Medical Ventilator Capable of Early Detecting and Recognizing Types of Pneumonia, Gas Recognition Chip, and Method for Recognizing Gas Thereof

• US 20130197384A1
• Filed: 04/27/2012
• Published: 08/01/2013
• Est. Priority Date: 02/01/2012
• Status: Active Grant

• Current Assignee: National Tsing Hua University
• Original Assignee: National Tsing Hua University
• Inventors: Yang, Chia-Min, Yao, Da-Jeng, Tang, Kea-Tiong, Shih, Chung-Hung, Wang, Li-Chun, Chen, Hsin, Liu, Yi-Wen, Shyu, Jyuo-Min
• Granted Patent: US 9,125,590 B2
• Time in Patent Office: Days
• US Class Current: 600/532
• CPC Class Codes: A61B 5/082 Evaluation by breath analysis, e.g. determination of the chemical composition of exhaled breath A61B5/083, A61B5/091 take precedence
• CPC Class Codes: A61B 5/7264 Classification of physiological signals or data, e.g. using neural networks, statistical classifiers, expert systems or fuzzy systems neural networks per se G06N3/00; expert systems per se G06N5/00
• CPC Class Codes: G01N 2033/4975 other than oxygen, carbon dioxide or alcohol, e.g. organic vapours
• CPC Class Codes: G01N 33/497 of gaseous biological material, e.g. breath
• CPC Class Codes: G06F 18/00 Pattern recognition
• CPC Class Codes: G16H 50/20 for computer-aided diagnosis, e.g. based on medical expert systems

Abstract

A medical ventilator capable of early detecting and recognizing types of pneumonia, a gas recognition chip, and a method for recognizing gas thereof are disclosed. The gas recognition chip of the medical ventilator comprises a sensor array, a sensor interface circuit, a stochastic neural network chip, a memory and a microcontroller. The sensor array receives a plurality of multiple types of gases to produce odor signals corresponding to each type of gas. The sensor interface circuit analyzes the odor signals to produce gas pattern signals corresponding to each type of gas. The stochastic neural network chip amplifies the differences between the gas pattern signals and performs dimensional reduction on the gas pattern signals to aid the analysis. The memory stores training data. The microcontroller performs a mixed gas recognizing algorithm to early detect and recognize the type of the pneumonia according to the gas training data.

Claims

1. A gas recognition chip, comprising:

a sensor array, including a plurality of sensors and a sensing film, and the sensing film being provided for adsorbing plural types of gases, such that each of the sensors produces an odor signal corresponding to each of the respective gases;

a sensor interface circuit, for reading and analyzing the odor signal of each of the gases to generate a gas pattern signal corresponding to each of the respective gases;

a stochastic neural network chip, for amplifying the difference between the gas pattern signals and reducing the dimensions of each gas pattern signal to produce an analysis result;

a memory, for storing gas training data; and

a microcontroller, for receiving the analysis result, and executing a mixed gas recognizing algorithm according to the analysis result to recognize the type of the gas, and classify an unknown gas not existed in the gas training data, and producing a recognition result according to the gas training data.

2. The gas recognition chip of claim 1, wherein the sensing film is made of a nanoporous carbon material and provided for growing a polymer with gas adsorbability in pores of the nanoporous carbon material.

3. The gas recognition chip of claim 1, wherein the microcontroller transmits data of the unknown gas to the stochastic neural network chip and the memory when the microcontroller detects the unknown gas, so that the gas recognition chip has a self learning ability.

4. The gas recognition chip of claim 3, wherein the mixed gas recognizing algorithm includes a K nearest neighbor (KNN) algorithm, a linear least squares regression algorithm and a median-threshold K nearest neighbor (MTKNN) classification algorithm, and the median-threshold K nearest neighbor classification algorithm is used to find a distance between every two data in the gas training data first, and then find a median of the distances to determine whether the gas is the unknown gas.

5. The gas recognition chip of claim 1, wherein the sensor array is comprised of conducting polymer (CP) sensors.

6. The gas recognition chip of claim 5, wherein the sensor interface circuit stores electric charges by a single polysilicon floating gate element to reduce electric leakage and circuit power.

7. The gas recognition chip of claim 1, wherein the sensor array is comprised of surface acoustic wave (SAW) sensors.

8. The gas recognition chip of claim 7, wherein the sensor interface circuit includes an adjustable oscillator circuit capable of adjusting an oscillation frequency according to different surface acoustic wave sensors to enhance the flexibility of usage.

17. A medical ventilator capable of early detecting and recognizing types of pneumonia, comprising an aspiration pipeline and a gas recognizing device, wherein the gas recognizing device uses a gas recognition chip to analyze a gas aspired by a patient from the aspiration pipeline to identify the type of pneumonia, the gas recognition chip is the one recited in the claim 1.

18. The medical ventilator capable of early detecting and recognizing types of pneumonia as recited in claim 17, wherein the gas recognizing device is coupled directly to the aspiration pipeline outside the patient'"'"'s body, and the gas aspired by the patient is collected to perform a recognition.

9. A method for recognizing gas, comprising the steps of:

using a sensing film of a sensor array to adsorb plural types of gases, such that each sensor of the sensor array generates an odor signal corresponding to each of the respective gases;

using a sensor interface circuit to read and analyze the odor signal of each of the gases to generate a gas pattern signal corresponding to each of the respective gases;

using a stochastic neural network chip to amplify the difference between the gas pattern signals and reduce the dimensions of each of the gas pattern signals to produce an analysis result;

storing gas training data in a memory; and

using a microcontroller to receive the analysis result, and execute a mixed gas recognizing algorithm to identify the type of the gas according to the analysis result, and classify an unknown gas not existed in the gas training data, and then produce a recognition result according to the gas training data.

10. The method for recognizing gas as recited in claim 9, wherein the sensing film is made of a nanoporous carbon material and a polymer with gas adsorbability is grown in pores of the nanoporous carbon material.

11. The method for recognizing gas as recited in claim 9, further comprising the step of:

using the microcontroller to transmit data of the unknown gas to the stochastic neural network chip and the memory when the unknown gas is detected, so that the gas recognition chip has a self-learning ability.

12. The method for recognizing gas as recited in claim 11, wherein the mixed gas recognizing algorithm includes a K nearest neighbor algorithm, a linear least squares regression algorithm and a median-threshold K nearest neighbor classification algorithm, wherein the median-threshold K nearest neighbor classification algorithm is used to find a distance between every two data in the gas training data, and then find a median of the distances, and the median is used to determine whether the gas is the unknown gas.

13. The method for recognizing gas as recited in claim 9, wherein the sensor array is comprised of conducting polymer sensors.

14. The method for recognizing gas as recited in claim 13, wherein the sensor interface circuit stores electric charges by a single polysilicon floating gate element to reduce electric leakage and circuit power.

15. The method for recognizing gas as recited in claim 9, wherein the sensor array is comprised of surface acoustic wave sensors.

16. The method for recognizing gas as recited in claim 15, wherein the sensor interface circuit includes an adjustable oscillator circuit for adjusting an oscillation frequency according to different surface acoustic wave sensors to enhance the flexibility of usage.