Untethered mobile microrobots controlled by an external magnetic gradient field can be employed as advanced biomedical applications inside the human body such as cell therapy, micromanipulation, and noninvasive surgery. Haptic technology and telecommunication, on the other hand, can extend the potentials of untethered microrobot applications. In those applications, users can communicate with the robot operating system remotely to manipulate microrobots with haptic feedback. Haptic sensations artificially constructed by the wirelessly communicated information can assist human operators to experience forces while controlling the microrobots. The proposed system is composed of a haptic device and a magnetic tweezer system, both of which are integrated through a teleoperation technique based on network communication. Users can control the microrobots remotely and feel the haptic interactions with the remote environment in real-time. The 3D haptic environment is reconstructed dynamically by a model-free haptic rendering algorithm using a 2D planar image input of the microscope. The interaction between microrobots and environmental objects is haptically rendered as 3D objects to achieve spatial haptic operation with obstacle avoidance. Moreover, path generation and path guidance forces provide virtual interaction for human users to manipulate the microrobot by following the near-optimal path in path-following tasks. The potential applications of the presented system are medical remote treatment in different sites, remote drug delivery by avoiding physically penetrating through the skin, remotely-controlled cell manipulations, and biopsy without a biopsy needle.

This paper presents a forward kinematic model of a wire-driven surgical robot arm with an articulated joint structure and path generation algorithms with solutions of inverse kinematics. The proposed methods were applied to a wire-driven surgical robot for single-port surgery. This robot has a snake-like robotic arm with double segments to fit the working space in a single port and a joint structure to secure stiffness. The accuracy of the model is highly important because small surgical robot arms are usually controlled by open-loop control. A curvature model is widely used to describe and control a continuum robotic body. However, the model is quite different from a continuum robotic arm with a joint structure and can lead to slack of the driving wires or decreased stiffness of the joints. An accurate forward kinematic model was derived to fit the actual hardware structure via the frame transformation method. An inverse kinematic model from the joint space to the wire-length space was determined from an asymmetric model for the joint structure as opposed to a symmetric curvature model. The path generation algorithm has to generate a command to send to each actuator in open-loop control. Two real-time path generation algorithms that solve for inverse kinematics from the task space to the joint space were designed and compared using simulations and experiments. One of the algorithms is an optimization method with sequential quadratic programming (SQP), and the other uses differential kinematics with a PID (Proportional-Integral-Derivative) control algorithm. The strengths and weaknesses of each algorithm are discussed.

Needles are widely used in medicine for minimally invasive procedures. A steerable flexible needle was first introduced about 15 years ago, which was a new type of needle and could follow three-dimensional curved trajectory during medical procedures. The flexible needle has the limit of a single and low curvature. In this paper, we overcome this limit by designing mechanisms for tube-wire type flexible needles. We also provide a systematic planning method for an automated operation of the needle insertion using the mechanisms. Using the new system, we can achieve high and multiple curvatures from needle trajectories. The proposed design consists of an inner prebent wire and an outer tube, which are connected to two special mechanisms, an extension switch and a friction cart. It allows the trajectory of the needle to have high and multiple curvatures, which will enable the needle to easily reach target positions while efficiently avoiding obstacles. Users can efficiently control the needle device with simple inputs (insertion and rotation) using the special operation mechanism, which achieves three system functions (insertion/retraction, rotation, curvature changes) using only two actuation motors. Compared to prebent needles or duty-cycled spinning, this needle design causes less tissue damage. We build an automatic system to operate the new design of the steerable needle and test it. The performance of the new needle is verified by experiments with ballistic gelatin and animal tissue samples.

BACKGROUND The Military Healthcare System (MHS) mandates medical documentation at all echelons of care, however care providers in high-intensity combat situations must prioritize lifesaving measures over record-keeping, leading to information gaps across the care continuum. Effective human machine teaming (HMT) solutions designed to autonomously document care delivery will serve as future force multipliers in tactical combat casualty care (TCCC) environments. OBJECTIVE To address this challenge, the United States Army Institute of Surgical Research (USAISR) commenced an effort to prototype HMT systems designed to passively document care delivery within the TCCC environment. However, common artificial intelligence (AI) performance evaluation methods do not adequately represent the temporal, repetitive and context dependent nature of real-world TCCC delivery. Therefore, it is essential to conduct comprehensive assessments to ascertain that the AI tools function in a timely, synchronized manner within operational workflows. During the initial prototyping phase, five algorithm developers were provided with annotated datasets from seventy-five TCCC simulations and given six months to develop their algorithms. METHODS To assess the algorithms the research team leveraged a reserved dataset to perform evaluations. In the first phase of the assessment a standardized, repeatable performance methodology and framework was leveraged to evaluate individual algorithms that detect: (1) injury location on a casualty; (2) medical objects visible in the scene, and (3) treatments administered by the care provider. Detection effectiveness included four metrics: modified accuracy; precision; recall; and F1 scores. Algorithm processing efficiency was also evaluated by calculating lag time scores. A final composite score was used to quantify performance differences among the algorithms within a specific detection category. The second phase of the evaluation integrated multiple algorithms into a centralized orchestration framework to enable synchronized execution and consolidated outputs. System-level resource usage and throughput metrics were evaluated to characterize the computational efficiencies. Quantified memory consumption and central processing unit (CPU) and graphics processing unit (GPU) resource utilization were assessed, followed by benchmarking the edge compute orchestration framework utilization. RESULTS Results are presented for a representative algorithm in each category. Medical Object Detection achieved the highest performance (mean F1≈0.42, range 0–0.71). Injury Detection and Localization showed lower performance (mean F1≈0.27, range 0–0.60), with higher recall than precision. Medical Procedure Detection yielded procedure-level mean F1 scores from 0.00 to 0.31 and simulation-level means from 0.00 to 0.33. Stronger results were observed for Nasopharyngeal Airway (NPA) and Chest Seal Application (mean F1≈0.28, 0.31) medical procedures. The results to date are preliminary and serve as illustrative examples of the evaluation framework outputs. CONCLUSIONS The preliminary results highlight the evaluation framework’s end-to-end, standardized results across core algorithm functions. While the algorithm performance to date is modest, the framework demonstrates its capacity to capture both variability and recurring patterns across simulations, thereby highlighting strengths, limitations, and areas requiring refinement. It enables reproducible, cross data set comparisons, allowing evaluators to quantify algorithm performance. By leveraging both simulation-level evaluations and detection specific performance aggregated across simulations, the framework enables targeted identification of underperforming areas, supporting iterative and strategic AI model refinement.