and stability in procedures, such as pedicle screw placement, reducing error rates, and improving patient outcomes even when experienced surgeons are scarce. AI-driven diagnostic and planning tools can rapidly analyze medical images, identify spinal pathologies with expert-level accuracy, and assist in developing personalized surgical plans, effectively extending specialist-level support to local clinicians. The integration of these technologies helps bridge gaps in healthcare delivery by ensuring more timely, accurate diagnoses and safer, more effective spine surgeries for patients in remote areas. Implementing robotics and AI in high-altitude and rural environments, however, requires overcoming challenges. Continued research and development efforts are needed to tailor these systems to local needs, alongside training programs to develop skilled personnel. Looking ahead, the synergy of robotic surgery and AI in spine care represents a transformative direction for improving surgical outcomes and narrowing healthcare disparities in China’s remote communities.

Introduction

Spinal disorders have become a growing health concern worldwide, particularly in remote regions of China [1–4]. These conditions, including degenerative diseases, traumatic spinal injuries, and deformities, are increasingly prevalent due to the aging population and changes in lifestyle [5, 6]. In remote and high-altitude areas, the diagnosis and treatment of spinal conditions face additional challenges, as these regions suffer from limited medical infrastructure, technological resources, and specialized professionals. In such regions, the ability to diagnose and design personalized treatment plans for spinal diseases is often constrained by the lack of advanced medical equipment and expertise [7]. To address these issues, the use of robotic spinal surgery and artificial intelligence (AI) has emerged as a promising approach to enhancing diagnostic accuracy and improving treatment outcomes [8]. Both robotic systems and AI technologies have been progressively integrated into clinical practice, showing significant benefits, especially for the treatment of complex conditions [9].

Robotic-assisted spinal surgery provides an advanced solution to the challenge of achieving high precision in spinal procedures [10]. The ability to guide surgeons with greater accuracy during procedures like spinal fixation and pedicle screw insertion reduces the chances of errors, leading to improved surgical outcomes. Furthermore, robotic technology helps to decrease intraoperative radiation exposure and ease the physical demands on the surgical team, which is critical in ensuring the safety and efficiency of these procedures [11, 12]. In parallel, artificial intelligence is revolutionizing the diagnosis and treatment by supporting early detection, image analysis, and surgical planning [13]. AI-based diagnostic tools can swiftly analyze spinal imaging, identifying abnormalities with great accuracy [14]. This is particularly valuable in areas with limited access to specialized care, as it provides crucial diagnostic support and allows healthcare providers to make informed decisions more quickly. AI can also assist in tailoring surgical strategies to individual patients, optimizing the likelihood of successful outcomes [15].

For regions like Qinghai Province, the combination of robotic surgery and AI presents a transformative opportunity to overcome challenges caused by limited resources and expertise [16]. AI’s ability to process and analyze medical images can significantly improve diagnostic accuracy, enabling local practitioners to receive expert-level support remotely. Similarly, robotic surgery enhances the quality of complex procedures by providing high-precision support, an essential factor in areas where trained specialists are scarce. Together, these technologies can enhance local medical capabilities, ensuring that patients in remote locations receive timely and effective spinal care [16].

While both robotic spinal surgery and AI technologies have seen notable advancements in lowland areas, their implementation in high-altitude regions like Qinghai presents unique challenges. These technologies must be adapted to the specific conditions of high-altitude environments to be effective. Future research should focus on refining these systems to better suit the needs of high-altitude healthcare settings and overcoming the obstacles to their widespread adoption. This paper explores the potential of robotic spinal surgery and artificial intelligence in diagnosing spinal diseases and formulating surgical plans, particularly in remote areas of China such as Qinghai Province. By evaluating the benefits and challenges of these technologies, this work aims to provide a theoretical foundation for advancing spinal care in underserved regions and offer practical insights into their future application.

Current status of medical and health care in remote areas of China and the development needs of spinal surgery

Regional disparities in healthcare resources and the current status of spinal disease diagnosis and treatment

The western regions of China, particularly Qinghai Province, have long faced significant challenges in the distribution of healthcare resources [17]. Geographic isolation and economic imbalance contribute to the underdevelopment of healthcare infrastructure in these areas. Spinal surgery, being a highly specialized field, requires advanced equipment and skilled surgeons [18]. However, many hospitals in remote areas lack both the necessary technological resources and experienced spinal surgeons, which severely limits their capacity to perform complex spinal surgeries [19]. Research indicates that there is a considerable disparity in the number and types of spinal surgeries performed between provincial-level and county-level hospitals in Qinghai, with provincial hospitals being better equipped and staffed to handle complex spinal conditions [20].

Additionally, the prevalence of spinal diseases in these regions exhibits some unique characteristics. In addition to degenerative diseases, traumatic and developmental spinal disorders are also common. Notably, the demand for screening and treatment of adolescent idiopathic scoliosis is substantial, yet many patients present at advanced stages where surgical intervention is required due to insufficient early diagnosis [21]. Furthermore, the lack of an effective referral system and difficulties in transportation often result in delayed treatment, making it difficult for patients to receive specialized care in a timely manner.

The value and potential application of intelligent technologies in spinal surgery

With the advancement of medical technology, robotic-assisted surgery systems and artificial intelligence (AI) diagnostic platforms are beginning to reshape traditional spinal surgery practices. These technologies hold significant promise, especially in remote regions where medical resources are limited [22].

In spinal surgery, robotic-assisted surgery has proven to enhance precision, particularly in procedures such as pedicle screw insertion. Studies have demonstrated that robotic systems, such as TiRobot, significantly improve the accuracy of screw placement, reducing the risk of misplacement and enhancing surgical outcomes. In clinical applications, TiRobot-assisted percutaneous screw insertion has achieved an accuracy rate of 97.67%, which is a considerable improvement over traditional methods [23]. For regions like Qinghai, where specialized spinal surgeons may be scarce, robotic technology provides an opportunity to perform complex procedures with high precision, even in the absence of highly experienced surgeons.

Similarly, AI technology has shown considerable promise in spinal imaging analysis [24]. Deep learning-based algorithms enable AI systems to automatically detect anatomical structures and pathological changes in spinal MRI and CT scans [25]. These systems can diagnose common spinal conditions, such as herniated discs, spinal stenosis, and degenerative disc disease, with accuracy comparable to that of specialist physicians [26]. AI algorithms also assist in scoliosis assessment by automatically measuring the Cobb angle, providing highly efficient and accurate results that rival manual measurements performed by orthopedic specialists [27].

These technological advancements offer new solutions to address the shortage of medical resources in remote areas. By leveraging 5G networks and telemedicine platforms, experts from large hospitals can provide remote guidance in surgical planning and even participate in surgeries [28]. AI-powered diagnostic systems further enhance the imaging capabilities of primary healthcare providers, reducing diagnostic errors and ensuring that patients receive timely and accurate diagnoses.

Policy support and feasibility of technology promotion

The Chinese government has placed a strong emphasis on the balanced distribution of healthcare resources, particularly in western regions. The implementation of the “Internet Plus Health” strategy has provided strong policy support for the development of telemedicine, facilitating the sharing of high-quality medical resources across regions. A network of remote consultation services has been established in many areas, allowing for the seamless exchange of medical imaging data between provincial, municipal, and county-level hospitals.

In terms of technology promotion, several national and provincial-level science and technology programs have supported the introduction of intelligent medical devices in remote regions. For example, programs such as the “Central Government’s Special Fund for Local Science and Technology Development” have provided financial support for the introduction of digital surgical equipment in healthcare institutions in the western regions. These policy initiatives have created favorable conditions for the introduction and expansion of robotic-assisted surgery and AI diagnostic technologies in spinal surgery, making them increasingly accessible to underserved regions.

With continued policy support and the expansion of remote healthcare infrastructure, these technologies hold the potential to significantly improve spinal disease diagnosis and treatment in remote areas like Qinghai. The introduction of robotic surgery and AI-powered diagnostics offers an effective solution to address the disparities in spinal healthcare between rural and urban areas, ultimately improving patient outcomes in these underserved regions.

Application of digital and intelligent technologies in spinal surgery

Digital imaging technology in spinal diagnosis and treatment

Digital medical imaging is the foundation of modern spinal surgery, enabling precise evaluation and planning [29]. In the past, plateau regions were limited by equipment and environment, and imaging modalities were relatively underdeveloped. Today, digital X-ray, computed tomography (CT), and magnetic resonance imaging (MRI) are gradually being adopted in Qinghai and have been modified for high-altitude conditions. Digital imaging devices integrated via Picture Archiving and Communication Systems (PACS) allow hospital-wide networked access, enabling doctors to retrieve high-definition 3D reconstructed images at any time and facilitating precise preoperative evaluation. High-definition digital X-ray, computed tomography (CT), and magnetic resonance imaging (MRI) provide detailed views of spinal anatomy. Advanced intraoperative imaging systems (e.g. O-arm 3D scanners) can capture 3D spinal images during surgery and integrate with navigation to enhance implant placement accuracy [30]. For example, intraoperative 3D imaging combined with navigation significantly improves pedicle screw placement precision. Digital subtraction angiography (DSA) is also used preoperatively in spinal tumor cases to embolize vessels and reduce intraoperative bleeding [31]. These digital imaging tools serve as a “clairvoyant eye” for surgeons, and even in remote hospitals, high-quality imaging with telecommunication allows experts to accurately assess a patient’s anatomy and pathology from afar. This greatly improves access to specialist input in underserved areas.

Three-dimensional (3D) visualization and planning are critical for complex spinal cases (Fig. 1). Patient CT/MRI data can be reconstructed into a 3D model of the spine, showing bones, neural elements, and vessels in realistic spatial relationships. By interactively examining a 3D model, surgeons can intuitively evaluate deformity angles, tumor extent, or fracture configurations, and then select the optimal surgical approach [32]. Research confirms that 3D visualization in preoperative planning can reduce operative time and intraoperative blood loss while improving surgical precision [33]. These models also facilitate multidisciplinary discussions; experts can jointly review a 3D spine reconstruction via teleconference to help plan surgeries at distant hospitals. In summary, digital imaging coupled with 3D planning has enabled a “what-you-see-is-what-you-get” approach in spine surgery, laying the groundwork for precision medicine.

Fig. 1.

Perlove medical PLX series 3D C-arm

Another transformative imaging application is 3D printing for patient-specific models and implants (Fig. 2). Using the reconstructed imaging data, surgeons can create physical 3D models of a patient’s spine or print custom surgical guides and implants. Patient-specific 3D-printed drill guides have been shown to greatly enhance intraoperative accuracy. For instance, in multi-level pedicle screw insertion, a 3D-printed guide can significantly increase the one-pass success rate of screw placement while also reducing radiation exposure from repeated fluoroscopy [34, 35]. A clinical study in pediatric scoliosis found that template-guided pedicle screw placement achieved an excellent accuracy of 96.1% versus 88.6% with freehand technique (p = 0.007), and notably, the guided group had 0% postoperative complications compared to 12.1% in the freehand group [36]. Beyond guides, 3D-printed personalized implants (such as artificial vertebral bodies) have shown good outcomes in reconstructing spinal stability after complex tumor resections. Studies indicate that custom implants can effectively restore physiological spinal function and reduce postoperative complication rates [37]. This means spinal patients can receive implants tailored exactly to their anatomy, which improves fit and outcomes. Importantly, such technology can be deployed locally—rather than sending patients to distant centers for specialized implants, local hospitals with 3D printing capability can produce patient-specific hardware on site, greatly enhancing the timeliness and accessibility of treatment. In sum, advances in digital imaging, 3D visualization, and printing allow truly personalized spinal surgery with higher precision and better outcomes, which is especially valuable in regions where referral to specialized centers is difficult.

Fig. 2.

3D-printed spinal demonstrator

Intraoperative navigation technology

Computer-assisted intraoperative navigation systems are a cornerstone of spinal surgery digitalization, providing real-time guidance for implant placement [38](Fig. 3). The spine’s anatomy is complex and surrounded by critical neural structures, so even small deviations during freehand techniques (e.g., pedicle screw insertion) can risk neural injury. Navigation works by aligning the patient’s preoperative imaging data with their anatomy in the operating room, essentially giving the surgeon a “GPS for the spine” that displays instrument position and trajectory inside the vertebrae in real time. A typical navigation setup includes an intraoperative 3D imaging device (such as CT or O-arm), tracking cameras with infrared or electromagnetic sensors attached to surgical instruments, and a workstation that fuses the imaging and tracking information. As the surgeon moves an instrument, its path is visualized on the patient’s scanned anatomy, enabling highly accurate guidance [35, 39, 40].

Fig. 3.

Surgical robotic navigation system—spinal imaging analysis interface (with planning function)

Accuracy and safety benefits: a large body of research has demonstrated that navigation significantly improves the accuracy of pedicle screw placement. Compared to conventional freehand or fluoroscopy-guided techniques, navigation greatly reduces screw misplacement rates, with reported accuracy often above 95% in the thoracolumbar spine [41]. A recent meta-analysis including over 24,000 pedicle screws confirmed superior outcomes with navigation: the rate of clinically acceptable screw positions was 96.2% with navigation vs. 94.2% with the traditional methods, alongside statistically significant reductions in intraoperative blood loss, length of hospital stay, and complication rates in the navigated group [42]. Navigation also substantially decreases reliance on intraoperative X-ray, thereby lowering radiation exposure for both patients and surgical staff. Studies have shown that using navigation can cut a surgeon’s radiation dose by nearly two-thirds compared to repeated fluoroscopy [38, 43]. This is an important occupational safety improvement, as spine surgeons typically face high cumulative radiation exposure. Notably, navigation’s advantages become even more pronounced in complex or high-risk scenarios. In upper cervical spine surgery, where bony landmarks are small and the margin for error is minimal, navigation can guide the safe insertion of C1–C2 screws, avoiding neurovascular injury [44]. During en bloc resection of spinal tumors, navigation helps define osteotomy planes with precision, increasing the likelihood of complete tumor removal with negative margins [45]. In minimally invasive spine surgery—where the surgeon operates through small incisions without direct line-of-sight—navigation provides a “see-through” view of hidden anatomy, enabling accurate instrument placement and reducing the risk of inadvertently breaching cortical bone or injuring nerves [46].

Improved training and consistency: Navigation technology effectively lowers the learning curve for less-experienced surgeons tackling complex spinal procedures [47]. By relying on real-time guidance rather than solely on anatomical intuition, surgeons in training can achieve high placement accuracy from early on, gaining confidence in difficult cases more quickly. This is especially valuable when specialist expertise is limited, as it allows a broader range of surgeons to perform advanced procedures safely. Navigation standardizes certain aspects of spine surgery, which can lead to more consistent outcomes across different centers and surgeons. Furthermore, the digital records of instrument trajectories and screw positions can be archived, facilitating postoperative review or future training.

In clinical practice, navigation systems have been adopted widely in tertiary spine centers around the world. The main barrier to broader use is cost and setup complexity, which can be challenging for smaller or resource-limited hospitals. However, innovative solutions are emerging to extend navigation’s reach. One approach is a “remote navigation” model: the patient is locally equipped with a mobile navigation unit, while expert guidance is provided remotely by specialists viewing the navigation feed in real time. With modern high-bandwidth networks, it is becoming feasible for an expert in a distant city to virtually assist a local surgical team by observing the navigation monitor and providing real-time input on instrument placement. In the near future, as 5G telecommunication technologies mature, such cross-regional collaborative surgeries may become routine, effectively democratizing access to navigated spinal surgery. Even before full telesurgery is realized, navigation systems combined with telemedicine mean that complex spine cases in remote areas can benefit from expert planning and oversight, thereby improving surgical safety and reducing the need for patients to travel to referral centers.

Robotic-assisted surgery

Surgical robots represent the cutting edge of modern intelligent surgery (Fig. 4). In spinal surgery, robotic systems use precision-controlled robotic arms to assist in tasks like implant placement, offering high accuracy and repeatability. As modern surgical concepts and technology continuously advance, spinal surgical robots have rapidly developed since their debut around 2000 [48]. They have evolved from the earliest mechanically static navigation systems to fourth-generation products that feature intraoperative image guidance, preoperative–intraoperative image fusion, automatic path planning, and wireless guidewire insertion for precise screw placement, fully entering an era of visualized and intelligent surgery [49–52].

Fig. 4.

Perlove medical surgical robot and associated equipment illustration

Early first-generation spinal robots (e.g., Mazor Spine Assist) primarily provided static positioning via a robotic arm. They relied on preoperative CT data for trajectory planning and lacked intraoperative imaging feedback or the ability to make dynamic adjustments [53]. With the development of intraoperative 3D imaging (such as the O-arm) and navigation technology, second- and third-generation systems achieved coordinated intraoperative image reconstruction and robotic guidance, significantly improving surgical precision [54]. Currently, third-generation robotic systems (such as the Mazor X Stealth Edition, ROSA Spine, and China’s TiRobot) not only support real-time intraoperative image updates and automatic screw trajectory alignment, but also incorporate advanced features like wireless guidewire delivery, dynamic collision detection, and integration of the robotic arm with navigation. These advancements provide huge clinical advantages in complex multi-level screw insertions, upper cervical procedures, pelvic and sacroiliac fixation, and other challenging spinal surgeries [55, 56]. Research reports that robot-assisted pedicle screw placement can achieve accuracy above 98% [57], and that surgery duration, intraoperative blood loss, and number of fluoroscopic images are all significantly reduced (Fig. 5).

Fig. 5.

Robot-assisted pedicle screw insertion with navigation system

China’s domestically developed TiRobot orthopedic robot has been widely adopted in many tertiary hospitals in recent years. It supports a wide range of procedures, including cervical, thoracic, and lumbar spine instrumentation, sacroiliac screws, and spinal tumor reconstruction. Its cost is lower than imported systems, making it suitable for widespread adoption in regional hospitals and even in remote areas, including high-altitude regions like Qinghai. The combination of robotic systems with intraoperative CT and navigation technologies has become a key driver for the intelligent and standardized development of spinal surgery in remote, high-altitude regions.

Types and functions of spinal surgical robots (Table 1): The spinal robots currently used in clinical practice are mainly master–slave positioning robotic arms. Typical examples include Mazor X and ROSA from abroad, as well as China’s independently developed TiRobot. Their basic principle generally involves using the patient’s preoperative CT to plan screw trajectories, and then, during surgery, the robotic arm moves to the predetermined positions under navigation guidance, providing the surgeon with an exact pathway for screw insertion. The surgeon then drills and inserts the screws through guide sleeves attached to the robotic arm. In essence, the robot serves as a high-precision positioning assistant, eliminating human hand tremor and achieving sub-millimeter accuracy. Numerous studies have confirmed that robot-assisted pedicle screw accuracy can reach over 98%, comparable to or even better than navigation, with excellent consistency [58]. Additionally, because the robotic arm takes on the heavy repetitive tasks, the surgeon’s physical strain is reduced, which is especially beneficial in long, multi-level surgeries. A systematic review of the past decade’s literature by Chinese researchers summarized new advances in various spinal procedures using robots. They found that robotic assistance has been extended to many operations—including cervical lateral mass screws, sacroiliac screws, percutaneous endoscopic trajectories, en bloc resection of spinal tumors, and laminoplasty—all demonstrating good accuracy and safety [59]. Moreover, the incidence of robot-related complications (such as mechanical failures or positioning errors) is low, and the cost-effectiveness of robotic surgery is gaining attention and being progressively optimized [60–62].

Table 1.

Generational evolution of orthopedic surgical robots

Generation	Representative product	Technical features	Core advance	Clinical application	Drawbacks
First generation (1990s–2010)	ROBODOC (Integrated Surgical Systems, USA)	(1) CT-based preoperative planning (2) Rigid bone milling control, (3) Pneumatic drive system	Pioneered automated robotic operations in orthopedic surgery	Total hip arthroplasty (femoral component only)	(1) No soft-tissue handling capability, (2) cannot adjust plans intraoperatively (trajectory is fixed from pre-op CT data), (3) bulky size (footprint ~ 8 m2)
Second generation (2011–2018)	MAKO RIO (Stryker, USA)	(1) Haptic guidance technology, (2) real-time force feedback (0.1 N resolution), (3) optical tracking navigation	(1) Achieved “human–machine cooperative” operation, (2) incorporated error-compensation algorithms (accuracy ± 1.2 mm), (3) expanded indications to knee joint surgeries	Unicondylar knee arthroplasty; total hip arthroplasty	(1) Compatible only with specific implant brands (Triathlon series), (2) cannot handle cases with severe bone defects, (3) system calibration time > 15 min
Third generation (2019–2023)	Tianji 3.0 (China)	(1) Electromagnetic + optical dual-mode navigation (improved anti-interference capability), (2) AI bone density adaptive algorithm, (3) 5G remote surgery module	(1) Stable performance in high-altitude or complex electromagnetic environments, (2) learning curve reduced by 50% (from ~ 80 cases to ~ 40 cases), (3) supports multi-department use (spine + joint surgery)	Anterior cervical fusion, percutaneous pelvic screw insertion	(1) Delay in dynamic soft-tissue tracking (> 200 ms), (2) disposable instrument costs ~ 30% higher than traditional surgery, (3) lack of a pediatric bone development model database
Fourth generation (2024 +)	HURWA (Huake Precision, China)	(1) Multi-modal sensing (vision + force feedback + acoustic emission), (2) real-time digital twin simulation, (3) nano-scale vibration control (< 10 μm)	(1) Real-time monitoring of bone condition during surgery (with early warning for micro-fractures), (2) self-evolving learning algorithms (reduce surgery time by ~ 20%), (3) compatibility with 3D-printed custom implants	Precision resection of bone tumors; minimally invasive joint revision surgery	(1) Not yet FDA-approved, (2) software only compatible with Windows systems, (3) very high cost per unit (over ¥25 million RMB)

The significance of introducing surgical robots in remote regions: For plateau provinces like Qinghai, the introduction of orthopedic surgical robots can bring multiple benefits. First, robots can help less-experienced surgeons perform complex operations, lowering the learning curve and enabling primary hospitals to conduct complicated spinal internal fixation surgeries [63]. Second, in multi-segment spinal deformity corrections requiring the implantation of numerous screws, robots can significantly save time, reduce fluoroscopy frequency and intraoperative bleeding, thereby improving surgical safety in the high-altitude environment where patients have limited tolerance [55]. Third, China’s independently developed robotic systems, such as the Tinavi system, have been extensively clinically validated and are already in use in many Chinese hospitals [64]. These systems have relatively lower costs and Chinese-language interfaces, making them more suitable for promotion in primary healthcare settings in China (Fig. 6). If Qinghai Province can deploy spinal robots and implement training programs, it could rapidly elevate its spinal surgery capabilities. Additionally, robotic technology continues to advance, with features like haptic feedback and collision avoidance being added, and future possibilities include autonomous execution of certain surgical steps by robots [65]. Remote robotic surgery using 5G communication has become a reality: The spinal surgery team at Beijing Jishuitan Hospital successfully performed remote robotic surgeries on 12 patients using a 5G-enabled orthopedic surgical robot, achieving satisfactory outcomes [66]. This process also involved “one-station-to-multiple-site” spinal surgeries, where the main control room at Beijing Jishuitan Hospital alternately operated on and guided patients in different regional operating rooms. For remote areas like Qinghai, this model allows external experts to perform surgeries without being physically present, greatly expanding the coverage of high-quality medical resources. Overall, the application prospects of orthopedic surgical robots in high-altitude regions are vast. Current efforts should prioritize pilot programs in provincial hospitals, followed by phased expansion to qualified regions based on accumulated experience, ultimately enabling more plateau patients to benefit from robot-assisted precision minimally invasive surgery.

Fig. 6.

Robotic-assisted spinal surgery procedure demo

AI-assisted diagnosis and decision-making

Artificial intelligence (AI) technology, especially deep learning algorithms, has shown tremendous potential in medical image analysis and clinical decision support. For spinal surgery, AI can assist in tasks such as image-based diagnosis, surgical planning, and prognosis prediction, thereby reducing physician workload and improving accuracy. In high-altitude regions with limited medical resources, deploying AI systems could partially alleviate the shortage of specialists, allowing primary healthcare providers to obtain diagnostic insights approaching expert level.

Intelligent imaging diagnosis: the diagnosis of spinal diseases largely depends on imaging—X-rays for assessing deformities and degeneration, MRI for evaluating disc and spinal cord compression, CT for detailed bony changes, etc. Traditional image interpretation is time-consuming and requires experience. In recent years, deep learning has achieved breakthroughs in medical image analysis, enabling automatic recognition of imaging features for tasks like disease screening and grading. For instance, Kai-Yu L et al. trained a model using region-of-interest detection combined with a convolutional neural network classifier, enabling automatic diagnosis of lumbar spinal stenosis from CT scans; the model’s sensitivity and specificity both exceeded 80% [25]. Other AI models can automatically measure the Cobb angle on X-rays for adolescent idiopathic scoliosis, with accuracy approaching that of manual measurement while significantly reducing the physician’s time [27]. For the extensive screening required in plateau regions (e.g., scoliosis checks in schoolchildren), AI can quickly identify suspicious cases from back photographs or X-rays, flagging those needing further examination [67]. Most deep learning research in spine imaging has focused on disease detection and diagnosis, with many models showing performance comparable to or even better than human experts in trials [26]. For example, Dapeng Wang et al. developed a deep learning model that can detect early spinal metastases on MRI, improving diagnostic sensitivity [68]; machine learning algorithms have also been used to identify intervertebral disc protrusions and spondylolisthesis and to predict which patients will require surgical intervention [69]. In clinical practice, AI is already helping radiologists improve the efficiency and consistency of MRI assessments for cervical spondylosis [70]. In a region like Qinghai, if a cloud platform for spinal imaging were established and AI diagnostic software deployed, images taken at primary hospitals could be initially read by AI to mark suspicious areas, and then remotely confirmed by a small number of experts, thereby greatly expanding diagnostic coverage. For common degenerative conditions with large case volumes (e.g., lumbar spondylolisthesis, cervical spondylosis), AI could even directly provide preliminary diagnoses and grading suggestions for clinicians’ reference.

Surgical decision support: beyond image interpretation, AI can leverage large datasets of clinical information to help formulate treatment plans and predict outcomes. Using machine learning algorithms on retrospective data of past spinal surgery cases, one can discover which combinations of patient characteristics and surgical strategies yield the best outcomes. Based on factors like a patient’s age, bone density, and deformity severity, AI could recommend whether surgery is needed and which surgical approach might be optimal. In the rehabilitation context for spinal cord injuries, AI models analyzing data from exoskeleton-assisted training can easily provide patient-specific, precisely calibrated high-dose training regimens, helping to develop individualized rehabilitation plans [71, 72]. For scoliosis surgery, researchers have used deep learning to facilitate surgical decision-making and predict surgical results, aiding surgeons in optimizing preoperative plans [73].

AI technology is also reshaping intraoperative decision-making systems in spinal surgery, with its core value reflected in real-time analysis of multi-dimensional data and intelligent augmentation. In neurophysiological monitoring, AI systems dynamically analyze multimodal electrophysiological signals (such as somatosensory evoked potentials, motor evoked potentials, and EMG). Dougho Park et al. suggest that machine learning can achieve objective and reliable neurophysiological monitoring, overcoming issues that human experts may encounter, such as interference from non-surgical factors, ambiguity in defining false positives, and inter-observer variability [74]. In minimally invasive surgical navigation, AI’s computer vision-based recognition—using topological segmentation algorithms—can interpret endoscopic video in real time, not only labeling key anatomical structures like discs and nerve roots [75], but also quantifying, via an Augmented Reality (AR) navigation system, the spatial relationships between instruments and at-risk structures [76]. Additionally, a well-trained AI consultation system could provide preliminary diagnostic and treatment advice based on a patient’s symptoms and test results, which is especially valuable for primary doctors in remote areas of China. Although currently AI serves mainly as an assistive tool and final decisions still rest with physicians, its value is increasingly being recognized, and AI is expected to play a more substantial role in spinal surgery.

Challenges and future prospects: despite its great potential, implementing AI in spinal surgery comes with challenges. One issue is the need for large, high-quality datasets—algorithms trained on data from one population may not generalize well to another. If most AI models are trained on data from urban, low-altitude hospitals, their performance might be suboptimal for unique populations (e.g., high-altitude residents or certain ethnic groups) unless additional local data are used. There are also concerns about the “black box” nature of some deep learning models—they may provide an output (e.g., “high risk of complication”) without a transparent rationale, which can make surgeons hesitant to trust them. Data security and patient privacy for AI services delivered via cloud platforms must be ensured, especially when transmitting medical images. To address these issues, collaborative efforts are needed: for instance, hospitals and research institutions can work together to build robust spinal surgery databases and to train AI models that are tailored to specific demographics or regional needs. Clinicians also need training on how to appropriately use AI outputs—understanding the limitations and knowing that AI is an aid, not a replacement for clinical judgment. With supportive policies, including regulatory frameworks for AI in healthcare, and continuous technological improvements, many of these challenges are likely to be overcome.

Looking ahead, AI is expected to become an integral part of the digital spine surgery ecosystem. It will work in synergy with other technologies: for example, AI-driven diagnostic screening can identify patients who then get treated with the help of navigation and robotic surgery, and later AI might help monitor their rehabilitation progress. Ultimately, the vision is that AI will help provide round-the-clock expert-level support even at primary care levels—a physician in a small clinic could get instant recommendations based on global best practices, and patients anywhere could benefit from early detection of spinal problems through AI screening programs. As these tools mature, they have the potential to significantly narrow the gap in healthcare quality between well-resourced centers and remote or underserved regions. In essence, AI can act as a force multiplier for spinal healthcare, elevating the precision, efficiency, and accessibility of diagnosis and treatment.

Telemedicine and 5G remote surgical assistance

Delivering high-quality spinal care to patients in remote or resource-poor regions has always been a challenge due to geographical barriers and the uneven distribution of specialists. Telemedicine—the use of telecommunications technology to provide clinical healthcare at a distance—offers a powerful solution to bridge this gap [77, 78]. In recent years, telemedicine initiatives for orthopedics and spine have expanded, enabling everything from remote consultations to real-time surgical mentoring and even robot-assisted remote surgery (Fig. 7).

Fig. 7.

Operational view of Qinghai Provincial Telemedicine Center (plateau-region multifunctional consultation unit)

Telemedicine networks and consultation: Many regions have established multi-tier telemedicine networks connecting top-tier hospitals with smaller rural hospitals and clinics. For example, a provincial telemedicine network in one high-altitude region of China links a central tertiary hospital with dozens of county and township hospitals. Through this network, over 12,000 remote consultations are conducted annually, and more than 90% of critically ill patients at county-level hospitals can now receive timely treatment on site with expert input. The core components of such a system typically include: (1) Teleconsultation platforms for multidisciplinary case discussions—imaging (using DICOM standards) and other investigations from a remote patient can be transmitted to specialists at a central hospital in real time for joint review. This allows complex spinal cases at primary hospitals to be evaluated via video conference by expert spine surgeons and radiologists, who work with the local doctors to formulate an optimal treatment plan [79]. (2) Real-time surgical guidance (telementoring)—using high-definition video streaming (e.g., 4 K ultra-HD) with minimal latency, an expert can virtually “join” a surgery being performed in a remote operating room. The mentor surgeon can observe the operative field on live video and the patient’s vital data, and then provide step-by-step guidance or feedback to the local surgical team via audio or even by drawing annotations on a screen. This has been successfully employed in spine surgery and neurosurgery, allowing less-experienced surgeons to perform new or complex procedures under the supervision of a distant expert [80]. (3) Tele-education and training—regular teleconferences and virtual workshops are conducted to train doctors at primary hospitals. For instance, spine specialists might provide weekly lectures, case reviews, or skill demonstrations (sometimes using virtual reality simulation) that remote practitioners can attend, greatly increasing their ongoing education opportunities [81]. These components work together to elevate the standard of care: teleconsultation ensures that patients get expert diagnoses and surgical planning advice, telementoring improves surgical technique and confidence in real time, and tele-education builds local capacity long-term.

Improving access and outcomes: Telemedicine has measurably improved the accessibility of spinal healthcare. With teleconsultation, patients in small community hospitals no longer need to travel hundreds of kilometers for a specialist opinion—their imaging and records can be reviewed by, say, a spine team in a national center within hours. This reduces delays in diagnosis and treatment. In fact, practical experience has shown that a robust telemedicine system can even reverse patient referral patterns, with fewer patients needing transfer out and more being treated locally. After implementing tele-spine services, some regions report that many patients who previously would have bypassed local hospitals now choose to receive care locally, knowing that those hospitals are backed by telemedicine support. In one report, telemedicine usage led to a significant drop in unnecessary patient transfers and fostered a “flow-back” of patients to regional hospitals, as quality of care there improved. Moreover, outcomes for emergency spine cases like acute traumatic injuries have improved: establishing a telemedicine link for emergency spinal consultations ensures that within minutes, a spine surgeon can guide a generalist at a remote hospital through stabilization steps or surgical decisions that are crucial in the first “golden hour” after injury. Telemedicine thus helps to optimize care pathways, ensuring that patients are managed at the right place and right time. It effectively brings specialized expertise to the patient’s bedside regardless of location, thereby enhancing medical resource equity.

5G remote surgery and robotic telesurgery: One of the most futuristic applications of telemedicine is remote surgery, made possible by modern telecommunication infrastructure. High-speed 5G networks, with their ultra-low latency and high bandwidth, allow the transmission of surgical signals (video, audio, and control commands) in near real time. This technology has enabled experienced surgeons to physically be in one location and operate on a patient in another location using robotic surgical systems [82]. In spine surgery, this concept has been proven in pilot projects. For example, in a landmark 2020 study, 12 patients underwent telerobotic spinal instrumentation where the surgeon in Beijing controlled a robot in an operating room several thousand kilometers away; all surgeries were completed successfully with outcomes comparable to conventional surgery [66]. During such procedures, multiple cameras and sensors provide the remote surgeon with a comprehensive view of the operative field and patient vitals, while the robotic system precisely executes the surgeon’s hand movements on site. Remote robotic surgery is still in an early stage and mostly limited to trial settings or simple procedures, but it holds tremendous promise. For remote and high-altitude areas that lack senior spine surgeons, this could be game-changing—a patient could receive, for instance, a complex scoliosis correction in their hometown hospital performed by a top surgeon who is controlling the robot from a major city. This completely eliminates geographic barriers for surgical care. Of course, to implement this widely, several conditions must be met: reliable 5G network coverage in hospitals, standard protocols for telesurgery, as well as legal frameworks to address licensure and liability when the surgeon is not physically present. Those challenges are actively being worked on, and as they are resolved, we expect remote surgical assistance to become an important mode of delivering care in spine surgery.

Even short of full remote surgery, intermediate forms of tele-assistance are already routine. Tele-mentoring in surgery (with the mentor not physically scrubbed in, but available via video link) has been successfully utilized in numerous spine cases. Studies show that this approach increases the confidence and skill of surgeons in peripheral hospitals and can prevent complications by timely expert intervention when a less experienced surgeon encounters difficulty [83]. Virtual reality and augmented reality are enhancing this experience: an expert wearing an AR headset can visualize the remote operative field with superimposed holographic anatomy or instrument trajectories and guide the on-site surgeon through complex maneuvers in a very intuitive way. For example, the mentor could draw a line on their screen to indicate a correct incision or pin placement, and that annotation appears in the surgeon’s AR display as if drawn on the patient. These technologies are actively being studied and have shown positive impacts on surgical training and outcomes [84].

Outlook: Telemedicine is becoming an indispensable part of spinal surgery practice [85]. To fully leverage its benefits, healthcare systems are investing in infrastructure (such as dedicated telemedicine centers and 5G networks in hospitals) and developing protocols for routine use. Key areas of focus include establishing regular tele-consultation clinics for spine (so that any complex case in the region is reviewed by a spine tumor board or deformity board via teleconference) and setting up on-call tele-support for emergencies like cauda equina syndrome or cervical spine trauma, where a specialist can guide initial management remotely. Additionally, policy support is needed to integrate telemedicine into insurance reimbursement and medicolegal statutes. Encouragingly, systematic reviews have found telemedicine to be generally effective and cost-efficient for patient management in remote and austere environments. If implemented well, tele-spine programs can substantially elevate the level of care available in underserved areas. One concrete measure of success is resource redistribution: as noted, a functioning telemedicine network can allow over 90% of critical emergency patients to be treated in local hospitals who previously would have required transfer. This alleviates the burden on tertiary centers and brings care closer to where patients live, which is ultimately the goal of an equitable healthcare system. In the coming years, telemedicine and remote surgical assistance—powered by digital technology and high-speed connectivity—will likely be pillars of delivering specialized spine care to every corner, ensuring that patients on the periphery receive quality treatment on par with those in metropolitan centers [86].

Intelligent postoperative rehabilitation and follow-up

Postoperative rehabilitation is crucial for functional recovery and preventing complications after spinal surgery. In high-altitude regions, rehabilitation services are relatively weak; patients often cannot receive standardized rehabilitation training after surgery due to lack of guidance or the difficulty of traveling long distances. The introduction of intelligent rehabilitation technology offers a potential solution, including wearable exoskeleton robots, Internet-of-Things (IoT) rehabilitation monitoring, and remote rehabilitation platforms, which help postoperative patients recover better.

Exoskeleton robots for assisted walking: For patients with spinal cord injury who are paraplegic or have incomplete paralysis, traditional rehabilitation methods often have limited effectiveness. Exoskeleton robots, as an emerging technology, can help paraplegic patients stand up again and undergo gait training [87, 88]. Wearable exoskeletons use motors to drive the knee and hip joints, simulating walking movements while also providing postural support. Studies have shown that task-specific overground gait training significantly improves motor recovery in moderate-to-severe incomplete thoracic spinal cord injury in animal models [89]. Whether using exoskeletons for gait training in humans can stimulate neural pathway remodeling and enhance patients’ use of remaining function—thereby markedly improving walking independence—has become a hot topic for future research.

A randomized-controlled trial reported that spinal cord injury patients in an exoskeleton training group had significantly higher walking ability scores than those in a control group, with good tolerance to the device and no major adverse events [90]. For patients in remote areas, exoskeleton robots can effectively improve motor function and prevent complications associated with prolonged bed rest, such as pressure ulcers, osteoporosis, and muscle atrophy. Currently, numerous exoskeleton robot products are in use, including Japan’s HAL, the U.S. Ekso system, and China’s own flexible exoskeletons, and these devices have been widely adopted in rehabilitation institutions [91–93]. However, for plateau patients, rehabilitation training should follow a gradual, stepwise principle to avoid excessive fatigue under hypoxic conditions. Overall, exoskeleton robots bring new hope for paraplegic patients to achieve self-care in daily life, while also reducing caregiving burdens. In primary hospitals where professional rehabilitation therapists are lacking, standardized training programs with exoskeletons are especially practical and can serve as an important supplement to traditional manual rehabilitation.

Remote and home-based rehabilitation: Most spinal surgery patients require ongoing functional exercise and follow-up guidance after discharge. In remote areas of China, many patients live in rural or pastoral areas and find it difficult to return to the hospital regularly for check-ups [93]. By leveraging “Internet + ” technology, rehabilitation guidance can be delivered to patients at home. For example, patients can use a smartphone app to watch video tutorials of rehabilitation exercises. By wearing motion-capture sensors that record movements of the lower limbs or trunk, the patient’s exercise data are uploaded to the cloud, where AI algorithms evaluate whether the movements meet the targets; alternatively, remote rehabilitation therapists review the data and provide feedback or corrections. Such remote rehabilitation training systems have been implemented in some cities and are especially beneficial for patients needing long-term exercise therapy, such as those after spinal fusion or scoliosis surgery [94]. In the future, a provincial postoperative remote management platform could be established for major spinal surgery patients to register. Rehabilitation physicians and nurses could conduct regular follow-ups via phone or video to monitor patient progress, guide exercises, and adjust medications. Patients wearing specialized braces or orthoses could also upload photos or videos for doctors to assess their usage [95]. Remote follow-up not only makes it convenient for patients, but also helps clinicians accumulate valuable long-term outcome data.

Smart rehabilitation devices and assessment: Besides exoskeletons, there are many intelligent devices in the rehabilitation field that help spinal patients regain function. Examples include smart assistive cycling devices for lower limbs, balance training robots, and posture-monitoring electronic vests. These devices use sensor technology and interactive feedback to make rehabilitation training more engaging and quantifiable. A posture-monitoring vest, for instance, can continuously record the standing/sitting angles of a post-spinal surgery patient, reminding them to avoid prolonged poor posture and to move regularly, thereby preventing reinjury to the spine. Intelligent electrical stimulation devices can automatically adjust stimulation intensity as the patient performs muscle exercises, promoting muscle strength recovery.

In view of the specific conditions in remote areas of China, rehabilitation programs that incorporate elements of the traditional pastoral lifestyle can be developed. For example, wearable devices can record parameters such as step counts and carrying loads, to quantify rehabilitation outcomes in daily activities. These training methods not only aid patients in regaining function but also help integrate rehabilitation into the local way of life. Intelligent rehabilitation equipment and follow-up systems provide a continuous “closed loop” of care for patients in remote areas, helping to ensure that treatment results are sustained. All of these tools fall under the umbrella of digital health and are an integral part of modern spinal surgery care. Once high-altitude patients are discharged and return to their daily lives, smart rehabilitation devices and follow-up systems serve as the link between the hospital and the patient, ensuring that the loop of care is completed.

Discussion

Technological efficacy

Robotic spine surgery and AI-assisted tools have demonstrated clear benefits for spinal care, even in China’s remote regions. Clinical studies show that robot-assisted techniques can markedly improve the accuracy of pedicle screw placement compared to freehand methods [96]. For example, China’s domestically developed TiRobot system achieved pedicle screw accuracy rates approaching 95%, significantly higher than conventional fluoroscopy-guided methods [66]. Early deployments of 5G-enabled telerobotic spine surgery further underscore the technical feasibility and safety of these innovations. Artificial intelligence (AI) applications are likewise emerging as valuable adjuncts—from automated analysis of spinal imaging to machine learning models that assist in surgical planning. These technologies hold particular promise for improving care quality and consistency in under-resourced areas by leveraging digital precision. In short, the efficacy of robotic and AI platforms in spine surgery has been proven in principle, laying a strong foundation for wider adoption. The challenge ahead is translating this success to routine practice in China’s vast rural and high-altitude regions, where multiple real-world constraints must be addressed.

Resource integration

Despite their potential, robotic and AI technologies face significant accessibility and integration hurdles in China’s remote and primary hospitals. A fundamental issue is the uneven distribution of medical resources—tertiary spine centers with advanced equipment are concentrated in major cities, whereas most county-level and township hospitals lack surgical navigation systems or robots altogether. High cost is a major barrier. Surgical robots entail steep upfront purchase prices and ongoing maintenance expenses that are often prohibitive for mid-size or rural hospitals [97]. Until recently, Chinese patients undergoing robot-assisted procedures also shouldered high out-of-pocket costs, since national insurance coverage for robotic surgery has been limited. These economic constraints hinder widespread adoption. However, steps are being taken to improve affordability. Domestic manufacturers have introduced orthopedic robots at lower cost than imported systems, and cities like Beijing have started including orthopedic robotic surgery in insurance reimbursement schemes. Such measures are expected to reduce the financial burden on patients and hospitals, gradually making robots more accessible to less-developed regions.

Even if equipment is acquired, maintenance and technical support in remote areas pose another challenge. Surgical robots and advanced imaging devices require regular servicing, software updates, and prompt repairs in case of faults. Yet, far-flung hospitals may not have on-site engineers, and sending specialists from metropolitan centers can be slow and costly. This underscores the need for robust support networks—for example, manufacturers establishing regional service centers or using IoT telemetry to monitor and troubleshoot devices remotely. Training and retention of skilled personnel are equally vital. Robotic spine surgery has a learning curve; surgeons, nurses, and technicians must be specifically trained to operate and troubleshoot these systems. In China’s rural west, very few surgeons are formally trained in robot-assisted spine procedures, and many cannot independently perform complex spinal instrumentation. Bridging this gap will require investment in training programs and mentorship. Strategies include short-term fellowships for rural surgeons at high-volume urban centers, on-site proctoring during initial robotic cases, and continuous education via simulation labs. Notably, telemedicine can facilitate knowledge transfer: through 5G teleconferencing, experts in top hospitals can guide local surgical teams in case discussions and even live procedures. Provincial telemedicine networks in China (such as Qinghai’s three-tier system covering all county hospitals) are already enabling thousands of remote consultations and radiologic diagnoses, which could be expanded into surgical tele-mentoring. By integrating tele-assistance with robotics, a remote hospital could receive real-time expert guidance during a spine operation, improving safety and outcomes even when local experience is limited.

Another critical aspect of integration is data quality and AI performance. AI-driven diagnostic algorithms for spine conditions rely on large volumes of high-quality imaging and clinical data. Unfortunately, data from under-resourced rural hospitals can be limited or inconsistent—imaging may be lower resolution, and record-keeping may lack the detail found in top-tier hospitals. If AI models are trained mostly on urban patient datasets, they may not generalize well to rural populations or uncommon pathologies, risking bias or decreased accuracy [98]. Ensuring that AI tools truly benefit remote communities requires improving data collection and representation. This could involve creating federated learning networks that securely incorporate data from diverse hospitals (including plateau regions) to train algorithms on local patterns. Additionally, ongoing performance monitoring is needed whenever an AI model is deployed clinically. Healthcare providers must validate AI predictions against expert opinions and patient outcomes in the local setting, so that errors or drifts in performance can be caught and addressed. Despite these hurdles—infrastructure costs, maintenance logistics, training deficits, and data gaps—the integration of robotics and AI into China’s remote spine care is gradually progressing. With targeted investments in infrastructure and human capital, remote hospitals can leverage these tools to narrow the healthcare gap. The experience so far suggests that a combination of government support (e.g. funding and insurance coverage), industry innovation (affordable devices, remote support capability), and workforce development will be required to fully integrate intelligent spine surgery into China’s hinterlands.

Future directions

Looking forward, a multi-pronged strategy is needed to overcome current limitations and fully realize the promise of robotic and AI-assisted spine surgery in China’s rural and high-altitude regions. Multi-modal technology integration will be at the forefront of this effort. The expansion of 5G networks—and eventually 6G—into remote areas is crucial [66, 99]. Ubiquitous high-bandwidth, low-latency connectivity allows not only telesurgery, but also routine telepresence and mentoring, effectively bringing expert care “to the doorstep” of patients in isolated communities. In parallel, satellite communication offers a means to reach areas beyond terrestrial network coverage. China’s recent success with satellite-enabled telesurgery from Lhasa demonstrated that ultra-remote operations are possible even over 36,000 km distances, after solving latency issues with innovative compensation algorithms [100]. Future systems will likely adopt hybrid communication models—for instance, using 5G as primary linkage and satellite as backup—to ensure a reliable connection for critical surgical data. Advanced network solutions, like edge computing and multi-access edge computing (MEC), are expected to play a key role in reducing effective latency and improving reliability. By processing data locally at the network edge, large imaging or navigation datasets can be analyzed on-site (or nearby) with only the crucial results transmitted to remote surgeons (101). This approach decreases bandwidth needs and mitigates delays, enabling smoother robot control and real-time AI decision support even when cloud connectivity is suboptimal. Edge computing combined with AI also enhances data privacy and security by limiting the transfer of sensitive patient information, which is an important consideration as telehealth expands.

Another important future direction is the deeper integration of AI with surgical robotics. Currently, most orthopedic robots act as precision positioning tools under surgeon control, and most clinical AI applications (such as imaging diagnostics) function separately. In coming years, these technologies will converge. We anticipate the development of AI-guided robot navigation, where machine learning algorithms assist with intraoperative decision-making—for example, autonomously suggesting optimal screw trajectories or detecting impending complications from sensor data. Early steps toward this are evident in experimental systems using neural networks to predict and compensate for robotic arm error under latency conditions. Fully autonomous surgery remains a future vision, but semi-autonomous functionalities may gradually augment the surgeon’s capabilities. Ensuring safety and building trust in such AI-driven automation will require extensive validation and likely a stepwise regulatory approach.

Conclusion

The digital and intelligent transformation of spine surgery in China’s remote and high-altitude regions is both a necessity and an attainable goal. Telerobotic surgery and AI-assisted diagnosis offer a pathway to bridge the urban–rural healthcare divide, but success will depend on addressing practical challenges of cost, training, maintenance, data quality, and environmental suitability. By improving technological access and robustness, and by integrating next-generation tools like 5G/6G networks, edge computing, and telepresence, remote hospitals can be empowered to deliver state-of-the-art spinal care. Continued commitment to research, infrastructure, and human capital development will be required to realize these advancements on a broad scale. The early successes—from high-accuracy robotic surgeries to the pioneering ultra-remote operations—provide optimism that spine surgery in China’s hinterlands can be elevated to a new standard of precision and safety. With a coordinated effort, patients in distant plateau communities may increasingly receive timely, cutting-edge surgical treatment at their local hospitals, fulfilling the ultimate vision of health equity through technological innovation.

Abbreviations

3D

Three-dimensional

AI

Artificial intelligence

PPS

Percutaneous pedicle screw

VTE

Venous thromboembolism

CT

Computed tomography

MRI

Magnetic resonance imaging

PS

Picture archiving and communication systems

DSA

Digital Subtraction angiography

EMG

Electromyography

AR

Augmented reality

MEC

Multi-access edge computing

ICT

Information and communication technologies

VR

Virtual reality

IoT

Internet-of-things

UV

Ultraviolet

R&D

Research and development

Author contributions

All the authors materially participated in the research and article preparation. The roles of all the authors are as follows: Zhibin Liu wrote the main manuscript text. Junlong Huang, Hao Zhang, Shuzhuo Zhang, Honghao Dai, and Yuexin Jiang: drawing. Hongtao Bi: writing—review and editing. Zhongshu Shan: project administration and funding acquisition.

Funding

This work was supported by the Qinghai Provincial Central Government Guided Local Science and Technology Development Fund (2024ZY032).

Data availability

No datasets were generated or analyzed during the current study.

Declarations

Conflict of interest

The authors declare no conflict of interest.

Ethical approval and consent to participate

Not applicable.

Consent for publication

Not applicable.