Stand-alone apparatus and methods for in vivo detection of tissue malignancy using laser spectroscopy
According to an embodiment of the present disclosure, there is provided a laser spectroscopy-based independent device, including: a spectrometer configured to measure a spectrum of generated light which is generated by a laser projected onto a sample; and a disease analysis module configured to determine whether there is lesion tissue by applying a lesion tissue detection learning model to a result of non-discrete spectrum measurement, which is measured by the spectrometer, wherein the spectrometer is configured to measure spectra of all generated light that is generated from a time when the laser is projected onto the sample.
1. A Laser-Induced Breakdown Spectroscopy (LIBS) based diagnostic device, the diagnostic device comprising:
a spectrometer configured to:
receive plasma light generated due to plasma ablation from after a pulsed laser beam applied on a target tissue by a non-gated manner, wherein the plasma ablation is induced at the target tissue by the pulsed laser beam and the pulsed laser beam is outputted from a laser generation module and is projected onto a skin of a subject,
obtain spectrum data related to the plasma light received from after the pulsed laser beam applied on the target tissue, wherein the spectrum data includes both of 1) a first spectrum data related to the continuum emission and having overall intensity values within a specific wavelength band, and 2) a second spectrum information related to the element specific emission and having intensity values of spectral peaks within the specific wavelength band, and
a processor configured to:
obtain, from the spectrometer, the spectrum data,
extract a feature data from the spectrum data, wherein the feature data includes a first set of intensity values corresponding to a first plurality of wavelengths related to the related continuum emission and a second set of intensity values corresponding to a second plurality of wavelengths related to the spectral peaks, and the first plurality of wavelengths includes wavelengths disposed between each wavelengths reflecting the spectral peaks within the specific wavelength band
obtain input data for a deep learning algorithm based on the feature data, wherein the input data includes both the at least part of the first set of intensity values and the at least part of the second set of intensity values, and the each intensity value of each wavelength included in the input data is for inputted each nodes of the deep learning algorithm and the number of intensity values of the input data corresponds to the number of the input nodes of the deep learning algorithm, and
provide, using the deep learning algorithm, information on whether the target tissue is a lesion tissue related to a skin cancer based on the input data, wherein the deep learning algorithm has been trained with training-data, which is related to plasma ablation induced at a cancerous sample and is labeled with a value indicating cancer, wherein the training data includes both of 1) a third spectrum data related to a continuum emission related to the plasma ablation induced at the cancerous sample and having overall intensity values within the specific wavelength band, and 2) a fourth spectrum data information related to element specific emission to the plasma ablation induced at the cancerous sample and having intensity values of spectral peaks within the specific wavelength band.
2. The diagnostic device of claim 1 , wherein the specific wavelength band is set to be between 200 nm and 1000 nm.
3. The diagnostic device of claim 1 , wherein a wavelength of the pulsed laser beam applied by the laser generation module is set to be a specific wavelength out of the specific wavelength band.
4. The diagnostic device of claim 1 , wherein the skin cancer includes at least one of squamous cell carcinoma, basal cell carcinoma, or melanoma.
5. A Laser-Induced Breakdown Spectroscopy (LIBS) based diagnostic device, the diagnostic device comprising:
a laser generation module configured to apply a pulsed laser beam onto a target object to induce plasma ablation at the target object, wherein the target object is a part of a body of a subject;
a spectrometer configured to:
receive light generated due to the plasma ablation from after the pulsed laser beam applied on the target object by non-gated manner, and
obtain target spectrum data related to the light received from after the pulsed laser beam applied on the target object, wherein the target spectrum data includes both of 1) a first spectrum data related to the continuum emission and having overall intensity values within a specific wavelength band, and 2) a second spectrum data related to the element specific emission and having intensity values of spectral peaks within the specific wavelength band, and
a processor configured to:
obtain, from the spectrometer, the target spectrum data,
extract a feature data from the target spectrum data, wherein the feature data includes a first set of intensity values corresponding to a first plurality of wavelengths related to the related continuum emission and a second set of intensity values corresponding to a second plurality of wavelengths related to the spectral peaks, and the first plurality of wavelengths are except the spectral peaks within the specific wavelength band,
obtain input data for a deep learning algorithm based on the feature data, wherein the input data includes both the at least of the first set of intensity values and the at least part of the second set of intensity values, and the each intensity value of each wavelength included in the input data is for inputted input nodes of the deep learning algorithm and the number of intensity values of the input data corresponds to the number of the input nodes of the deep learning algorithm, and
provide, using the deep learning algorithm, diagnostic information on whether the target object is a lesion tissue related to a skin cancer based on the target input data, wherein the deep learning algorithm has been trained with training-data, which is related to plasma ablation induced at a cancerous sample and is labeled with a value indicating cancer, wherein the training data includes both of 1) a third spectrum data related to a continuum emission related to the plasma ablation induced at the cancerous sample and having overall intensity values within the specific wavelength band for the predetermined wavelength range and 2) a fourth spectrum data related to element specific emission related to the plasma ablation induced at the cancerous sample and having intensity values of spectral peaks within the specific wavelength band.
6. The diagnostic device of claim 5 , wherein the specific wavelength band is set to be between 200 nm and 1000 nm.
7. The diagnostic device of claim 5 , wherein a wavelength of the pulsed laser beam applied by the laser generation module is set to be a specific wavelength out of the specific wavelength band.
8. The diagnostic device of claim 5 , and wherein the skin cancer includes at least one of squamous cell carcinoma, basal cell carcinoma, or melanoma.