Systems and methods for dynamic Raman analysis of biological diseases and disorders and methods of feature engineering thereof

Abstract

The present disclosure provides methods and systems for predicting a diagnostic status of a subject with respect to a disease or condition. The method may include exposing a biological sample of the subject to a laser, acquiring a plurality of Raman spectra from the exposed biological sample, processing the plurality of Raman spectra to generate a spatial map of the plurality of Raman spectra, and predicting a diagnostic status of the subject with respect to a disease or condition based at least in part on the spatial map of the plurality of Raman spectra. The analysis may include determining temporal dynamics of the potential biological process.

Description

[0001] CROSS-REFERENCE TO RELATED APPLICATIONS

[0002] This application claims priority to U.S. Provisional Patent Application Serial No. 63 / 350,090, filed on June 8, 2022, which is hereby incorporated by reference in its entirety. Background Art

[0003] Dynamic biological responses may indicate potential biological processes that have structural and functional significance to humans. For example, abnormal or abnormal dynamic biological responses may be associated with many biological conditions, such as diseases and disorders. Examples of such biological conditions may include neurological conditions (e.g., autism spectrum disorders, schizophrenia, or attention deficit / hyperactivity disorder (ADHD)), neurodegenerative diseases (e.g., amyotrophic lateral sclerosis (ALS), Alzheimer's disease, Parkinson's disease, and Huntington's disease) and cancer (e.g., pediatric cancer). Summary of the invention

[0004] In view of the above background, accurate methods and systems for diagnosis of biological conditions, and especially non-invasive diagnosis, are needed. Such diagnosis can be based on accurate analysis of biomarkers detectable by non-invasive methods for diagnosis of biological conditions. The present disclosure provides improved systems and methods for accurate diagnosis of biological conditions, which are based on analysis of dynamic biological response data of biological samples obtained non-invasively from subjects. Such improved systems and methods for accurate diagnosis of biological conditions can be based on a combination of Raman analysis of biological samples and artificial intelligence data analysis. The present disclosure addresses these needs, for example, by providing biological sample biomarkers for diagnosis of biological conditions. Biological samples include human biological samples associated with incremental growth. Such biological samples can be hair shafts, teeth, and nails. The non-invasive biomarkers of the present disclosure can be used for diagnosis of young children, even infants less than one year old.

[0005] In one aspect, the present disclosure provides a method for predicting a subject's diagnostic status regarding a disease or condition of the subject, the method comprising: (a) exposing a biological sample of the subject to a light source, wherein the biological sample comprises a tooth sample, a hair sample, or a nail sample; (b) acquiring a plurality of Raman spectra from the exposed biological sample; (c) processing the plurality of Raman spectra to generate a spatial map of the plurality of Raman spectra; and (d) predicting the subject's diagnostic status regarding a disease or condition based at least in part on the spatial map of the plurality of Raman spectra. In some embodiments, the light source comprises a laser.

[0006] In some embodiments, the analysis determines the temporal dynamics of the underlying biological process. In some embodiments, the analysis includes reducing the dimensionality of multiple Raman spectra before processing (e.g., by independent component analysis). In some embodiments, the light signal is generated by a light source (e.g., a laser). In some embodiments, the biological sample includes a tooth sample. In some embodiments, the method further includes detecting or monitoring changes in a time stress curve (e.g., one or more traces), which indicate the temporal response of the subject. In some embodiments, the time response includes a biochemical response. In some embodiments, the time response includes a biological response, a physiological response, an anatomical response, a therapeutic response, a stress-related response, or a combination thereof. In some embodiments, a plurality of Raman spectra include about 200 to about 3700 wavenumbers. In some embodiments, the acquisition includes using a Raman spectroscopic microscope. In some embodiments, the Raman spectroscopic microscope includes a 50X air-coupled objective lens, a 63X water-immersion coupled objective lens, or any combination thereof. In some embodiments, the laser includes a wavelength of about 785nm, a wavelength of about 532nm, or any combination thereof. In some embodiments, an integration time of about 0.2 seconds to about 0.3 seconds is used for acquisition. In some embodiments, acquiring includes moving the biological sample in a step size of about 2 microns to about 5 microns after acquiring the Raman spectrum in the plurality of Raman spectra.

[0007] In some embodiments, the disease or condition includes: autism spectrum disorder (ASD), attention deficit / hyperactivity disorder (ADHD), amyotrophic lateral sclerosis (ALS), schizophrenia, irritable bowel disease (IBD), pediatric kidney disease, renal transplant rejection, pediatric cancer, or any combination thereof. In some embodiments, the disease or condition includes ASD. In some embodiments, the subject is a human. In some embodiments, the subject is an adult. In some embodiments, the age of the subject is between about 12 years old and about 5 years old. In some embodiments, the subject is less than about 12, 11, 10, 9, 8, 7, 5, 4, 3, 2 or 1 years old. In some embodiments, the subject is at least about 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11 or 12 years old. In some embodiments, at least a portion of the temporal Raman curve corresponds to the prenatal period of the subject.

[0008] In some embodiments, predicting the diagnostic status of a subject with respect to a disease or condition comprises processing a spatial graph using a trained model. In some embodiments, processing comprises extracting features from the spatial graph (e.g., by recursive quantitative analysis), and analyzing the features using a trained model. In some embodiments, processing comprises, for example, computationally analyzing the temporal dynamics derived from the spatial graph by applying dimensionality reduction techniques (including independent component analysis (ICA) and / or principal component analysis (PCA)), followed by applying recursive quantitative analysis (RQA) to extract computational features describing the dimensions derived from the ICA / PCA.

[0009] In some embodiments, processing includes determining one or more features of the temporal dynamics of one or more traces. In some embodiments, the temporal dynamics of the one or more traces are determined by a data analysis method. In some embodiments, the data analysis method applies one or more of the following operations and / or methods to the one or more traces: determination of a linear slope, determination of multiple nonlinear parameters describing the curvature of the one or more traces, determination of a sudden change in the intensity of the one or more traces, determination of one or more changes in the baseline intensity of the one or more traces, determination of a change in the frequency domain representation of the one or more traces, determination of a change in the power spectrum domain representation of the one or more traces, determination of one or more recursive quantitative analysis parameters, determination of one or more cross-recursive quantitative analysis parameters, determination of one or more joint recursive quantitative analysis parameters, determination of one or more multidimensional recursive quantitative analysis parameters, estimation of a Lyapunov spectrum, or determination of a maximum Lyapunov exponent.

[0010] In some embodiments, the trained model is selected from the group consisting of: a neural network algorithm, a support vector machine algorithm, a decision tree algorithm, an unsupervised clustering algorithm, a supervised clustering algorithm, a regression algorithm, a gradient boosting algorithm (e.g., a gradient boosting implementation of a machine learning algorithm, such as a gradient boosting decision tree), and any combination thereof. In some embodiments, the trained model includes a gradient boosting ensemble model. In some embodiments, the trained model is configured to process one or more features selected from the group consisting of: recursion rate, certainty, average diagonal length, maximum diagonal length, divergence, Shannon entropy of diagonal length, recursion trend, hierarchical, capture time, maximum vertical line length, Shannon entropy of vertical line length, average recursion time, Shannon entropy of recursion time, the number of most likely recursions, and / or any combination thereof. In some embodiments, the trained model is configured to process two or more features selected from the group consisting of: recursion rate, certainty, average diagonal length, maximum diagonal length, divergence, Shannon entropy of diagonal length, recursion trend, stratification, capture time, maximum vertical line length, Shannon entropy of vertical line length, average recursion time, Shannon entropy of recursion time, the number of most likely recursions, and / or any combination thereof.

[0011] In some embodiments, the trained model is configured to process one or more features of the temporal dynamics of one or more traces. In some embodiments, the temporal dynamics of the one or more traces are determined by a data analysis method. In some embodiments, the data analysis method applies one or more of the following operations and / or methods to the one or more traces: determination of a linear slope, determination of multiple nonlinear parameters describing the curvature of the one or more traces, determination of a sudden change in the intensity of the one or more traces, determination of one or more changes in the baseline intensity of the one or more traces, determination of a change in the frequency domain representation of the one or more traces, determination of a change in the power spectrum domain representation of the one or more traces, determination of one or more recursive quantitative analysis parameters, determination of one or more cross-recursive quantitative analysis parameters, determination of one or more joint recursive quantitative analysis parameters, determination of one or more multidimensional recursive quantitative analysis parameters, estimation of a Lyapunov spectrum, or determination of a maximum Lyapunov exponent.

[0012] In some embodiments, the method further comprises predicting the subject's diagnostic status with respect to the disease or condition using a model having a sensitivity of at least about 70%, 75%, 80%, 85% or 90% in predicting diagnostic status with respect to the disease or condition across a suitable cohort population (e.g., such as provided in the Examples section below).

[0013] In some embodiments, the method further comprises predicting the subject's diagnostic status with respect to the disease or condition using a model having a sensitivity of up to about 70%, 75%, 80%, 85% or 90% in predicting the diagnostic status with respect to the disease or condition across a suitable cohort population.

[0014] In some embodiments, the method further comprises predicting the subject's diagnostic status with respect to the disease or condition using a model having a specificity of at least about 70%, 75%, 80%, 85% or 90% in predicting the diagnostic status with respect to the disease or condition across a suitable cohort population.

[0015] In some embodiments, the method further comprises predicting the subject's diagnostic status with respect to the disease or condition using a model having a specificity of up to about 70%, 75%, 80%, 85% or 90% in predicting the diagnostic status with respect to the disease or condition across a suitable cohort population.

[0016] In some embodiments, the method further comprises predicting the subject's diagnostic status with respect to the disease or condition using a model having a positive predictive value of at least about 70%, 75%, 80%, 85% or 90% when predicting diagnostic status with respect to the disease or condition across a suitable cohort population.

[0017] In some embodiments, the method further comprises predicting the subject's diagnostic status with respect to the disease or condition using a model having a positive predictive value of up to about 70%, 75%, 80%, 85% or 90% when predicting diagnostic status with respect to the disease or condition across a suitable cohort population.

[0018] In some embodiments, the method further comprises predicting the subject's diagnostic status with respect to the disease or condition using a model having a negative predictive value of at least about 70%, 75%, 80%, 85% or 90% when predicting diagnostic status with respect to the disease or condition across a suitable cohort population.

[0019] In some embodiments, the method further comprises predicting the subject's diagnostic status with respect to the disease or condition using a model having a negative predictive value of up to about 70%, 75%, 80%, 85% or 90% when predicting diagnostic status with respect to the disease or condition across a suitable cohort population.

[0020] In some embodiments, the method further comprises predicting the subject's diagnostic status with respect to a disease or condition using a model that predicts the diagnostic status with respect to a disease or condition using an area under the receiver operating characteristic (AUROC) of at least about 0.65, at least about 0.70, at least about 0.75, at least about 0.80, at least about 0.82, at least about 0.84, at least about 0.86, at least about 0.88, or at least about 0.90 for a suitable cohort population.

[0021] In another aspect, the present disclosure provides an apparatus comprising one or more processors and a memory storing one or more programs for execution by the one or more processors, the one or more programs comprising instructions for: (a) sampling each corresponding position of a plurality of positions along a reference line on a biological sample of a subject associated with a Raman feature of the subject, thereby obtaining a plurality of Raman spectra, each Raman spectrum of the plurality of Raman spectra corresponding to a different position of the plurality of positions, and each position of the plurality of positions representing a different growth period of the biological sample of the subject associated with the Raman feature; (b) analyzing each Raman spectrum of the plurality of Raman spectra across the reference line on the biological sample, thereby obtaining a first data set; (c) deriving a corresponding second data set from the corresponding plurality of Raman spectral measurements, each corresponding feature in the corresponding feature set being determined by a sequential change in the Raman spectrum; and (d) processing the features using a trained model to predict a diagnostic status of the subject with respect to a disease or condition associated with the Raman feature. In some embodiments, the corresponding second data set is derived by applying a recursive quantitative analysis or correlation method to the corresponding plurality of Raman spectral measurements. In some embodiments, analysis of the Raman spectrum includes cosmic ray removal, background correction, normalization, peak fitting, or any combination thereof.

[0022] In some embodiments, the biological sample comprises a tooth sample, a hair sample, a nail sample, or any combination thereof. In some embodiments, the instructions further comprise detecting or monitoring changes in Raman spectra across multiple positions, which indicate a temporal response of the subject. In some embodiments, the temporal response comprises a biological response, a physiological response, an anatomical response, a therapeutic response, a stress-related response, or a combination thereof. In some embodiments, a plurality of Raman spectra comprise about 200 to about 3700 wave numbers. In some embodiments, sampling comprises using a Raman spectroscopic microscope. In some embodiments, the Raman spectroscopic microscope comprises a 50X air-coupled objective, a 63X water-immersion coupled objective, or any combination thereof. In some embodiments, sampling comprises exposing the biological sample to a light source to generate a Raman spectrum in a plurality of Raman spectra at a plurality of positions. In some embodiments, the light source comprises a laser, wherein the laser comprises a wavelength of about 785nm, a wavelength of about 532nm, or any combination thereof. In some embodiments, the instructions further comprise translation, wherein the translation comprises moving the biological sample from a first position in a plurality of positions to a second position in a step of about 2 microns to about 5 microns after collecting the Raman spectra in the plurality of Raman spectra. In some embodiments, the translation is performed using an integration time of about 0.2 seconds to about 0.3 seconds. In some embodiments, the disease or condition includes: autism spectrum disorder (ASD), attention deficit / hyperactivity disorder (ADHD), amyotrophic lateral sclerosis (ALS), schizophrenia, irritable bowel disease (IBD), pediatric kidney disease, renal transplant rejection, pediatric cancer or any combination thereof. In some embodiments, the disease or condition includes ASD. In some embodiments, the prediction of the diagnostic state of the subject about the disease or condition includes using a trained model to process the change of the Raman spectrum across multiple positions. In some embodiments, the trained model is selected from the group consisting of the following: neural network algorithm, support vector machine algorithm, decision tree algorithm, unsupervised clustering algorithm, supervised clustering algorithm, regression algorithm, gradient boosting algorithm and any combination thereof. In some embodiments, the trained model includes a gradient boosting integrated model. In some embodiments, the trained model is configured to process one or more features selected from the group consisting of the following: hierarchical, entropy, capture time (TT), mean diagonal length (MDL), recursive time (RT), Vmax, certainty, Lmax and any combination thereof. In some embodiments, the trained model is configured to process two or more features selected from the group consisting of: hierarchicality, entropy, capture time (TT), mean diagonal length (MDL), recursion time (RT), Vmax, certainty, Lmax, and any combination thereof.

[0023] In some embodiments, the trained model is configured to process one or more features of the temporal dynamics of one or more traces. In some embodiments, the temporal dynamics of the one or more traces are determined by a data analysis method. In some embodiments, the data analysis method applies one or more of the following operations and / or methods to the one or more traces: determination of a linear slope, determination of multiple nonlinear parameters describing the curvature of the one or more traces, determination of a sudden change in the intensity of the one or more traces, determination of one or more changes in the baseline intensity of the one or more traces, determination of a change in the frequency domain representation of the one or more traces, determination of a change in the power spectrum domain representation of the one or more traces, determination of one or more recursive quantitative analysis parameters, determination of one or more cross-recursive quantitative analysis parameters, determination of one or more joint recursive quantitative analysis parameters, determination of one or more multidimensional recursive quantitative analysis parameters, estimation of a Lyapunov spectrum, or determination of a maximum Lyapunov exponent.

[0024] In another aspect, the present disclosure provides a non-transitory computer-readable storage medium and one or more computer programs embedded therein for classification, the one or more computer programs comprising instructions, which when executed by a computer system, cause the computer system to perform a method comprising: (a) sampling each corresponding position of a plurality of positions along a reference line on a biological sample of a subject associated with a Raman feature of the subject, thereby obtaining a plurality of Raman spectra, each Raman spectrum of the plurality of Raman spectra corresponding to a different position of the plurality of positions, and each position of the plurality of positions represents a different growth period of the biological sample of the subject associated with the Raman feature; (b) analyzing each Raman spectrum of the plurality of Raman spectra across the reference line on the biological sample, thereby obtaining a first data set; (c) deriving a corresponding second data set from the corresponding plurality of Raman spectral measurement results, each corresponding feature in the corresponding feature set being determined by a sequential change in the Raman spectrum; and (d) processing the features using a trained model to predict a diagnostic status of the subject with respect to a disease or condition associated with the Raman feature. In some embodiments, the corresponding second data set is derived by applying a recursive quantitative analysis or correlation method to the corresponding plurality of Raman spectral measurement results. In some embodiments, analysis of the Raman spectrum includes cosmic ray removal, background correction, normalization, peak fitting, or any combination thereof.

[0025] In some embodiments, the biological sample comprises a tooth sample, a hair sample, a nail sample, or any combination thereof. In some embodiments, the method further comprises detecting or monitoring changes in Raman spectra across multiple positions, which indicate a temporal response (i.e., one or more traces) of the subject. In some embodiments, the temporal response comprises a biological response, a physiological response, an anatomical response, a therapeutic response, a stress-related response, or a combination thereof. In some embodiments, a plurality of Raman spectra comprise about 200 to about 3700 wavenumbers. In some embodiments, sampling comprises using a Raman spectroscopic microscope. In some embodiments, the Raman spectroscopic microscope comprises a 50X air-coupled objective lens, a 63X water-immersion coupled objective lens, or any combination thereof. In some embodiments, sampling comprises exposing the biological sample to a light source to generate a Raman spectrum in a plurality of Raman spectra at a plurality of positions. In some embodiments, the light source comprises a laser, wherein the laser comprises a wavelength of about 785nm, a wavelength of about 532nm, or any combination thereof. In some embodiments, the instruction further comprises translation, wherein the translation comprises moving the biological sample from a first position in a plurality of positions to a second position in a step of about 2 microns to about 5 microns after collecting the Raman spectra in the plurality of Raman spectra. In some embodiments, the translation is performed using an integration time of about 0.2 seconds to about 0.3 seconds. In some embodiments, the disease or condition includes: autism spectrum disorder (ASD), attention deficit / hyperactivity disorder (ADHD), amyotrophic lateral sclerosis (ALS), schizophrenia, irritable bowel disease (IBD), pediatric kidney disease, renal transplant rejection, pediatric cancer or any combination thereof. In some embodiments, the disease or condition includes ASD. In some embodiments, the prediction of the diagnostic state of the subject about the disease or condition includes using a trained model to process the change of the Raman spectrum across multiple positions. In some embodiments, the trained model is selected from the group consisting of the following: neural network algorithm, support vector machine algorithm, decision tree algorithm, unsupervised clustering algorithm, supervised clustering algorithm, regression algorithm, gradient boosting algorithm and any combination thereof. In some embodiments, the trained model includes a gradient boosting integrated model. In some embodiments, the trained model is configured to process one or more features selected from the group consisting of the following: hierarchical, entropy, capture time (TT), mean diagonal length (MDL), recursive time (RT), Vmax, certainty, Lmax and any combination thereof. In some embodiments, the trained model is configured to process two or more features selected from the group consisting of: hierarchicality, entropy, capture time (TT), mean diagonal length (MDL), recursion time (RT), Vmax, certainty, Lmax, and any combination thereof.

[0026] In some embodiments, the trained model is configured to process one or more features of the temporal dynamics of one or more traces. In some embodiments, the temporal dynamics of the one or more traces are determined by a data analysis method. In some embodiments, the data analysis method applies one or more of the following operations and / or methods to the one or more traces: determination of a linear slope, determination of multiple nonlinear parameters describing the curvature of the one or more traces, determination of a sudden change in the intensity of the one or more traces, determination of one or more changes in the baseline intensity of the one or more traces, determination of a change in the frequency domain representation of the one or more traces, determination of a change in the power spectrum domain representation of the one or more traces, determination of one or more recursive quantitative analysis parameters, determination of one or more cross-recursive quantitative analysis parameters, determination of one or more joint recursive quantitative analysis parameters, determination of one or more multidimensional recursive quantitative analysis parameters, estimation of a Lyapunov spectrum, or determination of a maximum Lyapunov exponent.

[0027] In another aspect, the present disclosure provides a method for training a model, the method comprising: at a computer system having one or more processors and a memory storing one or more programs for execution by the one or more processors: (a) for each corresponding training subject in a plurality of training subjects, wherein a first subset of the training subjects in the plurality of training subjects have a first diagnostic state corresponding to a first biological condition associated with a Raman feature, and a second subset of the training subjects in the plurality of training subjects have a second diagnostic state corresponding to the first biological condition not associated with the Raman feature: (i) sampling each corresponding position in a plurality of positions along a reference line on a biological sample of the subject associated with the Raman feature of the subject, thereby obtaining a plurality of Raman spectra, each of the plurality of Raman spectra corresponding to a different position in the plurality of positions, and the plurality of positions Each position in represents a different growth period of the biological sample associated with the Raman feature; (ii) analyzing each Raman spectrum across a reference line on the biological sample to obtain a first data set; and (iii) deriving a corresponding second data set from the corresponding plurality of Raman spectra, each corresponding feature in the corresponding feature set being determined by a sequential change in the Raman spectrum; and (b) training an untrained or partially untrained model with (i) the corresponding feature set of each corresponding second data set of each training subject in the plurality of training subjects and (ii) the corresponding diagnostic state of each training subject in the plurality of training subjects selected from the first diagnostic state and the second diagnostic state, thereby obtaining a trained model, the trained model providing an indication of whether the test subject has the first biological condition associated with the Raman feature based on the value of the feature in the feature set obtained from the biological sample associated with the Raman feature of the test subject. In some embodiments, the corresponding second data set is derived by applying a recursive quantitative analysis or related method to the corresponding plurality of Raman spectral measurements. In some embodiments, the analysis of the Raman spectra includes cosmic ray removal, background correction, normalization, peak fitting, or any combination thereof.

[0028] In some embodiments, the trained model is a neural network algorithm, a support vector machine algorithm, a decision tree algorithm, an unsupervised clustering model algorithm, a supervised clustering model algorithm, a regression model or a gradient boosting algorithm (e.g., a gradient boosting implementation of a machine learning algorithm, such as a gradient boosting decision tree). In some embodiments, the trained model is a polynomial classifier. In some embodiments, the trained model is a binomial classifier. In some embodiments, the trained model is a regressor. In some embodiments, the first biological condition is selected from the group consisting of: autism spectrum disorder (ASD), attention deficit / hyperactivity disorder (ADHD), amyotrophic lateral sclerosis (ALS), schizophrenia, irritable bowel disease (IBD), pediatric kidney disease, renal transplant rejection and pediatric cancer.

[0029] In some embodiments, evaluating the first biological condition associated with the Raman feature of the test subject further includes distinguishing between the presence of the first biological condition associated with the Raman feature and the absence of the first biological condition associated with the Raman feature. In some embodiments, evaluating the test subject for the first biological condition associated with the Raman feature further includes: distinguishing the first biological condition associated with the Raman feature from the second biological condition associated with the Raman feature that is different from the first biological condition associated with the Raman feature. In some embodiments, the first biological condition is autism spectrum disorder, and the second biological condition is neurotypical development; that is, there is no neurodevelopmental disorder. In some embodiments, the first biological condition is autism spectrum disorder, and the second biological condition is attention deficit / hyperactivity disorder. In some embodiments, the test subject is a person. In some embodiments, the test subject is an adult. In some embodiments, the age of the person is between about 12 years old and about 5 years old. In some embodiments, the subject is less than about 12, 11, 10, 9, 8, 7, 5, 4, 3, 2 or 1 years old. In some embodiments, the subject is at least about 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, or 12 years old. In some embodiments, at least a portion of the temporal profile of the Raman curve corresponds to the prenatal period of the subject.

[0030] In some embodiments, the corresponding biological sample associated with the Raman signature of the corresponding training subject is selected from the group consisting of a hair shaft, a tooth, and a nail. In some embodiments, the corresponding biological sample associated with the Raman signature of the corresponding training subject is a hair shaft, and wherein the reference line corresponds to a longitudinal direction of the hair shaft. In some embodiments, the corresponding biological sample associated with the Raman signature of the corresponding training subject is a tooth, and wherein the reference line corresponds to a direction across a growth zone, including a new growth line of a tooth. In some embodiments, the corresponding plurality of positions are ordered such that a first position along the corresponding plurality of positions of the corresponding biological sample of the corresponding training subject corresponds to a position closest to the tip of the corresponding biological sample of the corresponding training subject. In some embodiments, each trace in the corresponding plurality of Raman spectroscopy measurements includes a plurality of data points, each data point being an instance of a corresponding position in the plurality of positions. In some embodiments, the corresponding feature set is selected from the group consisting of: recursion rate, certainty, average diagonal length, maximum diagonal length, divergence, Shannon entropy of diagonal length, recursion trend, hierarchical, capture time, maximum vertical line length, Shannon entropy of vertical line length, average recursion time, Shannon entropy of recursion time, number of most likely recursions, and / or any combination thereof. In some embodiments, the corresponding plurality of positions includes at least 1000, 1500, 2000, 2500, 3000, 3500, 4000, 4500, 5000, 5500, 6000, 6500, 7000, 7500, 8000, 8500, 9000, 9500, 10000 or more than 10000 positions.

[0031] In some embodiments, the corresponding feature set is selected from a set of time-dynamic features of one or more traces. In some embodiments, the time-dynamic features of the one or more traces are determined by a data analysis method. In some embodiments, the data analysis method applies one or more of the following operations and / or methods to one or more traces: determination of a linear slope, determination of multiple nonlinear parameters describing the curvature of the one or more traces, determination of a sudden change in the intensity of the one or more traces, determination of one or more changes in the baseline intensity of the one or more traces, determination of a change in the frequency domain representation of the one or more traces, determination of a change in the power spectrum domain representation of the one or more traces, determination of one or more recursive quantitative analysis parameters, determination of one or more cross-recursive quantitative analysis parameters, determination of one or more joint recursive quantitative analysis parameters, determination of one or more multidimensional recursive quantitative analysis parameters, estimation of a Lyapunov spectrum, or determination of a maximum Lyapunov exponent.

[0032] Another aspect of the present disclosure provides a non-transitory computer-readable medium comprising machine-executable code that, when executed by one or more computer processors, implements any of the methods described above or elsewhere herein.

[0033] Another aspect of the present disclosure provides a system comprising one or more computer processors and a computer memory coupled thereto. The computer memory includes machine executable code that, when executed by the one or more computer processors, implements any of the methods described above or elsewhere herein.

[0034] For those skilled in the art, additional aspects and advantages of the present disclosure will become apparent from the following detailed description, wherein only illustrative embodiments of the present disclosure are shown and described. As will be appreciated, the present disclosure is capable of other and different embodiments, and its several details can be modified in various obvious aspects, all without departing from the present disclosure. Therefore, the drawings and description will be regarded as illustrative rather than restrictive in nature.

[0035] Incorporated by Reference

[0036] All publications, patents, and patent applications mentioned in this specification are incorporated herein by reference to the same extent as if each individual publication, patent, or patent application was specifically and individually indicated to be incorporated by reference. To the extent publications and patents or patent applications incorporated by reference contradict the disclosure contained in the specification, the specification is intended to supersede and / or take precedence over any such contradictory material. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] The novel features of the invention are set forth with particularity in the appended claims. A better understanding of the features and advantages of the invention will be obtained by reference to the following detailed description which sets forth illustrative embodiments in which the principles of the invention are utilized, and in the accompanying drawings (also referred to herein as "Figures" and "FIG."):

[0038] Figure 1 An example of a block diagram of a computing device 100 of the present disclosure is shown.

[0039] FIG. 2A to FIG. 2C Hair samples of subjects are shown ( Figure 2A )、tooth samples( Figure 2B ) and nail samples ( Figure 2C ) icon.

[0040] Figure 3 A flow chart of a method 300 for assessing a biological condition in a subject is shown.

[0041] Figure 4A computer system is shown that is programmed or otherwise configured to implement the methods provided herein.

[0042] Figure 5 An example of model accuracy for predicting diagnostic status for autism spectrum disorder (ASD) using features derived from ICA-derived dimensions applied to Raman waveforms by RQA is shown, as shown by an experimental receiver operating characteristic (ROC) curve for evaluating the accuracy of the disclosed method for assessing autism spectrum disorder in a subject. Device performance is measured by calculating the area under the curve (AUC) of the ROC plot, which provides a measure of performance at different classification thresholds; here, the AUC is 0.86, indicating very accurate predictive performance.

[0043] Figure 6 An example of model accuracy for predicting diagnostic status for amyotrophic lateral sclerosis (ALS) using features derived from ICA-derived dimensions of RQA applied to Raman waveforms is shown, as shown in an experimental receiver operating characteristic (ROC) curve for evaluating the accuracy of the disclosed method for assessing autism spectrum disorder in a subject. Device performance is measured by calculating the area under the curve (AUC) of the ROC plot, which provides a measure of performance at different classification thresholds; here, the AUC is 0.88, indicating very accurate predictive performance. DETAILED DESCRIPTION

[0044] Although various embodiments of the present invention have been shown and described herein, it will be apparent to those skilled in the art that such embodiments are provided as examples only. Without departing from the present invention, those skilled in the art may appreciate that many variations, changes and substitutions may be made. It should be understood that various alternatives to the embodiments of the present invention described herein may be employed.

[0045] Dynamic biological responses may indicate potential biological processes that have structural and functional significance to humans. For example, abnormal or abnormal dynamic biological responses may be associated with many biological conditions, such as diseases and disorders. Examples of such biological conditions may include neurological conditions (e.g., autism spectrum disorders, schizophrenia, or attention deficit / hyperactivity disorder (ADHD)), neurodegenerative diseases (e.g., amyotrophic lateral sclerosis (ALS), Alzheimer's disease, Parkinson's disease, and Huntington's disease) and cancer (e.g., pediatric cancer).

[0046] In view of the above background, accurate methods and systems for diagnosis of biological conditions and especially non-invasive diagnosis are needed. Such diagnosis can be based on accurate analysis of biomarkers that can be detected by non-invasive methods for diagnosis of biological conditions. The present disclosure provides improved systems and methods for accurate diagnosis of biological conditions, which are based on analysis of dynamic biological response data of biological samples obtained non-invasively from subjects. In some embodiments, such improved systems and methods for accurate diagnosis of biological conditions are based on a combination of Raman analysis of biological samples and artificial intelligence data analysis. The present disclosure addresses these needs, for example, by providing biological sample biomarkers for diagnosis of biological conditions. Biological samples include human biological samples associated with incremental growth. Such biological samples can be hair shafts, teeth and nails. The non-invasive biomarkers of the present disclosure can be used for diagnosis of young children, even infants less than one year old. In some cases, the age of the child is between about 12 years old and about 5 years old. In some embodiments, the child is less than about 12, 11, 10, 9, 8, 7, 5, 4, 3, 2 or 1 years old. In some embodiments, the child is at least about 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, or 12 years old.

[0047] In one aspect, the present disclosure provides a method for predicting a subject's diagnostic status with respect to a disease or condition, the method comprising: (a) exposing a biological sample of the subject to a light source, wherein the biological sample comprises a tooth sample, a hair sample, or a nail sample; (b) acquiring a plurality of Raman spectra from the exposed biological sample; (c) processing the plurality of Raman spectra to generate a spatial map of the plurality of Raman spectra; and (d) predicting the subject's diagnostic status with respect to the disease or condition based at least in part on the spatial map of the plurality of Raman spectra. In some embodiments, the light source comprises a laser.

[0048] In some embodiments, the analysis determines the temporal dynamics of the underlying biological process. In some embodiments, the analysis includes reducing the dimension of multiple Raman spectra before processing (e.g., by independent component analysis). In some embodiments, the light signal is generated by a light source (e.g., a laser). In some embodiments, the biological sample includes a tooth sample. In some embodiments, the method further includes detecting or monitoring changes in a time stress curve (e.g., one or more traces described elsewhere herein), which indicate the temporal response of the subject. In some embodiments, the time response includes a biochemical response. In some embodiments, the time response includes a biological response, a physiological response, an anatomical response, a therapeutic response, a stress-related response, or a combination thereof. In some embodiments, a plurality of Raman spectra include about 200 to about 3700 wavenumbers. In some embodiments, the acquisition includes using a Raman spectroscopic microscope. In some embodiments, the Raman spectroscopic microscope includes a 50X air-coupled objective lens, a 63X water-immersion coupled objective lens, or any combination thereof. In some embodiments, the laser includes a wavelength of about 785nm, a wavelength of about 532nm, or any combination thereof. In some embodiments, an integration time of about 0.2 seconds to about 0.3 seconds is used for acquisition. In some embodiments, acquiring includes moving the biological sample in a step size of about 2 microns to about 5 microns after acquiring the Raman spectrum in the plurality of Raman spectra.

[0049] In some embodiments, the systems and methods disclosed herein use a single Raman spectrum, or use it in combination with other techniques. In some embodiments, such techniques include laser ablation inductively coupled plasma mass spectrometry (LA-ICP-MS), C-reactive immunohistochemical fluorescence staining, etc. In some embodiments, the combined technology improves the diagnostic accuracy or precision of a given technique alone. In some embodiments, the addition of LA-ICP-MS provides a plurality of non-invasive metal metabolism biomarkers for a given biological sample, which supplement the diagnostic capabilities of Raman spectroscopy. In some embodiments, metal metabolism biomarkers include zinc, tin, magnesium, copper, iodide, lithium, aluminum, phosphorus, sulfur, calcium, chromium, manganese, iron, cobalt, nickel, arsenic, strontium, cadmium, tin, iodine, barium, mercury, lead, bismuth, molybdenum, or any combination thereof. In some embodiments, the addition of C-reactive protein immunohistochemical fluorescence provides temporal fluctuations of inflammation to supplement the diagnostic capabilities of Raman spectroscopy. In some embodiments, a plurality of metal metabolism biomarkers include at least 2, at least 5, or at least 10 metal metabolism biomarkers. In some embodiments, a plurality of metal metabolism biomarkers include no more than 20, no more than 10 or no more than 5 metal metabolism biomarkers. In some embodiments, a plurality of metal metabolism biomarkers are composed of 2 to 5, 3 to 10 or 8 to 20 metal metabolism biomarkers. In some embodiments, a plurality of metal metabolism biomarkers fall into another range starting from no less than 2 metal metabolism biomarkers to no more than 20 metal metabolism biomarkers. In some embodiments, the plurality of Raman spectra include at least 2, at least 5 or at least 10 Raman spectra. In some embodiments, the plurality of Raman spectra include no more than 20, no more than 10 or no more than 5 Raman spectra. In some embodiments, the plurality of Raman spectra are composed of 2 to 5, 3 to 10 or 8 to 20 Raman spectra. In some embodiments, the plurality of Raman spectra fall into another range starting from no less than 2 spectra to no more than 20 Raman spectra.

[0050] In some embodiments, the disease or condition includes: autism spectrum disorder (ASD), attention deficit / hyperactivity disorder (ADHD), amyotrophic lateral sclerosis (ALS), schizophrenia, irritable bowel disease (IBD), pediatric kidney disease, renal transplant rejection, pediatric cancer, or any combination thereof. In some embodiments, the disease or condition includes ASD. In some embodiments, the subject is a human. In some embodiments, the subject is an adult. In some embodiments, the age of the subject is between about 12 years old and about 5 years old. In some embodiments, the subject is less than about 12, 11, 10, 9, 8, 7, 5, 4, 3, 2 or 1 years old. In some embodiments, the subject is at least about 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11 or 12 years old. In some embodiments, at least a portion of the temporal Raman curve corresponds to the prenatal period of the subject.

[0051] In some embodiments, predicting the diagnostic status of a subject with respect to a disease or condition comprises processing a spatial graph using a trained model. In some embodiments, processing comprises extracting features from the spatial graph (e.g., by recursive quantitative analysis), and analyzing the features using a trained model. In some embodiments, processing comprises, for example, computationally analyzing the temporal dynamics derived from the spatial graph by applying dimensionality reduction techniques (including independent component analysis (ICA) and / or principal component analysis (PCA)), followed by applying recursive quantitative analysis (RQA) to extract computational features describing the dimensions derived from the ICA / PCA.

[0052] In some embodiments, the trained model includes a plurality of parameters, where the term "parameter" refers to any coefficient in the model or similarly, any value of an internal or external element (e.g., weight and / or hyperparameter) (e.g., where the model is a regressor or classifier) ​​that can affect (e.g., modify, customize and / or adjust) one or more inputs, outputs and / or functions in the model. For example, in some embodiments, a parameter of a model refers to any coefficient, weight and / or hyperparameter that can be used to control, modify, customize and / or adjust the behavior, learning and / or performance of the model. In some cases, a parameter is used to increase or decrease the influence of an input (e.g., feature) on a model. As a non-limiting example, in some embodiments, a parameter is used to increase or decrease the influence of a node (e.g., of a neural network), where the node includes one or more activation functions. The assignment of parameters to specific inputs, outputs and / or functions of a model is not limited to any one paradigm of a given model, but can be used for any suitable model to achieve the desired performance. In some embodiments, a parameter has a fixed value. In some embodiments, the value of a parameter can be adjusted manually and / or automatically. In some embodiments, the values ​​of the parameters are modified through a validation and / or training process of the model (eg, through error minimization and / or back propagation methods). In some embodiments, the models of the present disclosure include multiple parameters. In some embodiments, the plurality of parameters associated with a model (e.g., an untrained, partially trained, or fully trained model) is n parameters, where: n≥2; n≥5; n≥10; n≥25; n≥40; n≥50; n≥75; n≥100; n≥125; n≥150; n≥200; n≥225; n≥250; n≥350; n≥500; n≥600; n≥750; n≥1,000; n≥2,000; n≥4,000; n≥5,000; n≥7,500; n≥10,000; n≥20,000; n≥40,000; n≥75,000; n≥100,000; n≥200,000; n≥500,000, n≥1×10 6 , n≥5×10 6 or n ≥ 1 × 10 7 In some embodiments, n is between 10,000 and 1×10 7 Between 100,000 and 5×10 6 or between 500,000 and 1×10 6 In some embodiments, the plurality of parameters includes at least 10, at least 100, at least 1000, at least 10,000, at least 100,000, at least 1×10 6 At least 5×10 6 At least 1×10 7 At least 5×10 7 or at least 1×108 In some embodiments, the plurality of parameters includes 1×10 9 No more than 1×10 8 No more than 1×10 7 No more than 1×10 6 In some embodiments, the plurality of parameters ranges from 10 to 10,000, 100 to 100,000, 1000 to 1×10 6 100,000 to 1×10 7 pcs, 1×10 6 Up to 1×10 8 or 1×10 7 Up to 1×10 9 In some embodiments, the plurality of parameters falls within a range from no less than 10 parameters to no more than 1×10 9 The next parameter ends in another scope.

[0053] In some embodiments, the trained model is selected from the group consisting of: a neural network algorithm, a support vector machine algorithm, a decision tree algorithm, an unsupervised clustering algorithm, a supervised clustering algorithm, a regression algorithm, a gradient boosting algorithm (e.g., a gradient boosting implementation of a machine learning algorithm, such as a gradient boosting decision tree), and any combination thereof. In some embodiments, the trained model includes a gradient boosting ensemble model. In some embodiments, the trained model is configured to process one or more features selected from the group consisting of: recursion rate, certainty, average diagonal length, maximum diagonal length, divergence, Shannon entropy of diagonal length, recursion trend, hierarchical, capture time, maximum vertical line length, Shannon entropy of vertical line length, average recursion time, Shannon entropy of recursion time, the number of most likely recursions, and / or any combination thereof. In some embodiments, the trained model is configured to process two or more features selected from the group consisting of: recursion rate, certainty, average diagonal length, maximum diagonal length, divergence, Shannon entropy of diagonal length, recursion trend, stratification, capture time, maximum vertical line length, Shannon entropy of vertical line length, average recursion time, Shannon entropy of recursion time, the number of most likely recursions, and / or any combination thereof.

[0054] In some embodiments, the trained model is configured to process one or more features of the temporal dynamics of one or more traces. In some embodiments, the temporal dynamics of the one or more traces are determined by a data analysis method. In some embodiments, the data analysis method applies one or more of the following operations and / or methods to the one or more traces: determination of a linear slope, determination of multiple nonlinear parameters describing the curvature of the one or more traces, determination of a sudden change in the intensity of the one or more traces, determination of one or more changes in the baseline intensity of the one or more traces, determination of a change in the frequency domain representation of the one or more traces, determination of a change in the power spectrum domain representation of the one or more traces, determination of one or more recursive quantitative analysis parameters, determination of one or more cross-recursive quantitative analysis parameters, determination of one or more joint recursive quantitative analysis parameters, determination of one or more multidimensional recursive quantitative analysis parameters, estimation of a Lyapunov spectrum, or determination of a maximum Lyapunov exponent.

[0055] In some embodiments, the method further comprises predicting the diagnostic status of the subject with a sensitivity of at least about 80% about the disease or condition. In some embodiments, the method further comprises predicting the diagnostic status of the subject with a sensitivity of at most about 80% about the disease or condition. In some embodiments, the method further comprises predicting the diagnostic status of the subject with a specificity of at least about 80% about the disease or condition. In some embodiments, the method further comprises predicting the diagnostic status of the subject with a specificity of at most about 80% about the disease or condition. In some embodiments, the method further comprises predicting the diagnostic status of the subject with a positive predictive value of at least about 80% about the disease or condition. In some embodiments, the method further comprises predicting the diagnostic status of the subject with a positive predictive value of at most about 80% about the disease or condition. In some embodiments, the method further comprises predicting the diagnostic status of the subject with a negative predictive value of at least about 80% about the disease or condition. In some embodiments, the method further comprises predicting the diagnostic status of the subject with a negative predictive value of at most about 80% about the disease or condition. In some embodiments, the method further comprises predicting the diagnostic status of the subject with a negative predictive value of at most about 80% about the disease or condition. In some embodiments, the method further comprises predicting the diagnostic status of the subject with a receiver operating characteristic area under (AUROC) of at least about 0.80 about the disease or condition.

[0056] In another aspect, the present disclosure provides an apparatus comprising one or more processors and a memory storing one or more programs for execution by the one or more processors, the one or more programs comprising instructions for: (a) sampling each corresponding position of a plurality of positions along a reference line on a biological sample of a subject associated with a Raman feature of the subject, thereby obtaining a plurality of Raman spectra, each Raman spectrum of the plurality of Raman spectra corresponding to a different position of the plurality of positions, and each position of the plurality of positions representing a different growth period of the biological sample of the subject associated with the Raman feature; (b) analyzing each Raman spectrum of the plurality of Raman spectra across the reference line on the biological sample, thereby obtaining a first data set; (c) deriving a corresponding second data set from the corresponding plurality of Raman spectral measurements, each corresponding feature in the corresponding feature set being determined by a sequential change in the Raman spectrum; and (d) processing the features using a trained model to predict a diagnostic status of the subject with respect to a disease or condition associated with the Raman feature. In some embodiments, the corresponding second data set is derived by applying a recursive quantitative analysis or correlation method to the corresponding plurality of Raman spectral measurements. In some embodiments, analysis of the Raman spectrum includes cosmic ray removal, background correction, normalization, peak fitting, or any combination thereof.

[0057] In another aspect, the present disclosure provides a non-transitory computer-readable storage medium and one or more computer programs embedded therein for classification, the one or more computer programs comprising instructions, which when executed by a computer system, cause the computer system to perform a method comprising: (a) sampling each corresponding position of a plurality of positions along a reference line on a biological sample of a subject associated with a Raman feature of the subject, thereby obtaining a plurality of Raman spectra, each Raman spectrum of the plurality of Raman spectra corresponding to a different position of the plurality of positions, and each position of the plurality of positions represents a different growth period of the biological sample of the subject associated with the Raman feature; (b) analyzing each Raman spectrum of the plurality of Raman spectra across the reference line on the biological sample, thereby obtaining a first data set; (c) deriving a corresponding second data set from the corresponding plurality of Raman spectral measurement results, each corresponding feature in the corresponding feature set being determined by a sequential change in the Raman spectrum; and (d) processing the features using a trained model to predict a diagnostic status of the subject with respect to a disease or condition associated with the Raman feature. In some embodiments, the corresponding second data set is derived by applying a recursive quantitative analysis or correlation method to the corresponding plurality of Raman spectral measurement results. In some embodiments, analysis of the Raman spectrum includes cosmic ray removal, background correction, normalization, peak fitting, or any combination thereof.

[0058] In another aspect, the present disclosure provides a method for training a model, the method comprising: at a computer system having one or more processors and a memory storing one or more programs for execution by the one or more processors: (a) for each corresponding training subject in a plurality of training subjects, wherein a first subset of the training subjects in the plurality of training subjects have a first diagnostic state corresponding to a first biological condition associated with a Raman feature, and a second subset of the training subjects in the plurality of training subjects have a second diagnostic state corresponding to the first biological condition not associated with the Raman feature: (i) sampling each corresponding position in a plurality of positions along a reference line on a biological sample of the subject associated with the Raman feature of the subject, thereby obtaining a plurality of Raman spectra, each of the plurality of Raman spectra corresponding to a different position in the plurality of positions, and the plurality of positions Each position in represents a different growth period of the biological sample associated with the Raman feature; (ii) analyzing each Raman spectrum across a reference line on the biological sample to obtain a first data set; and (iii) deriving a corresponding second data set from the corresponding plurality of Raman spectra, each corresponding feature in the corresponding feature set being determined by a sequential change in the Raman spectrum; and (b) training an untrained or partially untrained model with (i) the corresponding feature set of each corresponding second data set of each training subject in the plurality of training subjects and (ii) the corresponding diagnostic state of each training subject in the plurality of training subjects selected from the first diagnostic state and the second diagnostic state, thereby obtaining a trained model, the trained model providing an indication of whether the test subject has the first biological condition associated with the Raman feature based on the value of the feature in the feature set obtained from the biological sample associated with the Raman feature of the test subject. In some embodiments, the corresponding second data set is derived by applying a recursive quantitative analysis or related method to the corresponding plurality of Raman spectral measurements. In some embodiments, the analysis of the Raman spectra includes cosmic ray removal, background correction, normalization, peak fitting, or any combination thereof.

[0059] In certain embodiments, corresponding subject (for example, test subject) is selected from a plurality of subjects. In certain embodiments, the plurality of subjects include at least 2, at least 5, at least 10, at least 20, at least 50, at least 100 or at least 500 subjects. In certain embodiments, the plurality of subjects include no more than 1000, no more than 500, no more than 100, no more than 50, no more than 20 or no more than 10 subjects. In certain embodiments, the plurality of subjects are composed of 2 to 10, 5 to 20, 10 to 100 or 100 to 1000 subjects. In certain embodiments, the plurality of subjects fall into another range starting from no less than 2 subjects to no more than 1000 subjects.

[0060] In certain embodiments, the plurality of training subjects include at least 2, at least 5, at least 10, at least 20, at least 50, at least 100, at least 500, at least 1000, at least 5000 or at least 100,000 training subjects. In certain embodiments, the plurality of training subjects include no more than 1,000,000, no more than 100,000, no more than 10,000, no more than 100, no more than 500, no more than 100, no more than 20 or no more than 10 training subjects. In certain embodiments, the plurality of training subjects are composed of 2 to 1000, 500 to 10,000, 10,000 to 100,000 or 100,000 to 1,000,000 training subjects. In certain embodiments, the plurality of training subjects fall into another range starting from no less than 2 training subjects to no more than 1,000,000 training subjects.

[0061] In some embodiments, the corresponding subset of the training subjects in the multiple training subjects (e.g., the first subset and / or the second subset) includes at least 2, at least 5, at least 10, at least 20, at least 50, at least 100, at least 500, at least 1000, at least 5000 or at least 10,000 training subjects. In some embodiments, the corresponding subset of the training subjects includes no more than 500,000, no more than 10,000, no more than 1000, no more than 500, no more than 100, no more than 50, no more than 20 or no more than 10 training subjects. In some embodiments, the corresponding subset of the training subjects is composed of 2 to 100, 50 to 2000, 1000 to 10,000 or 10,000 to 500,000 training subjects. In some embodiments, the corresponding subset of the training subjects falls into another range starting from no less than 2 training subjects to no more than 500,000 training subjects.

[0062] In some embodiments, the feature set (e.g., determined by sequential changes in Raman spectra) includes at least 1, at least 2, at least 3, at least 5, at least 8, at least 10, at least 15, or at least 20 features. In some embodiments, the feature set includes no more than 50, no more than 20, no more than 10, no more than 5, or no more than 3 features. In some embodiments, the feature set consists of 1 to 10, 4 to 15, 8 to 20, or 15 to 50 features. In some embodiments, the feature set falls within another range starting from no less than 1 feature and ending at no more than 50 features.

[0063] In some embodiments, the trained model is a neural network algorithm, a support vector machine algorithm, a decision tree algorithm, an unsupervised clustering model algorithm, a supervised clustering model algorithm, a regression model, or a gradient boosting algorithm (e.g., a gradient boosting implementation of a machine learning algorithm, such as a gradient boosting decision tree). In some embodiments, the trained model is a polynomial classifier. In some embodiments, the trained model is a binomial classifier. In some embodiments, the first biological condition is selected from the group consisting of: autism spectrum disorder (ASD), attention deficit / hyperactivity disorder (ADHD), amyotrophic lateral sclerosis (ALS), schizophrenia, irritable bowel disease (IBD), pediatric kidney disease, renal transplant rejection, and pediatric cancer.

[0064] In some embodiments, evaluating the first biological condition associated with the Raman feature of the test subject further includes distinguishing between the presence of the first biological condition associated with the Raman feature and the absence of the first biological condition associated with the Raman feature. In some embodiments, evaluating the test subject for the first biological condition associated with the Raman feature further includes: distinguishing the first biological condition associated with the Raman feature from the second biological condition associated with the Raman feature that is different from the first biological condition associated with the Raman feature. In some embodiments, the first biological condition is autism spectrum disorder, and the second biological condition is neurotypical development; that is, there is no neurodevelopmental disorder. In some embodiments, the first biological condition is autism spectrum disorder, and the second biological condition is attention deficit / hyperactivity disorder. In some embodiments, the test subject is a person. In some embodiments, the test subject is an adult. In some embodiments, the age of the person is between about 12 years old and about 5 years old. In some embodiments, the subject is less than about 12, 11, 10, 9, 8, 7, 5, 4, 3, 2 or 1 years old. In some embodiments, the subject is at least about 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11 or 12 years old. In some embodiments, at least a portion of the time profile of the Raman curve corresponds to the prenatal period of the subject. In some embodiments, the human is at least 13 years old, at least 14 years old, at least 15 years old, at least 18 years old, or at least 30 years old. In some embodiments, the human is no more than 40 years old, no more than 30 years old, no more than 18 years old, or no more than 15 years old. In some embodiments, the age of the human is between about 1 year old and about 6 years old, between about 5 years old and about 10 years old, or between about 10 years old and about 40 years old. In some embodiments, the human falls within another age range starting from no less than about 1 year old and ending no more than about 40 years old.

[0065] In some embodiments, the first biological condition and / or the second biological condition are selected from a plurality of biological conditions. In some embodiments, the plurality of biological conditions include at least 2, at least 5 or at least 10 biological conditions. In some embodiments, a plurality of biological conditions include no more than 20, no more than 10 or no more than 5 biological conditions. In some embodiments, the plurality of biological conditions consist of 2 to 5, 3 to 10 or 8 to 20 biological conditions. In some embodiments, the plurality of biological conditions fall within another range starting from no less than 2 biological conditions to no more than 20 biological conditions. In some embodiments, the first diagnostic state and / or the second diagnostic state are selected from a plurality of diagnostic states. In some embodiments, the plurality of diagnostic states include at least 2, at least 5 or at least 10 diagnostic states. In some embodiments, the plurality of diagnostic states include no more than 20, no more than 10 or no more than 5 diagnostic states. In some embodiments, the plurality of diagnostic states consist of 2 to 5, 3 to 10 or 8 to 20 diagnostic states. In some embodiments, the plurality of diagnostic states fall within another range starting from no less than 2 diagnostic states to no more than 20 diagnostic states.

[0066] In some embodiments, the corresponding biological sample associated with the Raman signature of the corresponding subject (e.g., the test subject and / or the training subject) is selected from the group consisting of hair shaft, teeth and nails. In some embodiments, the corresponding biological sample associated with the Raman signature of the corresponding subject (e.g., the test subject and / or the training subject) is a hair shaft, and wherein the reference line corresponds to the longitudinal direction of the hair shaft. In some embodiments, the corresponding biological sample associated with the Raman signature of the corresponding subject (e.g., the test subject and / or the training subject) is a tooth, and wherein the reference line corresponds to the direction across the growth zone, including the newborn line of the tooth. In some embodiments, the corresponding plurality of positions are sorted so that the first position in the corresponding plurality of positions of the corresponding biological sample along the corresponding subject (e.g., the test subject and / or the training subject) corresponds to the position of the tip of the corresponding biological sample closest to the corresponding training subject. In some embodiments, each trace in the corresponding plurality of Raman spectroscopy measurements includes a plurality of data points, each data point being an instance of a corresponding position in the plurality of positions. In some embodiments, the corresponding feature set is selected from the group consisting of: hierarchical, entropy, capture time (TT), mean diagonal length (MDL), recursion time (RT), Vmax, certainty, Lmax, and any combination thereof. In some embodiments, the corresponding plurality of locations includes at least 1000, 1500, 2000, 2500, 3000, 3500, 4000, 4500, 5000, 5500, 6000, 6500, 7000, 7500, 8000, 8500, 9000, 9500, 10000 or more locations.

[0067] In some embodiments, the plurality of locations along the reference line on the biological sample comprises at least 50, at least 100, at least 500, at least 1000, at least 2000, at least 5000, at least 10,000, at least 50,000, at least 100,000, at least 500,000, or at least 1×10 6 In some embodiments, the plurality of locations includes no more than 1×10 7 No more than 1×10 6 , no more than 100,000, no more than 10,000, no more than 1000, or no more than 100 positions. In some embodiments, the plurality of positions ranges from 50 to 1000, 500 to 50,000, 10,000 to 1×10 6 or 1×10 6 Up to 1×10 7 In some embodiments, the plurality of positions falls within a range from no less than 50 positions to no more than 1×10 7In some embodiments, each corresponding growth period of the biological sample is selected from a plurality of growth periods. In some embodiments, the plurality of growth periods includes at least 50, at least 100, at least 500, at least 1000, at least 2000, at least 5000, at least 10,000, at least 50,000, at least 100,000, at least 500,000, or at least 1×10 6 In some embodiments, the plurality of growth periods includes no more than 1×10 7 No more than 1×10 6 In some embodiments, the plurality of growth periods ranges from 50 to 1000, 500 to 50,000, 10,000 to 1×10 6 or 1×10 6 to 1×10 7 In some embodiments, the plurality of growth periods falls within a range from no less than 50 growth periods to no more than 1×10 7 Another range at the end of a growth period.

[0068] In some embodiments, the plurality of Raman spectroscopy measurements comprises at least 50, at least 100, at least 500, at least 1000, at least 2000, at least 5000, at least 10,000, at least 50,000, at least 100,000, at least 500,000, or at least 1×10 6 In some embodiments, the plurality of Raman spectroscopy measurements include no more than 1×10 7 No more than 1×10 6 In some embodiments, the plurality of Raman spectroscopy measurements ranges from 50 to 1000, 500 to 50,000, 10,000 to 1×10 6 or 1×10 6 Up to 1×10 7 In some embodiments, the plurality of Raman spectroscopy measurement results fall within a range from no less than 50 Raman spectroscopy measurement results to no more than 1×10 7In some embodiments, the corresponding Raman spectrum measurement results include one or more traces of fluorescence intensity. In some embodiments, the corresponding Raman spectrum measurement results include at least 1, at least 2, at least 3, at least 4, at least 5, or at least 10 traces. In some embodiments, the corresponding Raman spectrum measurement results include no more than 50, no more than 10, no more than 5, or no more than 3 traces. In some embodiments, the corresponding Raman spectrum measurement results consist of 1 to 5, 2 to 10, or 10 to 20 traces. In some embodiments, the corresponding Raman spectrum measurement results include another range of traces starting from no less than 1 trace and ending at no more than 20 traces.

[0069] In some embodiments, for a corresponding trace, the plurality of data points includes at least 2, at least 5, at least 10, at least 20, at least 50, at least 100, at least 500, at least 1000, at least 5000, or at least 10,000 data points. In some embodiments, the plurality of data points includes no more than 500,000, no more than 10,000, no more than 1000, no more than 500, no more than 100, no more than 50, no more than 20, or no more than 10 data points. In some embodiments, the plurality of data points consists of 2 to 100, 50 to 2000, 1000 to 10,000, or 10,000 to 500,000 data points. In some embodiments, the plurality of data points falls within another range starting from no less than 2 data points to no more than 500,000 data points.

[0070] In some embodiments, the corresponding feature set is selected from a set of time-dynamic features of one or more traces. In some embodiments, the time-dynamic features of the one or more traces are determined by a data analysis method. In some embodiments, the data analysis method applies one or more of the following operations and / or methods to one or more traces: determination of a linear slope, determination of multiple nonlinear parameters describing the curvature of the one or more traces, determination of a sudden change in the intensity of the one or more traces, determination of one or more changes in the baseline intensity of the one or more traces, determination of a change in the frequency domain representation of the one or more traces, determination of a change in the power spectrum domain representation of the one or more traces, determination of one or more recursive quantitative analysis parameters, determination of one or more cross-recursive quantitative analysis parameters, determination of one or more joint recursive quantitative analysis parameters, determination of one or more multidimensional recursive quantitative analysis parameters, estimation of a Lyapunov spectrum, or determination of a maximum Lyapunov exponent.

[0071] Combination Figure 1Describing the details of an exemplary system, the figure shows an example of a block diagram of a computing device 100 of the present disclosure. In some embodiments, the device 100 includes one or more processing units CPU102 (also referred to as processors), one or more network interfaces 104, a user interface 106, non-persistent memory 111, persistent memory 112, and one or more communication buses 114 for interconnecting these components. One or more communication buses 114 optionally include circuits (sometimes referred to as chipsets) that interconnect and control communication between system components. Non-persistent memory 111 typically includes high-speed random access memory, such as DRAM, SRAM, DDR RAM, ROM, EEPROM, flash memory, while persistent memory 112 typically includes CD-ROM, digital versatile disk (DVD) or other optical storage, tape cassettes, tapes, magnetic disk storage or other magnetic storage devices, magnetic disk storage devices, optical disk storage devices, flash memory devices or other non-volatile solid-state storage devices. Persistent memory 112 optionally includes one or more storage devices remote from CPU102. Persistent memory 112 and non-volatile storage devices within non-persistent memory 112 contain non-transitory computer-readable storage media. In some embodiments, non-persistent memory 111 or alternatively non-transitory computer-readable storage media stores the following programs, modules, and data structures, or subsets thereof, sometimes in combination with persistent memory 112: an optional operating system 116, which includes processes for handling various basic system services and performing hardware-related tasks; an optional network communication module (or instructions) 118 for connecting system 100 to other devices and / or communication networks 104; an optional classifier training module 120 for training models (e.g., classifiers, regressors, etc.) to assess biological conditions of a subject; and an optional data storage 122 for data from training subjects. a dataset of biological samples of the subject, including feature data of one or more training subjects 124, wherein the feature data includes parameters associated with each feature 126, and a diagnostic status 128 (e.g., indicating that the corresponding training subject has been diagnosed with a biological condition or has not been diagnosed with a biological condition); an optional classifier validation module 130 for validating a model for distinguishing biological conditions; an optional data store 132 for a dataset of biological samples from the validation subjects; and an optional patient classification module 134 for classifying a subject as having a biological condition, for example, as trained using the classifier training module 120.

[0072] In various embodiments, one or more of the above elements are stored in one or more previously mentioned memory devices, and correspond to an instruction set for performing the above functions. The modules, data or programs (e.g., a set of instructions) identified above do not need to be implemented as separate software programs, processes, data sets or modules, and therefore various subsets of these modules and data can be combined or otherwise rearranged in various embodiments. In some embodiments, non-persistent memory 111 optionally stores a subset of the modules and data structures identified above. In addition, in some embodiments, memory stores additional modules and data structures not described above. In some embodiments, one or more of the above identified elements are stored in a computer system other than visualization system 100, which can be addressed by visualization system 100 so that visualization system 100 can retrieve all or part of these data when needed.

[0073] In some embodiments, system 100 is connected to or includes one or more analytical devices for performing chemical analysis. For example, optional network communication module (or instructions) 118 is configured to connect system 100 to one or more analytical devices, for example, via communication network 104. In some embodiments, one or more analytical devices include a laser ablation inductively coupled plasma mass spectrometer (LA-ICP-MS), a fluorescence image sensor, or a Raman spectrometer.

[0074] although Figure 1 A "system 100" is depicted, but the diagram is intended more as a functional description of various features that may be present in a computer system than as a schematic diagram of the architecture of the implementations described herein. In practice, and as recognized by one of ordinary skill in the art, items shown separately may be combined and some items may be separated. Furthermore, although Figure 1 Certain data and modules are described in non-persistent storage 111 , but some or all of these data and modules may be in persistent storage 112 .

[0075] In some embodiments, the methods of the present disclosure include obtaining a biological sample (e.g., a strand of hair comprising a hair shaft). In some embodiments, the subject is a human. In some embodiments, the subject is a child of 12 years of age or younger (e.g., a child of 5 years of age, 4 years of age, 3 years of age, 2 years of age, 1 year of age, 9 months, 6 months, 3 months, or 1 month of age or younger). In some embodiments, the child is between about 12 years of age and about 5 years of age. In some embodiments, the subject is less than about 12, 11, 10, 9, 8, 7, 5, 4, 3, 2, or 1 year of age. In some embodiments, the subject is at least about 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, or 12 years of age. In some embodiments, the subject is an adult. In some embodiments, the subject is an adult. Figure 2A An example of a hair sample of a subject including a hair shaft is shown. In some embodiments, the hair sample is cut from the subject (e.g., with the aid of scissors). In some embodiments, the method of obtaining the hair sample is non-invasive. In some embodiments, the hair sample obtained has a length of at least 1 cm (e.g., the hair sample is 1 cm, 2 cm, 3 cm, 4 cm, or 5 cm long). In some embodiments, the hair sample obtained is at least 1 cm, at least 2 cm, at least 3 cm, at least 4 cm, at least 5 cm, at least 6 cm, at least 10 cm, or at least 20 cm long. In some embodiments, the hair sample obtained is no more than 40 cm, no more than 20 cm, no more than 10 cm, or no more than 5 cm long. In some embodiments, the hair sample obtained is 1 to 5 cm, 4 to 10 cm, 8 to 20 cm, or 15 to 40 cm long. In some embodiments, the hair sample obtained falls within another range starting from no less than 1 cm long to no more than 40 cm long. In some embodiments, the hair sample includes any portion of the hair (e.g., the tip or the portion between the tip and the hair follicle). In particular, there is no particular requirement that the hair sample contain hair follicles. Figure 2B Examples of dental samples from subjects are shown. Figure 2C An example of a nail sample of a subject is shown. In some embodiments, in the case of a nail or hair, obtaining a biological sample refers to positioning the subject so that the nail or hair is sampled. In some embodiments, the nail sample comprises a complete nail or nail clippings.

[0076] In some embodiments, the obtained biological sample is pretreated, such as by washing the biological sample with one or more solvents and / or surfactants and drying. In some embodiments where the biological sample is hair, the hair sample is subjected to a pretreatment in a TRITON and ultrapure metal-free water (e.g. In some embodiments, the hair shafts are pretreated by washing in a solution of 1% ethanol (e.g., water) and drying in an oven overnight (e.g., at 60 degrees Celsius). In some embodiments, the pretreatment further comprises preparing the hair shafts for measurement by placing the hair shafts on a slide (e.g., a microscope slide) with an adhesive film (e.g., double-sided tape). In some embodiments, the hair shafts are positioned so that the hair shafts are substantially straight. In some embodiments, the slide with the hair shafts is placed in or near a measurement system (e.g., a laser ablation inductively coupled plasma mass spectrometer (LA-ICP-MS), a fluorescence image sensor, or a Raman spectrometer) for performing analysis. In some embodiments in which the biological sample is a tooth or nail, the surface of the biological sample is cleaned (e.g., with a surfactant, water, or one or more solvents). In some embodiments, the sample is sliced ​​and then placed in or near a measurement system (e.g., a laser ablation inductively coupled plasma mass spectrometer (LA-ICP-MS), a fluorescence image sensor, or a Raman spectrometer) for performing analysis.

[0077] Figure 3 A flow chart of a method 300 for assessing a biological condition of a subject, such as a method for predicting a subject's diagnostic status regarding a disease or condition, is shown. In some embodiments, method 300 includes exposing a biological sample of the subject to a light source (as in operation 302). In some cases, the light source includes a laser. In some embodiments, the temporal dynamics of the underlying biological process are determined by analysis. In some embodiments, the analysis includes reducing the dimensions of multiple Raman spectra before processing (e.g., by independent component analysis). In some embodiments, the light signal is generated by a light source (e.g., a laser). In some embodiments, the biological sample includes a tooth sample, a hair sample, or a nail sample. Next, in some embodiments, method 300 includes collecting multiple Raman spectra from the exposed biological sample (as in operation 304). Next, in some embodiments, method 300 includes processing multiple Raman spectra to generate a spatial map of multiple Raman spectra (as in operation 306). Next, in some embodiments, method 300 includes predicting a subject's diagnostic status regarding a disease or condition based at least in part on a spatial map of multiple Raman spectra (as in operation 308).

[0078] In some embodiments, a plurality of Raman spectra are collected using a Raman spectroscopic microscope comprising a 50X air-coupled objective lens or a 63X water-immersion coupled objective lens. In some embodiments, the laser comprises a wavelength of about 785 nm or a wavelength of about 532 nm. In some embodiments, the collection is performed using an integration time of about 0.2 seconds to about 0.3 seconds. In some embodiments, the collection comprises moving the biological sample with a step size of about 2 microns to about 5 microns after the Raman spectra in the plurality of Raman spectra are collected.

[0079] In some embodiments, the analyzing includes generating a temporal Raman curve based at least in part on the acquired Raman spectra, and analyzing the temporal curve for variability in the Raman spectra. In some embodiments, at least a portion of the temporal Raman curve corresponds to a prenatal period of the subject.

[0080] In some embodiments, measurement data is collected sequentially from a biological sample at multiple locations along the biological sample. In some embodiments, the corresponding multiple locations include at least 1000, 1500, 2000, 2500, 3000, 3500, 4000, 4500, 5000, 5500, 6000, 6500, 7000, 7500, 8000, 8500, 9000, 9500, 10000 or more than 10000 locations. In some embodiments, the corresponding locations are adjacent to each other. In this way, each area corresponding to a different location on the biological sample can therefore be associated with a dynamic (e.g., time-varying) abundance measurement result. In some embodiments, the corresponding locations are separated by a predetermined distance. In some embodiments, sampling is performed starting from a corresponding location closest to the tip of a biological sample such as a hair sample along a reference line of the biological sample (e.g., at a location corresponding to the youngest age of the subject). In general, sampling can be performed starting from the corresponding position closest to the tip or root, as long as the sampling direction is known and analyzed using an appropriate trained model.

[0081] In some embodiments, sampling produces a set of data points. In some embodiments, each set of data points corresponds to a measurement result (e.g., abundance or concentration) of a substance, which indicates a dynamic biological response measured at multiple locations along the biological sample. In some embodiments, each location on a reference line of the biological sample corresponds to a specific growth time of the biological sample. In some embodiments, in the example of a hair shaft, each location corresponds to a hair growth period of approximately 20 minutes (e.g., a hair growth period calculated using a 5 micron laser step size and an average hair growth rate of 1 cm per month). By associating multiple locations along the reference line of the biological sample with corresponding growth time periods, a first data set including multiple traces is obtained. Each trace includes a time-dependent abundance measurement result (e.g., abundance or concentration) of a substance, indicating a dynamic biological response measured from the biological sample. For example, the distance between the locations can correspond to an estimated growth of the biological sample (e.g., biological time). For example, the abundance of the hair sample can be measured along a 1.2 cm distance, which corresponds to a biological time of approximately 35 days. In some embodiments, the biological time is estimated by using the average hair growth rate (e.g., 1 cm per month).

[0082] In some embodiments, data analysis of the Raman spectra includes cosmic ray removal, background correction, spectral normalization, peak fitting, or any combination thereof.

[0083] In some embodiments, data analysis can be performed on a trace corresponding to a time-dependent abundance (e.g., time-dependent concentration) of a substance indicative of a dynamic biological response measured from a biological sample. This can include customized operations to clean the data (e.g., smoothing the data over a time span, and / or removing data points that are above or below a predetermined threshold). In some embodiments, the data analysis includes removing data points from the trace where the mean absolute difference between adjacent data points is at least one, two, or three times the standard deviation of the mean absolute difference between adjacent points.

[0084] In some embodiments, the data analysis further includes a dimensionality reduction step, whereby the high-dimensional array of Raman spectra is decomposed into a lower-dimensional array of derived time-varying components. Dimensionality reduction methods include independent component analysis (ICA), principal component analysis (PCA), non-negative matrix factorization (NNMF), and related unsupervised and supervised methods.

[0085] In some embodiments, the data analysis further includes performing recursive quantitative analysis (RQA) on the time-dependent traces or on components derived from a dimensionality reduction technique (ICA / PCA) applied to the time-dependent traces to obtain a set of features describing the dynamic periodic properties of the traces. RQA measures the variability of components in or derived from the time-dependent traces. RQA involves the estimation of features describing periodic characteristics in a given waveform, including recurrence rate, certainty, average diagonal length, maximum diagonal length, divergence, Shannon entropy of diagonal length, recurrence trend, stratification, capture time, maximum vertical line length, Shannon entropy of vertical line length, average recurrence time, Shannon entropy of recurrence time, number of most likely recurrences, determination of linear slope, determination of multiple nonlinear parameters describing the curvature of the one or more traces, determination of sudden changes in the intensity of the one or more traces, determination of one or more changes in the baseline intensity of the one or more traces, determination of changes in the frequency domain representation of the one or more traces, determination of changes in the power spectrum domain representation of the one or more traces, determination of one or more recurrence quantitative analysis parameters, determination of one or more cross-recurrence quantitative analysis parameters, determination of one or more joint recurrence quantitative analysis parameters, determination of one or more multi-dimensional recurrence quantitative analysis parameters, estimation of the Lyapunov spectrum or determination of the maximum Lyapunov exponent and / or any combination thereof. The methods and features of RQA are described, for example, by Webber et al. in "Simpler Methods Do It Better: Success of Recurrence Quantification Analysis as a General Purpose Data Analysis Tool," Physics Letters A373, 3753-3756 (2009) and by Marwan et al. in "Recurrence Plots for the Analysis of Complex Systems," Physics Reports 438, 237-239 (2007), the contents of each of which are hereby incorporated by reference in their entirety. In some embodiments, time-dependent traces are analyzed using other analysis methods such as Fourier transforms, wavelet analysis, and cosine analysis. Such techniques can be applied to derive similar indicators, including spectral analysis of frequency components and their associated power. These indicators and related derived metrics can be used to replace the features derived from RQA to analyze time-dependent traces obtained from biological samples for the purpose of predicting classification.

[0086] RQA includes the construction of recurrence diagrams that visualize and analyze the dynamic temporal structure in the corresponding obtained traces. Such recurrence diagrams can illustrate the stage process in sequential measurements by plotting a diagram derived from the time lag of a given sequence relative to that sequence. From the one-dimensional trace measured from the hair shaft, an additional dimension is calculated to embed the trace in a higher dimensional space, which is called a phase diagram, where t refers to the value of the original trace, and the dimensions (t+τ) and (t+2τ) are derived from lagging the original time series by an interval of τ. The embedded phase diagram is then subsequently analyzed to construct the recurrence diagram and perform a recursive quantitative analysis. By applying a threshold function to each point in the phase diagram, a recursive quantitative diagram can be derived from the phase diagram; on the corresponding recurrence diagram, consisting of a square binary matrix, usually represented as a white or black space, a given point is assigned a value of 1 at each time interval, where another point in the phase diagram shares the spatial limit of the assigned threshold boundary. The RQA method is applied to the recurrence diagram to examine the interval of delays between states in a given system, and the black points reflect the time interval when the system revisits the same state. Periodic processes where the system continuously repeats a given state pattern will appear as black diagonal lines in the recurrence diagram, while stable periods will appear as square structures, spurious repetitions as black dots, and unique events as blanks.

[0087] In some embodiments, a recurrence plot is constructed for the traces of a single substance or a combination of two substances (e.g., to visualize the interactive periodic pattern of two substances; this may be referred to as cross-recurrence quantitative analysis, or joint recurrence quantitative analysis). In some embodiments, a recurrence plot is constructed for a combination of three or more substances.

[0088] In some embodiments, the data analysis includes analyzing the recurrence graph to obtain a set of features associated with the recurrence graph. These features, which may be interchangeably referred to as "rhythmic features" or "dynamic features," provide quantitative measures describing the periodicity, predictability, and transitivity present in the plurality of traces. The features are selected from the set comprising: recursion rate, certainty, average diagonal length, maximum diagonal length, divergence, Shannon entropy of diagonal length, recursion trend, stratification, capture time, maximum vertical line length, Shannon entropy of vertical line length, average recursion time, Shannon entropy of recursion time, number of most likely recursions, determination of linear slope, determination of multiple nonlinear parameters describing the curvature of the one or more traces, determination of sudden changes in the intensity of the one or more traces, determination of one or more changes in the baseline intensity of the one or more traces, determination of changes in the frequency domain representation of the one or more traces, determination of changes in the power spectrum domain representation of the one or more traces, determination of one or more recursive quantitative analysis parameters, determination of one or more cross-recursive quantitative analysis parameters, determination of one or more joint recursive quantitative analysis parameters, determination of one or more multidimensional recursive quantitative analysis parameters, estimation of the Lyapunov spectrum or determination of the maximum Lyapunov exponent and / or any combination thereof.

[0089] In some embodiments, the data analysis further comprises inputting the obtained feature set into a trained model. In some embodiments, the trained model comprises a predictive computational algorithm to obtain the probability of a subject having a biological condition. In some embodiments, the predictive computational algorithm performs the following calculations:

[0090]

[0091] where p(subject) is the probability that the subject has the first biological condition, e is the Euler number, and α is the β 1 x 1 +…+β k x k The calculated parameter, x, that is related to the probability that the subject has the biological condition when it is equal to zero 1 ,…,x k corresponds to the value derived for each feature in the feature set, which includes features from 1 to k, and β 1 ,…,β k corresponds to a weight parameter associated with each feature in a feature set including features from 1 to k.

[0092] In some embodiments, the weight parameter β 1 ,…,β k Defined based on model training.In some embodiments, the probability p(subject) is provided as a number ranging from 0 to 1, where 1 corresponds to a 100% probability that the subject has the biological condition.

[0093] In some embodiments, data analysis includes applying a threshold to the probability p (subject) obtained. If the probability p (subject) obtained is higher than a predetermined threshold, the subject is assessed to have a biological condition. If the probability obtained is lower than a threshold, the subject is assessed to not have a biological condition. In some embodiments, the threshold is between about 0.3 and 0.6 (for example, the predetermined threshold is about 0.3, 0.35, 0.4, 0.45, 0.5, 0.55 or 0.6). In some embodiments, the threshold is at least 0.1, at least 0.2, at least 0.3, at least 0.4, at least 0.5, at least 0.6 or at least 0.7. In some embodiments, the threshold is no more than 0.9, no more than 0.8, no more than 0.7, no more than 0.6, no more than 0.5 or no more than 0.4. In some embodiments, the threshold falls within another range from not less than 0.1 to not more than 0.9. In some embodiments, the value assigned to the probability threshold is predetermined or estimated during model training using a receiver operating characteristic (ROC) chart, wherein the optimal threshold used corresponds to the value that produces the largest area under the curve (ROC-AUC). In some embodiments, the probability obtained is expressed as an associated probability (e.g., an odds ratio (OR), which can be derived from the probability such that OR=p / (1-p)). For example, the assessment includes assessing the probability that the subject has a biological condition.

[0094] In some embodiments, data analysis includes distinguishing the first biological condition and alternative condition, for example, the second biological condition.In some embodiments, alternative condition is related to unknown condition (for example, neurotypical condition (NT)).In some embodiments, the first biological condition is related to autism spectrum disorder (ASD), and alternative condition is related to attention deficit / hyperactivity disorder (ADHD).In some embodiments, alternative condition is any other neurodevelopmental condition, or the comorbidity diagnosis of two neurodevelopmental conditions.Therefore, in some embodiments, data analysis can distinguish between two neurodevelopmental conditions (for example, between autism spectrum disorder and ADHD, or between ASD and comorbidity (CM) cases diagnosed as both ASD and ADHD).

[0095] Healthcare providers, such as physicians and the patient's treatment team, may have access to patient data (e.g., dynamic biological response data or other health data) and / or have predictions or assessments generated from such data. Based on the data analysis results, the healthcare provider may determine a clinical decision or outcome.

[0096] For example, based at least in part on the predicted disease or condition of the experimenter, the doctor may instruct the patient to undergo one or more clinical tests at a hospital or other clinical location. In some embodiments, these instructions are provided when a predetermined criterion (e.g., a minimum threshold value of the probability of a disease or condition) is met.

[0097] In some embodiments, such minimum thresholds include, for example, at least about 5% probability, at least about 10% probability, at least about 20% probability, at least about 25% probability, at least about 30% probability, at least about 35% probability, at least about 40% probability, at least about 45% probability, at least about 50% probability, at least about 55% probability, at least about 60% probability, at least about 65% probability, at least about 70% probability, at least about 75% probability, at least about 80% probability, at least about 85% probability, at least about 90% probability, at least about 95% probability, at least about 96% probability, at least about 97% probability, at least about 98% probability, or at least about 99% probability. In some embodiments, the minimum threshold is no more than 99%, no more than 90%, no more than 80%, no more than 70%, no more than 60%, no more than 50%, or no more than 40%. In some embodiments, the minimum threshold is 5% to 20%, 10% to 50%, 30% to 70%, or 60% to 99%. In some embodiments, the minimum threshold falls within another range starting from no less than 5% and ending no more than 99%.

[0098] As another example, a physician may prescribe a therapeutically effective dose of a treatment (e.g., a drug) based at least in part on the predicted disease or condition of the subject, a clinical course, or further clinical testing to administer to the patient. For example, a physician may prescribe an anti-inflammatory treatment drug in response to an indication of inflammation in a patient.

[0099] Model

[0100] In some embodiments, the methods and systems of the present disclosure utilize or acquire external capabilities of artificial intelligence technology to develop features for various diseases or conditions. In some embodiments, these features are used to accurately predict diseases or conditions (e.g., months or years ahead of the standard of clinical care). Using such predictive capabilities, healthcare providers (e.g., doctors) can make informed, accurate risk-based decisions, thereby improving the quality of care and monitoring provided to patients.

[0101] In some embodiments, the methods and systems of the present disclosure analyze the collected dynamic biological response data from the subject (patient) to generate the possibility that the subject suffers from a disease or condition. For example, in some embodiments, the system applies a trained (e.g., prediction) algorithm to the collected dynamic biological response data to generate the possibility that the subject suffers from a disease or condition. In some embodiments, the trained algorithm includes an artificial intelligence-based model, such as a machine learning-based classifier, which is configured to process the collected dynamic biological response data to generate the possibility that the subject suffers from a disease or condition. In some embodiments, a clinical data set from one or more patient cohorts is used to train the model, for example, the patient's clinical health data and / or dynamic biological response data is used as input to the model, and the patient's known clinical health results (e.g., disease or condition) are used as output.

[0102] In some embodiments, the model includes one or more machine learning algorithms. Examples of machine learning algorithms include, but are not limited to, support vector machines (SVM), naive Bayes classification, random forests, neural networks (such as deep neural networks (DNNs), recurrent neural networks (RNNs), deep RNNs, long short-term memory (LSTM) recurrent neural networks (RNNs) or gated recurrent units (GRUs), or other supervised learning algorithms or unsupervised machine learning, statistical or deep learning algorithms for classification and regression. In some embodiments, the model also involves the estimation of an integrated model consisting of multiple prediction models, and utilizes techniques such as gradient enhancement, such as in the construction of gradient enhanced decision trees. In some embodiments, the model is trained using one or more training data sets corresponding to patient data.

[0103] In some embodiments, the training data set is generated from one or more patient cohorts, for example, having common clinical characteristics (features) and clinical outcomes (labels). In some embodiments, the training data set includes a feature set and a label corresponding to the feature. In some embodiments, the feature corresponds to an algorithm input, which includes dynamic biological response data, patient demographic information derived from an electronic medical record (EMR), and medical observations. In some embodiments, the feature includes clinical characteristics, such as certain ranges or categories of dynamic biological response data, for example. In some embodiments, the feature includes patient information, such as patient age, patient medical history, other medical conditions, current or past medications, and time since the last observation. For example, a feature set collected from a given patient at a given point in time can be used together as a feature that can indicate the health status or state of the patient at a given point in time.

[0104] For example, in some embodiments, ranges of dynamic bio-response data and other health measurements are represented as multiple disjoint continuous ranges of continuous measurements, and categories of dynamic bio-response data and other health measurements can be represented as multiple disjoint sets of measurements (e.g., {"high", "low"}, {"high", "normal"}, {"low", "normal"}, {"high", "high to near critical value", "normal", "low"}, etc.). In some embodiments, clinical features include clinical labels indicating a patient's health history, such as a diagnosis of a disease or condition, previous administration of clinical treatment (e.g., medication, surgical treatment, chemotherapy, radiotherapy, immunotherapy, etc.), behavioral factors, or other health conditions (e.g., hypertension or high blood pressure, hyperglycemia or high blood sugar, hypercholesterolemia or high blood cholesterol, history of allergic reactions or other adverse reactions, etc.).

[0105] In some embodiments, label includes clinical results, such as the presence, absence, diagnosis or prognosis of a disease or illness in a subject (e.g., a patient). In some embodiments, clinical results include the time characteristics related to the presence, absence, diagnosis or prognosis of a disease or illness with a patient. For example, time characteristics can indicate that a patient has the occurrence of a disease or illness in a certain time period after a previous clinical result (e.g., being discharged from hospital, receiving such as drug therapy, experiencing clinical procedures such as surgery, etc.). In some embodiments, such time period includes, for example, about 1 hour, about 2 hours, about 3 hours, about 4 hours, about 6 hours, about 8 hours, about 10 hours, about 12 hours, about 14 hours, about 16 hours, about 18 hours, about 20 hours, about 22 hours, about 24 hours, about 2 days, about 3 days, about 4 days, about 5 days, about 6 days, about 7 days, about 10 days, about 2 weeks, about 3 weeks, about 4 weeks, about 1 month, about 2 months, about 3 months, about 4 months, about 6 months, about 8 months, about 10 months, about 1 year or more than about 1 year. In some embodiments, the time period is no more than 5 years, no more than 1 year, no more than 6 months, no more than 3 months, no more than 1 month, no more than 2 weeks, no more than 1 week, no more than 1 day, or no more than 12 hours. In some embodiments, the time period is 1 hour to 12 hours, 12 hours to 24 hours, 1 day to 7 days, 1 week to 4 weeks, 1 month to 12 months, or 1 year to 5 years. In some embodiments, the time period falls within another range starting from no less than 1 hour and ending no more than 5 years.

[0106] In some embodiments, input features are constructed by binning or alternatively using one-hot encoding. In some embodiments, input also includes feature values ​​or vectors derived from the aforementioned inputs, such as cross-correlations calculated between separate dynamic bioreaction data or other measurements within a fixed time period, and discrete derivatives or finite differences between continuous measurements. In some embodiments, such time periods include, for example, about 1 hour, about 2 hours, about 3 hours, about 4 hours, about 6 hours, about 8 hours, about 10 hours, about 12 hours, about 14 hours, about 16 hours, about 18 hours, about 20 hours, about 22 hours, about 24 hours, about 2 days, about 3 days, about 4 days, about 5 days, about 6 days, about 7 days, about 10 days, about 2 weeks, about 3 weeks, about 4 weeks, about 1 month, about 2 months, about 3 months, about 4 months, about 6 months, about 8 months, about 10 months, about 1 year or more than about 1 year. In some embodiments, the time period is no more than 5 years, no more than 1 year, no more than 6 months, no more than 3 months, no more than 1 month, no more than 2 weeks, no more than 1 week, no more than 1 day, or no more than 12 hours. In some embodiments, the time period is 1 hour to 12 hours, 12 hours to 24 hours, 1 day to 7 days, 1 week to 4 weeks, 1 month to 12 months, or 1 year to 5 years. In some embodiments, the time period falls within another range starting from no less than 1 hour and ending no more than 5 years.

[0107] In some embodiments, training records are constructed from observation sequences. In some embodiments, such sequences include fixed lengths that facilitate data processing. For example, sequences can be zero-padded or selected as independent subsets of a single patient record.

[0108] In some embodiments, the model processes input features to generate output values ​​including one or more classifications, one or more predictions, or a combination thereof. For example, such classifications or predictions may include: a binary classification of a healthy / normal health state (e.g., absence of a disease or condition) or an unhealthy state (e.g., presence of a disease or condition), a classification between groups of classification labels (e.g., 'no disease or condition', 'obvious disease or condition', and 'possible disease or condition'), the likelihood of developing a particular disease or condition (e.g., relative likelihood or probability), a score indicating the presence of a disease or condition, a score indicating the level of systemic inflammation experienced by the patient, a 'risk factor' for the likelihood of death of the patient, a prediction of the time when the patient is expected to have developed the disease or condition, and confidence intervals for any numerical predictions. In some embodiments, various machine learning techniques are cascaded so that the output of the machine learning technique can also be used as an input feature for subsequent layers or sub-portions of the model.

[0109] In order to train the model (e.g., by determining the weights and relevance of the model) to generate real-time classifications or predictions, the model can be trained using a data set. In some embodiments, such a data set is large enough to generate statistically significant classifications or predictions. For example, the data set can include: a database of de-identified data including dynamic biological response data and other measurements, and dynamic biological response data and other measurements from a hospital or other clinical setting.

[0110] In some embodiments, the data set is divided into subsets (e.g., discrete or overlapping), such as a training data set, a development data set, and a test data set. For example, the data set can be divided into a training data set including 80% of the data set, a development data set including 10% of the data set, and a test data set including 10% of the data set. In some embodiments, the training data set accounts for about 10%, about 20%, about 30%, about 40%, about 50%, about 60%, about 70%, about 80% or about 90% of the data set. In some embodiments, the development data set accounts for about 10%, about 20%, about 30%, about 40%, about 50%, about 60%, about 70%, about 80% or about 90% of the data set. In some embodiments, the test data set accounts for about 10%, about 20%, about 30%, about 40%, about 50%, about 60%, about 70%, about 80% or about 90% of the data set. In some embodiments, the training set (e.g., training data set) is selected by randomly sampling the data set corresponding to one or more patient cohorts to ensure the independence of sampling. Alternatively, in some embodiments, a training set (eg, a training data set) is selected by sampling proportionally from data sets corresponding to one or more patient cohorts to ensure independence of the sampling.

[0111] In some embodiments, in order to improve the accuracy of model predictions and reduce overfitting of the model, the data set is augmented to increase the number of samples in the training set. For example, data augmentation can include rearranging the order of observations in the training records. To accommodate data sets with missing observations, methods for interpolating missing values, such as forward filling, backward filling, linear interpolation, and multi-task Gaussian processes, can be used. The data set can be filtered to remove confounding factors. For example, within a database, a subset of patients can be excluded.

[0112] The model may include one or more neural networks, such as a neural network, a convolutional neural network (CNN), a deep neural network (DNN), a recurrent neural network (RNN), or a deep RNN. The recurrent neural network may include a unit that may be a long short-term memory (LSTM) unit or a gated recurrent unit (GRU). For example, the model may include an algorithmic architecture that includes a neural network with a set of input features (such as vital signs and other measurements, patient medical history, and / or patient demographics). Neural network techniques, such as discarding or regularization, may be used during training the model to prevent overfitting. The neural network may include multiple subnetworks, each of which is configured to generate a classification or prediction of a different type of output information (e.g., which may be combined to form the overall output of the neural network). The machine learning model may alternatively utilize statistical or related algorithms, including random forests, classification and regression trees, support vector machines, discriminant analysis, regression techniques, and their ensemble and gradient boosting variants.

[0113] When the model generates a classification or prediction of a disease or condition, a notification (e.g., an alert or alarm) can be generated and transmitted to a healthcare provider, such as a doctor, nurse, or other member of the patient's treatment team within a hospital. The notification can be transmitted via an automated phone call, a short message service (SMS) or multimedia messaging service (MMS) message, an email, or an alert within a dashboard. The notification can include output information such as a prediction of a disease or condition, a predicted likelihood of a disease or condition, a time until the expected onset of a disease or condition, a confidence interval for the likelihood or time, or a recommended treatment regimen for the disease or condition.

[0114] To validate the performance of the model, different performance metrics can be generated. For example, the area under the receiver operating curve (AUROC) can be used to determine the diagnostic ability of the model. For example, the model can use an adjustable classification threshold so that specificity and sensitivity are tunable, and a receiver operating curve (ROC) can be used to identify different operating points corresponding to different values ​​of specificity and sensitivity.

[0115] In some cases, such as when the dataset is not large enough, cross-validation can be performed to assess the robustness of the model across different training and test datasets.

[0116] In order to calculate performance indicators, such as sensitivity, specificity, accuracy, positive predictive value (PPV), negative predictive value (NPV), AUPRC, AUROC or similar indicators, the following definitions can be used. "False positive" can refer to the result of generating a positive result or consequence by mistake or prematurely (for example, before the actual onset of a disease or condition or without any onset of a disease or condition). "True positive" can refer to the result of correctly generating a positive result or consequence when a patient suffers from a disease or condition (for example, the patient shows the symptom of a disease or condition, or the patient's record shows a disease or condition). "False negative" can refer to the result of generating a negative result or consequence, but the patient suffers from the result of a disease or condition (for example, the patient shows the symptom of a disease or condition, or the patient's record shows a disease or condition). "True negative" can refer to the result of generating a negative result or consequence (for example, before the actual onset of a disease or condition or without any onset of a disease or condition).

[0117] Models can be trained until certain predetermined conditions of accuracy or performance are met, such as having a minimum expected value corresponding to a diagnostic accuracy measurement. For example, a diagnostic accuracy measurement can correspond to a prediction of the likelihood of a disease or condition occurring in a subject. As another example, a diagnostic accuracy measurement can correspond to a prediction of the likelihood of a disease or condition that a subject has previously been treated for deteriorating or recurring. Examples of diagnostic accuracy metrics can include sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), accuracy, area under the precision-recall curve (AUPRC), and the area under the curve (AUC) (AUROC) of a receiver operating characteristic (ROC) curve corresponding to the diagnostic accuracy of detecting or predicting a disease or condition.

[0118] For example, in some embodiments, such predetermined conditions can be that the sensitivity of predicting a disease or condition includes, for example, a value of at least about 50%, at least about 55%, at least about 60%, at least about 65%, at least about 70%, at least about 75%, at least about 80%, at least about 85%, at least about 90%, at least about 95%, at least about 96%, at least about 97%, at least about 98%, or at least about 99%. In some embodiments, the predetermined condition is that the sensitivity includes a value of no more than 100%, no more than 99%, no more than 90%, no more than 80%, no more than 70%, or no more than 60%. In some embodiments, the predetermined condition is that the sensitivity includes a value of 50% to 70%, 60% to 80%, 70% to 90%, or 90% to 100%. In some embodiments, the predetermined condition is that the sensitivity includes a value falling within another range starting from no less than 50% and ending at no more than 100%.

[0119] As another example, in some embodiments, such predetermined conditions are that the specificity of predicting a disease or condition includes, for example, a value of at least about 50%, at least about 55%, at least about 60%, at least about 65%, at least about 70%, at least about 75%, at least about 80%, at least about 85%, at least about 90%, at least about 95%, at least about 96%, at least about 97%, at least about 98%, or at least about 99%. In some embodiments, the predetermined condition is that the specificity includes a value of no more than 100%, no more than 99%, no more than 90%, no more than 80%, no more than 70%, or no more than 60%. In some embodiments, the predetermined condition is that the specificity includes a value of 50% to 70%, 60% to 80%, 70% to 90%, or 90% to 100%. In some embodiments, the predetermined condition is that the specificity includes a value falling within another range starting from no less than 50% and ending no more than 100%.

[0120] As another example, in some embodiments, such a predetermined condition is that the positive predictive value (PPV) for predicting a disease or condition includes, for example, at least about 50%, at least about 55%, at least about 60%, at least about 65%, at least about 70%, at least about 75%, at least about 80%, at least about 85%, at least about 90%, at least about 95%, at least about 96%, at least about 97%, at least about 98%, or at least about 99%. In some embodiments, the predetermined condition is that the PPV includes a value of no more than 100%, no more than 99%, no more than 90%, no more than 80%, no more than 70%, or no more than 60%. In some embodiments, the predetermined condition is that the PPV includes a value of 50% to 70%, 60% to 80%, 70% to 90%, or 90% to 100%. In some embodiments, the predetermined condition is that the PPV includes a value falling within another range starting from no less than 50% to no more than 100%.

[0121] As another example, in some embodiments, such a predetermined condition is that the negative predictive value (NPV) for predicting a disease or condition includes, for example, at least about 50%, at least about 55%, at least about 60%, at least about 65%, at least about 70%, at least about 75%, at least about 80%, at least about 85%, at least about 90%, at least about 95%, at least about 96%, at least about 97%, at least about 98%, or at least about 99%. In some embodiments, the predetermined condition is that the NPV includes a value of no more than 100%, no more than 99%, no more than 90%, no more than 80%, no more than 70%, or no more than 60%. In some embodiments, the predetermined condition is that the NPV includes a value of 50% to 70%, 60% to 80%, 70% to 90%, or 90% to 100%. In some embodiments, the predetermined condition is that the NPV includes a value falling within another range starting from no less than 50% to no more than 100%.

[0122] As another example, in some embodiments, such a predetermined condition is that the area under the curve (AUC) of the receiver operating characteristic (ROC) curve predicting the disease or condition (AUROC) comprises a value of at least about 0.50, at least about 0.55, at least about 0.60, at least about 0.65, at least about 0.70, at least about 0.75, at least about 0.80, at least about 0.85, at least about 0.90, at least about 0.95, at least about 0.96, at least about 0.97, at least about 0.98, or at least about 0.99. In some embodiments, the predetermined condition is that the AUROC comprises a value of no more than 1, no more than 0.99, no more than 0.90, no more than 0.80, no more than 0.70, or no more than 0.60. In some embodiments, the predetermined condition is that the AUROC comprises a value of 0.50 to 0.70, 0.60 to 0.80, 0.70 to 0.90, or 0.90 to 1. In some embodiments, the predetermined condition is that the AUROC includes a value falling within another range starting from no less than 0.50 and ending no greater than 1.

[0123] As another example, in some embodiments, such a predetermined condition is that the area under the precision-recall curve (AUPRC) for predicting a disease or condition comprises a value of at least about 0.10, at least about 0.15, at least about 0.20, at least about 0.25, at least about 0.30, at least about 0.35, at least about 0.40, at least about 0.45, at least about 0.50, at least about 0.55, at least about 0.60, at least about 0.65, at least about 0.70, at least about 0.75, at least about 0.80, at least about 0.85, at least about 0.90, at least about 0.95, at least about 0.96, at least about 0.97, at least about 0.98, or at least about 0.99. In some embodiments, the predetermined condition is that the AUPRC comprises a value of no more than 1, no more than 0.99, no more than 0.90, no more than 0.80, no more than 0.70, no more than 0.60, or no more than 0.50. In some embodiments, the predetermined condition is that AUPRC includes values ​​of 0.10 to 0.40, 0.30 to 0.70, 0.60 to 0.90, or 0.80 to 1. In some embodiments, the predetermined condition is that AUPRC includes values ​​falling within another range starting from no less than 0.10 and ending no more than 1.

[0124] In some embodiments, the trained model is trained or configured to predict a disease or condition with a sensitivity of at least about 50%, at least about 55%, at least about 60%, at least about 65%, at least about 70%, at least about 75%, at least about 80%, at least about 85%, at least about 90%, at least about 95%, at least about 96%, at least about 97%, at least about 98%, or at least about 99%. In some embodiments, the model is trained or configured to predict a disease or condition with a sensitivity of no more than 100%, no more than 99%, no more than 90%, no more than 80%, no more than 70%, or no more than 60%. In some embodiments, the model is trained or configured to predict a disease or condition with a sensitivity of 50% to 70%, 60% to 80%, 70% to 90%, or 90% to 100%. In some embodiments, the model is trained or configured to predict a disease or condition with a sensitivity falling within another range starting from no less than 50% and ending at no more than 100%.

[0125] In some embodiments, the trained model is trained or configured to predict a disease or condition with a specificity of at least about 50%, at least about 55%, at least about 60%, at least about 65%, at least about 70%, at least about 75%, at least about 80%, at least about 85%, at least about 90%, at least about 95%, at least about 96%, at least about 97%, at least about 98%, or at least about 99%. In some embodiments, the model is trained or configured to predict a disease or condition with a specificity of no more than 100%, no more than 99%, no more than 90%, no more than 80%, no more than 70%, or no more than 60%. In some embodiments, the model is trained or configured to predict a disease or condition with a specificity of 50% to 70%, 60% to 80%, 70% to 90%, or 90% to 100%. In some embodiments, the model is trained or configured to predict a disease or condition with a specificity falling within another range starting from no less than 50% and ending at no more than 100%.

[0126] In some embodiments, the trained model is trained or configured to predict a disease or condition with a positive predictive value (PPV) of at least about 50%, at least about 55%, at least about 60%, at least about 65%, at least about 70%, at least about 75%, at least about 80%, at least about 85%, at least about 90%, at least about 95%, at least about 96%, at least about 97%, at least about 98%, or at least about 99%. In some embodiments, the model is trained or configured to predict a disease or condition with a PPV of no more than 100%, no more than 99%, no more than 90%, no more than 80%, no more than 70%, or no more than 60%. In some embodiments, the model is trained or configured to predict a disease or condition with a PPV of 50% to 70%, 60% to 80%, 70% to 90%, or 90% to 100%. In some embodiments, the model is trained or configured to predict a disease or condition with a PPV falling within another range starting from no less than 50% to no more than 100%.

[0127] In some embodiments, the trained model is trained or configured to predict a disease or condition with a negative predictive value (NPV) of at least about 50%, at least about 55%, at least about 60%, at least about 65%, at least about 70%, at least about 75%, at least about 80%, at least about 85%, at least about 90%, at least about 95%, at least about 96%, at least about 97%, at least about 98%, or at least about 99%. In some embodiments, the model is trained or configured to predict a disease or condition with an NPV of no more than 100%, no more than 99%, no more than 90%, no more than 80%, no more than 70%, or no more than 60%. In some embodiments, the model is trained or configured to predict a disease or condition with an NPV of 50% to 70%, 60% to 80%, 70% to 90%, or 90% to 100%. In some embodiments, the model is trained or configured to predict a disease or condition with an NPV falling within another range starting from no less than 50% to no more than 100%.

[0128] In some embodiments, the trained model is trained or configured to predict a disease or condition with an area under the curve (AUC) (AUROC) of a receiver operating characteristic (ROC) curve of at least about 0.50, at least about 0.55, at least about 0.60, at least about 0.65, at least about 0.70, at least about 0.75, at least about 0.80, at least about 0.85, at least about 0.90, at least about 0.95, at least about 0.96, at least about 0.97, at least about 0.98, or at least about 0.99. In some embodiments, the trained model is trained or configured to predict a disease or condition with an AUROC of no more than 1, no more than 0.99, no more than 0.90, no more than 0.80, no more than 0.70, or no more than 0.60. In some embodiments, the trained model is trained or configured to predict a disease or condition with an AUROC of 0.50 to 0.70, 0.60 to 0.80, 0.70 to 0.90, or 0.90 to 1. In some embodiments, the trained model is trained or configured to predict a disease or condition with an AUROC that falls within another range starting from no less than 0.50 and ending no greater than 1.

[0129] In some embodiments, the trained model is trained or configured to predict a disease or condition with an area under the precision-recall curve (AUPRC) of at least about 0.10, at least about 0.15, at least about 0.20, at least about 0.25, at least about 0.30, at least about 0.35, at least about 0.40, at least about 0.45, at least about 0.50, at least about 0.55, at least about 0.60, at least about 0.65, at least about 0.70, at least about 0.75, at least about 0.80, at least about 0.85, at least about 0.90, at least about 0.95, at least about 0.96, at least about 0.97, at least about 0.98, or at least about 0.99. In some embodiments, the model is trained or configured to predict a disease or condition with an AUPRC of no more than 1, no more than 0.99, no more than 0.90, no more than 0.80, no more than 0.70, no more than 0.60, or no more than 0.50. In some embodiments, the model is trained or configured to predict a disease or condition with an AUPRC of 0.10 to 0.40, 0.30 to 0.70, 0.60 to 0.90, or 0.80 to 1. In some embodiments, the model is trained or configured to predict a disease or condition with an AUPRC that falls within another range starting from no less than 0.10 and ending no greater than 1.

[0130] In some embodiments, the training data set is collected from training subjects (e.g., people). Each training has a diagnosis status indicating that they have either been diagnosed with a biological condition or have not been diagnosed with a biological condition. In some embodiments, the training subject is a child whose age is equal to or less than 12 years old (e.g., equal to or less than 5 years old, 4 years old, 3 years old, 2 years old, 1 year old, 9 months, 6 months, 3 months or 1 month). In some embodiments, the age of the child is between about 12 years old and about 5 years old. In some embodiments, the subject is less than about 12, 11, 10, 9, 8, 7, 5, 4, 3, 2 or 1 years old. In some embodiments, the subject is at least about 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11 or 12 years old. In some embodiments, the following training process is performed for each training subject in a plurality of training subjects. In some embodiments, the training subject is no older than 18 years old, no older than 15 years old, no older than 12 years old, no older than 11 years old, no older than 10 years old, no older than 9 years old, no older than 8 years old, no older than 7 years old, no older than 5 years old, no older than 4 years old, no older than 3 years old, no older than 2 years old, or no older than 1 year old.

[0131] In some embodiments, multiple positions of a reference line on a biological sample of a training subject are sampled to obtain multiple dynamic biological reaction samples. Each of the corresponding multiple dynamic biological reaction samples is for a different position in the corresponding multiple positions, and each position in the corresponding multiple positions represents a different growth period of the corresponding biological sample. Next, each corresponding dynamic biological reaction sample is analyzed (for example, using a laser ablation inductively coupled plasma mass spectrometer (LA-ICP-MS), a fluorescent image sensor, or a Raman spectrometer) to obtain multiple traces. Each of the corresponding multiple traces corresponds to an abundance measurement result of a corresponding substance, which is jointly determined from the corresponding multiple dynamic biological reaction samples over time.

[0132] Next, a corresponding second data set is obtained from the corresponding plurality of traces including a corresponding feature set, each corresponding feature in the corresponding feature set being determined by a change in abundance of one or more substances in the corresponding plurality of traces, as assessed by applying recursive quantitative analysis or a related method to the Raman waveform or to a dimension derived from the Raman waveform by ICA / PCA or a related dimensionality reduction technique.

[0133] Next, an untrained or partially untrained model is generated with (i) a corresponding feature set of each corresponding second data set of each training subject in a plurality of training subjects and (ii) a corresponding diagnostic state of each training subject in a plurality of training subjects selected from a first diagnostic state and a second diagnostic state, thereby obtaining a trained model. The trained model provides an indication of whether the test subject has a first biological condition based on the value of the feature in the feature set collected from the biological sample of the test subject. In some embodiments, the trained model is a neural network algorithm, a support vector machine algorithm, a decision tree algorithm, an unsupervised clustering algorithm, a supervised clustering algorithm, a regression algorithm, or any combination or variant thereof, particularly including a gradient boosting implementation of the described algorithm, such as a gradient boosting decision tree. In some embodiments, the trained machine learning model utilizes a gradient boosting ensemble algorithm. In some embodiments, the trained model is a polynomial or binomial classifier. In some embodiments, the trained model can be used to make a binary prediction about whether the sample is derived from a subject with a first biological condition; alternatively, it can be polynomial, distinguishing subjects without a diagnosis from subjects with a first biological condition or a second biological condition, wherein the second biological condition is different from the first biological condition.

[0134] In some embodiments, the model is a neural network or a convolutional neural network. See Vincent et al., 2010, "Stacked denoising autoencoders: Learning useful representations in a deep network with a local denoising criterion," J Mach Learn Res 11, pp. 3371-3408; Larochelle et al., 2009, "Exploring strategies for training deep neural networks," J Mach Learn Res 10, pp. 1-40; and Hassoun, 1995, Fundamentals of Artificial Neural Networks, Massachusetts Institute of Technology, each of which is hereby incorporated by reference.

[0135] Independent component analysis (ICA), such as that described herein in Unsupervised Dimensionality Reduction of Raman Waveforms, is described in Lee, T.-W. (1998): Independent component analysis: Theory and applications, Boston, Mass: Kluwer Academic Publishers, ISBN 0-7923-8261-7; and A.; Karhunen, J.; Oja, E. (2001): Independent Component Analysis, New York: Wiley, ISBN 978-0-471-40540-5, which references are hereby incorporated by reference in their entirety.

[0136] Principal component analysis (PCA), such as that described herein in Unsupervised Dimensionality Reduction of Raman Waveforms, is described in Jolliffe, IT (2002). Principal Component Analysis. Springer Series in Statistics. New York: Springer-Verlag. doi: 10.1007 / b98835. ISBN 978-0-387-95442-4, which is hereby incorporated by reference in its entirety.

[0137] SVMs are described in Cristianini and Shawe-Taylor, 2000, “An Introduction to Support Vector Machines,” Cambridge University Press, Cambridge; Boser et al., 1992, “A training algorithm for optimal margin classifiers,” in Proceedings of the 5 th Annual ACM Workshop on Computational Learning Theory, ACM Press, Pittsburgh, Pa., pp. 142-152; Vapnik, 1998, Statistical Learning Theory, Wiley, New York; Mount, 2001, Bioinformatics: sequence and genome analysis, Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY; Duda, Pattern Classification, 2nd ed., 2001, John Wiley & Sons, Inc., pp. 259, 262-265; and Hastie, 2001, The Elements of Statistical Learning, Springer, New York; and Furey et al., 2000, Bioinformatics 16, 906-914, each of which is hereby incorporated by reference in its entirety. When used for classification, SVM separates a given binary labeled data set with a hyperplane that is maximally away from the labeled data. For cases where there is no linear separation, the SVM can be run in conjunction with kernel techniques that automatically implement a nonlinear mapping to the feature space. The hyperplane found by the SVM in the feature space corresponds to a nonlinear decision boundary in the input space.

[0138] Decision trees are generally described in Duda, 2001, Pattern Classification, John Wiley & Sons, Inc., New York, pp. 395-396, which is hereby incorporated by reference. Tree-based methods divide the feature space into a set of rectangles and then fit a model (such as a constant) in each rectangle. In some embodiments, the decision tree is a random forest regression. One specific algorithm that can be used is classification and regression tree (CART). Other specific decision tree algorithms include, but are not limited to, ID3, C4.5, MART, and random forests. CART, ID3, and C4.5 are described in Duda, 2001, Pattern Classification, John Wiley & Sons, Inc., New York. pp. 396-408 and pp. 411-412, which are hereby incorporated by reference. CART, MART, and C4.5 are described in Hastie et al., 2001, The Elements of Statistical Learning, Springer-Verlag, New York, Chapter 9, which is hereby incorporated by reference in its entirety. Random forests are described in Breiman, 1999, "Random Forests—Random Features," Technical Report 567, Statistics Department, UC Berkeley, September 1999, which is hereby incorporated by reference in its entirety.

[0139] Clustering (e.g., unsupervised clustering model algorithms and supervised clustering model algorithms) is described in Duda and Hart, Pattern Classification and Scene Analysis, 1973, John Wiley & Sons, Inc., New York, pages 211-256 (hereinafter referred to as "Duda 1973"), which is hereby incorporated by reference in its entirety. As described in Section 6.7 of Duda 1973, the clustering problem is described as one of finding natural groupings in a data set. In order to identify natural groupings, two problems are solved. First, determine a way to measure the similarity (or dissimilarity) between two samples. This metric (similarity measure) is used to ensure that the samples in one cluster are more similar than the samples in other clusters. Second, determine a mechanism for partitioning data into clusters using a similarity measure. Similarity measures are discussed in Section 6.7 of Duda 1973, where it is stated that one way to start a clustering study is to define a distance function and calculate a distance matrix between all pairs of samples in a training set. If distance is a good measure of similarity, the distance between reference entities in the same cluster will be significantly smaller than the distance between reference entities in different clusters. However, as described on page 215 of Duda 1973, clustering does not need to use a distance metric. For example, a non-metric similarity function s(x, x') can be used to compare two vectors x and x'. Typically, s(x, x') is a symmetric function that has a larger value when x and x' are "similar" in some way. An example of a non-metric similarity function s(x, x') is provided on page 218 of Duda 1973. Once a method for measuring the "similarity" or "dissimilarity" between points in a data set is selected, clustering requires a standard function that measures the clustering quality of any partition of the data. The data is clustered using the partition of the data set that maximizes the standard function. See page 217 of Duda 1973. Duda 1973, section 6.8, discusses standard functions. Recently, Duda et al., Pattern Classification, 2nd edition, John Wiley & Sons, Inc. New York, has been published. Clustering is described in detail on pages 537-563.More information about clustering techniques can be found in the following references: Kaufman and Rousseeuw, 1990, Finding Groups in Data: An Introduction to Cluster Analysis, Wiley, New York, NY; Everitt, 1993, Cluster analysis (3rd edition), Wiley, New York, NY; and Backer, 1995, Computer-Assisted Reasoning in Cluster Analysis, Prentice Hall, Upper Saddle River, New Jersey, each of which is hereby incorporated by reference. Specific exemplary clustering techniques that can be used in the present disclosure include, but are not limited to, hierarchical clustering (agglomerative clustering using the nearest neighbor algorithm, the farthest neighbor algorithm, the average connection algorithm, the centroid algorithm, or the sum of squares algorithm), k-means clustering, fuzzy k-means clustering algorithm, and Jarvis-Patrick clustering. In some embodiments, clustering includes unsupervised clustering, in which no preconceived notion of what clusters should be formed when the training set is clustered is imposed.

[0140] Regression models, such as multi-class logit models, are described in Agresti, An Introduction to Categorical Data Analysis, 1996, John Wiley & Sons, Inc., New York, Chapter 8, which is hereby incorporated by reference in its entirety. In some embodiments, the model utilizes the regression model disclosed in Hastie et al., 2001, The Elements of Statistical Learning, Springer-Verlag, New York, which is hereby incorporated by reference in its entirety. In some embodiments, gradient boosting models are used for classification algorithms such as those described herein; these gradient boosting models are described in the following references: Boehmke, Bradley; Greenwell, Brandon (2019). "Gradient Boosting". Hands-On Machine Learning with R. Chapman & Hall. Pages 221–245. ISBN 978-1-138-49568-5., which is hereby incorporated by reference in its entirety. In some embodiments, integrated modeling techniques are used, for example, in the classification algorithms described herein; these integrated modeling techniques are described in the implementation of the classification model herein, which is described in Zhou Zhihua (2012). Ensemble Methods: Foundations and Algorithms. Chapman and Hall / CRC. ISBN 978-1-439-83003-1, which is hereby incorporated by reference in its entirety.

[0141] In some embodiments, machine learning analysis is performed by executing one or more programs (e.g., stored in a computer program) that include instructions for performing data analysis. Figure 1 In some embodiments, the data analysis is performed by a system including at least one processor (e.g., processor core 102) and a memory containing instructions for performing the data analysis (e.g., one or more programs stored in non-persistent memory 111 or persistent memory 112).

[0142] Computer Systems

[0143] The present disclosure provides a computer system programmed to implement the method of the present disclosure. Figure 4A computer system 401 is shown that is programmed or otherwise configured to, for example: obtain Raman signatures of a dental sample, analyze Raman spectra across the dental sample spatially, generate temporal Raman curves, process data using a trained model, and predict a subject's diagnostic status with respect to a disease or condition. The computer system 401 can accommodate various aspects of the sensor data analysis of the present disclosure, such as, for example: staining a dental sample, obtaining a fluorescence image of the stained dental sample, analyzing fluorescence intensity across the stained dental sample spatially, generating temporal Raman curves, measuring the dynamics of the temporal curves, processing data using a trained model, and predicting a subject's diagnostic status with respect to a disease or condition. The computer system 401 can be a user's electronic device or a computer system remotely located with respect to the electronic device. The electronic device can be a mobile electronic device.

[0144] The computer system 401 includes a central processing unit (CPU, also referred to herein as a "processor" and "computer processor") 405, which may be a single-core or multi-core processor, or may be a plurality of processors for parallel processing. The computer system 401 also includes a memory or storage location 410 (e.g., random access memory, read-only memory, flash memory), an electronic storage unit 415 (e.g., a hard disk), a communication interface 420 (e.g., a network adapter) for communicating with one or more other systems, and peripheral devices 425 (such as cache, other memory, data storage, and / or an electronic display adapter). The memory 410, the storage unit 415, the interface 420, and the peripheral devices 425 communicate with the CPU 405 via a communication bus (solid lines), such as a motherboard. The storage unit 415 may be a data storage unit (or data repository) for storing data. The computer system 401 may be operably coupled to a computer network ("network") 430 with the aid of the communication interface 420. The network 430 may be the Internet, an internetwork and / or an extranet, or an intranet and / or an extranet in communication with the Internet. Network 430 is a telecommunications and / or data network in some cases. Network 430 may include one or more computer servers that may implement distributed computing, such as cloud computing. In some cases, with the help of computer system 401, network 430 may implement a peer-to-peer network that may enable devices coupled to computer system 401 to act as clients or servers.

[0145] The CPU 405 may execute a series of machine-readable instructions, which may be embodied in a program or software. The instructions may be stored in a memory location, such as the memory 410. The instructions may be directed to the CPU 405, which may then program or otherwise configure the CPU 405 to implement the methods of the present disclosure. Examples of operations performed by the CPU 405 may include fetching, decoding, executing, and writing back.

[0146] CPU 405 may be part of a circuit, such as an integrated circuit. One or more other components of system 401 may be included in the circuit. In some cases, the circuit is an application specific integrated circuit (ASIC).

[0147] Storage unit 415 can store files, such as drivers, libraries, and saved programs. Storage unit 415 can store user data, such as user preferences and user programs. In some cases, computer system 401 can include one or more additional data storage units external to computer system 401, such as located on a remote server that communicates with computer system 401 via an intranet or the Internet.

[0148] Computer system 401 can communicate with one or more remote computer systems via network 430. For example, computer system 401 can communicate with a remote computer system of a user (e.g., a healthcare provider). Examples of remote computer systems include a personal computer (e.g., a portable PC), a notebook computer, or a tablet computer (e.g., iPad, Galaxy Tab), phones, smartphones (e.g. iPhone, Android-supported devices, ) or a personal digital assistant. A user can access the computer system 401 via the network 430.

[0149] The methods described herein may be implemented by machine (e.g., computer processor) executable code stored in an electronic storage location of computer system 401, such as, for example, memory 410 or electronic storage unit 415. Machine executable or machine readable code may be provided in the form of software. During use, the code may be executed by processor 405. In some cases, the code may be retrieved from storage unit 415 and stored on memory 410 for access by processor 405 at any time. In some cases, electronic storage unit 415 may be excluded, and machine executable instructions may be stored on memory 410.

[0150] The code may be precompiled and configured for use with a machine having a processor suitable for executing the code, or may be compiled during runtime. The code may be provided in a programming language, which may be selected so that the code can be executed in a precompiled or compiled manner.

[0151] Aspects of the systems and methods provided herein, such as computer system 401, can be embodied in programming. Various aspects of the technology can be considered as "products" or "articles" generally in the form of machine (or processor) executable code and / or related data, which are carried or embodied in a type of machine-readable medium. The machine executable code can be stored on an electronic storage unit, such as a memory (e.g., read-only memory, random access memory, flash memory) or a hard disk. "Storage" type media can include any or all tangible memories of a computer, processor, etc. or its related modules, such as various semiconductor memories, tape drives, disk drives, etc., which can provide non-transitory storage for software programming at any time. All or part of the software can sometimes communicate through the Internet or various other telecommunications networks. For example, such communications can enable software to be loaded from one computer or processor to another computer or processor, such as from a management server or host computer to a computer platform of an application server. Therefore, another type of medium that can carry software elements includes light waves, radio waves, and electromagnetic waves, such as through wired and optical landline networks and through various air links on the physical interface between local devices. Physical elements that carry such waves, such as wired or wireless links, optical links, etc., can also be considered to be the medium that carries the software. As used herein, unless limited to non-transitory, tangible "storage" media, terms such as computer or machine "readable media" refer to any medium that participates in providing instructions to a processor for execution.

[0152] Therefore, machine-readable media such as computer executable code can take a variety of forms, including but not limited to tangible storage media, carrier media, or physical transmission media. Non-volatile storage media include, for example, optical or magnetic disks, such as any of the storage devices in any computer, etc., such as can be used to implement the database shown in the accompanying drawings, etc. Volatile storage media include dynamic memory, such as the main memory of such a computer platform. Tangible transmission media include coaxial cables; copper wires and optical fibers, including wires that constitute the bus within a computer system. Carrier transmission media can take the form of electrical or electromagnetic signals or sound or light waves, such as those generated during radio frequency (RF) and infrared (IR) data communications. Therefore, common forms of computer-readable media include, for example: floppy drives, floppy disks, hard disks, magnetic tapes, any other magnetic media, CD-ROMs, DVDs or DVD-ROMs, any other optical media, punched card tapes, any other physical storage media with hole patterns, RAMs, ROMs, PROMs and EPROMs, FLASH-EPROMs, any other storage chips or boxes, carriers that transmit data or instructions, cables or links that transmit such carriers, or any other media from which a computer can read program code and / or data. Many of these forms of computer readable media may be involved in carrying one or more sequences of one or more instructions to a processor for execution.

[0153] Computer system 401 may include or communicate with an electronic display 435 including a user interface (UI) 440 for providing, for example, Raman image data, Raman spectral data, temporal Raman curves, and models. Examples of UIs include, but are not limited to, graphical user interfaces (GUIs) and web-based user interfaces.

[0154] The methods and systems of the present disclosure may be implemented by one or more algorithms. The algorithms may be implemented by software when executed by the central processing unit 405. The algorithms may, for example, obtain Raman images of tooth samples, analyze Raman spectra across the tooth samples spatially, generate temporal Raman curves, process data using trained models, and predict the diagnostic status of the subject with respect to a disease or condition.

[0155] Examples

[0156] Example 1: Dynamic Raman spectroscopy in dental samples to determine autism spectrum disorder (ASD) disease risk

[0157] Using the method and system disclosed herein, a dynamic Raman spectroscopy curve of a tooth sample is generated and subsequently analyzed to determine the disease risk of a subject. In general, the temporal dynamics of a biological response (e.g., an inflammatory response) is found to be printed in a sample (e.g., a tooth sample), and can be analyzed to determine the disease risk of a subject. Dynamic Raman spectroscopy curves during the time period including fetal (prenatal) development and early childhood of two groups of children (a first group suffers from autism spectrum disorder and a second group does not suffer from autism spectrum disorder (ASD)) are generated. The dynamic Raman spectroscopy curve is analyzed to reveal new features therein, which accurately distinguish autism cases and controls. For example, it is found that the spectral features in early life can reveal the ASD disease risk in later life. In contrast, the clinical diagnosis of autism is usually determined at around 3 to 4 years old.

[0158] A deciduous tooth sample was obtained from each child subject. The tooth sample was cut open and the Raman spectral signal on the tooth sample was measured to develop a temporal Raman spectral curve indicating physiological responses during the prenatal and postnatal periods. The temporal spectrum was analyzed using a machine learning algorithm of the present disclosure to train a highly accurate classifier to determine disease risk (e.g., autism).

[0159] Figure 5An example of the accuracy of a classifier for diagnosing autism spectrum disorder (ASD) using features derived from ICA-derived dimensions applied to Raman waveforms by RQA is shown, as shown in an experimental receiver operating characteristic (ROC) curve for evaluating the accuracy of the disclosed method for evaluating autism spectrum disorder in a subject. ROC curves can be used to evaluate the performance of binary classifiers. ROC curves are plotted as sensitivity (also known as true positive rate) relative to specificity (also known as true negative rate). A perfect classifier can have 100% sensitivity and 100% specificity and an area under the curve (AUC) of 1.0. As Figure 5 As shown, a classifier configured to determine the presence of ASD in a subject based on a dynamic Raman RQA dynamic curve has a receiver operating characteristic (ROC) area under the curve (AUC) of 0.861, with a 95% confidence interval (CI) of 0.769 to 0.954. The receiver operating characteristic (ROC) shows how the sensitivity and specificity values ​​of the classifier change with changes in the threshold assigned to the probability prediction.

[0160] Thus, analysis of Raman spectral signatures using the methods and systems of the present disclosure successfully determined disease risk for autism with an AUC greater than 0.86, using only spectral signatures measured on biological samples (e.g., dental samples) obtained non-invasively from pediatric subjects. These results suggest that the dynamics of human physiology early in life are associated with later disease that can be accurately detected and analyzed using the methods and systems of the present disclosure.

[0161] Example 2: Dynamic Raman spectroscopy in dental samples to determine ALS disease risk

[0162] Using the methods and systems of the present disclosure, dynamic Raman spectroscopy curves of dental samples are generated and subsequently analyzed to determine the disease risk of a subject. In general, the temporal dynamics of biological responses (e.g., inflammatory responses) are found to be imprinted in samples (e.g., dental samples) and can be analyzed to determine the disease risk of a subject. Dynamic Raman spectra were generated during a time period including early childhood and adolescence for two groups of adults (the first group had amyotrophic lateral sclerosis (ALS) and the second group did not have ALS). The dynamic Raman spectroscopy curves were analyzed to reveal new features therein that accurately distinguished ALS cases from controls. For example, it was found that spectral features early in life could reveal ALS disease risk later in life.

[0163] Permanent tooth samples were obtained from each adult subject. The tooth samples were cut open and Raman spectroscopic signals on the tooth samples were measured to develop temporal Raman spectroscopic curves indicative of physiological responses during early childhood and adolescence. The temporal curves were analyzed using the machine learning algorithm of the present disclosure to train a highly accurate classifier to determine disease risk (e.g., ALS).

[0164] Figure 6 An example of the accuracy of a classifier for diagnosing ALS using features derived from ICA-derived dimensions of RQA applied to Raman waveforms is shown, as shown in an experimental receiver operating characteristic (ROC) curve for evaluating the accuracy of the disclosed method for evaluating autism spectrum disorder in a subject. ROC curves can be used to evaluate the performance of a binary classifier. ROC curves are plotted as sensitivity (also known as true positive rate) versus specificity (also known as true negative rate). A perfect classifier can have 100% sensitivity and 100% specificity and an area under the curve (AUC) of 1.0. As Figure 6 As shown, a classifier configured to determine the presence of ASD in a subject based on a dynamic Raman RQA dynamic curve has a receiver operating characteristic (ROC) area under the curve (AUC) of 0.880, with a 95% confidence interval (CI) of 0.658 to 1.000. The receiver operating characteristic (ROC) shows how the sensitivity and specificity values ​​of the classifier change with changes in the threshold assigned to the probability prediction.

[0165] Thus, analysis of Raman spectral signatures using the methods and systems of the present disclosure successfully determined disease risk for ALS with an AUC greater than 0.88, using only spectral signatures measured on biological samples (e.g., dental samples) from adults. These results suggest that the dynamics of human physiology early in life are associated with later disease, which can be accurately detected and analyzed using the methods and systems of the present disclosure.

[0166] Although the methods described elsewhere herein illustrate sets of steps or operations according to embodiments, those of ordinary skill in the art will recognize many variations based on the teachings described herein. The steps may be performed in a different order. Steps may be added or omitted. Some steps may include sub-steps. Many steps may be repeated multiple times as long as it is beneficial.

[0167] One or more steps of each of the methods or sets of operations may be performed using circuit systems described herein, for example, one or more processors or logic circuit systems, such as programmable array logic or field programmable gate arrays. The circuits may be programmed to provide one or more steps of each method or set of operations, and the program may include program instructions stored on a computer readable memory or programming steps of logic circuits, such as programmable array logic or, for example, field programmable gate arrays.

[0168] Although the preferred examples of the present invention have been shown and described herein, it should be clear to those skilled in the art that such embodiments are provided only as examples. The present invention is not intended to be limited by the specific examples provided in the specification. Although the present invention has been described with reference to the above description, the description and illustration of the embodiments herein are not intended to be understood in a limiting sense. Without departing from the present invention, those skilled in the art will now be aware of many variations, changes and substitutions. In addition, it should be understood that all aspects of the present invention are not limited to the specific depictions, configurations or relative proportions depending on various conditions and variables set forth herein. It should be understood that each alternative of the embodiments of the present invention described herein can be used to practice the present invention. Therefore, it is contemplated that the present invention should also cover any such substitutions, modifications, variations or equivalents. The following claims are intended to define the scope of the present invention, and accordingly cover methods and structures within the scope of these claims and their equivalents.

[0169] Example

[0170] Embodiment 1. A method for predicting a subject's diagnostic status with respect to a disease or condition, the method comprising: (a) exposing a biological sample of the subject to a light source; (b) acquiring a plurality of Raman spectra from the biological sample; (c) processing the plurality of Raman spectra to generate a spatial map of the plurality of Raman spectra; and (d) predicting the subject's diagnostic status with respect to the disease or condition based at least in part on the spatial map of the plurality of Raman spectra.

[0171] Embodiment 2. The method according to embodiment 1, wherein the biological sample comprises a tooth sample, a hair sample, a nail sample or any combination thereof.

[0172] Embodiment 3. The method according to embodiment 1 or 2, further comprising detecting or monitoring changes in the temporal stress curve of the spatial diagram, the changes being indicative of the temporal response of the subject.

[0173] Embodiment 4. The method according to embodiment 3, wherein the temporal response comprises a biological response, a physiological response, an anatomical response, a therapeutic response, a stress-related response, or a combination thereof.

[0174] Embodiment 5. The method of any one of embodiments 1 to 4, wherein the plurality of Raman spectra comprises from about 200 to about 3700 wavenumbers.

[0175] Embodiment 6. The method according to any one of embodiments 1 to 5, wherein collecting comprises using a Raman spectroscopic microscope.

[0176] Embodiment 7. The method according to embodiment 6, wherein the Raman spectroscopic microscope comprises a 50X air-coupled objective lens, a 63X water-immersion coupled objective lens, or any combination thereof.

[0177] Embodiment 8. The method of any one of Embodiments 1 to 7, wherein the light source comprises a laser, wherein the laser comprises a wavelength of about 785 nm, a wavelength of about 532 nm, or any combination thereof.

[0178] Embodiment 9. The method of any one of embodiments 1 to 8, wherein acquisition is performed using an integration time of about 0.2 seconds to about 0.3 seconds.

[0179] Embodiment 10. The method of any one of embodiments 1 to 9, wherein acquiring comprises moving the biological sample in a step size of about 2 microns to about 5 microns after acquiring the Raman spectrum in the plurality of Raman spectra.

[0180] Embodiment 11. The method of any one of embodiments 1 to 10, wherein the disease or condition comprises: autism spectrum disorder (ASD), attention deficit / hyperactivity disorder (ADHD), amyotrophic lateral sclerosis (ALS), schizophrenia, irritable bowel disease (IBD), pediatric kidney disease, renal transplant rejection, pediatric cancer, or any combination thereof.

[0181] Embodiment 12. The method of any one of embodiments 1 to 10, wherein the disease or disorder comprises ASD.

[0182] Embodiment 13. A method according to any one of embodiments 1 to 12, wherein predicting the subject's diagnostic status with respect to a disease or condition comprises processing the spatial map using a trained model.

[0183] Embodiment 14. The method according to embodiment 13, wherein the trained model is selected from the group consisting of: a neural network algorithm, a support vector machine algorithm, a decision tree algorithm, an unsupervised clustering algorithm, a supervised clustering algorithm, a regression algorithm, a gradient boosting algorithm and any combination thereof.

[0184] Embodiment 15. A method according to embodiment 13, wherein the trained model comprises a gradient boosting ensemble model.

[0185] Embodiment 16. A method according to embodiment 13, wherein the trained model is configured to process one or more features selected from the group consisting of: stratification, entropy, capture time (TT), mean diagonal length (MDL), recursion time (RT), Vmax, certainty, Lmax, determination of the linear slope of the time-stress curve, determination of multiple nonlinear parameters describing the curvature of the time-stress curve, determination of sudden changes in the intensity of the time-stress curve, determination of one or more changes in the baseline intensity of the time-stress curve, determination of changes in the frequency domain representation of the time-stress curve, determination of changes in the power spectrum domain representation of the time-stress curve, determination of one or more recursive quantitative analysis parameters, determination of one or more cross-recursive quantitative analysis parameters, determination of one or more joint recursive quantitative analysis parameters, determination of one or more multidimensional recursive quantitative analysis parameters, estimation of the Lyapunov spectrum, determination of the maximum Lyapunov exponent, and any combination thereof.

[0186] Embodiment 17. A method according to embodiment 3, wherein the trained model is configured to process two or more features selected from the group consisting of: stratification, entropy, capture time (TT), mean diagonal length (MDL), recursion time (RT), Vmax, certainty, Lmax, determination of the linear slope of the time-stress curve, determination of multiple nonlinear parameters describing the curvature of the time-stress curve, determination of sudden changes in the intensity of the time-stress curve, determination of one or more changes in the baseline intensity of the time-stress curve, determination of changes in the frequency domain representation of the time-stress curve, determination of changes in the power spectrum domain representation of the time-stress curve, determination of one or more recursive quantitative analysis parameters, determination of one or more cross-recursive quantitative analysis parameters, determination of one or more joint recursive quantitative analysis parameters, determination of one or more multidimensional recursive quantitative analysis parameters, estimation of the Lyapunov spectrum, determination of the maximum Lyapunov exponent, and any combination thereof.

[0187] Embodiment 18. The method of any one of embodiments 1 to 17, wherein the trained model predicts a diagnostic status regarding a disease or condition with a sensitivity of at least about 80%.

[0188] Embodiment 19. The method of embodiment 1, wherein the trained model predicts a diagnostic status regarding a disease or condition with a specificity of at least about 80%.

[0189] Embodiment 20. The method of embodiment 1, wherein the trained model predicts the diagnostic status regarding the disease or condition with a positive predictive value of at least about 80%.

[0190] Embodiment 21. The method of embodiment 1, wherein the trained model predicts a diagnostic status regarding a disease or condition with a negative predictive value of at least about 80%.

[0191] Embodiment 22. A method according to embodiment 1, wherein the trained model predicts the diagnostic status of the disease or condition with an area under the receiver operating characteristic (AUROC) of at least about 0.80.

[0192] Embodiment 23. A device comprising one or more processors and a memory storing one or more programs for execution by the one or more processors, the one or more programs comprising instructions for: (a) sampling each corresponding position of a plurality of positions along a reference line on a biological sample of a subject associated with the subject's Raman signature, thereby obtaining a plurality of Raman spectra, each Raman spectrum of the plurality of Raman spectra corresponding to a different position of the plurality of positions, and each position of the plurality of positions representing a different growth period of the biological sample of the subject associated with the Raman signature; (b) analyzing each Raman spectrum of the plurality of Raman spectra across the reference line on the biological sample, thereby obtaining a first data set; (c) deriving a corresponding second data set from the corresponding plurality of Raman spectral measurement results, each corresponding feature in the corresponding feature set being determined by a sequential change in the Raman spectrum; and (d) processing the features using a trained machine learning model to predict the subject's diagnostic status with respect to a disease or condition associated with the Raman signature.

[0193] Embodiment 24. The device of embodiment 23, wherein the biological sample comprises a tooth sample, a hair sample, a nail sample, or any combination thereof.

[0194] Embodiment 25. The device of Embodiment 23 or 24, wherein the instructions further comprise detecting or monitoring changes in the Raman spectrum across the plurality of locations, the changes being indicative of a temporal response of the subject.

[0195] Embodiment 26. A device according to embodiment 25, wherein the temporal response comprises a biological response, a physiological response, an anatomical response, a therapeutic response, a stress-related response, or a combination thereof.

[0196] Embodiment 27. The device of any one of Embodiments 23 to 26, wherein the plurality of Raman spectra comprises from about 200 to about 3700 wavenumbers.

[0197] Embodiment 28. The apparatus of any one of embodiments 23 to 27, wherein sampling comprises using a Raman spectroscopic microscope.

[0198] Embodiment 29. The apparatus of Embodiment 28, wherein the Raman spectroscopic microscope comprises a 50X air-coupled objective lens, a 63X water-immersion-coupled objective lens, or any combination thereof.

[0199] Embodiment 30. The apparatus of Embodiment 23, wherein sampling comprises exposing the biological sample to a light source to generate a Raman spectrum among a plurality of Raman spectra at a plurality of locations.

[0200] Embodiment 31. The device of Embodiment 30, wherein the light source comprises a laser, wherein the laser comprises a wavelength of approximately 785 nm, a wavelength of approximately 532 nm, or any combination thereof.

[0201] Embodiment 32. An apparatus according to any one of Embodiments 23 to 31, wherein the instructions further include translation, wherein the translation includes moving the biological sample from a first position among a plurality of positions to a second position in a step size of about 2 microns to about 5 microns after acquiring the Raman spectrum in the plurality of Raman spectra.

[0202] Embodiment 33. A device according to embodiment 32, wherein the translation is performed using an integration time of about 0.2 seconds to about 0.3 seconds.

[0203] Embodiment 34. A device according to any one of embodiments 23 to 33, wherein the disease or condition comprises: autism spectrum disorder (ASD), attention-deficit / hyperactivity disorder (ADHD), amyotrophic lateral sclerosis (ALS), schizophrenia, irritable bowel disease (IBD), pediatric kidney disease, renal transplant rejection, pediatric cancer, or any combination thereof.

[0204] Embodiment 35. The device of any one of Embodiments 23 to 33, wherein the disease or condition comprises autism spectrum disorder ASD.

[0205] Embodiment 36. The device of any one of embodiments 23 to 35, wherein predicting the diagnostic status of the subject with respect to a disease or condition comprises processing variations in the Raman spectra across multiple locations using a trained model.

[0206] Embodiment 37. A device according to embodiment 36, wherein the trained model is selected from the group consisting of: a neural network algorithm, a support vector machine algorithm, a decision tree algorithm, an unsupervised clustering algorithm, a supervised clustering algorithm, a regression algorithm, a gradient boosting algorithm and any combination thereof.

[0207] Embodiment 38. A device according to embodiment 36, wherein the trained model includes a gradient boosting ensemble model.

[0208] Embodiment 39. A device according to embodiment 36, wherein the trained model is configured to process one or more features selected from the group consisting of: stratification, entropy, capture time (TT), mean diagonal length (MDL), recursion time (RT), Vmax, certainty, Lmax, determination of linear slopes of the multiple Raman spectra across reference lines, determination of multiple nonlinear parameters describing the curvature of the multiple Raman spectra across reference lines, determination of sudden changes in the intensity of the multiple Raman spectra across reference lines, determination of one or more changes in baseline intensity of the multiple Raman spectra across reference lines, determination of changes in frequency domain representations of the multiple Raman spectra across reference lines, determination of changes in power spectrum domain representations of the multiple Raman spectra across reference lines, determination of one or more recursive quantitative analysis parameters, determination of one or more cross-recursive quantitative analysis parameters, determination of one or more joint recursive quantitative analysis parameters, determination of one or more multidimensional recursive quantitative analysis parameters, estimation of Lyapunov spectrum, determination of maximum Lyapunov exponent, and any combination thereof.

[0209] Embodiment 40. A device according to embodiment 36, wherein the trained model is configured to process two or more features selected from the group consisting of: stratification, entropy, capture time (TT), mean diagonal length (MDL), recursion time (RT), Vmax, certainty, Lmax, determination of linear slopes of the multiple Raman spectra across reference lines, determination of multiple nonlinear parameters describing the curvature of the multiple Raman spectra across reference lines, determination of sudden changes in the intensity of the multiple Raman spectra across reference lines, determination of one or more changes in baseline intensity of the multiple Raman spectra across reference lines, determination of changes in frequency domain representations of the multiple Raman spectra across reference lines, determination of changes in power spectrum domain representations of the multiple Raman spectra across reference lines, determination of one or more recursive quantitative analysis parameters, determination of one or more cross-recursive quantitative analysis parameters, determination of one or more joint recursive quantitative analysis parameters, determination of one or more multidimensional recursive quantitative analysis parameters, estimation of Lyapunov spectrum, determination of maximum Lyapunov exponent, and any combination thereof.

[0210] Embodiment 41. The device of Embodiment 23, wherein the trained model predicts a diagnostic status regarding a disease or condition with a sensitivity of at least about 80%.

[0211] Embodiment 42. The device of Embodiment 23, wherein the trained model predicts a diagnostic status regarding a disease or condition with a specificity of at least about 80%.

[0212] Embodiment 43. The apparatus of Embodiment 23, wherein the trained model predicts a diagnostic status regarding a disease or condition with a positive predictive value of at least about 80%.

[0213] Embodiment 44. The apparatus of Embodiment 23, wherein the trained model predicts a diagnostic status regarding a disease or condition with a negative predictive value of at least about 80%.

[0214] Embodiment 45. The apparatus of Embodiment 23, wherein the trained model predicts a diagnostic status regarding a disease or condition with an area under the receiver operating characteristic (AUROC) of at least about 0.80.

[0215] Embodiment 46. A non-transitory computer-readable storage medium and one or more computer programs embedded therein for classification, the one or more computer programs comprising instructions that, when executed by a computer system, cause the computer system to perform a method comprising: (a) sampling each corresponding position of a plurality of positions along a reference line on a biological sample of a subject associated with the subject's Raman signature, thereby obtaining a plurality of Raman spectra, each Raman spectrum of the plurality of Raman spectra corresponding to a different position of the plurality of positions, and each position of the plurality of positions representing a different growth period of the biological sample of the subject associated with the Raman signature; (b) analyzing each Raman spectrum of the plurality of Raman spectra across the reference line on the biological sample, thereby obtaining a first data set; (c) deriving a corresponding second data set from the corresponding plurality of Raman spectral measurement results, each corresponding feature in the corresponding feature set being determined by a sequential change in the Raman spectrum; and (d) processing the features using a trained model to predict the subject's diagnostic status with respect to a disease or condition associated with the Raman signature.

[0216] Embodiment 47. The non-transitory computer-readable storage medium of Embodiment 46, wherein the biological sample comprises a tooth sample, a hair sample, a nail sample, or any combination thereof.

[0217] Embodiment 48. A non-transitory computer-readable storage medium according to embodiment 46 or 47, wherein the method further comprises detecting or monitoring changes in the Raman spectrum across multiple locations, the changes being indicative of the temporal response of the subject.

[0218] Embodiment 49. A non-transitory computer-readable storage medium according to embodiment 48, wherein the temporal response comprises a biological response, a physiological response, an anatomical response, a therapeutic response, a stress-related response, or a combination thereof.

[0219] Embodiment 50. The non-transitory computer-readable storage medium of any one of Embodiments 46 to 49, wherein the plurality of Raman spectra comprises about 200 to about 3700 wavenumbers.

[0220] Embodiment 51. The non-transitory computer-readable storage medium of any one of Embodiments 46 to 50, wherein sampling comprises using a Raman spectroscopic microscope.

[0221] Embodiment 52. A non-transitory computer-readable storage medium according to Embodiment 51, wherein the Raman spectroscopic microscope comprises a 50X air-coupled objective lens, a 63X water-immersion coupled objective lens, or any combination thereof.

[0222] Embodiment 53. A non-transitory computer-readable storage medium according to any one of embodiments 46 to 52, wherein sampling comprises exposing the biological sample to a light source to generate a Raman spectrum in a plurality of Raman spectra at a plurality of locations.

[0223] Embodiment 54. The non-transitory computer-readable storage medium of Embodiment 53, wherein the light source comprises a laser, wherein the laser comprises a wavelength of approximately 785 nm, a wavelength of approximately 532 nm, or any combination thereof.

[0224] Embodiment 55. A non-transitory computer-readable storage medium according to any one of embodiments 46 to 54, wherein the instructions further include translation, wherein the translation includes moving the biological sample from a first position among a plurality of positions to a second position in a step size of about 2 microns to about 5 microns after acquiring a Raman spectrum in a plurality of Raman spectra.

[0225] Embodiment 56. A non-transitory computer-readable storage medium according to embodiment 55, wherein the translation is performed using an integration time of about 0.2 seconds to about 0.3 seconds.

[0226] Embodiment 57. A non-transitory computer-readable storage medium according to any one of embodiments 46 to 56, wherein the disease or condition comprises: autism spectrum disorder (ASD), attention deficit / hyperactivity disorder (ADHD), amyotrophic lateral sclerosis (ALS), schizophrenia, irritable bowel disease (IBD), pediatric kidney disease, renal transplant rejection, pediatric cancer, or any combination thereof.

[0227] Embodiment 58. A non-transitory computer-readable storage medium according to any one of Embodiments 46 to 56, wherein the disease or condition comprises autism spectrum disorder (ASD).

[0228] Embodiment 59. A non-transitory computer-readable storage medium according to any one of embodiments 46 to 58, wherein predicting the subject's diagnostic status with respect to a disease or condition comprises processing variations in the Raman spectra across multiple locations using a trained model.

[0229] Embodiment 60. A non-transitory computer-readable storage medium according to embodiment 59, wherein the trained model is selected from the group consisting of: a neural network algorithm, a support vector machine algorithm, a decision tree algorithm, an unsupervised clustering algorithm, a supervised clustering algorithm, a regression algorithm, a gradient boosting algorithm, and any combination thereof.

[0230] Embodiment 61. A non-transitory computer-readable storage medium according to embodiment 59, wherein the trained model includes a gradient boosting ensemble model.

[0231] Embodiment 62. A non-transitory computer-readable storage medium according to embodiment 59, wherein the trained model is configured to process one or more features selected from the group consisting of: stratification, entropy, capture time (TT), mean diagonal length (MDL), recursion time (RT), Vmax, certainty, Lmax, determination of linear slopes of the multiple Raman spectra across reference lines, determination of multiple nonlinear parameters describing the curvature of the multiple Raman spectra across reference lines, determination of sudden changes in the intensity of the multiple Raman spectra across reference lines, determination of one or more changes in baseline intensity of the multiple Raman spectra across reference lines, determination of changes in frequency domain representations of the multiple Raman spectra across reference lines, determination of changes in power spectrum domain representations of the multiple Raman spectra across reference lines, determination of one or more recursive quantitative analysis parameters, determination of one or more cross-recursive quantitative analysis parameters, determination of one or more joint recursive quantitative analysis parameters, determination of one or more multidimensional recursive quantitative analysis parameters, estimation of Lyapunov spectrum, determination of maximum Lyapunov exponent, and any combination thereof.

[0232] Embodiment 63. A non-transitory computer-readable storage medium according to embodiment 59, wherein the trained model is configured to process two or more features selected from the group consisting of: stratification, entropy, capture time (TT), mean diagonal length (MDL), recursion time (RT), Vmax, certainty, Lmax, determination of linear slopes of the multiple Raman spectra across reference lines, determination of multiple nonlinear parameters describing the curvature of the multiple Raman spectra across reference lines, determination of sudden changes in the intensity of the multiple Raman spectra across reference lines, determination of one or more changes in baseline intensity of the multiple Raman spectra across reference lines, determination of changes in frequency domain representations of the multiple Raman spectra across reference lines, determination of changes in power spectrum domain representations of the multiple Raman spectra across reference lines, determination of one or more recursive quantitative analysis parameters, determination of one or more cross-recursive quantitative analysis parameters, determination of one or more joint recursive quantitative analysis parameters, determination of one or more multidimensional recursive quantitative analysis parameters, estimation of Lyapunov spectrum, determination of maximum Lyapunov exponent, and any combination thereof.

[0233] Embodiment 64. The non-transitory computer-readable storage medium of Embodiment 46, wherein the trained model predicts a diagnostic status regarding a disease or condition with a sensitivity of at least about 80%.

[0234] Embodiment 65. A non-transitory computer-readable storage medium according to Embodiment 46, wherein the trained model predicts a diagnostic status regarding a disease or condition with a specificity of at least about 80%.

[0235] Embodiment 66. The non-transitory computer-readable storage medium of Embodiment 46, wherein the instructions further comprise predicting the subject's diagnostic status with respect to the disease or condition with a positive predictive value of at least about 80%.

[0236] Embodiment 67. A non-transitory computer-readable storage medium according to Embodiment 46, wherein the trained model predicts a diagnostic status regarding a disease or condition with a positive predictive value of at least about 80%.

[0237] Embodiment 68. A non-transitory computer-readable storage medium according to Embodiment 46, wherein the trained model predicts a diagnostic status regarding a disease or condition with a negative predictive value of at least about 80%.

[0238] Embodiment 69. A method for training a model, the method comprising: at a computer system having one or more processors and a memory storing one or more programs for execution by the one or more processors: (a) for each corresponding training subject in a plurality of training subjects, wherein a first subset of the training subjects in the plurality of training subjects have a first diagnostic state, the first diagnostic state corresponding to a first biological condition associated with a Raman feature, and a second subset of the training subjects in the plurality of training subjects have a second diagnostic state, the second diagnostic state corresponding to the first biological condition not associated with the Raman feature: (i) sampling each corresponding position in a plurality of positions along a reference line on a biological sample of the subject associated with the Raman feature of the subject, thereby obtaining a plurality of Raman spectra, each Raman spectrum in the plurality of Raman spectra corresponding to a different position in the plurality of positions, and wherein the plurality of positions have a first diagnostic state corresponding to a first biological condition not associated with the Raman feature; (a) analyzing each Raman spectrum across a reference line on the biological sample to obtain a first data set; and (b) deriving a corresponding second data set from the corresponding plurality of Raman spectra, each corresponding feature in a corresponding feature set being determined by a sequential change in the Raman spectrum; and (c) training an untrained or partially untrained model with (i) a corresponding feature set for each corresponding second data set of each training subject in the plurality of training subjects and (ii) a corresponding diagnostic state selected from a first diagnostic state and a second diagnostic state for each training subject in the plurality of training subjects to obtain a trained model that provides an indication of whether the test subject has the first biological condition associated with the Raman feature based on values ​​of features in the feature set obtained from the biological sample associated with the test subject's Raman feature.

[0239] Embodiment 70. A method according to embodiment 69, wherein the trained model is selected from the group consisting of: a neural network algorithm, a support vector machine algorithm, a decision tree algorithm, an unsupervised clustering algorithm, a supervised clustering algorithm, a regression algorithm, a gradient boosting algorithm and any combination thereof.

[0240] Embodiment 71. The method of embodiment 69, wherein the trained model is a polynomial classifier.

[0241] Embodiment 72. The method of embodiment 69, wherein the trained model is a binomial classifier.

[0242] Embodiment 73. A method according to embodiment 69, wherein the first biological condition is selected from the group consisting of: autism spectrum disorder (ASD), attention deficit / hyperactivity disorder (ADHD), amyotrophic lateral sclerosis (ALS), schizophrenia, irritable bowel disease (IBD), pediatric kidney disease, renal transplant rejection and pediatric cancer.

[0243] Embodiment 74. A method according to any one of Embodiments 69 to 73, wherein evaluating the test subject for a first biological condition associated with the Raman signature further comprises: distinguishing the first biological condition associated with the Raman signature from a second biological condition associated with the Raman signature that is different from the first biological condition associated with the Raman signature.

[0244] Embodiment 75. The method of embodiment 74, wherein the first biological condition is autism spectrum disorder and the second biological condition is attention-deficit / hyperactivity disorder.

[0245] Embodiment 76. The method of any one of Embodiments 69 to 75, wherein the test subject is a human.

[0246] Embodiment 77. The method of Embodiment 76, wherein the human is less than 12 years old.

[0247] Embodiment 78. The method of Embodiment 76, wherein the human is less than 1 year old.

[0248] Embodiment 79. The method of any one of embodiments 69 to 78, wherein the corresponding biological sample associated with the Raman signature of the corresponding training subject is selected from the group consisting of hair shafts, teeth, and nails.

[0249] Embodiment 80. The method of Embodiment 79, wherein the corresponding biological sample associated with the Raman signature of the corresponding training subject is a hair shaft, and wherein the reference line corresponds to a longitudinal direction of the hair shaft.

[0250] Embodiment 81. A method according to Embodiment 79, wherein the corresponding biological sample associated with the Raman signature of the corresponding training subject is a tooth, and wherein the reference line corresponds to a direction across the growth zone, including the new growth line of the tooth.

[0251] Embodiment 82. A method according to any one of Embodiments 69 to 81, wherein the corresponding multiple positions are ordered so that the first position among the corresponding multiple positions along the corresponding biological sample of the corresponding training subject corresponds to the position of the tip of the corresponding biological sample closest to the corresponding training subject.

[0252] Embodiment 83. A method according to any one of Embodiments 69 to 82, wherein each trace in the corresponding plurality of Raman spectroscopy measurements comprises a plurality of data points, each data point being an instance of a corresponding position in the plurality of positions.

[0253] Embodiment 84. A method according to any one of embodiments 69 to 83, wherein the corresponding feature set is selected from the group consisting of: hierarchicality, entropy, capture time (TT), mean diagonal length (MDL), recursion time (RT), Vmax, certainty, Lmax.

[0254] Embodiment 85. A method according to any one of embodiments 69 to 83, wherein the corresponding multiple positions include at least 1000, 1500, 2000, 2500, 3000, 3500, 4000, 4500 or 5000, 5500, 6000, 6500, 7000, 7500, 8000, 8500, 9000, 9500, 10000 or more than 10000 positions.

[0255] Embodiment 86. A method according to any one of embodiments 69 to 85, wherein the trained model is configured to process one or more features selected from the group consisting of: stratification, entropy, capture time (TT), mean diagonal length (MDL), recursion time (RT), Vmax, certainty, Lmax, determination of linear slopes of the multiple Raman spectra across reference lines, determination of multiple nonlinear parameters describing the curvature of the multiple Raman spectra across reference lines, determination of sudden changes in the intensity of the multiple Raman spectra across reference lines, determination of one or more changes in baseline intensity of the multiple Raman spectra across reference lines, determination of changes in frequency domain representations of the multiple Raman spectra across reference lines, determination of changes in power spectrum domain representations of the multiple Raman spectra across reference lines, determination of one or more recursive quantitative analysis parameters, determination of one or more cross-recursive quantitative analysis parameters, determination of one or more joint recursive quantitative analysis parameters, determination of one or more multidimensional recursive quantitative analysis parameters, estimation of Lyapunov spectrum, determination of maximum Lyapunov exponent, and any combination thereof.

[0256] Embodiment 87. A method according to any one of embodiments 69 to 85, wherein the trained model is configured to process two or more features selected from the group consisting of: stratification, entropy, capture time (TT), mean diagonal length (MDL), recursion time (RT), Vmax, certainty, Lmax, determination of linear slopes of the multiple Raman spectra across reference lines, determination of multiple nonlinear parameters describing the curvature of the multiple Raman spectra across reference lines, determination of sudden changes in the intensity of the multiple Raman spectra across reference lines, determination of one or more changes in baseline intensity of the multiple Raman spectra across reference lines, determination of changes in frequency domain representations of the multiple Raman spectra across reference lines, determination of changes in power spectrum domain representations of the multiple Raman spectra across reference lines, determination of one or more recursive quantitative analysis parameters, determination of one or more cross-recursive quantitative analysis parameters, determination of one or more joint recursive quantitative analysis parameters, determination of one or more multidimensional recursive quantitative analysis parameters, estimation of Lyapunov spectrum, determination of maximum Lyapunov exponent, and any combination thereof.

Claims

1. A method for predicting a subject's diagnostic status with respect to a disease or condition, the method comprising: (a) exposing a biological sample from a subject to a light source; (b) collecting a plurality of Raman spectra from the biological sample; (c) processing the plurality of Raman spectra to generate a spatial map of the plurality of Raman spectra; as well as (d) predicting a diagnostic status of the subject with respect to the disease or condition based at least in part on the spatial map of the plurality of Raman spectra.

2. The method of claim 1, wherein the biological sample comprises a tooth sample, a hair sample, a nail sample, or any combination thereof.

3. The method according to claim 1 or 2, further comprising detecting or monitoring changes in the temporal stress curve of the spatial diagram, the changes indicating the temporal response of the subject.

4. The method of claim 3, wherein the temporal response comprises a biological response, a physiological response, an anatomical response, a therapeutic response, a stress-related response, or a combination thereof.

5. The method of any one of claims 1 to 4, wherein the plurality of Raman spectra comprises from about 200 to about 3700 wavenumbers.

6. The method of any one of claims 1 to 5, wherein collecting comprises using a Raman spectroscopic microscope.

7. The method of claim 6, wherein the Raman spectroscopic microscope comprises a 50X air-coupled objective lens, a 63X water-immersion-coupled objective lens, or any combination thereof.

8. The method of any one of claims 1 to 7, wherein the light source comprises a laser, wherein the laser comprises a wavelength of about 785 nm, a wavelength of about 532 nm, or any combination thereof.

9. The method of any one of claims 1 to 8, wherein the acquisition is performed using an integration time of about 0.2 seconds to about 0.3 seconds.

10. The method of any one of claims 1 to 9, wherein the acquiring comprises moving the biological sample in a step size of about 2 microns to about 5 microns after acquiring the Raman spectrum in the plurality of Raman spectra.

11. The method according to any one of claims 1 to 10, wherein the disease or condition comprises: Autism spectrum disorder (ASD), attention-deficit / hyperactivity disorder (ADHD), amyotrophic lateral sclerosis (ALS), schizophrenia, irritable bowel disease (IBD), pediatric kidney disease, renal transplant rejection, pediatric cancer, or any combination thereof.

12. The method of any one of claims 1 to 10, wherein the disease or condition comprises ASD.

13. The method of any one of claims 1 to 12, wherein predicting the subject's diagnostic status with respect to the disease or condition comprises processing the spatial map using a trained model.

14. The method of claim 13, wherein the trained model is selected from the group consisting of a neural network algorithm, a support vector machine algorithm, a decision tree algorithm, an unsupervised clustering algorithm, a supervised clustering algorithm, a regression algorithm, a gradient boosting algorithm, and any combination thereof.

15. The method of claim 13, wherein the trained model comprises a gradient boosting ensemble model.

16. A method according to claim 3, wherein the trained model is configured to process one or more features selected from the group consisting of: hierarchicality, entropy, capture time (TT), mean diagonal length (MDL), recursion time (RT), Vmax, certainty, Lmax, determination of the linear slope of the time stress curve, determination of multiple nonlinear parameters describing the curvature of the time stress curve, determination of sudden changes in the intensity of the time stress curve, determination of one or more changes in the baseline intensity of the time stress curve, determination of changes in the frequency domain representation of the time stress curve, determination of changes in the power spectrum domain representation of the time stress curve, determination of one or more recursive quantitative analysis parameters, determination of one or more cross-recursive quantitative analysis parameters, determination of one or more joint recursive quantitative analysis parameters, determination of one or more multidimensional recursive quantitative analysis parameters, estimation of Lyapunov spectrum, determination of maximum Lyapunov exponent, and any combination thereof.

17. The method of claim 16, wherein the trained model is configured to process two or more features selected from the group consisting of: stratification, entropy, capture time (TT), mean diagonal length (MDL), recursion time (RT), Vmax, certainty, Lmax, determination of the linear slope of the time-stress curve, determination of multiple nonlinear parameters describing the curvature of the time-stress curve, determination of sudden changes in the intensity of the time-stress curve, determination of one or more changes in the baseline intensity of the time-stress curve, determination of changes in the frequency domain representation of the time-stress curve, determination of changes in the power spectrum domain representation of the time-stress curve, determination of one or more recursive quantitative analysis parameters, determination of one or more cross-recursive quantitative analysis parameters, determination of one or more joint recursive quantitative analysis parameters, determination of one or more multidimensional recursive quantitative analysis parameters, estimation of the Lyapunov spectrum, determination of the maximum Lyapunov exponent, and any combination thereof.

18. The method of any one of claims 1 to 17, wherein the trained model predicts a diagnostic status with respect to the disease or condition with a sensitivity of at least about 80%.

19. The method of claim 1, wherein the trained model predicts a diagnostic status with respect to the disease or condition with a specificity of at least about 80%.

20. The method of claim 1, wherein the trained model predicts a diagnostic status with respect to the disease or condition with a positive predictive value of at least about 80%.

21. The method of claim 1, wherein the trained model predicts a diagnostic status with respect to the disease or condition with a negative predictive value of at least about 80%.

22. The method of claim 1, wherein the trained model predicts a diagnostic status regarding the disease or condition with an area under the receiver operating characteristic (AUROC) of at least about 0.

80.

23. An apparatus comprising one or more processors and a memory storing one or more programs for execution by the one or more processors, the one or more programs comprising instructions for: (a) sampling each corresponding position of a plurality of positions along a reference line on a biological sample of a subject associated with a Raman feature of the subject, thereby obtaining a plurality of Raman spectra, each Raman spectrum of the plurality of Raman spectra corresponding to a different position of the plurality of positions, and each position of the plurality of positions representing a different growth period of the biological sample associated with the Raman feature; (b) analyzing each of the plurality of Raman spectra across a reference line on the biological sample to obtain a first data set; (c) deriving a corresponding second data set from a corresponding plurality of Raman spectral measurements, each corresponding feature in a corresponding feature set being determined by a sequential change in the Raman spectrum; and (d) processing the features using the trained model to predict a diagnostic status of the subject with respect to a disease or condition associated with the Raman features.

24. The device of claim 23, wherein the biological sample comprises a tooth sample, a hair sample, a nail sample, or any combination thereof.

25. The apparatus of claim 23 or 24, wherein the instructions further comprise detecting or monitoring changes in the Raman spectrum across the plurality of locations, the changes being indicative of a temporal response of the subject.

26. The device of claim 25, wherein the temporal response comprises a biological response, a physiological response, an anatomical response, a therapeutic response, a stress-related response, or a combination thereof.

27. The apparatus of any one of claims 23 to 26, wherein the plurality of Raman spectra comprises from about 200 to about 3700 wavenumbers.

28. An apparatus according to any one of claims 23 to 27, wherein sampling comprises using a Raman spectroscopic microscope.

29. The apparatus of claim 28, wherein the Raman spectroscopic microscope comprises a 50X air-coupled objective lens, a 63X water-immersion-coupled objective lens, or any combination thereof.

30. The apparatus of claim 23, wherein the sampling comprises exposing the biological sample to a light source to generate the Raman spectrum in the plurality of Raman spectra at the plurality of locations.

31. The apparatus of claim 30, wherein the light source comprises a laser, wherein the laser comprises a wavelength of approximately 785 nm, a wavelength of approximately 532 nm, or any combination thereof.

32. The apparatus of any one of claims 23 to 31, wherein the instructions further comprise translating, wherein translating comprises moving the biological sample from a first position in the plurality of positions to a second position in a step size of about 2 microns to about 5 microns after acquiring a Raman spectrum in the plurality of Raman spectra.

33. The apparatus of claim 32, wherein the translation is performed using an integration time of about 0.2 seconds to about 0.3 seconds.

34. The device of any one of claims 23 to 33, wherein the disease or condition comprises: Autism spectrum disorder (ASD), attention-deficit / hyperactivity disorder (ADHD), amyotrophic lateral sclerosis (ALS), schizophrenia, irritable bowel disease (IBD), pediatric kidney disease, renal transplant rejection, pediatric cancer, or any combination thereof.

35. The apparatus of any one of claims 23 to 33, wherein the disease or condition comprises autism spectrum disorder (ASD).

36. The apparatus of any one of claims 23 to 35, wherein predicting a diagnostic status of a subject with respect to the disease or condition comprises processing variations in the Raman spectrum across the plurality of locations using a trained model.

37. The apparatus of claim 36, wherein the trained model is selected from the group consisting of a neural network algorithm, a support vector machine algorithm, a decision tree algorithm, an unsupervised clustering algorithm, a supervised clustering algorithm, a regression algorithm, a gradient boosting algorithm, and any combination thereof.

38. The apparatus of claim 36, wherein the trained model comprises a gradient boosting ensemble model.

39. An apparatus according to claim 36, wherein the trained model is configured to process one or more features selected from the group consisting of: stratification, entropy, capture time (TT), mean diagonal length (MDL), recursion time (RT), Vmax, certainty, Lmax, determination of linear slopes of the multiple Raman spectra across reference lines, determination of multiple nonlinear parameters describing the curvature of the multiple Raman spectra across reference lines, determination of sudden changes in intensity of the multiple Raman spectra across reference lines, determination of one or more changes in baseline intensity of the multiple Raman spectra across reference lines, determination of changes in frequency domain representations of the multiple Raman spectra across reference lines, determination of changes in power spectrum domain representations of the multiple Raman spectra across reference lines, determination of one or more recursive quantitative analysis parameters, determination of one or more cross-recursive quantitative analysis parameters, determination of one or more joint recursive quantitative analysis parameters, determination of one or more multidimensional recursive quantitative analysis parameters, estimation of Lyapunov spectrum, determination of maximum Lyapunov exponent, and any combination thereof.

40. An apparatus according to claim 36, wherein the trained model is configured to process two or more features selected from the group consisting of: stratification, entropy, capture time (TT), mean diagonal length (MDL), recursion time (RT), Vmax, certainty, Lmax, determination of linear slopes of the multiple Raman spectra across reference lines, determination of multiple nonlinear parameters describing the curvature of the multiple Raman spectra across reference lines, determination of sudden changes in intensity of the multiple Raman spectra across reference lines, determination of one or more changes in baseline intensity of the multiple Raman spectra across reference lines, determination of changes in frequency domain representations of the multiple Raman spectra across reference lines, determination of changes in power spectral domain representations of the multiple Raman spectra across reference lines, determination of one or more recursive quantitative analysis parameters, determination of one or more cross-recursive quantitative analysis parameters, determination of one or more joint recursive quantitative analysis parameters, determination of one or more multidimensional recursive quantitative analysis parameters, estimation of Lyapunov spectrum, determination of maximum Lyapunov exponent, and any combination thereof.

41. The device of claim 23, wherein the trained model predicts a diagnostic status regarding the disease or condition with a sensitivity of at least about 80%.

42. The device of claim 23, wherein the trained model predicts a diagnostic status with respect to the disease or condition with a specificity of at least about 80%.

43. The device of claim 23, wherein the trained model predicts a diagnostic status with respect to the disease or condition with a positive predictive value of at least about 80%.

44. The device of claim 23, wherein the trained model predicts a diagnostic status with respect to the disease or condition with a negative predictive value of at least about 80%.

45. The device of claim 23, wherein the trained model predicts a diagnostic status regarding the disease or condition with an area under the receiver operating characteristic (AUROC) of at least about 0.

80.

46. ​​A non-transitory computer-readable storage medium and one or more computer programs embedded therein for classification, the one or more computer programs comprising instructions that, when executed by a computer system, cause the computer system to perform a method comprising: (a) sampling each corresponding position of a plurality of positions along a reference line on a biological sample of a subject associated with a Raman feature of the subject, thereby obtaining a plurality of Raman spectra, each Raman spectrum of the plurality of Raman spectra corresponding to a different position of the plurality of positions, and each position of the plurality of positions representing a different growth period of the biological sample associated with the Raman feature; (b) analyzing each of the plurality of Raman spectra across a reference line on the biological sample to obtain a first data set; (c) deriving a corresponding second data set from a corresponding plurality of Raman spectral measurements, each corresponding feature in a corresponding feature set being determined by a sequential change in the Raman spectrum; and (d) processing the features using the trained model to predict a diagnostic status of the subject with respect to a disease or condition associated with the Raman features.

47. The non-transitory computer-readable storage medium of claim 46, wherein the biological sample comprises a tooth sample, a hair sample, a nail sample, or any combination thereof.

48. The non-transitory computer-readable storage medium of claim 46 or 47, wherein the method further comprises detecting or monitoring changes in the Raman spectrum across the plurality of locations, the changes being indicative of a temporal response of the subject.

49. The non-transitory computer readable storage medium of claim 48, wherein the temporal response comprises a biological response, a physiological response, an anatomical response, a therapeutic response, a stress-related response, or a combination thereof.

50. The non-transitory computer readable storage medium of any one of claims 46 to 49, wherein the plurality of Raman spectra comprises about 200 to about 3700 wavenumbers.

51. The non-transitory computer readable storage medium of any one of claims 46 to 50, wherein sampling comprises using a Raman spectroscopic microscope.

52. The non-transitory computer-readable storage medium of claim 51, wherein the Raman spectroscopic microscope comprises a 50X air-coupled objective, a 63X water-immersion-coupled objective, or any combination thereof.

53. The non-transitory computer-readable storage medium of any one of claims 46 to 52, wherein sampling comprises exposing the biological sample to a light source to generate the Raman spectrum of the plurality of Raman spectra at the plurality of locations.

54. The non-transitory computer readable storage medium of claim 53, wherein the light source comprises a laser, wherein the laser comprises a wavelength of approximately 785 nm, a wavelength of approximately 532 nm, or any combination thereof.

55. The non-transitory computer-readable storage medium of any one of claims 46 to 54, wherein the instructions further comprise translating, wherein translating comprises moving the biological sample from a first position in the plurality of positions to a second position in a step size of about 2 microns to about 5 microns after acquiring a Raman spectrum in the plurality of Raman spectra.

56. The non-transitory computer readable storage medium of claim 55, wherein the translation is performed using an integration time of about 0.2 seconds to about 0.3 seconds.

57. The non-transitory computer-readable storage medium of any one of claims 46 to 56, wherein the disease or condition comprises: Autism spectrum disorder (ASD), attention-deficit / hyperactivity disorder (ADHD), amyotrophic lateral sclerosis (ALS), schizophrenia, irritable bowel disease (IBD), pediatric kidney disease, renal transplant rejection, pediatric cancer, or any combination thereof.

58. The non-transitory computer-readable storage medium of any one of claims 46 to 56, wherein the disease or condition comprises autism spectrum disorder (ASD).

59. The non-transitory computer-readable storage medium of any one of claims 46 to 58, wherein predicting a diagnostic status of a subject with respect to the disease or condition comprises processing variations in the Raman spectrum across the plurality of locations using a trained model.

60. The non-transitory computer-readable storage medium of claim 59, wherein the trained model is selected from the group consisting of a neural network algorithm, a support vector machine algorithm, a decision tree algorithm, an unsupervised clustering algorithm, a supervised clustering algorithm, a regression algorithm, a gradient boosting algorithm, and any combination thereof.

61. The non-transitory computer-readable storage medium of claim 59, wherein the trained model comprises a gradient boosting ensemble model.

62. The non-transitory computer-readable storage medium of claim 59, wherein the trained model is configured to process one or more features selected from the group consisting of: stratification, entropy, capture time (TT), mean diagonal length (MDL), recursion time (RT), Vmax, certainty, Lmax, determination of linear slopes of the multiple Raman spectra across reference lines, determination of multiple nonlinear parameters describing curvature of the multiple Raman spectra across reference lines, determination of abrupt changes in intensity of the multiple Raman spectra across reference lines, determination of one or more changes in baseline intensity of the multiple Raman spectra across reference lines, determination of changes in frequency domain representations of the multiple Raman spectra across reference lines, determination of changes in power spectral domain representations of the multiple Raman spectra across reference lines, determination of one or more recursive quantitative analysis parameters, determination of one or more cross-recursive quantitative analysis parameters, determination of one or more joint recursive quantitative analysis parameters, determination of one or more multidimensional recursive quantitative analysis parameters, estimation of a Lyapunov spectrum, determination of a maximum Lyapunov exponent, and any combination thereof.

63. The non-transitory computer-readable storage medium of claim 59, wherein the trained model is configured to process two or more features selected from the group consisting of: stratification, entropy, capture time (TT), mean diagonal length (MDL), recursion time (RT), Vmax, certainty, Lmax, determination of linear slopes of the multiple Raman spectra across reference lines, determination of multiple nonlinear parameters describing curvature of the multiple Raman spectra across reference lines, determination of abrupt changes in intensity of the multiple Raman spectra across reference lines, determination of one or more changes in baseline intensity of the multiple Raman spectra across reference lines, determination of changes in frequency domain representations of the multiple Raman spectra across reference lines, determination of changes in power spectral domain representations of the multiple Raman spectra across reference lines, determination of one or more recursive quantitative analysis parameters, determination of one or more cross-recursive quantitative analysis parameters, determination of one or more joint recursive quantitative analysis parameters, determination of one or more multidimensional recursive quantitative analysis parameters, estimation of a Lyapunov spectrum, determination of a maximum Lyapunov exponent, and any combination thereof.

64. The non-transitory computer-readable storage medium of claim 46, wherein the trained model predicts a diagnostic status regarding the disease or condition with a sensitivity of at least about 80%.

65. The non-transitory computer-readable storage medium of claim 46, wherein the trained model predicts a diagnostic status with respect to the disease or condition with a specificity of at least about 80%.

66. The non-transitory computer readable storage medium of claim 46, wherein the instructions further comprise predicting the subject's diagnostic status with respect to the disease or condition with a positive predictive value of at least about 80%.

67. The non-transitory computer-readable storage medium of claim 46, wherein the trained model predicts a diagnostic status with respect to the disease or condition with a positive predictive value of at least about 80%.

68. The non-transitory computer-readable storage medium of claim 46, wherein the trained model predicts a diagnostic status with respect to the disease or condition with a negative predictive value of at least about 80%.

69. A method for training a model, the method comprising: At a computer system having one or more processors and a memory storing one or more programs for execution by the one or more processors: (a) for each respective training subject in a plurality of training subjects, wherein a first subset of the training subjects in the plurality of training subjects have a first diagnostic status corresponding to a first biological condition having a correlation with the Raman signature, and a second subset of the training subjects in the plurality of training subjects have a second diagnostic status corresponding to not having the first biological condition having a correlation with the Raman signature: (i) sampling each corresponding position of a plurality of positions along a reference line on a biological sample of the subject associated with the Raman feature of the subject, thereby obtaining a plurality of Raman spectra, each Raman spectrum of the plurality of Raman spectra corresponding to a different position of the plurality of positions, and each position of the plurality of positions representing a different growth period of the biological sample of the subject associated with the Raman feature; (ii) analyzing each Raman spectrum across a reference line on the biological sample to obtain a first data set; as well as (iii) deriving a corresponding second data set from the corresponding plurality of Raman spectra, each corresponding feature in the corresponding feature set being determined by a sequential change in the Raman spectrum; as well as (b) training an untrained or partially untrained model with (i) the corresponding feature set of each corresponding second data set of each training subject in the plurality of training subjects and (ii) the corresponding diagnostic state selected from the first diagnostic state and the second diagnostic state of each training subject in the plurality of training subjects, thereby obtaining a trained model, wherein the trained model provides an indication of whether the test subject has the first biological condition associated with the Raman feature based on values ​​of features in the feature set obtained from a biological sample associated with the Raman feature of the test subject.

70. The method of claim 69, wherein the trained model is selected from the group consisting of a neural network algorithm, a support vector machine algorithm, a decision tree algorithm, an unsupervised clustering algorithm, a supervised clustering algorithm, a regression algorithm, a gradient boosting algorithm, and any combination thereof.

71. The method of claim 69, wherein the trained model is a polynomial classifier.

72. The method of claim 69, wherein the trained model is a binomial classifier.

73. The method of claim 69, wherein the first biological condition is selected from the group consisting of autism spectrum disorder (ASD), attention-deficit / hyperactivity disorder (ADHD), amyotrophic lateral sclerosis (ALS), schizophrenia, irritable bowel disease (IBD), pediatric kidney disease, renal transplant rejection, and pediatric cancer.

74. The method of any one of claims 69 to 73, wherein evaluating the test subject for the first biological condition associated with the Raman signature further comprises: The first biological condition associated with the Raman signature is distinguished from a second biological condition associated with the Raman signature that is different from the first biological condition associated with the Raman signature.

75. The method of claim 74, wherein the first biological condition is autism spectrum disorder and the second biological condition is attention-deficit / hyperactivity disorder.

76. The method of any one of claims 69 to 75, wherein the test subject is a human.

77. The method of claim 76, wherein the human is less than 12 years old.

78. The method of claim 76, wherein the human is less than 1 year old.

79. The method of any one of claims 69 to 78, wherein the corresponding biological sample associated with the Raman signature of the respective training subject is selected from the group consisting of hair shafts, teeth and nails.

80. The method of claim 79, wherein the corresponding biological sample associated with the Raman signature of the corresponding training subject is the hair shaft, and wherein the reference line corresponds to a longitudinal direction of the hair shaft.

81. The method of claim 79, wherein the corresponding biological sample associated with the Raman signature of the corresponding training subject is the tooth, and wherein the reference line corresponds to a direction across growth zones, including a new growth line of the tooth.

82. A method according to any one of claims 69 to 81, wherein the corresponding multiple positions are ordered so that the first position of the corresponding multiple positions along the corresponding biological sample of the corresponding training subject corresponds to the position of the tip of the corresponding biological sample closest to the corresponding training subject.

83. The method of any one of claims 69 to 82, wherein each trace in a corresponding plurality of Raman spectroscopy measurements comprises a plurality of data points, each data point being an instance of the respective position in the plurality of positions.

84. The method of any one of claims 69 to 83, wherein the corresponding feature set is selected from the group consisting of: hierarchicality, entropy, capture time (TT), mean diagonal length (MDL), recursion time (RT), Vmax, certainty, Lmax.

85. A method according to any one of claims 69 to 83, wherein the corresponding plurality of locations comprises at least 1000, 1500, 2000, 2500, 3000, 3500, 4000, 4500 or 5000, 5500, 6000, 6500, 7000, 7500, 8000, 8500, 9000, 9500, 10000 or more than 10000 locations.

86. A method according to any one of claims 69 to 85, wherein the trained model is configured to process one or more features selected from the group consisting of: stratification, entropy, capture time (TT), mean diagonal length (MDL), recursion time (RT), Vmax, certainty, Lmax, determination of linear slopes of the multiple Raman spectra across reference lines, determination of multiple nonlinear parameters describing the curvature of the multiple Raman spectra across reference lines, determination of sudden changes in the intensity of the multiple Raman spectra across reference lines, determination of one or more changes in baseline intensity of the multiple Raman spectra across reference lines, determination of changes in frequency domain representations of the multiple Raman spectra across reference lines, determination of changes in power spectrum domain representations of the multiple Raman spectra across reference lines, determination of one or more recursive quantitative analysis parameters, determination of one or more cross-recursive quantitative analysis parameters, determination of one or more joint recursive quantitative analysis parameters, determination of one or more multidimensional recursive quantitative analysis parameters, estimation of Lyapunov spectrum, determination of maximum Lyapunov exponent, and any combination thereof.

87. A method according to any one of claims 69 to 85, wherein the trained model is configured to process two or more features selected from the group consisting of: stratification, entropy, capture time (TT), mean diagonal length (MDL), recursion time (RT), Vmax, certainty, Lmax, determination of linear slopes of the multiple Raman spectra across reference lines, determination of multiple nonlinear parameters describing the curvature of the multiple Raman spectra across reference lines, determination of sudden changes in intensity of the multiple Raman spectra across reference lines, determination of one or more changes in baseline intensity of the multiple Raman spectra across reference lines, determination of changes in frequency domain representations of the multiple Raman spectra across reference lines, determination of changes in power spectrum domain representations of the multiple Raman spectra across reference lines, determination of one or more recursive quantitative analysis parameters, determination of one or more cross-recursive quantitative analysis parameters, determination of one or more joint recursive quantitative analysis parameters, determination of one or more multidimensional recursive quantitative analysis parameters, estimation of Lyapunov spectrum, determination of maximum Lyapunov exponent, and any combination thereof.

Images