Differences between image-based and imageless robotics for total knee arthroplasty – an overview

Hannes Vermue; Cécile Batailler; Sébastien Lustig

doi:10.1016/j.jor.2025.08.045

View/Download PDF

Buy Reprints

PDF

Translate this page into:

71 (); 23-28

doi:

10.1016/j.jor.2025.08.045

Differences between image-based and imageless robotics for total knee arthroplasty – an overview

Hannes Vermue^a,b,, Cécile Batailler^b, Sébastien Lustig^b

a Department of Orthopedic Surgery and Traumatology, Ghent University Hospital, Ghent, Belgium

b Department of Orthopedic Surgery, Hôpital Croix Rousse, Lyon, France

⁎Corresponding author: Hannes Vermue. Hannes_Vermue@hotmail.com

Received: 2025-8-18, Accepted: 2025-8-20, Published: 2025-9-29

Professor P K Surendran Memorial Education Foundation

Disclaimer:
This article was originally published by Reed Elsevier India Pvt. Ltd. and was migrated to Scientific Scholar after the change of Publisher.

Abstract

Robot-assisted total knee arthroplasty (TKA) has revolutionized implant positioning by enhancing surgical precision. However, significant differences exist between image-based and imageless robotic systems, influencing their accuracy, workflow, and potentially clinical outcomes. As such, this review aims to provide a structured comparison between image-based and imageless systems for TKA relevant to the surgeon. Image-based systems utilize preoperative imaging modalities such as computed tomography (CT) or magnetic resonance imaging (MRI) to construct three-dimensional (3D) anatomical models for surgical planning and intraoperative guidance. In contrast, imageless systems rely on intraoperative anatomical landmark registration, reducing costs and radiation exposure but yielding a potential higher inaccuracy to define the joint-specific coordinate system.

This current concepts review examines key differences between these technologies, focusing on coordinate system accuracy, anatomical landmark identification, resection precision, and clinical implications. Image-based systems may demonstrate superior accuracy in defining coordinate systems, particularly for femoral and tibial rotational axes, yet involve higher costs and logistical complexity. Imageless systems offer real-time adaptability and avoid preoperative imaging but may be more susceptible to anatomical registration errors. Comparative clinical studies suggest similar coronal plane alignment error between the two approaches, though image-based systems may offer advantages in tibial slope precision. Radiation exposure is an important consideration, as it varies significantly based on imaging protocols and may warrant greater attention from surgeons, especially for patients who receive frequent CT follow-up. As well, while both system types allow for assessing patellofemoral tracking, in addition patient-specific 3D models allow to restore the native trochlea orientation. Although both systems improve implant positioning compared to conventional techniques, evidence on mid- and long-term improved clinical outcomes compared to the conventional technique is still lacking.

Keywords

Image-based

Imageless

Robotics

Robot-assisted

Computer-assisted

Total knee arthroplasty

Show Related Articles from PubMed

1 Introduction

Robotics has introduced a new era in total knee arthroplasty (TKA), offering the promise of enhanced precision in the positioning of tibial and femoral implants. However, not all robotic systems are created equal and significant differences exist between image-based and imageless systems.¹ This distinction is critical, as it stresses the need for a nuanced evaluation of their capabilities and limitations.

Image-based robotic systems rely on preoperative imaging modalities including as computed tomography (CT), radiography, or magnetic resonance imaging (MRI) to reconstruct individualized three-dimensional (3D) models of the native anatomy.^2–9 These models serve as a basis for surgical planning and intraoperative guidance, allowing precise quantification of femoral and tibial cuts while integrating soft-tissue data gathered during surgery. In contrast, imageless systems instead rely on manual intraoperative mapping of anatomical landmarks to establish the surgical coordinate system, after which the joint surfaces are registered.^10–15 This approach avoids the added costs, potential radiation exposure, and preoperative time demands associated with image-based techniques but may raise questions about the accuracy and reliability of anatomical representation.¹⁶^,¹⁷

When evaluating the choice between image-based and imageless systems there should be no semantic confusion about the terms accuracy and precision. Precision is defined as the error of measurement, i.e. the deviation of a series of repeated measurements from a random value. Accuracy is defined as the closeness of a given measurement to the true value for the variable of interest. In the case of total knee arthroplasty, accuracy for implant position refers to the ideal implant position for an individual patient. The definition and construction of a reliable coordinate system are prerequisites for achieving optimal implant positioning and tibiofemoral kinematics. Surgical navigation and robotics have expanded the ability to assess these coordinate systems intraoperatively, yet debates persist regarding the added value of imaging modalities in improving the accuracy of the procedure.¹^,^17–20 Each approach carries inherent advantages and trade-offs, which must be critically appraised to guide system selection and future innovation.

While the long-term clinical benefits of robot-assisted TKA remain under investigation, it is essential to address the foundational differences between image-based and imageless systems.¹ This current concepts review aims to provide a comprehensive comparison of image-based and imageless technology for TKA, weighing their benefits and drawbacks (Table 1).

Table 1 Overview of the comparison between image-based and imageless systems for robot-assisted TKA as described in the current review. + describes favorable results for the assessed aspect, ± describes moderate results for the assessed aspect, - describes unfavorable results for the assessed aspect.

	Image-based	Imageless
Coordinate system	+	±
Precision	+	+
Radiation	–	+
Patellofemoral assessment	±	±
Clinical outcomes	One study in favor of imageless. One study without differences of complications/PROMs between imageless and image-based.
Soft-tissue evaluation	±	+
Footprint	–	±

2 Coordinate systems

The accuracy and precision of the coordinate system are crucial, particularly with the growing use of personalized alignment strategies (Fig. 1). For example, a malrotated coordinate system can result in significant cutting errors, especially when the goal is to achieve cuts that are not perpendicular to the mechanical axis. During imageless robot-assisted TKA, the anatomical landmarks which reconstruct the anatomical reference frame can be identified through:1.A pointer for landmarks that are accessible intraoperatively, or2.Kinematic analysis for landmarks that cannot be directly palpated during surgery.

Visualization of the coordinate systems used in total knee surgery.

For instance, the femoral hip center can be localized using imageless acquisition by circumduction the hip joint while avoiding secondary pelvic displacement. This motion generates a cone-shaped mechanical axis, with the apex of the cone corresponding to the femoral hip center. After the intraoperative acquisition of the coordinate system, the surfaces of the distal femur and proximal tibia can be registered using a probe. Notably, it is challenging to adequately register the posterior aspect of the tibia prior to performing any bony cuts. In contrast, image-based systems define the reference frame and bony volumes using preoperative CT scans, which are matched intraoperatively to the native anatomy.

2.1

2.1 Anatomical landmarks and their accuracy

2.1.1

2.1.1 Femoral hip center

A geometric assessment of the proximal femur can determine the hip center accurately by fitting a sphere to the femoral head.²¹^,²² A kinematic assessment, as described by Lustig et al., shows a spatial error range of 1.5–3.9 mm, leading to a corresponding angular error in the mechanical axis of 0.25–0.64°. Extreme pelvic movement increases this error to 11.7 mm and 1.9°.²³ Despite these errors, other studies report that the clinical impact of inaccuracies in determining the femoral hip center is minimal.^24–26

2.1.2

2.1.2 Femoral knee center

Defining the femoral knee center remains a debated topic due to a lack of consensus on its precise definition. In a study by Davis et al., five surgeons determined the femoral knee center as the distal femoral center, with a deviation of 0.1 mm (Standard Deviation (SD) 2.7) relative to a reference point established by the most experienced surgeon.²⁴ Victor et al. demonstrated low error rates in a CT-based study, with an average error of 0.5 mm (SD 0.3) corresponding to an angular mechanical axis error of 0.3° (SD 0.4).²²

2.1.3

2.1.3 Femoral rotational axes

Victor et al. established a reference frame for femoral rotation, showing inter- and intra-observer variability of 0.16°–1.15° for transepicondylar and posterior condylar axes. The trochlear axis exhibited higher inter-observer variability of 2°.¹⁹ While image-based techniques provide high precision for defining femoral rotational axes, intraoperative identification of the transepicondylar axis can be challenging.²⁷^,²⁸ Nevertheless, Hernandez-Vaquero et al. reported minimal differences in femoral rotational axes between image-based and imageless systems.²⁹

2.1.4

2.1.4 Tibial ankle center

Both image-based and imageless methods reliably determine the tibial ankle center.²⁵^,³⁰^,³¹ Due to the tibial shaft's length, angular errors in the mechanical axis are minimal. The tibial ankle center is approximated as either one of the following methods:1.The midpoint of a line connecting the most prominent points of both malleoli,2.A weighted transmalleolar line (54 % medial malleolus, 46 % lateral malleolus), or3.The midpoint of a line connecting the apices of both malleoli.

2.1.5

2.1.5 Tibial knee center

Victor et al. demonstrated high inter- and intra-observer precision for the tibial knee center in an image-based study, with an average error of 1.5 mm (SD 0.7).²² Corresponding angular errors in the tibial mechanical axis were low (average 0.7°, SD 0.6). In comparison, Perrin et al. observed slightly higher variability in manual acquisition, with standard deviations up to 2.2 mm.¹²

2.1.6

2.1.6 Tibial rotational axes

Numerous studies have evaluated tibial rotational axes based on preoperative imaging, emphasizing alignment of the femoral and tibial sagittal planes in full extension.²⁰ Techniques include:1.Projecting the femoral rotation axis onto the tibia, although full extension may not be possible in severe osteoarthritis cases.2.Defining a surrogate axis on the tibia resembling the femoral rotation axis in full extension. The Akagi line, connecting the posterior cruciate ligament insertion to the medial side of the tibial tuberosity, has been identified as the most reliable surrogate in a systematic review by Saffarini et al.²⁰ Notably, Vermue et al. directly compared image-based and imageless systems, reporting superior accuracy and precision with image-based techniques, particularly for the Akagi line and anatomical tibial axis.¹⁷

2.1.7

2.1.7 Kinematic analysis

The range of motion of the knee is similarly evaluated in image-based and imageless techniques. It involves assessing the angle between femoral and tibial mechanical axes in the sagittal plane during flexion and extension. The absolute values of flexion and extension are influenced by the anteroposterior positions of key landmarks (hip center, ankle center, and femoral/tibial knee center). Clinically, the difference between native and post-implant alignment in full extension holds greater importance than absolute values, as to minimize the effect of an error in the anteroposterior direction when selecting these landmarks.

3 Resection landmarks

When utilizing imageless techniques, the surgeon must account for healthy cartilage thickness and cartilage wear when selecting the location of the anatomical landmarks and when performing subsequent implant planning. Assessing cartilage wear, however, remains challenging as its location and extent can vary significantly depending on the severity of the osteoarthritic process.³² In contrast, image-based systems use a bony reference, thus they shift the focus to estimating the thickness of healthy cartilage while bypassing concerns about cartilage wear. In addition, bony wear must also be considered, especially in more severe osteoarthritis cases where significant bone loss or deformation is present. Up to date, estimating the native cartilage thickness remains difficult. A study analyzing 3910 MRI scans from the Osteoarthritis Initiative (Kellgren Lawrence 0–1) revealed considerable variability in cartilage thickness between patients (Table 1).³³ The 95 % confidence interval for the distal lateral condyle extends up to 3 mm, highlighting the challenges in estimating cartilage thickness accurately (See Table 2). For instance, relying on an average cartilage thickness of 2 mm could lead to an over- or underestimation of the true thickness by as much as 1.5 mm. Factors such as higher BMI and male sex were associated with thicker femoral cartilage, underscoring the need to individualize assessments.

Table 2 Overview of the thickness of the cartilage at the distal femur according to Shah et al.³³.

Location	Average (SD) (mm)	95 % Confidence Interval (mm)
Distal medial condyle	2.13 (0.62)	0.91–3.34
Distal lateral condyle	1.99 (0.74)	0.53–3.44
Posterior medial condyle	2.52 (0.56)	1.42–3.61
Posterior lateral condyle	2.30 (0.54)	1.24–3.35

Similarly, a systematic review by Giurazza et al., encompassing 27 studies, emphasizes on the significant variability in native cartilage thickness at the knee joint with similar results to the study of Shah et al.³⁴ These variations highlight the complexity of accurately reconstructing joint anatomy in (robot-assisted) total knee arthroplasty, regardless of the system employed.

4 Precision

The execution of bone resections differs significantly between robotic systems. Active image-based systems, such as TSolution One (THINK Surgical, California, USA) and CUVIS (Curexo Inc., Seoul, South-Korea) operate autonomously, performing cuts without direct surgeon involvement. Semi-active image-based systems, on the other hand, guide and restrict the surgeon's movements within the resection plane. For imageless systems requiring a saw, the surgeon performs cuts manually using a handheld motor. In contrast, the Navio/CORI (Smith&Nephew, London, United Kingdom) system employs a mill that automatically retracts when it is not in contact with the designated bone volume, adding a layer of safety and precision.

Industry-sponsored cadaveric studies have demonstrated high precision across all available robotic systems.²^,^4–11^,^13–15 These studies were conducted in highly controlled surgical environments by developer surgeons, presenting ideal circumstances. However, diverse surgical teams and varying degrees of knee osteoarthritis may limit the ability to construct reliable reference frames, thus impacting precision in a real-world setting.

There are few comparative clinical studies available comparing imageless and image-based systems. Yee et al. reported no significant difference in coronal implant alignment between imageless systems (Navio) and image-based systems (MAKO (Stryker, Michigan, USA)).³⁵ Both approaches achieved an average alignment error of 1° relative to the preoperative plan, though the imageless system tended to produce varus alignment and the image-based system valgus alignment. Tibial slope alignment, as measured on short-leg lateral radiographs, showed greater deviation in the imageless group, with an error of 2.5° (SD 1.9) compared to 1.4° (SD 1.6) in the image-based group. Rajgor et al. found no significant differences in surgical precision when comparing ROSA (Zimmer Biomet, Indiana, USA) to MAKO, further supporting the notion that both image-based and imageless systems can achieve comparable levels of accuracy under specific conditions.³⁶

5 Imaging cost and radiation

The cost of preoperative imaging for image-based robot-assisted TKA varies depending on the healthcare system and several contextual factors. For example, Abdelfadeel et al. reported a mean total cost of $446 for preoperative scans, but substantial variation exists due to geographical differences, negotiated contracts, and local healthcare dynamics.³⁷

Radiation exposure associated with preoperative imaging is another critical consideration. The exposure is often expressed as Effective Dose (ED), which provides a standardized measure of overall risk from ionizing radiation by factoring in the type of radiation and the varying sensitivity of different tissues and organs. It is quantified in Sieverts (Sv). Preoperative CT scans can result in significantly varied radiation exposure depending on the imaging center and protocol:¹⁶•Hip: Mean ED of 3.09 ± 1.37 mSv•Knee: Mean ED of 0.16 ± 0.12 mSv•Ankle: Mean ED of 0.07 ± 0.05 mSv

To reduce radiation exposure related to imaging, many centers use fewer CT slices when estimating the hip center, which lowers the ED compared to conventional CT scans of the hip. However, literature shows significant variability, as highlighted by Ponzio and Lonner, who reported preoperative CT scans for unicompartmental knee arthroplasty (UKA) delivering doses ranging from 3.0 to 8.5 mSv.³⁸

Radiation exposure should also be considered in the context of potential cancer risks. The background radiation exposure averages 3 mSv/year, and evidence suggests an increasing likelihood of developing malignancies at exposures exceeding 5 mSv/year.¹⁶ Below this level, the risk is not entirely eliminated but is significantly lower. In line with the As Low As Reasonably Achievable principle, clinicians may need to exercise caution for image-based TKA, especially for patients undergoing multiple CT scans within the same year.

The implementation of an image-based approach also requires logistical planning. Scheduling the CT scan and processing the images to make them ready for use during surgery requires time and resources, adding another layer of complexity to preoperative planning.

6 Patellofemoral assessment

In image-based systems, there is the capability to evaluate the alignment of the native trochlea compared to the implant's trochlea. This provides valuable insights into how well the anterior compartment is restored postoperatively. In contrast, only imageless systems such as CORI and Navio offer a similar option, although the 3D model relies on bone morphing procedure rather than the patient's actual bone anatomy. For other imageless systems, there is currently no effective method to evaluate the position of the patellofemoral joint relative to the native trochlea.

Two key aspects of the anterior compartment can be evaluated:1.Implant Orientation Compared to the Native Trochlea: Restoration of the anterior compartment has been linked to improved patient-reported outcomes, as highlighted by Kafelov et al.³ Over- or understuffing of the patellofemoral compartment can negatively impact patient satisfaction and function.2.Patellofemoral Tracking: All image-based systems and one imageless (Navio) system enable intraoperative assessment of patellofemoral tracking. According to studies by Batailler et al. and Shatrov et al., this is achieved by monitoring the location of a fixed point on the anterior aspect of the patella during flexion/extension movements.³⁹^,⁴⁰ This technique helps identify the relative position of the patella to the trochlea, serving as a measure of anterior offset and patellofemoral tracking when comparing the native and post-implant situations. However, it is worth noting that this relative positioning lacks precise quantification, highlighting an area for potential improvement in both system types.

7 Clinical outcomes

Studies comparing the clinical outcomes between imageless and image-based robotic-assisted TKA are limited and prone to significant confounding due to differences not related to the robot, with only two studies currently available. Kang et al. reported no significant differences in postoperative patient-reported outcome measures up to 1 year postoperatively (Oxford Knee Score, FS-12, Knee Injury and Osteoarthritis Outcome Score) and observed similar learning curves between the ROSA and MAKO systems.⁴¹ Yee et al. found higher Knee Society Scores at 1 year in favor of the imageless Navio/CORI system compared to the image-based MAKO (95 ± 9 vs. 92 ± 7).³⁵

Additional insights can be drawn from studies on computer-assisted TKA. Tabatabaee et al. conducted a registry-based study of over 13,000 navigated TKA procedures and found no differences in postoperative complications up to 90 days between imageless and image-based cohorts.⁴² However, a higher rate of blood transfusions was noted in the imageless group. Martin et al. evaluated navigated TKA in a small case-control study and reported no significant differences in postoperative outcomes up to 2 years, specifically regarding mechanical alignment, between imageless and image-based techniques.⁴³

These findings suggest minimal differences in clinical outcomes between imageless and image-based systems, although further research is needed to solidify these conclusions as they are biased by different implants, alignment strategies and soft tissue evaluation.

8 Soft tissues protection and evaluation

In TKA, protecting the soft-tissue envelope during surgery is of critical importance. Different robotic systems employ distinct mechanisms to ensure no unwanted soft-tissue damage occurs:•The MAKO, ROPA (Longwood Valley Medical Technology, Beijing, China) and Yuanhua (Yuanhua Robotics, Perception & AI Technologies Ltd., Shenzen, China) image-based systems utilize boundary control, which restricts the movement of the surgical saw to the predefined bony edges, thereby minimizing the risk of soft tissue damage.⁴⁴•Navio/CORI: These robots employ a surgical mill that activates only when near the target bone, further limiting the potential for inadvertent soft tissue injury.

Modern robotic systems also contribute to the evaluation and management of soft tissues, with varying capabilities:•OMNIBotics (Corin Group, Circencester, United Kingdom), and CORI: These systems incorporate tools for separate medial and lateral soft tissue assessment, applying a known force to provide consistent and standardized evaluations. Notably, current studies focused on the relationship between objectified soft tissue assessment and clinical outcomes primarily pertain to OMNIBotics.^45–48•MAKO: Although MAKO lacks a comparable objectified assessment system, an industry-sponsored cadaveric study has demonstrated low variability between analyzed gaps when soft tissue evaluations were conducted by different surgeons, suggesting reliable performance across users.⁴⁹

9 System size

The physical size of robotic systems can influence surgical workflows, particularly in constrained operating environments. Imageless systems such as Navio/CORI, OMNIBotics, and Velys (Johnson&Johnson, New Jersey, USA) are smaller and more compact compared to image-based platforms like MAKO and TSolution One, making them more suitable for space-limited surgical setups.⁵⁰

10 Conclusion

Both imageless and image-based robotic systems offer significant advantages in total knee arthroplasty, improving precision and reproducibility compared to conventional techniques. There are important differences between imageless and image-based systems regarding the reference frames used and the resection landmarks. Imageless systems streamline the surgical workflow by eliminating the need for preoperative imaging, reducing costs and radiation exposure while providing intraoperative adaptability and standardized soft-tissue assessments. In contrast, image-based systems offer highly detailed reference frames and preoperative planning of the implant position, including reproducing the native trochlea orientation. However, their reliance on imaging increases logistical demands and costs. While robot-assisted TKA is still aiming to find superiority over conventional TKA regarding clinical outcomes, in current literature no major differences have been reported between image-based and imageless systems.

Author's contributions according to CRediT taxonomy

HV: conceptualization, formal analysis, writing-original draft, writing-review and editing; CB: writing-review and editing, writing-original manuscript; SL: writing-review and editing, formal analysis.

Ethical statement

Not applicable.

Guardian or patients consent

Not applicable.

Funding statement

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

References

Vermue H., Batailler C., Monk P., Haddad F., Luyckx T., Lustig S., . The evolution of robotic systems for total knee arthroplasty, each system must be assessed for its own value: a systematic review of clinical evidence and meta-analysis. Arch Orthop Trauma Surg Ger. 2023;143(6):3369-3381.
[Google Scholar]
Hampp E.L., Chughtai M., Scholl L.Y., et al . Robotic-arm assisted total knee arthroplasty demonstrated greater accuracy and precision to plan compared with manual techniques. J Knee Surg. 2019;32(3):239-250.
[Google Scholar]
Kafelov M., Batailler C., Servien E., Lustig S., . Restoration of the anterior compartment after robotic total knee arthroplasty significantly improves functional outcome and range of motion at 1 year. Knee Surg Sports Traumatol Arthrosc Ger. 2025;33(1):319-328.
[Google Scholar]
Kreuzer S., Brar A., Campanelli V., . Dimensional accuracy of TKA cut surfaces with an active robotic system. Comput Assist Surg (Abingdon, England) Engl. 2022;27(1):41-49.
[Google Scholar]
Lee H.-J., Park K.K., Park Y.-B., Choi S.W., Kim B.-O., Kim S.H., . Accuracy of advanced active robot for total knee arthroplasty: a cadaveric study. J Knee Surg Ger. 2024;37(2):135-141.
[Google Scholar]
Li Z., Chen X., Zhang X., et al . Better precision of a new robotically assisted system for total knee arthroplasty compared to conventional techniques: a sawbone model study. Int J Med Robot Engl. 2021;17(4)
[Google Scholar]
Xia R., Tong Z., Hu Y., Kong K., Wu X., Li H., . ‘Skywalker’ surgical robot for total knee arthroplasty: an experimental sawbone study. Int J Med Robot Engl. 2021;17(5)
[Google Scholar]
Yi J., Gao Z., Huang Y., Liu Y., Zhang Y., Chai W., . Evaluating the accuracy of a new robotically assisted system in cadaveric total knee arthroplasty procedures. J Orthop Surg Res Engl. 2024;19(1):354.
[Google Scholar]
Scholl L.Y., Hampp E.L., Souza KM de, et al . How does robotic-arm assisted technology influence total knee arthroplasty implant placement for surgeons in fellowship training? J Knee Surg Ger. 2022;35(2):198-203.
[Google Scholar]
Figueroa F., Wakelin E., Twiggs J., Fritsch B., . Comparison between navigated reported position and postoperative computed tomography to evaluate accuracy in a robotic navigation system in total knee arthroplasty. Knee Netherlands 2019
[Google Scholar]
Doan G.W., Courtis R.P., Wyss J.G., Green E.W., Clary C.W., . Image-free robotic-assisted total knee arthroplasty improves implant alignment accuracy: a cadaveric study. J Arthroplasty U S. 2022;37(4):795-801.
[Google Scholar]
Perrin N., Stindel E., Roux C., . BoneMorphing versus freehand localization of anatomical landmarks: consequences for the reproducibility of implant positioning in total knee arthroplasty. Comput Aided Surg. 2005;10(5):301-309.
[Google Scholar]
Seidenstein A., Birmingham M., Foran J., Ogden S., . Better accuracy and reproducibility of a new robotically-assisted system for total knee arthroplasty compared to conventional instrumentation: a cadaveric study. Knee Surg Sports Traumatol Arthrosc Ger. 2021;29(3):859-866.
[Google Scholar]
Casper M., Mitra R., Khare R., et al . Accuracy assessment of a novel image-free handheld robot for total knee arthroplasty in a cadaveric study. Comput Assist Surg (Abingdon, England) Engl. 2018;23(1):14-20.
[Google Scholar]
Singh V., Teo G.M., Long W.J., . Versatility and accuracy of a novel image-free robotic-assisted system for total knee arthroplasty. Arch Orthop Trauma Surg Ger. 2021;141(12):2077-2086.
[Google Scholar]
Biswas D., Bible J.E., Bohan M., Simpson A.K., Whang P.G., Grauer J.N., . Radiation exposure from musculoskeletal computerized tomographic scans. J Bone Joint Surg Am U S. 2009;91(8):1882-1889.
[Google Scholar]
Vermue H., Stroobant L., Pringels L., Chevalier A., Victor J., . The definition of the tibial sagittal plane and the paradox of imageless navigation and robotics: a cadaveric study. J Arthroplasty U S. 2023;38(6S):S374-S378.
[Google Scholar]
Vermue H., Tack P., Gryson T., Victor J., . Can robot-assisted total knee arthroplasty be a cost-effective procedure? A Markov decision analysis. Knee Netherlands. 2021;29:345-352.
[Google Scholar]
Victor J., Doninck D Van, Labey L., Glabbeek F Van, Parizel P., Bellemans J., . A common reference frame for describing rotation of the distal femur: a ct-based kinematic study using cadavers. J Bone Joint Surg Br Engl. 2009;91(5):683-690.
[Google Scholar]
Saffarini M., Nover L., Tandogan R., et al . The original Akagi line is the most reliable: a systematic review of landmarks for rotational alignment of the tibial component in TKA. Knee Surg Sports Traumatol Arthrosc Ger. 2019;27(4):1018-1027.
[Google Scholar]
Meng L., Childs C., Buis A., . Evaluation of functional methods of joint centre determination for quasi-planar movement. PLoS One. 2019;14(1)
[Google Scholar]
Victor J., Doninck D Van, Labey L., Innocenti B., Parizel P.M., Bellemans J., . How precise can bony landmarks be determined on a CT scan of the knee? Knee Netherlands. 2009;16(5):358-365.
[Google Scholar]
Lustig S., Fleury C., Servien E., Demey G., Neyret P., Donell S.T., . The effect of pelvic movement on the accuracy of hip centre location acquired using an imageless navigation system. Int Orthop. 2011;35(11):1605-1610.
[Google Scholar]
Davis E.T., Pagkalos J., Gallie P.A.M., Macgroarty K., Waddell J.P., Schemitsch E.H., . Defining the errors in the registration process during imageless computer navigation in total knee arthroplasty: a cadaveric study. J Arthroplasty U S. 2014;29(4):698-701.
[Google Scholar]
Jenny J., Boeri C., Picard F., Leitner F., . Reproducibility of intra-operative measurement of the mechanical axes of the lower limb during total knee replacement with a non-image-based navigation system. Comput Aided Surg. 2004;9(4):161-165.
[Google Scholar]
Speirs A.D., Benoit D.L., Beaulieu M.L., Lamontagne M., Beaulé P.E., . The accuracy of the use of functional hip motions on localization of the center of the hip. HSS J. 2012;8(3):192-197.
[Google Scholar]
Valstar E.R., Molen AJ van der, Nelissen R.G.H.H., . Transepicondylar axis accuracy in computer assisted knee surgery: a comparison of the CT-based measured axis versus the CAS-determined axis. Comput aided Surg Off J Int Soc Comput Aided Surg Engl. 2008;13(4):200-206.
[Google Scholar]
Jenny J.-Y., Boeri C., . Low reproducibility of the intra-operative measurement of the transepicondylar axis during total knee replacement. Acta Orthop Scand Engl. 2004;75(1):74-77.
[Google Scholar]
Hernández-Vaquero D., Noriega-Fernandez A., Perez-Coto I., et al . Computed tomography is not necessary to assess rotation of the femoral component in navigation-assisted total knee replacement. J Int Med Res. 2016;44(6):1314-1322.
[Google Scholar]
Nofrini L., Slomczykowski M., Iacono F., Marcacci M., . Evaluation of accuracy in ankle center location for tibial mechanical axis identification. J Investig Surg Off J Acad Surg Res U S. 2004;17(1):23-29.
[Google Scholar]
Siston R.A., Daub A.C., Giori N.J., Goodman S.B., Delp S.L., . Evaluation of methods that locate the center of the ankle for computer-assisted total knee arthroplasty. Clin Orthop Relat Res U S. 2005;439:129-135.
[Google Scholar]
Colyn W., Bruckers L., Scheys L., Truijen J., Smeets K., Bellemans J., . Changes in coronal knee-alignment parameters during the osteoarthritis process in the varus knee. J ISAKOS Jt Disord Orthop Sport Med England. 2023;8(2):68-73.
[Google Scholar]
Shah R.F., Martinez A.M., Pedoia V., Majumdar S., Vail T.P., Bini S.A., . Variation in the thickness of knee cartilage. The use of a novel machine learning algorithm for cartilage segmentation of magnetic resonance images. J Arthroplasty U S. 2019;34(10):2210-2215.
[Google Scholar]
Giurazza G., Caria C., Campi S., et al . Femoral cartilage thickness measured on MRI varies among individuals: time to deepen one of the principles of kinematic alignment in total knee arthroplasty. A systematic review. Knee Surg Sports Traumatol Arthrosc Ger 2024
[Google Scholar]
Yee D.K.-H., Ng J.P., Lau C.T.-K., et al . Surgical accuracy of image-free versus image-based robotic-assisted total knee arthroplasty. Int J Med Robot Engl 2023
[Google Scholar]
Rajgor H.D., Mayne A., Munasinghe C., et al . Mako versus ROSA: comparing surgical accuracy in robotic total knee arthroplasty. J Robot Surg England. 2024;18(1):33.
[Google Scholar]
Abdelfadeel W., Houston N., Star A., Saxena A., Hozack W.J., . CT planning studies for robotic total knee arthroplasty. Bone Joint J Engl. 2020;102-B(6_Supple_A):79-84.
[Google Scholar]
Ponzio D.Y., Lonner J.H., . Preoperative mapping in unicompartmental knee arthroplasty using computed tomography scans is associated with radiation exposure and carries high cost. J Arthroplasty U S. 2015;30(6):964-967.
[Google Scholar]
Batailler C., Greiner S., Rekik H.-L., Olivier F., Servien E., Lustig S., . Intraoperative patellar tracking assessment during image-based robotic-assisted total knee arthroplasty: technical note and reliability study. SICOT-J Fr. 2024;10:44.
[Google Scholar]
Shatrov J., Khasian M., Lording T., Monk A.P., Parker D., Lustig S., . Robotic assessment of patella tracking in total knee arthroplasty. J ISAKOS Jt Disord Orthop Sport Med Engl. 2024;9(5)
[Google Scholar]
Kang L., Lee J., Geng X., Korostyshevskiy V., Boucher H., . Comparison of learning curves and short-term outcomes: ROSA versus MAKO robotic-assisted total knee arthroplasty. Curr Orthop Pract 2024
[Google Scholar]
Tabatabaee R.M., Rasouli M.R., Maltenfort M.G., Fuino R., Restrepo C., Oliashirazi A., . Computer-assisted total knee arthroplasty: is there a difference between image-based and imageless techniques? J Arthroplasty United States. 2018;33(4):1076-1081.
[Google Scholar]
Martin A., von Strempel A., . Two-year outcomes of computed tomography-based and computed tomography free navigation for total knee arthroplasties. Clin Orthop Relat Res (449):275-282.
[Google Scholar]
Kayani B., Konan S., Pietrzak J.R.T., Haddad F.S., . Iatrogenic bone and soft tissue trauma in robotic-arm assisted total knee arthroplasty compared with conventional jig-based total knee arthroplasty: a prospective cohort study and validation of a new classification system. J Arthroplasty U S. 2018;33(8):2496-2501.
[Google Scholar]
Keating T.C., Wakelin E.A., Plaskos C., et al . Patient specific variables impact sensitivity to association between joint balance and 2 year outcomes. J Orthop India. 2025;65:71-77.
[Google Scholar]
Keggi J.M., Wakelin E.A., Koenig J.A., et al . Impact of intra-operative predictive ligament balance on post-operative balance and patient outcome in TKA: a prospective multicenter study. Arch Orthop Trauma Surg Ger. 2021;141(12):2165-2174.
[Google Scholar]
Wakelin E.A., Ponder C.E., Randall A.L., et al . Intra-operative laxity and balance impact 2-year pain outcomes in TKA: a prospective cohort study. Knee Surg Sports Traumatol Arthrosc Ger. 2023;31(12):5535-5545.
[Google Scholar]
Wakelin E.A., Shalhoub S., Lawrence J.M., et al . Improved total knee arthroplasty pain outcome when joint gap targets are achieved throughout flexion. Knee Surg Sports Traumatol Arthrosc Ger. 2022;30(3):939-947.
[Google Scholar]
Scholl L.Y., Hampp E.L., Caba M., et al . Robotic-assisted total knee arthroplasty technology provides a repeatable and reproducible method of assessing soft tissue balance. J Knee Surg Ger. 2024;37(8):607-611.
[Google Scholar]
Walgrave S., Oussedik S., . Comparative assessment of current robotic-assisted systems in primary total knee arthroplasty. Bone Jt open Engl. 2023;4(1):13-18.
[Google Scholar]