A Clinical Dataset for the Evaluation of Motion Planners in Medical Applications Inbar Fried12 Jason A. Akulian3 and Ron Alterovitz1

2025-04-30 2 0 2.9MB 4 页 10玖币

侵权投诉

A Clinical Dataset for the Evaluation of Motion Planners in Medical

Applications

Inbar Fried1,2, Jason A. Akulian3, and Ron Alterovitz1

Abstract— The prospect of using autonomous robots to en-

hance the capabilities of physicians and enable novel procedures

has led to considerable efforts in developing medical robots and

incorporating autonomous capabilities. Motion planning is a

core component for any such system working in an environment

that demands near perfect levels of safety, reliability, and

precision. Despite the extensive and promising work that has

gone into developing motion planners for medical robots, a

standardized and clinically-meaningful way to compare existing

algorithms and evaluate novel planners and robots is not

well established. We present the Medical Motion Planning

Dataset (Med-MPD), a publicly-available dataset of real clinical

scenarios in various organs for the purpose of evaluating

motion planners for minimally-invasive medical robots. Our

goal is that this dataset serve as a ﬁrst step towards creating a

larger robust medical motion planning benchmark framework,

advance research into medical motion planners, and lift some

of the burden of generating medical evaluation data.

I. INTRODUCTION

Automation of medical robots for clinical procedures or

subtasks is increasingly being shown to be feasible. Achiev-

ing autonomy in interventional medical procedures has a lot

of potential beneﬁts for patient care and hospital efﬁciency.

Much like teleoperated medical robots, such as the da Vinci

(Intuitive Surgical Inc., Sunnyvale, CA), can compensate for

physician fatigue and hand instability, autonomous medical

robotics can further improve and standardize patient care

by accounting for inter- and intra-physician variability while

also focusing the physician’s time on sub-tasks that require

their expertise. However, beyond the technical challenges

that exist in hardware and software, a critical, if not the most

important, challenge in these systems is making them safe

and reliable. To address these challenges and still beneﬁt

from the advantages of automation, integrating motion plan-

ning into medical robots to ensure safe motions is essential.

This research was supported by the U.S. National Institutes of Health

(NIH) under awards R01EB024864 and F30CA265234, and by the National

Science Foundation (NSF) under awards 2008475 and 2038855.

The authors acknowledge the National Cancer Institute and the Foun-

dation for the National Institutes of Health, and their critical role in the

creation of the free publicly available LIDC/IDRI Database used in this

study.

The MR brain images from healthy volunteers used in this paper were

collected and made available by the CASILab at The University of North

Carolina at Chapel Hill and were distributed by the MIDAS Data Server at

Kitware, Inc.

1I. Fried and R. Alterovitz are with the Department of Computer Science,

University of North Carolina at Chapel Hill, Chapel Hill, NC 27599, USA.

{ifried01, ron}@cs.unc.edu

2I. Fried is also with the Medical Scientist Training Program, University

of North Carolina School of Medicine, Chapel Hill, NC, 27599, USA.

3J. A. Akulian is with the Division of Pulmonary Diseases and Critical

Care Medicine at the University of North Carolina at Chapel Hill, NC 27599,

USA. jason akulian@med.unc.edu

One class of medical robots that has been studied ex-

tensively over the past couple decades has been medical

continuum robots, which include, for example, concentric

tube robots and steerable needles [1]. Many mechanical

designs have been proposed for these devices, but at their

core, medical continuum robots can follow curvilinear tra-

jectories in 3D, allowing them to curve around obstacles and

access regions of the anatomy that are otherwise inaccessible

when using straight rigid tools. The potential beneﬁt of

these devices has been proposed in numerous organs and for

various medical procedures. The complex kinematics of these

devices in conjunction with the precision required for safe

medical procedures make manual operation of these devices

unintuitive and impractical. To overcome this challenge,

autonomous robots have been proposed that actuate the

medical continuum robot following a planned trajectory.

Despite the numerous motion planners that have been

proposed for medical continuum robots, to the best of our

knowledge, a benchmarking dataset to evaluate the perfor-

mance of these algorithms does not exist. The lack of a

shared benchmarking resource has lead each research group

to generate their own testing data, which is often a time

intensive effort. Since the motion planners have been tested

in various organs, in different anatomical models of those

organs, and likely with different obstacle resolutions, it is

difﬁcult to properly assess the beneﬁts and drawbacks of

each proposed motion planning approach and to compare

motion planners. To help evaluate the beneﬁts of robot

automation in medicine, it is important to have benchmarks

that can robustly and equitably evaluate the performance of

algorithms in clinically relevant scenarios.

In this work, we propose Med-MPD, a medical bench-

marking dataset consisting of real clinical motion planning

environments for assessing motion planners for medical

continuum robots and related minimally-invasive medical

robots. The data includes benchmark scenarios deﬁned by the

relevant anatomy and the clinical problem in the lungs, liver,

and brain. We make Med-MPD publicly available at https:

//github.com/UNC-Robotics/Med-MPD.

II. RELATED WORK

There are several robotics datasets and benchmarking

suites that have been developed speciﬁcally to allow robust

evaluation of motion planners [2], [3], [4], [5] (and citations

within). These works focus on non-medical robots. There

have also been several medical robotics datasets that have

been published, but these are mostly focused on computer

vision problems like tool or anatomy segmentation, physician

arXiv:2210.10834v1 [cs.RO] 19 Oct 2022

文档加载中……请稍候！
如果长时间未打开，您也可以点击刷新试试。

下载文档到电脑，查找使用更方便

10 玖币 0人已下载

立即下载

摘要：

AClinicalDatasetfortheEvaluationofMotionPlannersinMedicalApplicationsInbarFried1;2,JasonA.Akulian3,andRonAlterovitz1AbstractTheprospectofusingautonomousrobotstoen-hancethecapabilitiesofphysiciansandenablenovelprocedureshasledtoconsiderableeffortsindevelopingmedicalrobotsandincorporatingautonomousca...

展开>> 收起<<

A Clinical Dataset for the Evaluation of Motion Planners in Medical Applications Inbar Fried12 Jason A. Akulian3 and Ron Alterovitz1.pdf

共4页,预览1页

还剩页未读，继续阅读

声明：本站为文档C2C交易模式，即用户上传的文档直接被用户下载，本站只是中间服务平台，本站所有文档下载所得的收益归上传人(含作者)所有。玖贝云文库仅提供信息存储空间，仅对用户上传内容的表现方式做保护处理，对上载内容本身不做任何修改或编辑。若文档所含内容侵犯了您的版权或隐私，请立即通知玖贝云文库，我们立即给予删除！

A Clinical Dataset for the Evaluation of Motion Planners in Medical Applications Inbar Fried12 Jason A. Akulian3 and Ron Alterovitz1

相关推荐

开通VIP享超值会员特权

作者详情

相关内容

热门标签

举报选择: