SenSE: Smart Grasper for Intelligent Surgical Manipulation

SenSE：用于智能手术操作的智能抓取器

• 批准号: 2036255
• 负责人: Blake Hannaford
• 金额: $ 74.02万
• 依托单位: University of Washington
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2020
• 资助国家: 美国
• 起止时间: 2020-10-01 至 2024-06-30
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=2036255&HistoricalAwards=false
• 关键词: SenSE; Smart; Grasper; Intelligent; Surgical

This project will develop a novel innovative smart surgical instrument as a multimodal sensing system to advance surgical health care. Modern surgical paradigms, including minimally invasive laparoscopic surgery and robotic assisted surgery, have deprived surgeons of important tactile information that this project aims to partially restore and expand into new dimensions. This project builds a tiny multimodal system that includes optical, electrical, thermal, mechanical and ultrasound sensors that is affixed to a basic surgical device such as a grasper or scalpel (either hand-held or used by an advanced surgical robot). This will turn the device into a diagnostic instrument able to measure blood oxygenation (SpO2), physical damage, and potentially other tissue states like cancer. Advances in machine intelligence are required to process this raw, multidimensional, data into information that is clinically useful to the surgeon. This novel system will promote safer and more effective manipulation of tissues in complex surgical situations protecting delicate tissues easily damaged unintentionally or tissues that are hard to visually identify. The project will support graduate and undergraduate students, including from underrepresented groups, in their STEM studies, and support participation by the PIs and students in outreach events designed to promote and broaden participation in STEM educational programs. This project, proposed by a multidisciplinary team, aims to advance health discovery through combining advanced machine learning with a novel multimodal sensing system aimed at improved surgical outcomes. This project will obtain the first ever simultaneous in-vivo measurements of pulse oximetry (blood oxygen saturation, SO2), temperature, electrical impedance (1-100kHz), pressure, and acoustic properties at multiple sites in multiple functioning organs. It will develop, test, and demonstrate a novel, highly integrated, multimodal sensing surgical grasper suitable for endoscopic use. The instrument will generate a continuous stream of a new type of sensed data specific to surgical work with tissues. This project will generate the first such database of in-vitro and in-vivo data along with ground truth from surgical videos, imaging studies, expert evaluations, and post-operative pathological analysis of tissues. The new data set will be shared openly online via an achieved open repository. The new classification and clustering machine learning systems will correlate measured tissue properties with organ anatomy and physiology, especially at interfaces between tissues, aiding in surgical planning and execution, and potentially supporting new modes of clinical care.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

该项目将开发一种新型的创新智能手术器械，作为多模式传感系统，以推进外科医疗保健。现代手术范例，包括微创腹腔镜手术和机器人辅助手术，剥夺了外科医生的重要触觉信息，本项目旨在部分恢复和扩展到新的层面。该项目构建了一个微型多模式系统，其中包括光学、电气、热、机械和超声传感器，该系统固定在基本手术设备上，例如抓钳或手术刀（手持式或由高级手术机器人使用）。这将使该设备成为一种诊断仪器，能够测量血氧饱和度（SpO 2），物理损伤和潜在的其他组织状态，如癌症。机器智能的进步需要将这些原始的多维数据处理成对外科医生临床有用的信息。这种新型系统将促进在复杂的手术情况下更安全和更有效地操纵组织，保护容易意外损坏的脆弱组织或难以视觉识别的组织。该项目将支持研究生和本科生，包括代表性不足的群体，在他们的STEM研究，并支持PI和学生参与旨在促进和扩大STEM教育计划参与的外联活动。该项目由一个多学科团队提出，旨在通过将先进的机器学习与旨在改善手术结果的新型多模态传感系统相结合来推进健康发现。该项目将首次在多个功能器官的多个部位同时进行脉搏血氧饱和度（血氧饱和度，SO2）、温度、电阻抗（1- 100 kHz）、压力和声学特性的体内测量。它将开发、测试和展示一种适用于内窥镜使用的新型、高度集成、多模式传感手术抓钳。该仪器将产生一个连续的数据流的一种新的类型的传感器特定的手术工作与组织。该项目将生成第一个这样的体外和体内数据数据库，其中沿着有来自手术视频、成像研究、专家评估和组织的术后病理分析的基本事实。新的数据集将通过一个已实现的开放储存库在网上公开分享。新的分类和聚类机器学习系统将测量的组织特性与器官解剖学和生理学相关联，特别是在组织之间的界面，有助于手术计划和执行，并可能支持新的临床护理模式。该奖项反映了NSF的法定使命，并通过使用基金会的知识价值和更广泛的影响审查标准进行评估，被认为值得支持。

项目成果

Comparison of Deep Learning and Signal Processing Methods for Removing a Ringing Artifact from Ultrasound Signals

用于去除超声信号中振铃伪影的深度学习和信号处理方法的比较

• DOI: 10.1109/sas58821.2023.10254036
• 发表时间: 2023
• 期刊: IEEE
• 作者: Sosnovskaya, Yana;Shlizerman, Eli;Hannaford, Blake;Sinanan, Mika N.
• 通讯作者: Sinanan, Mika N.

Ultrasound Waveforms with and without Ringdown Artifacts

有或没有振铃伪影的超声波形

• DOI: 10.21227/z6v5-mf23
• 发表时间: 2022
• 期刊: IEEE DataPort
• 作者: Sosnovskaya, Yana Sosnovskaya

AL-SAR: Active Learning for Skeleton-Based Action Recognition

AL-SAR：基于骨架的动作识别的主动学习

• DOI: 10.1109/tnnls.2023.3297853
• 发表时间: 2023
• 期刊: IEEE Transactions on Neural Networks and Learning Systems
• 影响因子: 10.4
• 作者: Li, Jingyuan;Le, Trung;Shlizerman, Eli
• 通讯作者: Shlizerman, Eli

Sensor Fusion for Force and Position Calibration of a Motorized Surgical Smart Grasper

用于电动手术智能抓取器的力和位置校准的传感器融合

• DOI: 10.1109/ismr48346.2021.9661520
• 发表时间: 2021
• 期刊: International Symposium on Medical Robotics
• 作者: Kaplan, Jack;Sosnovskaya, Yana;Arnold, Matthew;Hannaford, Blake
• 通讯作者: Hannaford, Blake