IEEE Solid-States Circuits Magazine - Winter 2022 - 78

an application-based IEEE road map
that could unite multiple technology
disciplines to offer device, instrument,
and measurement solutions.
In 2022, IEEE Brain established a
Best Paper Award with each Sponsoring
society. The goal
is to promote
outstanding work in these
important cross-disciplinary areas.
In this inaugural year, the SSCS determined
that the winner would be
selected from brain- and neuroscience-related
papers published at all
2021 SSCS-sponsored conferences.
Nominations were solicited from
the IEEE Asian Solid-State Circuits
Conference, IEEE Custom Integrated
Circuits Conference, European Solid-State
Circuits and Devices Conference,
ISSCC, and IEEE Symposium
on VLSI Technology and Circuits,
and an independent committee was
formed to select the winner. The
committee chose one winning paper
and two for honorable mentions.
Congratulations to all the recipients!
Winner of the 2021 SSCS-Brain
Best Paper Award
The winning paper was " An RF-Ultrasound
Relay for Powering Deep
Implants Across Air-Tissue Interfaces
With a Multioutput Regulating
Rectifier and Ultrasound Beamforming, "
by Ernest So, Pyungwoo Yeon,
E.J. Chichilnisky, and Amin Arbabian.
The paper, from Stanford University,
proposed a novel approach
for wirelessly powering deep implants
by interfacing RF and ultrasound
(US) with high efficiency and
beamforming capability. Wireless
powering has the potential to enable
long-term, millimeter-sized
implantable medical devices (IMDs)
and BMIs that interface with the nervous
system at cellular resolution to
treat paralysis, stroke, epilepsy, and
vision loss. A bulky battery is exchanged
for a small transducer that
further miniaturizes and extends
the lifetime of an implant. This is
critical to reduce tissue damage, infection,
and surgical complications.
The retina and brain are regions
of interest for neural interfacing
because of the high density of neu78
WINTER 2022
rons. However, wirelessly powering
millimeter-sized IMDs in these areas
is challenging since the electricity
must cross multiple media. For
the retina, power will pass through
air before going through the eye because
weight limitations preclude
attaching a battery to the organ. In
the brain, power must pass through
the skull before entering brain tissue.
This makes it difficult for single-modality
solutions, such as RF,
optical, and US, since centimeters
of depth and a millimeter-sized receiver
are required.
Imagine we want to communicate
a message (power) to a diver in the
ocean from above the water. Shouting
(US) does not work because although
sound generated underwater
is audible, sound from above the
surface is not. Hand signals (optical/RF)
will be visible only if the
diver is near the surface. If we place
a buoy that could convert the hand
signals into a sound and transmit
it underwater, the diver will
hear the message even at greater
depths. The proposed relay acts as
the " buoy " to convert between the
RF and US domains for retinal and
neural applications.
The winning paper [3] describes
an RF-US relay that can be used to
power deep, millimeter-sized IMDs
in the retina and brain. As detailed
in Figure 1, an RF inductive link powers
a relay placed at the interface of
the two media. Because the separation
distance can be comparable to
the coil size, this link can be very
efficient (86%). The relay converts
the RF power to dc to drive an array
of piezos for beamforming US power
through the tissue to the implant
since US has low loss in tissue, good
coupling with transducers, and focusing
capability. It is the first relay
architecture that includes programmable
beamforming to adjust for
implant placement and movement.
Several key innovations are proposed
to reduce the cascaded power
loss when converting from RF to
electrical to US. First, a multioutput
regulating rectifier is introduced to
generate and regulate three voltage
IEEE SOLID-STATE CIRCUITS MAGAZINE
supplies (6-29, 4.5 and, 1.8 V) in a
single step to increase efficiency by
eliminating the extra voltage conversion
steps in traditional architectures.
Second, an adiabatic US PA
design increases the efficiency of
driving capacitive piezo loads without
using a large, bulky inductor. By
utilizing the principle of adiabatic
charging, the efficiency of driving
a capacitive piezo load can be increased
by up to 42% compared to
traditional class D PAs. Finally, the
adjustable, free-wheeling HV supply
and HV US PAs enable the transmitted
US power to be adjusted for different
applications and depths.
This technology can wirelessly
power millimeter-sized BMIs deep
in the brain, with milliwatt levels
to record and stimulate neurons at
cellular resolution, which could be
critical to decipher the neural code.
In the future, the authors plan to
add bidirectional communication to
the RF-US relay to enable stimulation
and recording for the implant.
The authors appear in Figure 2.
Their work was performed in collaboration
with the Stanford Artificial
Retina team, led by E.J. Chichilnisky,
with collaborators from Stanford
University, Delft University of Technology,
and the University of Washington
[7]. The team aims to restore
high-acuity vision by examining
neural code to design a better stimulus
and tackle challenges including
cell classification, stimulation algorithms,
packaging, electrode design,
wireless power and communication,
surgery, animal experiments, and
implant design.
2021 SSCS-Brain Best Paper
Award Honorable Mention
An honorable mention went to " An
Optically Addressed NanowireBased
Retinal Prosthesis With 73%
RF-to-Stimulation Power Efficiency
and 20nC-to-3μC Wireless Charge
Telemetering, " by Abraham Akinin,
Jeremy M. Ford, Jiajia Wu, Chul Kim,
Hiren D. Thacker, Patrick P. Mercier,
and Gert Cauwenberghs. The
awardees elegantly demonstrated
an inductive retinal stimulator SoC
IEEE Solid-States Circuits Magazine - Winter 2022

Table of Contents for the Digital Edition of IEEE Solid-States Circuits Magazine - Winter 2022
