Biosignal processing apparatus based on intelligent control

Patent metadata

Field	Value
Publication number	US-12102451-B2
Application number	US-202117394419-A
Country	US
Kind code	B2
Filing date	Aug 5, 2021
Priority date	Oct 21, 2020
Publication date	Oct 1, 2024
Grant date	Oct 1, 2024

Abstract

The present invention relates to an integrated circuit for processing biosignals, a biosignal processing apparatus, and a biosignal processing system, and the integrated circuit includes: a digital conversion unit for converting an analog biosignal input through a biosignal input terminal into digital biodata; and an AI block for processing a plurality of biodatas converted through the digital conversion unit according to an artificial intelligence processing flow, and outputting a result data according to processing of the plurality of biodatas.

First claim

What is claimed is:

1. An integrated circuit for processing biosignals, the integrated circuit comprising:
a digital conversion unit for converting an analog biosignal input through a biosignal input terminal into a digital biodatas; and
an Artificial Intelligence (AI) block for processing a plurality of biodatas converted through the digital conversion unit according to an artificial intelligence processing flow, and outputting result data according to processing of the plurality of biodatas,
wherein the digital conversion unit includes:
a demultiplexer for outputting an analog biosignal output from the biosignal input terminal to a first output line or a second output line according to a type signal that is input through a type input terminal;
a trans-impedance amplifier for amplifying the analog biosignal of the first output line and outputting the amplified biosignal;
a low-noise amplifier for amplifying the analog biosignal of the second output line and outputting the amplified biosignal; and
a first multiplexer for outputting a biosignal selected from the biosignal of the trans-impedance amplifier and the biosignal of the low-noise amplifier according to the type signal.

2. The integrated circuit according to claim 1 , wherein the digital conversion unit further includes: a temperature sensor;
a second multiplexer for outputting the biosignal output from the first multiplexer or a temperature signal output from the temperature sensor according to a selection signal received from the AI block; and
an Analog to Digital Converter (ADC) for converting an analog signal output from the second multiplexer into digital data.

3. The integrated circuit according to claim 2 , wherein the digital conversion unit includes the demultiplexer, the trans-impedance amplifier, the low-noise amplifier, the first multiplexer, the second multiplexer, and the ADC in plurality to independently convert analog biosignals input through a plurality of biosignal input terminals into a plurality of digital biodatas, and output the digital biodatas to the AI block.

4. The integrated circuit according to claim 1 , further comprising a power supply unit for supplying power to the digital conversion unit and the AI block from a supplied main power or a signal recognized from the outside, wherein the power supply unit outputs power generated from the signal recognized from the outside to the digital conversion unit and the AI block as a voltage of a specified level according to whether the main power is detected.

5. The integrated circuit according to claim 4 , wherein the power supply unit includes: a main power detector for detecting the main power from a main power input terminal, and outputting a detection signal indicating whether or not the main power is detected;
a harvester for generating power from a signal recognized from the outside of the integrated circuit for processing biosignals; and
a third multiplexer for selecting and outputting one among the power from the harvester and the main power of the main power input terminal according to the detection signal of the main power detector, wherein the power output from the third multiplexer is supplied to the digital conversion unit and the Al block.

6. The integrated circuit according to claim 5 , wherein the harvester includes:
a solar energy harvester for harvesting power from a solar signal output from a solar input terminal;
a radio frequency (RF) harvester for harvesting power from an RF signal output from an RF input terminal;
a vibration/electrostatic harvester for harvesting power from a vibration/electrostatic signal output from a vibration/static-electricity input terminal; and
a buck-boost converter for increasing or decreasing power received from one or more among the solar energy harvester, the RF harvester, and the vibration/electrostatic harvester, and outputting the increased or decreased power to the third multiplexer.

7. The integrated circuit according to claim 6 , wherein the solar energy harvester includes a Maximum Power Point Tracker (MPPT), the RF harvester includes a reconfigurable RF-DC converter, and the vibration/electrostatic harvester includes a rectifier, wherein the harvesting power output through the buck-boost converter is stored through a capacitor outside the integrated circuit for processing biosignals.

8. The integrated circuit according to claim 1 , wherein the AI block predicts infection of a disease from the biodatas and outputs a predicted result, or predicts an output signal of a neural network of a human body from the biodatas and outputs the predicted signal.

9. A biosignal processing apparatus comprising:
a digital conversion unit for converting an analog biosignal into a digital biodatas; and
an Artificial Intelligence (AI) block for processing a plurality of biodatas converted through the digital conversion unit according to an artificial intelligence processing flow and outputting result data according to processing of the plurality of biodatas using an analog Processing in Memory (PIM),
wherein the AI block includes:
a memory for storing a plurality of biodatas received from the digital conversion unit;
the analog PIM capable of storing a plurality of analog data of each of a plurality of kernels, updating the plurality of analog data of the plurality of kernels, and outputting a plurality of digital data corresponding to the plurality of kernels; and
an AI digital controller for generating predicted result data by applying a processing flow of an artificial intelligence algorithm using a plurality of corresponding digital data of the plurality of kernels received from the analog PIM, and the plurality of biodatas of the memory,
wherein the AI digital controller applying a Convolutional Neural Network (CNN) algorithm generates a plurality of feature maps by a convolution operation between the plurality of biodatas of the memory and the plurality of corresponding digital data of the plurality of kernels of the analog PIM, and stores the feature maps in the memory in a first convolution operation process, and generates a plurality of digital data of a plurality of kernels to be used in a second convolution operation process subsequent to the first convolution operation process, and stores a plurality of digital data of each of a plurality of kernels to be updated in the analog PIM.

10. The apparatus according to claim 9 , wherein the analog PIM includes:
M Digital to Analog Converters (DACs);
a Static Random Access Memory (SRAM) array including L of M SRAMs respectively connected the M DACs according to the number of the plurality of kernels, each SRAM storing analog data received from a connected DAC; and
L or more Analog to Digital Converters (ADCs) connected to the SRAMs of the SRAM array to convert an analog data of the SRAM into digital data of a specified number of bits, wherein M and L are equal to or greater than 2.

11. The apparatus according to claim 9 , wherein the biosignal processing apparatus is an integrated circuit for processing biosignals.

12. A biosignal processing system comprising the integrated circuit for processing biosignals of claim 1 .

13. The system according to claim 12 , further comprising:
a capacitor for storing harvesting power; and
a biosensor for generating an analog biosignal using power harvested by the integrated circuit for processing biosignals, and outputting the generated analog biosignal of current or voltage to a biosignal input terminal of the integrated circuit for processing biosignals, wherein the integrated circuit for processing biosignals including a digital conversion unit and an AI block further inclu

Assignees

• Skaichips Co Ltd
• Research & Business Found Sungkyunkwan Univ
• Research & Business Foundation Sungskyunkwan Univ

Inventors

• Lee Kang Yoon
• Jo Jong Wan
• Pu Young Gun
• Ali Imran
• Kim Dong Gyu
• Park Joon Hong
• Kim Dong Gyun
• Kim Yun Gwan
• Kim Jae Bin
• Park Dong Soo
• Byun Sung June

Classifications

• A61B2562/0271

  Thermal or temperature sensors · CPC title

• A61B2560/0462

  Apparatus with built-in sensors · CPC title

• A61B2560/0214

  of power generation or supply · CPC title

• A61B5/7275

  Determining trends in physiological measurement data; Predicting development of a medical condition based on physiological measurements, e.g. determining a risk factor · CPC title

• A61B5/7264 Primary

  Classification of physiological signals or data, e.g. using neural networks, statistical classifiers, expert systems or fuzzy systems · CPC title