Apparatus and method for discriminating biological tissue, surgical apparatus using the apparatus

Patent metadata

Field	Value
Publication number	US-10864037-B2
Application number	US-201615203872-A
Country	US
Kind code	B2
Filing date	Jul 7, 2016
Priority date	Jul 8, 2015
Publication date	Dec 15, 2020
Grant date	Dec 15, 2020

Abstract

The present disclosure relates to an apparatus and method for discriminating biological tissue, and a surgical apparatus using the same, the biological tissue discriminating method being capable of exactly discriminating the biological tissue by measuring an impedance value per frequency, teaching the measured impedance value per frequency in a single classifier according to learning algorithms that are different from one another having the measured impedance value per frequency as an input variable to discriminate the biological tissue, and re-teaching the biological tissue discriminated from each single classifier in a meta classifier to finally discriminate the biological tissue.

First claim

The invention claimed is:

1. An apparatus for discriminating biological tissue, the apparatus comprising:
an impedance measurer comprising a first electrode for applying a signal to the biological tissue and a second electrode for receiving the signal, the impedance measurer being configured to measure an impedance magnitude and an impedance phase from the received signal;
a base classifier comprising a plurality of single classifiers that are different from one another, each of the plurality of single classifiers being configured to discriminate the biological tissue according to a first machine learning algorithm, wherein the impedance magnitude and the impedance phase are used as input variables of the plurality of single classifiers, wherein the input variables of the plurality of single classifiers are selected to be different for each of the plurality of single classifiers using a genetic algorithm method, and wherein one or more classifiers among the plurality of single classifiers are selected based on discrimination performance, wherein the selected one or more classifiers is a subset of the plurality of single classifiers, the subset having the highest discrimination performance among the plurality of single classifiers; and
a meta classifier configured to finally discriminate the biological tissue according to a second machine learning algorithm, wherein results of the selected one or more classifiers are used as input variables of the second machine learning algorithm.

2. The apparatus according to claim 1 ,
wherein the impedance measurer is further configured to measure the impedance magnitude and the impedance phase with a frequency of the signal being adjusted by units of 10 kH between 10 kHz and 100 kHz.

3. The apparatus according to claim 1 ,
wherein the plurality of single classifiers comprises a classifier according to any one or any combination of any two or more of a support vector machine (SVM) algorithm, a k-nearest neighbors (k-NN) algorithm, a decision tree (DT) algorithm, a quadratic discriminant analysis (QDA) algorithm, and a random forest (RF) algorithm.

4. The apparatus according to claim 1 ,
wherein the base classifier comprises a classifier according to any one or any combination of any two or more of a support vector machine (SVM) algorithm, a quadratic discriminant analysis (QDA) algorithm, and a random forest (RF) algorithm.

5. The apparatus according to claim 1 ,
wherein the meta classifier comprises a classifier according to an artificial neural network (ANN) algorithm.

6. A method for discriminating biological tissue, the method comprising:
measuring an impedance magnitude and an impedance phase using a first electrode for applying a signal to the biological tissue and a second electrode for receiving the signal;
discriminating the biological tissue by a plurality of single classifiers that are different from one another and configured to discriminate the biological tissue according to a first machine learning algorithm, wherein the impedance magnitude and the impedance phase are used as input variables of the plurality of single classifiers, wherein the input variables of the plurality of single classifiers are selected to be different for each of the plurality of single classifiers using a genetic algorithm method, and wherein one or more classifiers among the plurality of single classifiers are selected based on discrimination performance, wherein the selected one or more classifiers is a subset of the plurality of single classifiers, the subset having the highest discrimination performance among the plurality of single classifiers; and
finally discriminating the biological tissue by a meta classifier according to a second machine learning algorithm, wherein results of the selected one or more classifiers are used as input variables of the second machine learning algorithm.

7. An apparatus for performing coagulation or cutting of biological tissue, the apparatus comprising:
a surgical device configured to output energy to a surgery area;
a biological tissue discriminating device configured to make a final discrimination of biological tissue of the surgery area; and
an output control device configured to automatically adjust the output of energy based on the final discrimination,
wherein the biological tissue discriminating device comprises
an impedance measurer comprising a first electrode for applying a signal to the biological tissue and a second electrode for receiving the signal, the impedance measurer being configured to measure an impedance magnitude and an impedance phase from the received signal,
a base classifier comprising a plurality of single classifiers that are different from one another, each of the plurality of single classifiers being configured to discriminate the biological tissue according to a first machine learning algorithm, wherein the impedance magnitude and the impedance phase are used as input variables of the plurality of single classifiers, wherein the input variables of the plurality of single classifiers are selected to be different for each of the plurality of single classifiers using a genetic algorithm method, and wherein one or more classifiers among the plurality of single classifiers are selected based on discrimination performance, wherein the selected one or more classifiers is a subset of the plurality of single classifiers, the subset having the highest discrimination performance among the plurality of single classifiers, and
a meta classifier configured to make the final discrimination of the biological tissue according to a second machine learning algorithm, wherein results of the selected one or more classifiers are used as input variables of the second machine learning algorithm.

8. The apparatus according to claim 7 ,
wherein the impedance measurer is further configured to measure the impedance magnitude and the impedance phase with a frequency of the signal being adjusted by units of 10 kH between 10 kHz and 100 kHz.

9. The apparatus according to claim 7 ,
wherein the plurality of single classifiers comprise a classifier according to any one or any combination of any two or more of a support vector machine (SVM) algorithm, a k-nearest neighbors (k-NN) algorithm, a decision tree (DT) algorithm, a quadratic discriminant analysis (QDA) algorithm, and a random forest (RF) algorithm.

10. The apparatus according to claim 7 ,
wherein the base classifier comprises a classifier according to any one or any combination of any two or more of a support vector machine (SVM) algorithm, a quadratic discriminant analysis (QDA) algorithm, and a random forest (RF) algorithm.

11. The apparatus according to claim 7 ,
wherein the meta classifier comprises a classifier according to an artificial neural network (ANN) algorithm.

12. The apparatus according to claim 7 ,
wherein the energy is generated by an ultrasonic wave signal.

13. The apparatus of claim 1 , wherein the second machine learning algorithm re-teaches classification results through the plurality of single classifiers to the meta classifier to finally discriminate the biological tissue.

14. The apparatus of claim 1 , wherein the impedance measurer is further configured to measure the impedance magnitude by incrementally changing a frequency of the signal between a minimum frequency and a maximum frequency.

Assignees

• Research & Business Found Sungkyunkwan Univ

Inventors

• Mun Joung Hwan
• Youn Su Hyun
• Choi Ahnryul

Classifications

• A61B2018/00869

  Phase · CPC title

• A61B18/1445 Primary

  at the distal end of a shaft, e.g. forceps or scissors at the end of a rigid rod · CPC title

• A61B2018/00904

  Automatic detection of target tissue · CPC title

• A61B2018/00994

  combining two or more different kinds of non-mechanical energy or combining one or more non-mechanical energies with ultrasound · CPC title

• A61B2018/0069

  using fuzzy logic · CPC title