CIMIT blog

Turning Thought Into Action: Brain Body Interfaces

Brain-Machine Interface or Brain-Body Interface? How to Best Exploit the Remarkable Adaptive Competence of the Brain

Robert Ajemian, PhD, Postdoctoral Research Fellow, Laboratory of Emilio Bizzi, McGovern Institute for Brain Research, Massachusetts Institute of Technology, ajemian@mit.edu

Jonathan Winograd, MD, Assistant Professor of Surgery, Harvard Medical School, Plastic and Reconstructive Surgery, Massachusetts General Hospital, jwinograd@partners.org  

 Moderator: Steven C. Schachter, MD, Program Leader, CIMIT Neurotechnology Program and CIMIT Site Miner, Beth Israel Deaconess Medical Center (BIDMC); Professor of Neurology, Harvard Medical School, Director of Research, Department of Neurology, BIDMC; Associate Director, Clinical Research, Harvard Medical School Osher Institute; Member, Board of the Epilepsy Therapy Development Project; Founder & Editor-in–Chief, Epilepsy & Behavior; Editor-in-Chief of Epilepsy.com, sschacht@bidmc.harvard.edu

One can imagine engineering two distinct types of brain interfaces:

1. Brain-Machine Interfaces   -- Brain signals are used to control an artificial device such as a   cursor on a computer screen or a robotic arm. 
2. Brain-Body Interfaces   -- The brain signals are used to innervate an individual’s own muscles   through an unnatural pathway, e.g., the brain signals control the activation   of input to Functional Electrical Stimulating (FES) electrodes implanted   in muscles.  In this way, the muscles are still the final common pathway   of actuation, but the usual route of innervation through the spinal   cord is bypassed.

Basic research efforts have focused on brain-machine interfacing, while brain-body interfacing has been largely ignored. This state of affairs seems somewhat counter-intuitive: would it not be potentially advantageous to interface the motor cortex with its original end-organ, muscle, rather than with a machine, since brain circuits have become calibrated for controlling the natural end-effectors of the body? Drs. Winograd and Ajemian will explore the technical problems and conceptual misunderstandings that have steered the field away from brain-body interfacing and towards brain-machine interfacing.  They will show how a fully autonomous brain-body interface can be constructed, indicate how such an interface can lead to performance gains in terms of superior generalization capacity and fully autonomous learning (i.e., no need for the always slippery “neural decoder”) and suggest how this technology could be incorporated into therapeutic devices built for humans.

Mechanical Cardiac Assist/Total Artificial Heart: Present and Future

Moderator: Frederick J. Schoen, MD, PhD, Professor of Pathology and Health Sciences and Technology, Harvard Medical School; Director of Cardiac Pathology and Executive Vice-Chairman, Department of Pathology, Brigham and Women's Hospital; Director, BWH Biomedical Research Institute (BRI) Technology in Medicine Initiative; CIMIT Site Miner, BWH, fschoen@partners.org

Challenges and Available Therapies for Heart Failure
Kimberly Parks, DO, FACC, Clinical and Research Fellow, Advanced Heart Failure and Cardiac Transplantation, Division of Cardiology, Harvard Medical School, Massachusetts General Hospital, kaparks@partners.org
Our understanding of the pathophysiologic process in the failing heart has greatly evolved over the past decade and has broadened the armamentarium of technology available. Cardiac assist devices have become a growing interest in management of symptoms of heart failure and may even provide a “cure” for this deadly disease.  This discussion will highlight the magnitude of the problem that heart failure creates, outline current advanced therapies available and illuminate future pathways of investigation.

Total Artificial Heart: Translation to Clinical Use, FDA Approval and Reimbursement
Marvin J. Slepian, MD, Director, Interventional Cardiology and Director, Tissue Engineering Laboratory, The University of Arizona Sarver Heart Center; Professor of Medicine (Cardiology), The University of Arizona in Tucson; Chairman and Chief Medical and Scientific Officer, SynCardia Systems Inc., slepian@email.arizona.edu
Congestive Heart Failure (CHF) is the final common pathway of cardiac pump function decline associated with all forms of heart disease. Despite major advances in pharmacologic and pacemaker therapy for CHF, unfortunately this disease syndrome is on the rise. It is anticipated that over the next two decades there will be more than 10 million patients with this condition in the United States alone. While current therapies to some degree reduce mortality, they “shift to the right” the progressive decline in pump function which ultimately fails, leading to a slippery slope of progressive decline and death of the patient. For these advanced CHF patients what they require is a pump! Towards this end the field of mechanical circulatory support has emerged. This presentation will review the basics of mechanical circulatory support with emphasis on the Total Artificial Heart, mechanisms of pump function and elements of design. Device selection for differing patient scenarios will be reviewed and clinical experience and outcomes with the TAH will be discussed. Advances in TAH driver technology, allowing for increased patient mobility and hospital discharge will be presented. Participants should gain a better understanding of the emerging and growing role for the TAH as a vital therapeutic tool the treatment of patients with advanced CHF.

Device Failure: What are the Issues?
Robert Padera, Jr., MD, PhD, Attending Pathologist, Brigham and Women’s Hospital; Assistant Professor of Pathology, Harvard Medical School; affiliated faculty within HST, rpadera@partners.org
Pathologic evaluation of medical devices at the time of explant or autopsy provides important information about mechanisms of failure. This information can be used to influence ongoing device development and clinical use. Mechanical circulatory support devices, including ventricular assist devices and total artificial hearts, and their interactions with the human body are associated with several pathologies including infection, thrombosis and thromboembolism, hemorrhage, mechanical failure and other local complications. These pathologies will be illustrated in this presentation, and highlighted as ongoing challenges for further device development.

Bio-inspired Design and Fluidic Logic

Moderator: Alex Slocum, PhD, Pappalardo Professor of Mechanical Engineering MacVicar Faculty Fellow, MIT, slocum@mit.edu

 

Bio-Inspired Robot Design with Compliant Mechanism Fabrication
Sangbae Kim, PhD, Post-doctorate fellow, Micro-robotics Laboratory, Harvard University, sangbae@mit.edu

Mobile robot designers are increasingly searching for inspiration and design cues from biological models. Biomechanical studies on running animals underscore the importance of the passive properties of muscles, tendons, and other elements of the musculoskeletal system. These elements play significant roles in self-stabilization and elastic energy storage, resulting in smoother and more efficient performance in natural environments. Through abstraction and simplification of biological inspiration, fundamental design principles are embodied in a number of bio-inspired robots. To demonstrate this process, the core design features of three legged robots are described, focusing on compliant under-actuated mechanism made by multi-material manufacturing process called shape deposition manufacturing. The first bio-inspired robot is iSprawl, a cockroach-inspired hexapod with compliant, under-actuated legs. Its passive hip joints and a light and flexible push-pull cable transmission allow it to run at 15 body-lengths per second. The second robot is Spinybot, a hexapod that uses its toes with microspines to climb rough surfaces, including stucco, concrete and brick walls. The last robot is Stickybot, a gecko-inspired quadruped that climbs smooth vertical surfaces using directional dry adhesion. Stickybot contains several types of under-actuated mechanisms in its body, legs and toes. At the smallest length scale, the undersides of the toes are covered with a unique material called directional polymeric stalk (DPS), inspired by the directional setae and lamellae of the gecko. The future direction of Sangbae Kim's research involves several design principles toward bio-inspired robots and a novel manufacturing technology that enables hybrid structures combining multiple materials, including multi-grade elastomeric/solid materials, liquid, and gas with embedded sensors and actuators.

Fluidic Logic_- Merging Chemistry and Computation with Microfluidics
Manu Prakesh, PhD, Elected Junior Fellow (Physics), Harvard Society of Fellows; Visiting Scholar, MIT, manup@mit.edu

Starting from genetic blueprints, biological materials turn into complex, multi-compartmental "living" things with algorithmic precision. One fundamental difference between biological and physical materials is their inherent computational ability. Though we have long understood that "Information representation is invariably physical", we are only now beginning to exploit this insight to shape, program and manipulate matter in engineered systems. Manu Prakesh will introduce a paradigm in computation where bits can simultaneously transport and manipulate both materials and information, similar to how integrated circuits allow us to control the flow of electrons. He will describe an entirely new digital logic family which implements universal Boolean logic in an all-fluidic system, exploiting purely hydrodynamic nonlinearities in low-Reynolds number two-phase flow. Such a physical implementation thus provides a flow control mechanism for sub-nanoliter droplets operating at KHz frequencies with no moving parts. A "lego-set" like toolbox comprising of microfluidic circuits including AND/OR/NOT gates, flip-flops, counters, ring-oscillators and synchronizers will be introduced. These show the nonlinearity, gain, bistability, synchronization, cascadability, feedback and programmability required for scalable universal computation and all-fluidic control. This platform technology represents initial steps towards modular design of sub-nanoliter droplet reactors with applications in high throughput screening for novel materials, combinatorial chemistry and implanted devices.