Virtual Needle Insertion with Enhanced Haptic Feedback for Guidance and Needle-Tissue Interaction Forces.

Interventional radiologists mainly rely on visual feedback via imaging modalities to steer a needle toward a tumor during biopsy and ablation procedures. In the case of CT-guided procedures, there is a risk of exposure to hazardous X-ray-based ionizing radiation. Therefore, CT scans are usually not used continuously, which increases the chances of a misplacement of the needle and the need for reinsertion, leading to more tissue trauma. Interventionalists also encounter haptic feedback via needle-tissue interaction forces while steering a needle. These forces are useful but insufficient to clearly perceive and identify deep-tissue structures such as tumors. The objective of this paper was to investigate the effect of enhanced force feedback for sensing interaction forces and guiding the needle when applied individually and simultaneously during a virtual CT-guided needle insertion task. We also compared the enhanced haptic feedback to enhanced visual feedback. We hypothesized that enhancing the haptic feedback limits the time needed to reach the target accurately and reduces the number of CT scans, as the interventionalist depends more on real-time enhanced haptic feedback. To test the hypothesis, a simulation environment was developed to virtually steer a needle in five degrees of freedom (DoF) to reach a tumor target embedded in a liver model. Twelve participants performed in the experiment with different feedback conditions where we measured their performance in terms of the following: targeting accuracy, trajectory tracking, number of CT scans required, and the time needed to finish the task. The results suggest that the combination of enhanced haptic feedback for guidance and sensing needle-tissue interaction forces significantly reduce the number of scans and the duration required to finish the task by 32.1% and 46.9%, respectively, when compared to nonenhanced haptic feedback. The other feedback modalities significantly reduced the duration to finish the task by around 30% compared to nonenhanced haptic feedback.

Finite-element modelling of interactions of needle with tympanic membrane and middle ear.

The tympanic membrane (TM) is one of the most common routes to access the middle ear and inner ear for the treatment of hearing and balance pathologies. Since the TM is a soft thin biological tissue with small dimensions, using needles seems to be among the most practical interventional approaches. In this study, we proposed a finite-element (FE) analysis of needle-TM interactions that combines a 3D model of the TM and other main middle-ear structures in gerbil, and a 2D model of needle insertion into the TM based on the cohesive zone method (CZM). The TM was modelled using a 1st-order Ogden hyperelastic material and its properties were obtained by fitting to the experimental force-displacement plots of large deformation in the TM under needle indentation. The cohesive parameters were also acquired by calibrating the puncture force against the experimental data of needle insertion into the TM. These FE models were then used to obtain the deformation behaviour of the TM and other middle-ear structures due to the insertion force applied at different locations on the TM. Moreover, we investigated the effect of the TM thickness, the geometry of the needle (i.e., diameter and tip angle), and needle material on the insertion of needles into the TM. We also studied the penetration success of deformable needles.

A method for predicting needle insertion deflection in soft tissue based on cutting force identification.

The deflection modeling during the insertion of bevel-tipped flexible needles into soft tissues is crucial for robot-assisted flexible needle insertion into specific target locations within the human body during percutaneous biopsy surgery. This paper proposes a mechanical model based on cutting force identification to predict the deflection of flexible needles in soft tissues. Unlike other models, this method does not require measuring Young's modulus (E) and Poisson's ratio (ν) of tissues, which require complex hardware to obtain. In the model, the needle puncture process is discretized into a series of uniform-depth puncture steps. The needle is simplified as a cantilever beam supported by a series of virtual springs, and the influence of tissue stiffness on needle deformation is represented by the spring stiffness coefficient of the virtual spring. By theoretical modeling and experimental parameter identification of cutting force, the spring stiffness coefficients are obtained, thereby modeling the deflection of the needle. To verify the accuracy of the proposed model, the predicted model results were compared with the deflection of the puncture experiment in polyvinyl alcohol (PVA) gel samples, and the average maximum error range predicted by the model was between 0.606 ± 0.167 mm and 1.005 ± 0.174 mm, which showed that the model can successfully predict the deflection of the needle. This work will contribute to the design of automatic control strategies for needles.

Experimental Study of Needle Insertion into Gerbil Tympanic Membrane.

The perforation characteristics and fracture-related mechanical properties of the tympanic membrane (TM) greatly affect surgical procedures like myringotomy and tympanostomy performed on the middle ear. We analyzed the most important features of the gerbil TM perforation using an experimental approach that was based on force measurement during a 2-cycle needle insertion/extraction process. Fracture energy, friction energy, strain energy, and hysteresis loss were taken into consideration for the analysis of the different stages of needle insertion and extraction. The results demonstrated that (1) although the TM shows viscoelastic behavior, the contribution of hysteresis loss was negligible compared to other irreversible dissipated energy components (i.e., fracture energy and friction energy). (2) The TM puncture force did not substantially change during the first hours after animal death, but interestingly, it increased after 1 week due to the drying effects of soft tissue. (3) The needle geometry affected the crack length and the most important features of the force-displacement plot for the needle insertion process (puncture force, puncture displacement, and jump-in force) increased with increasing needle diameter, whereas the insertion velocity only changed the puncture and jump-in forces (both increased with increasing insertion velocity) and did not have a noticeable effect on the puncture displacement. (4) The fracture toughness of the gerbil TM was almost independent of the needle geometry and was found to be around 0.33 ± 0.10 kJ/m2.

CT Angiography-Guided Needle Insertion for Interstitial Brachytherapy in Locally Advanced Cervical Cancer.

CT angiography might be a suitable procedure to avoid arterial puncture in combined intracavitary and interstitial brachytherapy for cervical cancer curatively treated with combined chemoradiation and brachytherapy boost. Data in the literature about this technique are scarce. We introduced this method and collected brachytherapy data from patients treated in our department between May 2021 and April 2024. We analyzed the applicator subtype, needle insertion (planned versus implanted), implanted depth and the role of CT angiography in selecting needle trajectories and insertion depths. None of the patients managed through this protocol experienced atrial puncture and consequent hemorrhage. Needle positions were accurately selected with the aid of CT angiography with proper coverage of brachytherapy targets and avoidance of organs at risk. CT angiography is a promising method for guiding needle insertion during interstitial brachytherapy.

Curvilinear catheter implantation in HDR prostate brachytherapy: feasibility study.

BACKGROUND: High-dose-rate (HDR) brachytherapy (BT) has been acknowledged as a widely utilized treatment for patients with intermediate- and high-risk prostate cancer, despite its side effects such as edema, incontinence, and impotence. Nevertheless, the treatment is consistently limited by the potential danger of excessive irradiation to organs-at-risk (OARs) like the urethra, bladder, and rectum. PURPOSE: This study aims to introduce curvilinear catheter implantation in the prostate gland for HDR treatment. The objective is to improve the radiation dose distribution by offering access channels conformal to the prostate anatomy. This approach seeks to minimize toxicity to nearby OARs while utilizing a reduced number of needles, potentially leading to improved clinical outcomes. METHODS: Curvilinear catheters were first pre-planned for an anonymized patient using Oncentra treatment planning system (TPS) and hybrid inverse planning optimization (HIPO) algorithm. The trajectories of the catheters were then analyzed using MATLAB to extract their radius of curvature. Tendon-driven active needles were then used to implant curvilinear catheters inside an anthropomorphic phantom. RESULTS: Proposed curvilinear catheter implantation resulted in significant improvement in terms of dosimetric constraints to the OARs and coverage to the prostate. Tendon-driven active needles were shown to be capable of realizing the required pre-planned curvatures inside prostate. It was shown that the active needle can realize a desired radius of curvature and a desired trajectory with an average accuracy of 9.1 ± 8.6  and 1.27 ± 0.50 mm in air and inside a tissue-mimicking phantom, respectively. CONCLUSION: This work demonstrates the feasibility of using tendon-driven active curvilinear catheter implantation in prostate to improve the outcomes of HDR-BT via improved radiation dose distribution to the prostate and reduced toxicity to the OARs.

Phantom study of a fully automatic radioactive seed placement robot for the treatment of skull base tumours.

BACKGROUND: Interstitial brachytherapy is a form of intensive local irradiation that facilitates the effective protection of surrounding structures and the preservation of organ functions, resulting in a favourable therapeutic response. As surgical robots can perform needle placement with a high level of accuracy, our team developed a fully automatic radioactive seed placement robot, and this study aimed to evaluate the accuracy and feasibility of fully automatic radioactive seed placement for the treatment of tumours in the skull base. METHODS: A fully automatic radioactive seed placement robot was established, and 4 phantoms of skull base tumours were built for experimental validation. All the phantoms were subjected to computed tomography (CT) scans. Then, the CT data were imported into the Remebot software to design the preoperative seed placement plan. After the phantoms were fixed in place, navigation registration of the Remebot was carried out, and the automatic seed placement device was controlled to complete the needle insertion and particle placement operations. After all of the seeds were implanted in the 4 phantoms, postoperative image scanning was performed, and the results were verified via image fusion. RESULTS: A total of 120 seeds were implanted in 4 phantoms. The average error of seed placement was (2.51 ± 1.44) mm. CONCLUSION: This study presents an innovative, fully automated radioactive particle implantation system utilizing the Remebot device, which can successfully complete automated localization, needle insertion, and radioactive particle implantation procedures for skull base tumours. The phantom experiments showed the robotic system to be reliable, stable, efficient and safe. However, further research on the needle-soft tissue interaction and deformation mechanism of needle puncture is still needed.

A Hollow Microneedle Equipped with a Micropillar for Improved Needle Insertion and Injection of Drug Solution.

PURPOSE: Hollow-type microneedles (hMNs) are a promising device for the effective administration of drugs into intradermal sites. Complete insertion of the needle into the skin and administration of the drug solution without leakage must be achieved to obtain bioavailability or a constant effect. In the present study, several types of hMN with or without a rounded blunt tip micropillar, which suppresses skin deformation, around a hollow needle, and the effect on successful needle insertion and administration of a drug solution was investigated. Six different types of hMNs with needle lengths of 1000, 1300, and 1500 µm with or without a micropillar were used. METHODS: Needle insertion and the disposition of a drug in rat skin were investigated. In addition, the displacement-force profile during application of hMNs was also investigated using a texture analyzer with an artificial membrane to examine needle factors affecting successful insertion and administration of a drug solution by comparing with in vivo results. RESULTS: According to the results with the drug distribution of iodine, hMN1300 with a micropillar was able to successfully inject drug solution into an intradermal site with a high success rate. In addition, the results of displacement-force profiles with an artificial membrane showed that a micropillar can be effective for depth control of the injected solution as well as the prevention of contact between the hMN pedestal and the deformed membrane. CONCLUSION: In the present study, hMN1300S showed effective solution delivery into an intradermal site. In particular, a micropillar can be effective for depth control of the injected solution as well as preventing contact between the hMN pedestal and the deformed membrane. The obtained results will help in the design and development of hMNs that ensure successful injection of an administered drug.

Effect of twisting of intravitreal injections on ocular bio-mechanics: a novel insight to ocular surgery.

Although intravitreal (IVT) injections provide several advantages in treating posterior segment eye diseases, several associated challenges remain. The current study uses the finite element method (FEM) to highlight the effect of IVT needle rotation along the insertion axis on the reaction forces and deformation inside the eye. A comparison of the reaction forces at the eye's key locations has been made with and without rotation. In addition, a sensitivity analysis of various parameters, such as the needle's angular speed, insertion location, angle, gauge, shape, and intraocular pressure (IOP), has been carried out to delineate the individual parameter's effect on reaction forces during rotation. Results demonstrate that twisting the needle significantly reduces the reaction forces at the penetration location and throughout the needle travel length, resulting in quicker penetration. Moreover, ocular biomechanics are influenced by needle insertion location, angle, shape, size, and IOP. The reaction forces incurred by the patient may be reduced by using a bevel needle of the higher gauge when inserted close to the normal of the local scleral surface toward the orra serrata within the Pars Plana region. Results obtained from the current study can deepen the understanding of the twisting needle's interaction with the ocular tissue.

Diffuse domain approach for flexible needle insertion and relaxation.

Needle insertion simulations play an important role in medical training and surgical planning. Most simulations require boundary conforming meshes, while the diffuse domain approach, currently limited to stiff needles, eliminates the need for meshing geometries. In this article the diffuse domain approach for needle insertion simulations is first extended to the use of flexible needles with bevel needle tips, which are represented by an Euler-Bernoulli beam. The model parameters are tuned and the model is evaluated on a real-world phantom experiment. Second, a new method for the relaxation of the needle-tissue system after the user releases the needle is introduced. The equilibrium state of the system is determined by minimizing the potential energy. The convergence rate of the coupled Laplace equations for solving the Euler-Bernoulli beam is 1.92 ± 0.14 for decreasing cell size. The diffuse penalty method for the application of Dirichlet boundary conditions results in a convergence rate of 0.73 ± 0.21 for decreasing phase field width. The simulated needle deviates on average by 0.29 mm compared to the phantom experiment. The error of the tissue deformation is below 1 mm for 97.5% of the attached markers. Two additional experiments demonstrate the feasibility of the relaxation process. The simulation method presented here is a valuable tool for patient-specific medical simulations using flexible needles without the need for boundary conforming meshing. To the best of the authors' knowledge this is the first work to introduce a relaxation model, which is a major step for simulating accurate needle-tissue positioning during realistic medical interventions.

In-Silico Analysis of Optimal Configurations for Rotational Bioinspired Bone Marrow Biopsy Needle Designs: An ANN Approach.

Medical needle innovations have utilized rotating motion to enhance tissue-cutting capabilities, reducing cutting force and improving clinical outcomes. This study analyzes the effects of six essential factors on insertion and extraction forces during bone marrow biopsy (BMB) procedures. The study uses Taguchi's L32 orthogonal array and numerically simulates the BMB process using the Lagrangian surface-based method on a three-dimensional (3D) heterogeneous Finite Element (FE) model of the human iliac crest. The study evaluates cutting forces in needle insertion and extraction using uni-directional (360° rotation) and bidirectional (180° clock and anti-clock rotation) bioinspired BMB needles. This work aims to create an AI tool that assists researchers and clinicians in selecting the most suitable and safe design parameters for a bio-inspired barbed biopsy needle. An efficient Graphical User Interface (GUI) has been developed for easy use and seamless interaction with the AI tool. With a remarkable accuracy rate exceeding 98%, the tool's predictions hold significant value in facilitating the development of environmentally conscious biopsy needles. The tool demonstrates significantly higher efficiency compared to Abaqus, rendering it a valuable asset for researchers and clinicians engaged in bio-inspired biopsy needle development.

A comprehensive review of haptic feedback in minimally invasive robotic liver surgery: Advancements and challenges.

BACKGROUND: Liver medical procedures are considered one of the most challenging because of the liver's complex geometry, heterogeneity, mechanical properties, and movement due to respiration. Haptic features integrated into needle insertion systems and other medical devices could support physicians but are uncommon. Additional training time and safety concerns make it difficult to implement in robot-assisted surgery. The main challenges of any haptic device in a teleoperated system are the stability and transparency levels required to develop a safe and efficient system that suits the physician's needs. PURPOSE: The objective of the review article is to investigate whether haptic-based teleoperation potentially improves the efficiency and safety of liver needle insertion procedures compared with insertion without haptic feedback. In addition, it looks into haptic technology that can be integrated into simulators to train novice physicians in liver procedures. METHODS: This review presents the physician's needs during liver interventions and the consequent requirements of haptic features to help the physician. This paper provides an overview of the different aspects of a teleoperation system in various applications, especially in the medical field. It finally presents the state-of-the-art haptic technology in robot-assisted procedures for the liver. This includes 3D virtual models of the liver and force measurement techniques used in haptic rendering to estimate the real-time position of the surgical instrument relative to the liver. RESULTS: Haptic feedback technology can be used to navigate the surgical tool through the desired trajectory to reach the target accurately and avoid critical regions. It also helps distinguish between various textures of liver tissue. CONCLUSION: Haptic feedback can complement the physician's experience to compensate for the lack of real-time imaging during Computed Tomography guided (CT-guided) liver procedures. Consequently, it helps the physician mitigate the destruction of healthy tissues and takes less time to reach the target.

Preliminary evaluation for ultrasound-guided targeted prostate biopsy using a portable surgical robot: Ex vivo results.

BACKGROUND: Robotic systems are increasingly used to enhance clinical outcomes in prostate intervention. To evaluate the clinical value of the proposed portable robot, the robot-assisted and robot-targeted punctures were validated experimentally. METHOD: The robot registration utilising the electromagnetic tracker achieves coordinate transformation from the ultrasound (US) image to the robot. Subsequently, Transrectal ultrasound (TRUS)-guided phantom trials were conducted for robot-assisted, free-hand, and robot-targeted punctures. RESULTS: The accuracy of robot registration was 0.95 mm, and the accuracy of robot-assisted, free-hand, and robot-targeted punctures was 2.38 ± 0.64 mm, 3.11 ± 0.72 mm, and 3.29 ± 0.83 mm sequentially. CONCLUSION: The registration method has been successfully applied to robot-targeted puncture. Current results indicate that the accuracy of robot-targeted puncture is slightly inferior to that of manual operations. Moreover, in manual operation, robot-assisted puncture improves the accuracy of free-hand puncture. Accuracy superior to 3.5 mm demonstrates the clinical applicability of both robot-assisted and robot-targeted punctures.

A standardized protocol for needle placement in the infraspinatus muscle: an anatomical perspective.

PURPOSE: This study aimed to evaluate the morphology of the three parts of the infraspinatus muscle based on surface landmarks for precise and effective access, and to propose the most effective fine-wire electrode insertion technique and sites. METHODS: Fifteen Asian fresh cadavers were used. We investigated the probability of the presence of the superior, middle, and inferior parts in each infraspinatus muscle based on surface landmarks. Based on the positional characteristics of the muscle, we determined the needle insertion method and confirmed its effectiveness by dissection. RESULTS: The superior part was mostly observed near the spine of the scapula. The middle part was broadly observed within the infraspinous fossa. The inferior part showed variable location within the infraspinous fossa. The injection accuracy of the superior, middle, and inferior parts in the infraspinatus muscle was 95.8%, 100%, and 91.7%, respectively. Targeting the superior and middle parts for injection of the infraspinatus muscle is relatively more straightforward than targeting the inferior part. Targeting the inferior part of the infraspinatus muscle in this study was more challenging than targeting the superior and middle parts. CONCLUSION: Needling for electromyography should be performed with special care to avoid unintended muscle parts, which could lead to inaccurate data acquisition and affect the conclusions about muscle function.

A novel computational fracture toughness model for soft tissue in needle insertion.

During the process of percutaneous puncture vascular intervention operation in endoscopic liver surgery, high precision needle manipulation requires the accurate needle tissue interaction model where the tissue fracture toughness is an important parameter to describe the tissue crack propagation, as well as to estimate tissue deformation and target displacement. However, the existing studies on fracture toughness estimation did not consider Young's modulus and the organ capsule structure. In this paper, a novel computational fracture toughness model is proposed considering insertion velocity, needle diameter and Young's modulus in insertion process, where the fracture toughness is determined by the tissue surface deformation, which was estimated through energy modeling using integrated shell element and three-dimensional solid element. The testbed is built to study the effect of different insertion velocities, needle diameters and Young's modulus on fracture toughness. The experiment result shows that the estimated result of computational fracture toughness model agrees well with the physical experimental data. In addition, the sensitivity analysis of different factors is conducted. Meanwhile, the model robustness analysis is investigated with different observation noises of Young's modulus and puncture displacement.

Mechanical characterisation of commercial artificial skin models.

Understanding of the mechanical properties of skin is crucial in evaluating the performance of skin-interfacing medical devices. Artificial skin models (ASMs) have rapidly gained attention as they are able to overcome the challenges in ethically sourcing consistent and representative ex vivo animal or human tissue models. Although some ASMs have become commercialised, a thorough understanding of the mechanical properties of the skin models is crucial to ensure that they are suitable for the purpose of the study. In the present study, skin and fat layers of ASMs (Simulab®, LifeLike®, SynDaver® and Parafilm®) were mechanically characterised through hardness, needle insertion, tensile and compression testing. Different boundary constraint conditions (minimally and highly constrained) were investigated for needle insertion testing, while anisotropic properties of the skin models were investigated through different specimen orientations during tensile testing. Analysis of variance (ANOVA) tests were performed to compare the mechanical properties between the skin models. Properties of the skin models were compared against literature to determine the suitability of the skin models based on the material property of interest. All skin models offer relatively consistent mechanical performance, providing a solid basis for benchtop evaluation of skin-interfacing medical device performance. Through prioritising models with mechanical properties that are consistent with human skin data, and with limited variance, researchers can use the data presented here as a toolbox to select the most appropriate ASM for their particular application.

A study on modeling the deflection of surgical needle during insertion into multilayer tissues.

The use of subcutaneous and percutaneous needle and catheter insertions is standard in modern clinical practice. However, a common issue with bevel tip surgical needles is their tendency to deflect, causing them to miss the intended target inside the tissue. This study aims to understand the interaction between the needle and soft tissue and develop a model to predict the deflection of a bevel tip needle during insertion into multi-layered soft tissues. The study examined the mechanics of needle-tissue interaction and modeled the forces involved during insertion. The force model includes cutting force, deformation force, and friction between the needle and tissue. There was an 8%-23% difference between the total analytical and experimental force measurements. A modified Euler-Bernoulli beam elastic foundation theory was used to create an analytical model to predict the needle tip deflection in soft tissue. To validate the results, the analytical deflection model was then compared to the deflection from needle insertion experiments on multi-layered phantom tissues, showing a 9%-21% error between the two. While there is a slight discrepancy between the analytical and experimental results, the study shows that the proposed model can accurately predict needle tip deflection during insertion.

Bimanual Intravenous Needle Insertion Simulation Using Nonhomogeneous Haptic Device Integrated into Mixed Reality.

In this study, we developed a new haptic-mixed reality intravenous (HMR-IV) needle insertion simulation system, providing a bimanual haptic interface integrated into a mixed reality system with programmable variabilities considering real clinical environments. The system was designed for nursing students or healthcare professionals to practice IV needle insertion into a virtual arm with unlimited attempts under various changing insertion conditions (e.g., skin: color, texture, stiffness, friction; vein: size, shape, location depth, stiffness, friction). To achieve accurate hand-eye coordination under dynamic mixed reality scenarios, two different haptic devices (Dexmo and Geomagic Touch) and a standalone mixed reality system (HoloLens 2) were integrated and synchronized through multistep calibration for different coordinate systems (real world, virtual world, mixed reality world, haptic interface world, HoloLens camera). In addition, force-profile-based haptic rendering proposed in this study was able to successfully mimic the real tactile feeling of IV needle insertion. Further, a global hand-tracking method, combining two depth sensors (HoloLens and Leap Motion), was developed to accurately track a haptic glove and simulate grasping a virtual hand with force feedback. We conducted an evaluation study with 20 participants (9 experts and 11 novices) to measure the usability of the HMR-IV simulation system with user performance under various insertion conditions. The quantitative results from our own metric and qualitative results from the NASA Task Load Index demonstrate the usability of our system.

Experimental and analytical study on insertion force of composite-coated needle in soft tissue material.

Medical interventions require control over surgical needle insertion to minimize tissue damage and target inaccuracies during percutaneous procedures. The composite coating of the needle using Polydopamine (PDA), Polytetrafluoroethylene (PTFE), and Activated Carbon (C) has been used to reduce the damaging needle insertion force. This research aims to further understand the interfacial mechanics of coated needle insertion by studying the forces at the needle and tissue interface and developing an analytical insertion force model through a combined experimental and numerical method. The proposed analytical force model is divided into two components: (1) Friction force on the needle shaft, modeled using a modified Karnopp model that includes an elastic force component; (2) Cutting force on the needle tip, modeled using a constant cutting coefficient for a given tissue and insertion speed. In this work, the analytical model was established by incorporating experiments conducted at a reasonable 35 mm insertion depth in tissues. In a bovine kidney with a 35 mm insertion depth, the insertion force evaluated through experimentation and modeling differed by 6.5% for a bare needle and 17.1% for a coated needle. It is important to note that this difference in the analytical insertion force model is anticipated when dealing with real tissues with a highly complex structured tissue. Prediction of the insertion force could potentially be utilized in robotic needle systems for needle control to improve the success of percutaneous procedures.

An autonomous X-ray image acquisition and interpretation system for assisting percutaneous pelvic fracture fixation.

PURPOSE: Percutaneous fracture fixation involves multiple X-ray acquisitions to determine adequate tool trajectories in bony anatomy. In order to reduce time spent adjusting the X-ray imager's gantry, avoid excess acquisitions, and anticipate inadequate trajectories before penetrating bone, we propose an autonomous system for intra-operative feedback that combines robotic X-ray imaging and machine learning for automated image acquisition and interpretation, respectively. METHODS: Our approach reconstructs an appropriate trajectory in a two-image sequence, where the optimal second viewpoint is determined based on analysis of the first image. A deep neural network is responsible for detecting the tool and corridor, here a K-wire and the superior pubic ramus, respectively, in these radiographs. The reconstructed corridor and K-wire pose are compared to determine likelihood of cortical breach, and both are visualized for the clinician in a mixed reality environment that is spatially registered to the patient and delivered by an optical see-through head-mounted display. RESULTS: We assess the upper bounds on system performance through in silico evaluation across 11 CTs with fractures present, in which the corridor and K-wire are adequately reconstructed. In post hoc analysis of radiographs across 3 cadaveric specimens, our system determines the appropriate trajectory to within 2.8 ± 1.3 mm and 2.7 ± 1.8[Formula: see text]. CONCLUSION: An expert user study with an anthropomorphic phantom demonstrates how our autonomous, integrated system requires fewer images and lower movement to guide and confirm adequate placement compared to current clinical practice. Code and data are available.