Contact-Adaptive Robotic Ultrasound Probe Control for Tissue Exploration and Continuous Task-Relevant Visualization Using Robot-Free Image-Motion Demonstration

Seong Jeong; Minsung Kim; Dongho Yee; Juahn Oh; Yechan Seo; Jiyul Lee; Jinseok Lee; Seonho Shim

cs.ROSep 27, 2026

Contact-Adaptive Robotic Ultrasound Probe Control for Tissue Exploration and Continuous Task-Relevant Visualization Using Robot-Free Image-Motion Demonstration

Authors: Seong Jeong, Minsung Kim, Dongho Yee, Juahn Oh, Yechan Seo, Jiyul Lee, Jinseok Lee, Seonho Shim, +3 more

Organizations: Department of Transdisciplinary Medicine, Seoul National University Hospital · Rosota Inc., Seoul, Republic of Korea · Department of Medicine, Seoul National University College of Medicine · Department of Mechanical Engineering, Seoul National University · Department of Computer Science and Engineering, Seoul National University · Eulji University College of Medicine, Daejeon, Republic of Korea · Department of Mechanical Engineering, Chungang University, Seoul, Republic of Korea · Interdisciplinary Program in Medical Informatics, Seoul National University College of Medicine · Institute of Convergence Medicine with Innovative Technology, Seoul National University Hospital

Abstract

Robotic ultrasound commonly targets standardized views, predefined scanning protocols, or expert-scan reproduction. We target a different role: while a clinician performs the primary procedure, a robotic assistant maintains a clinician-selected view so that changing tissue remains observable. We present contact-adaptive robotic ultrasound monitoring derived from robot-free demonstrations. Sonologger clamps a six-axis inertial sensor to the clinician-operated probe and records synchronized ultrasound and probe-frame relative rotations without a robot, camera, or tracker. Rather than imitating trajectories, the recordings measure which probe axis and sign the clinician uses to correct an image offset. A measured image latency is removed before still-move-still segmentation yields relative-rotation labels; translation is excluded from control because of inertial drift. These measurements parameterize the Demonstration-Informed Image-Motion Controller (DIMC), an interpretable, non-learned controller combined with beam-axis force regulation and an independent safety supervisor with 5 N forced retreat. On a dynamic bladder phantom, DIMC acquired and held a task-relevant view in 7 of 7 blind, pose-paired trials versus 1 of 7 for a static hold (McNemar p = 0.031), and retained the view during syringe-driven hydro-distension while the supervisor bounded the force excursion. The framework offers an operational representation of ultrasound context for contact-adaptive intraoperative monitoring; clinical validation remains future work.

Figures & tables

Fig. 2: Sonologger: robot-free image–motion acquisition. (a) BNO085 on an nRF52840 board clamped to the C10UR probe. (b) Freehand acquisition on phantom A with a laptop; no robot, camera or tracker is present. (c) Probe frame in which every relative-rotation label is expressed: $x$ lateral, $y$ elevational (image-plane normal), $z$ beam; $\theta_{y}$ is the in-plane rotation. (d) One lateral-search episode: gyroscope magnitude with the still mask (grey), the sector image at each still and the relative rotation between stills. (e) Effective image latency: correlation of image change with the box-integrated gyroscope magnitude against lag, pooled over 226 sessions (thick) and for 40 sessions (thin); the peak is at $200\text{\,}\mathrm{ms}$ . (f) Corpus composition: 268 sessions, $45\,902$ frames, $1818$ move segments, $4389$ chunks. (g) Behavioral measurements from the demonstrations: only $\theta_{y}$ of ( 1 ) correlates with the lateral centroid offset before the move (Spearman $\rho=$-0.46$$ , $n=$3920$$ chunks), and the fraction of corrective moves rises from chance (dashed) inside the dead band to $92\text{\,}\mathrm{\%}$ at $|dx|\geq$0.10$$ ( $n=518$ , 525, 649 and 536 per bin). — Fig. 2: Sonologger: robot-free image–motion acquisition. (a) BNO085 on an nRF52840 board clamped to the C10UR probe. (b) Freehand acquisition on phantom A with a laptop; no robot, camera or tracker is present. (c) Probe frame in which every relative-rotation label is expressed: $x$ lateral, $y$ elevational (image-plane normal), $z$ beam; $\theta_{y}$ is the in-plane rotation. (d) One lateral-search episode: gyroscope magnitude with the still mask (grey), the sector image at each still and the relative rotation between stills. (e) Effective image latency: correlation of image change with the box-integrated gyroscope magnitude against lag, pooled over 226 sessions (thick) and for 40 sessions (thin); the peak is at $200\text{\,}\mathrm{ms}$ . (f) Corpus composition: 268 sessions, $45\,902$ frames, $1818$ move segments, $4389$ chunks. (g) Behavioral measurements from the demonstrations: only $\theta_{y}$ of ( 1 ) correlates with the lateral centroid offset before the move (Spearman $\rho=$ -0.46 $, $n=$3920$ chunks), and the fraction of corrective moves rises from chance (dashed) inside the dead band to $92\text{\,}\mathrm{\%}$ at $|dx|\geq$ 0.10$$ ( $n=518$ , 525, 649 and 536 per bin).

Quantity / configuration	RMS / 95th pct.	Drift	Role in DIMC
Six-axis fusion, no magnetometer	$2.52\text{\,}\mathrm{\SIUnitSymbolDegree}$ / $3.61\text{\,}\mathrm{\SIUnitSymbolDegree}$	$0.23\text{\,}\mathrm{\SIUnitSymbolDegree}\text{\,}{\mathrm{min}}^{-1}$	Used for short relative rotation labels
Nine-axis on-chip rotation vector	$12.55\text{\,}\mathrm{\SIUnitSymbolDegree}$ / $18.91\text{\,}\mathrm{\SIUnitSymbolDegree}$	$8.29\text{\,}\mathrm{\SIUnitSymbolDegree}\text{\,}{\mathrm{min}}^{-1}$	Rejected: magnetic distortion near the robot
IMU translation integration	$6.3\text{\,}\mathrm{mm}$ at $5\text{\,}\mathrm{s}$	accum.	Recorded; excluded from DIMC

Table 2

Fig. 3: From robot-free demonstrations to DIMC and the force and safety stack. Offline (dashed) the demonstrations supply the behavioral parameters of the Demonstration-Informed Image–Motion Controller (DIMC), which axis is corrected, in which sign and from what offset, plus the U-Net labels; a $60\text{\,}\mathrm{s}$ robot self-probe supplies the gain magnitude, the image responses, the frame map and the safe search ranges. Online (solid, $10\text{\,}\mathrm{Hz}$ ) the sector image is segmented, reduced to area, centroid and largest-component features, and turned into a twist by the Find or Hold mode of DIMC; the beam axis is owned by the admittance loop and the independent safety supervisor.

Fig. 4: Force regulation and safety supervision hardware. (a) FR5 arm, PX6D force/torque sensor, printed mount and a probe replica on the force bench, with the console and the $500\text{\,}\mathrm{ml}$ syringe. (b) The same stack with the C10UR probe on the gel phantom. (c) Probe frame ( $x$ lateral, $y$ elevational, $z$ beam) and the safety supervisor’s envelope on $\lVert\mathbf{F}\rVert$ : $3\text{\,}\mathrm{N}$ hold target, $4.5\text{\,}\mathrm{N}$ warning, $5\text{\,}\mathrm{N}$ forced retreat. (d) Compensated probe force against a laboratory scale over $0.92\text{\,}\mathrm{N}9.85\text{\,}\mathrm{N}$ ( $n=19$ ; slope $1.019$ , one invalid trial open; dashed: identity, solid: fit).

Fig. 5: Platforms for dynamic contact conditions. (a) Force bench: a torso mannequin with an inflatable tube under the abdominal pad, driven by a $500\text{\,}\mathrm{ml}$ syringe, and a 3D-scanned, printed replica of the clinical C10UR probe. (b) Dynamic bladder phantom: an agar-gelatin block with a double water-balloon lumen filled and emptied through a syringe line, scanned by the robot-held C10UR. (c) Contact-stiffness measurement of the gel with the same replica (inset). (d) Equilibrium force against indentation on the mannequin bench from the force holds, $k=$0.374\text{\,}\mathrm{N}\text{\,}{\mathrm{mm}}^{-1}$$ , $R^{2}=$0.997$$ . — Fig. 5: Platforms for dynamic contact conditions. (a) Force bench: a torso mannequin with an inflatable tube under the abdominal pad, driven by a $500\text{\,}\mathrm{ml}$ syringe, and a 3D-scanned, printed replica of the clinical C10UR probe. (b) Dynamic bladder phantom: an agar-gelatin block with a double water-balloon lumen filled and emptied through a syringe line, scanned by the robot-held C10UR. (c) Contact-stiffness measurement of the gel with the same replica (inset). (d) Equilibrium force against indentation on the mannequin bench from the force holds, $k=$ 0.374\text{,}\mathrm{N}\text{,}{\mathrm{mm}}^{-1} $, $R^{2}=$0.997$ .

Target [N]	SD [N]	Mean error [N]	In band [%]	Peak [N]
0.5	0.023	$-0.017$	88.9	0.60
1.0	0.022	$-0.013$	83.9	1.10
1.5	0.022	$-0.012$	94.1	1.57
2.0	0.026	$-0.016$	89.9	2.09
2.5	0.029	$-0.031$	72.9	2.59
3.0	0.026	$-0.015$	89.7	3.10

Table 6

Fig. 6: Force regulation and safety response during syringe-induced surface displacement. Force holds, loop-on/off injection responses, and the $4.0\text{\,}\mathrm{N}$ safety-retreat run are shown.

Fig. 7: Blind task-relevant view acquisition and maintenance. Seven DIMC episodes and static-hold outcomes; DIMC maintained the view in 7/7 poses versus 1/7 for static hold.

Pose	DIMC [s]	Static hold [s]
1	50.7	no
2	10.9	35.4
3	13.0	no
4	7.8	no
5	25.4	no
6	16.4	no

Table 9

Fig. 8: Hydro-distension view maintenance. DIMC retains the target through the force peak; static hold lacks image-centering adaptation.

Explore similar work

Sep 28, 2026

Contact-Aware Impedance Controller for Robot-Assisted Ultrasound Imaging

Safe robot-assisted ultrasound imaging requires a reliable controller able to detect and localize probe--tissue interaction. In this paper, we present a B-mode ultrasound image-based contact perception method and a contact-aware impedance controller for robotic ultrasound imaging. The proposed method detects acoustic contact independently of force measurements, enabling contact-conditioned force/torque taring to reduce residual wrench bias. During contact, the method continuously estimates the effective contact location along the curved probe surface and uses it to update the controller interaction frame, enabling visual servoing of the physical probe--tissue contact point during imaging. Experiments on an agar phantom demonstrated a contact-localization RMSE of

\mathbf{1.46 \pm 0.14}

~mm over probe roll angles from

\mathbf{-15^\circ}

\mathbf{15^\circ}

. During static rolling, the proposed controller maintained task-space tracking accuracy comparable to a conventional fixed-frame impedance controller while reducing the maximum compressive interaction force from

\mathbf{31.56}

~N to

\mathbf{20.09}

~N, corresponding to a

\mathbf{36.3\%}

reduction. These results demonstrate the potential of ultrasound images as direct contact feedback for safe and accurate robot-assisted ultrasound imaging.

MD Miraj Arefin, M Efe Tiryaki

Center for Robotics and AI (ROMER), Middle East Technical University, Ankara, Turkey

Impedance Control Ultrasound Imaging Force-Controlled Robotic Manipulation Robotic Ultrasound Vision-Based Robot Control

Ask PDF

Aug 11, 2026

From Self-Normal-Positioning to Omni-Directional Tracking: Real-Time Surface Modeling Enabled Probe Tilt Control for Robotic Ultrasound Imaging

Ultrasound (US) provides real-time, radiation-free imaging, but the image quality depends strongly on how the probe is oriented against the patient body. Robotic US can reduce operator workload and improve acquisition consistency; however, most existing systems focus on normal positioning, where the probe is maintained perpendicular to the local surface. This constraint is inadequate for examinations like echocardiography, where obtaining a diagnostic view requires a non-normal probe angle. Consequently, a clinically useful robotic system must sense the local surface in real-time and preserve the desired probe orientation. Here, we propose an omni-directional probe-orientation control framework that integrates RGB-D perception, local-surface modeling, and task-space orientation control. The surface model fuses multi-view point clouds and provides a quadratic estimate of the local surface. A desired imaging direction is then encoded relative to the normal, enabling the probe to track arbitrary angles. The framework was evaluated through flat-surface tracking, phantom target-angle recovery, and in-vivo tracking of an expert selected view. Results show that the mean angular tracking error was 1.06 +- 0.66 deg. The system recovered a non-normal tilt angle of up to 44.39 +- 2.59 deg relative to the surface normal, and acquired the desired heart chamber view in the phantom and in-vivo experiments.

Xihan Ma, Haichong Zhang

Department of Robotics Engineering, Worcester Polytechnic Institute, 100 Institute Road, Worcester, MA 01609 USA

Robotic Control Ultrasound Imaging Robotic Ultrasound Vision-Based Robot Control

Ask PDF

Oct 5, 2026

Dual-Rate Force-Image Control with Model-Based Orientation Limits for Robotic Ultrasound

Robotic ultrasound couples a high-rate contact-force loop with slower, delayed image feedback, so image-guided ultrasound probe rotation can perturb contact force before the resulting image response is observed. We derive a closed-form orientation-rate limit that bounds the modeled rotation-induced estimated-force excursion over a finite horizon while accounting for disturbance rejection by the fast force loop. The limit depends on local contact stiffness, force-loop gains, a conservative rotation-to-force gain bound, the excursion budget, and the prediction horizon. We implement this model in a dual-rate controller with timestamp-based delay reconstruction and joint-torque-based force estimation, and evaluate it on a curved gelatin phantom using paired controller comparisons and component ablations. Relative to unconstrained image guidance, the proposed rate-limited controller reduced first-second root-mean-square (RMS) estimated-force error by 0.40 N while increasing cue-convergence time by 0.94 s. A fixed rate cap near the analytically predicted ceiling produced no resolvable difference in force error and converged 0.32 s faster, indicating that the principal practical value of the model is the rate-design rule rather than online prediction. Delay reconstruction had no resolvable effect at the tested latency. A single-subject popliteal scan demonstrated feasibility, although the image cue was noise-limited on heterogeneous tissue.

Tyler Foster, Qiang Zhang, A B M Tahidul Haque +1

Department of Mechanical Engineering, The University of Alabama, Tuscaloosa, AL 35487, USA · Department of Chemical & Biological Engineering, The University of Alabama, Tuscaloosa, AL 35487, USA · Department of Computer Science, The University of Alabama, Tuscaloosa, AL 35487, USA

Robotic Control Robotic Ultrasound Feedback Control

Ask PDF