Robotic Tissue Sampling for Safe Post-Mortem Biopsy in Infectious Corpses

Abstract

In pathology and legal medicine, the histopathological and microbiological analysis of tissue samples from infected deceased is a valuable information for developing treatment strategies during a pandemic such as COVID-19. However, a conventional autopsy carries the risk of disease transmission and may be rejected by relatives. We propose minimally invasive biopsy with robot assistance under CT guidance to minimize the risk of disease transmission during tissue sampling and to improve accuracy. A flexible robotic system for biopsy sampling is presented, which is applied to human corpses placed inside protective body bags. An automatic planning and decision system estimates optimal insertion point. Heat maps projected onto the segmented skin visualize the distance and angle of insertions and estimate the minimum cost of a puncture while avoiding bone collisions. Further, we test multiple insertion paths concerning feasibility and collisions. A custom end effector is designed for inserting needles and extracting tissue samples under robotic guidance. Our robotic post-mortem biopsy (RPMB) system is evaluated in a study during the COVID-19 pandemic on 20 corpses and 10 tissue targets, 5 of them being infected with SARS-CoV-2. The mean planning time including robot path planning is 5.72±167s. Mean needle placement accuracy is 7.19± 422mm.

Keywords: COVID-19; Collaborative robot; forensic medicine; medical robotics; path planning.

Fig. 1.

Minimally invasive tissue sampling with a robot. The robot drives a biopsy needle into the corpse to the desired target. The physician can then take a biopsy sample while the robot holds the needle with our specially designed needle holder. A protective body bag reduces the risk of disease transmission.

Fig. 2.

Steps to visualize feasible robot insertion positions. Illustration of iterative overlay generation. Every arrow represents an operation applied to the previous overlay. Skin is segmented, maximum CT density is estimated, margin is added, distance and insertion angle are estimated, and reachability is verified. Skin colored points are outside of needle range, dark red points are occluded by dense tissue, grey points cannot be reached, and blue points have low objective value.

Fig. 3.

Best path visualization module in 3D Slicer. The module segments the skin of the corpse and projects colormaps for various overlay types, e.g., bone density or distance to target onto the skin. The colors in this example indicate the maximum CT values along the needle insertion path. The module and its code can be downloaded from: www.github.com/StefanTUHH/robotic_needle_insertion.

Fig. 4.

Communication between workstations. Overview of communication and module setup. Three different application bundles are running (CT acquisition, 3D Slicer, MoveIt ecosystem. Note that applications can run on separate computers and multiple (N) instances of the MoveIt ecosystem can be executed to parallelize reachability evaluation—even across multiple computers.

Fig. 5.

RPMB System Setup. Left: To prevent disease contamination, we place infectious corpses in body bags. A robot (LBR Med 14, KUKA) inserts biopsy needles into the corpse through the body bag. Top right: [A] Custom robot end effector for mounting a hollow guide needle in which we insert a co-axial needle. [B] A screw prevents slipping along the needle axis of the co-axial needle during insertions and two additional screws (not visible) on the side tighten the clamping jaws. [C] CAD model with reflective tracking markers (orange) and inserted biopsy needle. Bottom right: Registration marker for estimating the position of the CT coordinate system relative to the robot’s world coordinate system. The marker orientation can be tracked with [D] the tracking camera and [E] in the CT imaging system with a steel ball checkerboard, [F] the inner transform can be estimated from the CAD model.

Fig. 6.

Biopsy workflow. The workflow for robotic biopsy sampling with the RPMB system. Note that system setup and calibration only has to be done once and can be reused after repositioning a corpse and for consecutive corpses.

Fig. 7.

Relevant transformations for calibration and registration. Fixed and variable transformations estimated for calibration and registration. Note the protective body bag which covers the patient to reduce the risk of disease transmission.

Fig. 8.

Mean computation time. Mean time for each segment of the algorithm for estimating an optimal insertion point. We subdivide possible insertion points in a grid of 30x30mm and check robot reachability for each grid point.

Fig. 9.

Feasible surface area for insertions. Boxplots for all patients per target are shown and sorted by mean area of manual insertion (a). In blue the surface area without reachability check is shown, i.e., the surface area for feasible needle insertion by hand. In red the surface area for feasible robotic needle insertion is shown. The distribution of the objective value at feasible insertion points for all patients and targets is shown on the right (b). Here, the relative objective value, normalized between extrema for the particular patient and target is shown.

Fig. 10.

Needle insertion colormaps. Example of colormaps showing the feasible insertion points and insertion point quality for case 18. Grey points are not reachable by the robot, red points are obstructed by dense structures, blue points represent favourable insertion points.

Fig. 11.

Planning vs. Evaluation CT. Large tissue deformation as a result of a pneumothorax (red dashed line) that causes the anatomy to no longer correlate to the annotated targets.

Fig. 12.

Needle placement accuracy. Box plots of the absolute error from the annotated target to the center of the sampled biopsy. The 3D deviation is shown in blue and the lateral error normal to the needle axis is displayed in red.

Fig. 13.

Histological examination. For each needle insertion, the acquired sample is characterized and compared to the desired target tissue in a histological examination (a). The employed biopsy sampling needle and biopsy samples are shown in (b).