A digital nervous system aiming toward personalized IoT healthcare

Body area networks (BANs), cloud computing, and machine learning are platforms that can potentially enable advanced healthcare outside the hospital. By applying distributed sensors and drug delivery devices on/in our body and connecting to such communication and decision-making technology, a system for remote diagnostics and therapy is achieved with additional autoregulation capabilities. Challenges with such autarchic on-body healthcare schemes relate to integrity and safety, and interfacing and transduction of electronic signals into biochemical signals, and vice versa. Here, we report a BAN, comprising flexible on-body organic bioelectronic sensors and actuators utilizing two parallel pathways for communication and decision-making. Data, recorded from strain sensors detecting body motion, are both securely transferred to the cloud for machine learning and improved decision-making, and sent through the body using a secure body-coupled communication protocol to auto-actuate delivery of neurotransmitters, all within seconds. We conclude that both highly stable and accurate sensing—from multiple sensors—are needed to enable robust decision making and limit the frequency of retraining. The holistic platform resembles the self-regulatory properties of the nervous system, i.e., the ability to sense, communicate, decide, and react accordingly, thus operating as a digital nervous system.

Data from three participants wearing the gesture capture glove and performing repetitive motions of counting on one to four fingers while making a closed fist in between each movement. An optimized threshold based on the results from the complete study group was used to determine if a finger was up or down (i.e., opened or closed).

Comments on the MicroChip BodyCom kit used in the experiments
The BCC system uses the base components of the BodyCom kit defined by Microchip (https://www.microchip.com/design-centers/embedded-security/technology/bodycom-tradetechnology, May 2019) but has since been redesigned both from a hardware and software perspective.
The system works in the following manner: there is a base unit connected to the body, and a set of "tags"/slave units. A tag could contain either a sensor or an actuator. The base unit is considered the gateway of the body. In our experiments we only used one.
The base unit can be set to poll tags for response by individually calling them using their unique IDs.
Only tags with the polled ID will respond/listen to the commands sent to the base unit allowing for no interference between the units. Time-out functions are implemented to abort waiting for lack of response from malfunctioning tags, etc. The number of sensors will be limited, and time delays are not critical. Also broadcast options are possible from the base unit such that all tags can take actions based on the instruction transmitted by the base unit.
The communication frequency of the downlink (from base unit to tag) differs from the uplink (other direction), and thus downlink communication can never interfere. Instructions sent from the phone to the base unit for further transportation to the tags will be queued based on time of arrival. In each tag and base unit, the receiver and transmitter are tuned to their respective frequency bands of 125 kHz and 8 MHz. Microchip's BodyCom kit then uses a PIC16LF1829 20-pin Flash in the base unit and a PIC16LF1827 18-pin Flash MCU device in the tag to control the modulation of the signal in these bands. The de-/modulated signals are further filtered and amplified. In the transmit path, the voltage of an electrode in close proximity of the body is affected, and the variation of the electrical field can be detected by the other unit. The challenge is that the signal is typically heavily attenuated due to the properties of the human as a communication channel and the receivers have to react on very weak signals.
The base unit can also be set in "listen mode" and wait for tags that themselves decide to transmit (through the detection of a user interaction, passing a certain threshold, etc.). In this scenario, there is a potential risk for interference between signals from tags towards the base unit. However, messages are short (their duration is in the order of a few ms) thus unlikely to collide. If there is a collision, due to a cyclic redundancy check (CRC) mechanism, the message will be discarded by the base unit. Upon detection of a corrupted message in "listen mode" the base unit could decide to go back into polling mode and poll all tags for their latest message. In "listen mode" a future system could involve more controlled turn-based communication between tag and base unit where each tag is allocated a certain time slot. The base board can also call for resubmission of an incomprehensive message and allocate random time delays to the sensors forcing them to not submit their responses repeatedly.