"The future of surgery is not about blood and guts; the future of surgery is about bits and bytes.”
/Dr. Richard Satava/

Friday, November 20, 2015

Updates on Frauhofer's new systems

As we reported before, Fraunhofer announced some great new systems (new catheter robot and robot assistant). Here are the updates after their introduction at Medica:
  • Further developments of needle-based therapies
  • Human-machines-team in the operating room:
  • Endoscopic biopsy and inline cell processing:
  • Novel solutions for human-machine interactions:

Wednesday, November 18, 2015

Video Wednesday

A good old concept video from the early 1990s, how DARPA imagined the outcome of the TraumaPod project.

Monday, November 16, 2015

Cal-MR: Center for Automation and Learning for Medical Robotics

Time to introduce another great lab with visionary projects. See also our earlier post.
"UC Berkeley's Center for Automation and Learning for Medical Robotics (Cal-MR) advances research in automation and machine learning to improve robots and devices for medicine and surgery.  Cal-MR is co-directed by Profs. Pieter Abbeel (EECS) and Ken Goldberg (IEOR, EECS, School of Information, Art Practice at UC Berkeley, and Department of Radiation Oncology at UCSF) and Researcher Dr. Sachin Patil. Cal-MR, established in Nov 2014, is based in the College of Engineering at UC Berkeley and is funded by a major grant from the National Science Foundation (NSF), from companies including Intuitive Surgical and Google, and from private donors. Cal-MR supports research with postdocs, grads, and undergraduate students and collaborations with leading physicians and researchers at UC San Francisco, UC Davis, UC Santa Cruz, University of Washington, Johns Hopkins, and Stanford.

Robot-Assisted Surgery  
We are developing algorithms and control methods to enable the automation of surgical subtasks such as tissue debridement, cutting, and suturing. This project integrates automatic planning algorithms, learning skills from demonstrations, robot control, and human in the loop guidance. This project has the potential to improve patient health by enhancing surgeon performance, reduce tedium and thus medical errors, and reduce costs by reducing operation time. The objective is to improve speed, accuracy, and precision of existing procedures and enable new classes of surgical procedures that require dexterity and control beyond the capability of a human operator. We are evaluating our techniques on a da Vinci robot (dVRK) donated by Intuitive Surgical. Innovations in Radiation Oncology  Brachytherapy is a widely-used treatment modality for fighting cancer in many parts of the body in which small radioactive sources are positioned proximal to cancerous tumors. Working with the Radiation Oncology team at UCSF, we have developed a series of dose planning algorithms and a novel method for robot-assisted insertion of “skew-line” needles. In current practice for intracavitary brachytherapy, standardized applicators with internal channels are inserted into body cavities to guide the sources. These standardized implants are prone to shifting inside the body, resulting in suboptimal dosages that might cause radiation damage to healthy organs and tissues. We are developing 3D printed customized implants (patent pending) containing customized curvature-constrained internal channels that fit securely, minimize air gaps, and precisely guide radioactive sources through printed channels and aim to significantly improve dose coverage and treatment options. Steerable Needles  This project encompasses a decade of pioneering research into needles that can be steered from outside the body to reach internal targets for biopsy or to deliver treatment. This work integrates adaptive modeling and planning algorithms, and real-time image-guided intraoperative control of thin, flexible needles that move through deformable tissues around critical anatomical structures to reach specified targets. This project was funded under funded under a NIH R01 EB006435 grant and the steerable needle technology has been patented (US patent no. 7,822,458) and licensed. 

Planned Projects  
Cloud-Based Surgical Video Analysis for Evaluation and Training: The rising popularity of the use of minimally invasive surgical robots such as the Intuitive Surgical da Vinci robot in operating rooms (over 2000 such systems deployed worldwide) has created new opportunities to record videos of the surgical workspace and motions performed by the surgeon during surgery. This project will explore how large numbers of recordings can be analyzed in the Cloud for evaluating and training surgeons, producing quantitative measures of surgical proficiency, and for automatically annotating surgical recordings for training purposes. [In collaboration with Johns Hopkins University and Stanford University].  
Cloud-Based Collaborative Robot Learning of Surgical Subtasks: The Cloud has the potential to facilitate collective learning across all such systems deployed worldwide in terms of sharing trajectories, control policies, and task execution outcomes. This project will explore how to leverage recorded data (vision, motion data) for learning how to better perform surgical subtasks such as tissue debridement, cutting, and suturing. In preliminary work, we have developed an approach that is able to perform such generalizations using non-rigid registration to find detailed correspondences between the demonstration environment and the new environment. This project will investigate how to efficiently index into this data to find demonstrations that best match the current situation using computational and data storage resources available in the Cloud.  
Human-in-the-Loop Robot Guidance: This project will investigate strategies for using expert human guidance and skills for annotating scenes and providing a high-level task description for surgical subtasks, analyzing, annotating and classifying video and motion data for learning surgical subtasks, and for facilitating error recovery during task execution. [In collaboration with Stanford University, John Hopkins University and University of Washington]."

Press coverage: New York Times, Medgadget


Sunday, November 15, 2015

Quiz - contd

We still have not received a correct answer, qualifying someone for the valuable prizes. Here is another clue for the system presented in the original Quiz #17 post.

Friday, November 13, 2015

Systematic Design of Medical Capsule Robots

The Science and Technology of Robotics in Medicine (STORM) group at Vanderbilt decided that its medical capsule robots’ hardware, software may become open-source. In their recent paper in IEEE Design & Test, they described Vanderbilt’s capsule robots that could be made small enough eventually to be swallowed, could be used for preventative screenings and to diagnose and treat a number of internal diseases.

"Now research groups with hypotheses about how to use the capsules won’t have to redesign boards and interfaces from scratch, which means they can get to the prototyping stage faster.  Medical capsule robots differ from the PillCam, put on the market in 2001, because they can be manipulated to perform internal tasks rather than just passing through the body and recording video.  The paper explains the hardware modules available, which handle computation, wireless communication, power, sensing and actuation. Each is designed to interface easily with new modules contributed from other research groups.  On the software side, Vanderbilt engineers used TinyOS — a free, open-source, flexible operating system — to develop reusable components.  Ledeczi, senior research scientist at Vanderbilt’s Institute for Software Integrated Systems, said a medical capsule robot is the ideal example of a cyber-physical system.  
It must work inside the challenging physical environment of the human body, sense its environment and move through it effectively, and then complete tasks such as release a drug, take a tissue sample or deploy a clamp. Finally, it must constantly communicate with a base station through the entire process.  “Our focus is the design environment, not the software per se, with the goal of easing the learning curve for new researchers and engineers who start in this field,” Ledeczi said. “Designing a capsule from scratch requires deep hardware, software and domain expertise.”  By providing a hardware and software component library and the tools to make their composition easy, Vanderbilt opens up the field of medical capsule robots to engineers and scientists who have great ideas but aren’t hardware or software experts, plus makes development costs far more affordable, he said."

Source: Vanderbilt