HMD-EgoPose: head-mounted display-based egocentric marker-less tool and hand pose estimation for augmented surgical guidance

Enhanced 2D Hand Pose Estimation for Gloved Medical Applications: A Preliminary Model

(1) Background: As digital health technology evolves, the role of accurate medical-gloved hand tracking is becoming more important for the assessment and training of practitioners to reduce procedural errors in clinical settings. (2) Method: This study utilized computer vision for hand pose estimation to model skeletal hand movements during in situ aseptic drug compounding procedures. High-definition video cameras recorded hand movements while practitioners wore medical gloves of different colors. Hand poses were manually annotated, and machine learning models were developed and trained using the DeepLabCut interface via an 80/20 training/testing split. (3) Results: The developed model achieved an average root mean square error (RMSE) of 5.89 pixels across the training data set and 10.06 pixels across the test set. When excluding keypoints with a confidence value below 60%, the test set RMSE improved to 7.48 pixels, reflecting high accuracy in hand pose tracking. (4) Conclusions: The developed hand pose estimation model effectively tracks hand movements across both controlled and in situ drug compounding contexts, offering a first-of-its-kind medical glove hand tracking method. This model holds potential for enhancing clinical training and ensuring procedural safety, particularly in tasks requiring high precision such as drug compounding.

Portable Head-Mounted System for Mobile Forearm Tracking

Computer vision (CV)-based systems using cameras and recognition algorithms offer touchless, cost-effective, precise, and versatile hand tracking. These systems allow unrestricted, fluid, and natural movements without the constraints of wearable devices, gaining popularity in human-system interaction, virtual reality, and medical procedures. However, traditional CV-based systems, relying on stationary cameras, are not compatible with mobile applications and demand substantial computing power. To address these limitations, we propose a portable hand-tracking system utilizing the Leap Motion Controller 2 (LMC) mounted on the head and controlled by a single-board computer (SBC) powered by a compact power bank. The proposed system enhances portability, enabling users to interact freely with their surroundings. We present the system's design and conduct experimental tests to evaluate its robustness under variable lighting conditions, power consumption, CPU usage, temperature, and frame rate. This portable hand-tracking solution, which has minimal weight and runs independently of external power, proves suitable for mobile applications in daily life.

Evaluation of single-stage vision models for pose estimation of surgical instruments

PURPOSE

Multiple applications in open surgical environments may benefit from adoption of markerless computer vision depending on associated speed and accuracy requirements. The current work evaluates vision models for 6-degree of freedom pose estimation of surgical instruments in RGB scenes. Potential use cases are discussed based on observed performance.

METHODS

Convolutional neural nets were developed with simulated training data for 6-degree of freedom pose estimation of a representative surgical instrument in RGB scenes. Trained models were evaluated with simulated and real-world scenes. Real-world scenes were produced by using a robotic manipulator to procedurally generate a wide range of object poses.

RESULTS

CNNs trained in simulation transferred to real-world evaluation scenes with a mild decrease in pose accuracy. Model performance was sensitive to input image resolution and orientation prediction format. The model with highest accuracy demonstrated mean in-plane translation error of 13 mm and mean long axis orientation error of 5[Formula: see text] in simulated evaluation scenes. Similar errors of 29 mm and 8[Formula: see text] were observed in real-world scenes.

CONCLUSION

6-DoF pose estimators can predict object pose in RGB scenes with real-time inference speed. Observed pose accuracy suggests that applications such as coarse-grained guidance, surgical skill evaluation, or instrument tracking for tray optimization may benefit from markerless pose estimation.

On-Device Execution of Deep Learning Models on HoloLens2 for Real-Time Augmented Reality Medical Applications

The integration of Deep Learning (DL) models with the HoloLens2 Augmented Reality (AR) headset has enormous potential for real-time AR medical applications. Currently, most applications execute the models on an external server that communicates with the headset via Wi-Fi. This client-server architecture introduces undesirable delays and lacks reliability for real-time applications. However, due to HoloLens2's limited computation capabilities, running the DL model directly on the device and achieving real-time performances is not trivial. Therefore, this study has two primary objectives: (i) to systematically evaluate two popular frameworks to execute DL models on HoloLens2-Unity Barracuda and Windows Machine Learning (WinML)-using the inference time as the primary evaluation metric; (ii) to provide benchmark values for state-of-the-art DL models that can be integrated in different medical applications (e.g., Yolo and Unet models). In this study, we executed DL models with various complexities and analyzed inference times ranging from a few milliseconds to seconds. Our results show that Unity Barracuda is significantly faster than WinML (p-value < 0.005). With our findings, we sought to provide practical guidance and reference values for future studies aiming to develop single, portable AR systems for real-time medical assistance.

Augmenting Performance: A Systematic Review of Optical See-Through Head-Mounted Displays in Surgery

We conducted a systematic review of recent literature to understand the current challenges in the use of optical see-through head-mounted displays (OST-HMDs) for augmented reality (AR) assisted surgery. Using Google Scholar, 57 relevant articles from 1 January 2021 through 18 March 2022 were identified. Selected articles were then categorized based on a taxonomy that described the required components of an effective AR-based navigation system: data, processing, overlay, view, and validation. Our findings indicated a focus on orthopedic (n=20) and maxillofacial surgeries (n=8). For preoperative input data, computed tomography (CT) (n=34), and surface rendered models (n=39) were most commonly used to represent image information. Virtual content was commonly directly superimposed with the target site (n=47); this was achieved by surface tracking of fiducials (n=30), external tracking (n=16), or manual placement (n=11). Microsoft HoloLens devices (n=24 in 2021, n=7 in 2022) were the most frequently used OST-HMDs; gestures and/or voice (n=32) served as the preferred interaction paradigm. Though promising system accuracy in the order of 2–5 mm has been demonstrated in phantom models, several human factors and technical challenges—perception, ease of use, context, interaction, and occlusion—remain to be addressed prior to widespread adoption of OST-HMD led surgical navigation.