Diagnosis method using laser induced breakdown spectroscopy and diagnosis device performing the same

1. A method for diagnosing a disease of a patient by using Laser-Induced Breakdown Spectroscopy (LIBS), the method comprising:

projecting a first pulsed laser beam to a suspicious tissue to induce plasma ablation at the suspicious tissue of the patient;

receiving first light generated by the plasma ablation induced at the suspicious tissue;

obtaining first spectrum data related to the first light;

projecting a second pulsed laser beam to a reference tissue to induce plasma ablation at the reference tissue of the patient;

receiving second light generated by the plasma ablation induced at the reference tissue;

obtaining second spectrum data related to the second light; and

determining whether the suspicious tissue is a disease tissue based on an output of an artificial neural network to which input data including comparison data of the first spectrum data and the second spectrum data is inputted,

wherein the artificial neural network has been trained with training-data set including first data and second data, the first data includes first comparison data of two spectrum data related to light generated by plasma ablation induced respectively at two different non-disease tissues of a single subject and the second data includes second comparison data of two spectrum data related to light generated by plasma ablation induced respectively at a non-disease tissue and a disease tissue of a single subject, and

wherein the first data is labeled with a first class value indicating normal status and the second data is labeled with a second class value indicating disease status.

2. The method according to claim 1 , wherein the comparison data is related to a difference of the first and the second spectrum data.

3. The method according to claim 1 , wherein the comparison data is related to a ratio of the first and the second spectrum data.

4. The method according to claim 1 , wherein the tissue includes a skin, and wherein the disease includes a skin cancer.

5. The method according to claim 1 , wherein the suspicious tissue and the reference tissue are included in the same body part of the patient.

6. The method according to claim 1 , wherein the reference tissue is located within a predetermined distance from the suspicious tissue.

7. The method according to claim 1 , further comprising: normalizing the first spectrum data and the second spectrum data, and obtaining the comparison data based on the normalized first spectrum data and the normalized second spectrum data.

8. The method according to claim 7 , wherein the normalizing comprises:

processing the first spectrum data based on an energy of the first spectrum data, and

processing the second spectrum data based on an energy value of the second spectrum data.

9. The method according to claim 7 , wherein the normalizing comprises:

adjusting intensity values of at least one of the first spectrum data and the second spectrum data so that a first energy value corresponding to the first spectrum data and a second energy value corresponding to the second spectrum data have the same value.

10. The method according to claim 1 , further comprising:

when a first energy value corresponding to the first spectrum data is less than a first threshold, outputting a message indicating re-projecting the first pulsed laser beam to the suspicious tissue; and

when a second energy value corresponding to the second spectrum data is less than the first threshold, outputting a message indicating re-projecting the second pulsed laser beam to the reference tissue.

11. The method according to claim 1 , further comprising:

outputting a message indicating re-projecting the first pulsed laser beam to the suspicious tissue when a max peak value of the first spectrum data is above a second threshold; and

outputting a message indicating re-projecting the first pulsed laser beam to the reference tissue when a max peak value of the second spectrum data is above the second threshold.

12. The method according to claim 1 , further comprising:

adjusting the number of the intensity values of the first spectrum data and the number of the intensity values of the second spectrum data to correspond to the number of the input nodes of the artificial neural network, respectively.

13. The method according to claim 12 , wherein the adjusting comprises:

interpolating a first intensity value and a second intensity value respectively corresponding to a first measured wavelength and a second measured wavelength of the first spectrum data,

selecting, from the interpolated intensity values of the first spectrum data, one of more intensity values corresponding to one of more predetermined standard wavelengths,

interpolating a third intensity value and a fourth intensity value respectively corresponding to the first measured wavelength and the second measured wavelength of the second spectrum data, and

selecting, from the interpolated intensity values of the second spectrum data, one or more intensity values corresponding to one or more predetermined standard wavelengths.

14. The method according to claim 13 ,

wherein the adjusted number of intensity values of the first spectrum data is greater than the number of intensity values of the first spectrum data, and

wherein the adjusted number of intensity values of the second spectrum data is greater than the number of intensity values of the second spectrum data.

15. The method according to claim 13 , wherein

each of the predetermined standard wavelengths corresponds to each of the input nodes of the artificial neural network, and

wherein each of the intensity values corresponding to the predetermined standard wavelengths of the comparison data is inputted to each of the input nodes of the artificial neural network.

16. A method for diagnosing a disease of a patient by using Laser-Induced Breakdown Spectroscopy (LIBS), the method comprising:

obtaining first spectrum data related to first light generated by plasma ablation induced at a target tissue of the patient;

obtaining second spectrum data related to second light generated by plasma ablation induced at a reference tissue of the patient; and

determining whether the target tissue is a disease tissue based on an output of an artificial neural network to which input data including comparison data of the first spectrum data and the second spectrum data is inputted,

wherein the artificial neural network has been trained with a training-data set including first data and second data, the first data includes first comparison data of two spectrum data related to light generated by plasma ablation induced respectively at two different non-disease tissues of a single subject and the second data includes second comparison data of two spectrum data related to light generated by plasma ablation induced respectively at a non-disease tissue and a disease tissue of a single subject, and

wherein the first data is labeled with a first class value indicating normal status and the second data is labeled with a second class value indicating disease status.

17. A diagnostic device comprising:

a laser projection module configured to project a pulsed laser beam to a tissue for inducing plasma ablation;

a light receiving module configured to receive light generated by the plasma ablation induced at the tissue by the pulsed laser beam, wherein the light generated by the plasma ablation includes continuum emission light having continuous spectrum characteristic and element specific emission light having discontinuous spectrum characteristic;

a spectrum measurement module configured to obtain spectrum data from the received light; and

a controller configured to:

obtain first spectrum data related to first light generated by the first plasma ablation induced at a target tissue,

obtain second spectrum data related to second light generated by second plasma ablation induced at a reference tissue, and

determine whether the target tissue is a disease tissue based on an output of an artificial neural network to which input data including comparison data of the first spectrum data and the second spectrum data is inputted,

wherein the artificial neural network has been trained with a training-data set including first data and second data, the first data includes first comparison data of two spectrum data related to light generated by plasma ablation induced respectively at two different non-disease tissues of a single subject and the second data includes second comparison data of two spectrum data related to light generated by plasma ablation induced respectively at a non-disease tissue and a disease tissue of a single subject, and

wherein the first data is labeled with a first class value indicating normal status and the second data is labeled with a second class value indicating disease status.

18. The diagnostic device according to claim 17 ,

wherein a pulse duration of the laser projection module is set to be one of 1) a nanosecond duration, 2) a picosecond duration, or 3) a femtosecond duration.

19. The diagnostic device according to claim 17 ,

wherein the spectrum measurement module receives the light during a predetermined time period between a first time point and a second time point,

wherein the first time point is within 100 ns after pulsed laser beam has been projected, and the second time point is later than 1 us after the pulsed laser beam has been projected.

20. The diagnostic device according to claim 17 ,

wherein the spectrum measurement module measures spectrum-related to a predetermined wavelength range, wherein the predetermined wavelength range is between a first wavelength less than 300 nm and a second wavelength larger than 700 nm.