Apparatus, system and method for topographical analysis of biological surfaces

By using a system with light sources and sensors to generate multidimensional data sets and applying predictive models, the method addresses the lack of continuous predictive insights in non-invasive diagnostics, enabling accurate phenotypic trait prediction.

JP2024541506AInactive Publication Date: 2024-11-08LINUS BIOTECHNOLOGY INC
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
JP2024531233
Authority / Receiving Office
JP · JP
Patent Type
Applications
Current Assignee / Owner
Priority Date
2021-11-24
Filing Date
2022-11-23
Publication Date
2024-11-08
Estimated Expiration
Not applicable · inactive patent

AI Technical Summary

Technical Problem

Non-invasive diagnostic techniques provide rich datasets but lack the ability to leverage them for continuous, predictive insights beyond human capabilities, necessitating innovations for integrating and analyzing biological surface topography for clinical predictions.

Method used

A method and system for determining the topography of biological surfaces using a compressible body optically coupled with light sources and sensors, generating multidimensional data sets, and employing predictive models like machine learning algorithms to analyze and predict phenotypic characteristics.

Benefits of technology

Enables accurate prediction of phenotypic traits such as disease presence, anthropometric features, and elemental concentrations by analyzing biological surface topography, enhancing diagnostic capabilities.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure 2024541506000001_ABST
    Figure 2024541506000001_ABST
Patent Text Reader

Abstract

Provided herein are devices, systems and methods for measuring the topography of a biological surface and obtaining a probabilistic prediction of the presence or absence of a phenotypic characteristic thereof.
Need to check novelty before this filing date? Find Prior Art

Description

[Technical field]

[0001] cross reference This application claims the benefit of U.S. Provisional Application No. 63 / 283,139, filed November 24, 2021, which is incorporated herein by reference. [Background technology]

[0002] Non-invasive diagnostic technologies have advanced dramatically over the past decade, enabling the incorporation of rich representative data sets of complex biological and physiological systems. Although such technologies have made appropriate advances in improving signal-to-noise ratios and molecular specificity, they are still only snapshots at a given time. Diagnostic platforms are at a tipping point in their evolution where integration of predictive models, machine learning or artificial intelligence is required to leverage and interpret rich data sets beyond what humans can obtain to provide insightful and actionable predictions. Thus, there is a clinical unmet need for innovation in diagnostic platforms to leverage such rich data sets to provide paradigm-shifting predictions of biological or physiological conditions. Summary of the Invention

[0003] Aspects of the disclosure provided herein include a method for determining a topography of a surface of a biological sample, the method comprising: (a) receiving a biological sample; (b) mounting the biological sample in a fixture; (c) contacting the biological sample with a compressible body optically coupled to at least one light source and at least one sensor, the at least one light source generating a multidimensional dataset of the surface of the biological sample detected by the at least one sensor; and (d) determining a surface profile of the biological sample from the multidimensional dataset. In some embodiments, the biological sample comprises an ex-vivo or in-vivo tooth. In some embodiments, the compressible body comprises a gel cartridge. In some embodiments, the fixture comprises a putty, a mechanical attachment, an adhesive, or any combination thereof. In some embodiments, the method further comprises calibrating the compressible body on a standard glass, a ball grid array, a groove target, or any combination thereof. In some embodiments, the at least one light source comprises at least one coherent laser, incoherent laser, pulsed laser, light emitting diode, superluminescent diode, or any combination thereof. In some embodiments, the method further comprises washing the biological sample with water. In some embodiments, the spatial position of the compressible body can be releasably locked. In some embodiments, the at least one light source and the at least one sensor are in electrical communication with a processor via a button, switch, trigger, or any combination thereof configured to initiate illumination of the biological sample and detection of the multidimensional dataset of the surface. In some embodiments, the multidimensional dataset comprises at least one dimension. In some embodiments, the surface profile is measured from the incisal edge to the cervical end of the crown of the ex-vivo or in-vivo tooth.

[0004] Aspects of the disclosure provided herein include a method for determining a topography of a surface of a biological sample, the method comprising: (a) receiving a biological sample; (b) mounting the biological sample in a fixture; (c) illuminating the biological sample with at least one light source and at least one sensor, the at least one light source generating a multidimensional dataset of the surface of the biological sample detected by the at least one sensor; and (d) determining a surface profile of the biological sample from the multidimensional dataset. In some embodiments, the biological sample comprises an ex-vivo or in-vivo tooth. In some embodiments, the fixture comprises a putty, a mechanical attachment, a stabilizer, an adhesive, or any combination thereof. In some embodiments, the at least one light source comprises at least one coherent laser, an incoherent laser, a pulsed laser, a light emitting diode, a superluminescent diode, or any combination thereof. In some embodiments, the method further comprises washing the biological sample with water. In some embodiments, the at least one light source and the at least one sensor are in electrical communication with a processor via a button, switch, trigger, or any combination thereof configured to initiate illumination of the biological sample and detection of the multi-dimensional dataset of the surface. In some embodiments, the multi-dimensional dataset includes at least one dimension. In some embodiments, the surface profile is measured from the incisal edge to the cervical end of the crown of the ex-vivo or in-vivo tooth.

[0005] Aspects of the disclosure provided herein include a system for determining a topography of a surface of a biological specimen, the system including: (a) a fixture mechanically configured to constrain a three-dimensional position of the biological specimen; (b) a compressible body optically coupled to at least one light source and at least one sensor, the at least one light source generating a multidimensional data set of the surface of the biological specimen detected by the at least one sensor; and (c) a processor electrically coupled to the at least one light source and the at least one sensor, the processor including a set of programmatic instructions stored in a non-transitory storage medium configured to cause the processor to determine a topography of the surface of the biological specimen from the multidimensional data set. In some embodiments, the biological specimen includes a tooth. In some embodiments, the compressible body includes a gel cartridge. In some embodiments, the fixture includes a putty, a mechanical attachment, an adhesive, or any combination thereof. In some embodiments, the at least one light source comprises at least one coherent laser, incoherent laser, pulsed laser, light emitting diode, superluminescent diode, or any combination thereof. In some embodiments, the spatial position of the compressible body can be releasably locked. In some embodiments, the at least one light source and the at least one sensor are in electrical communication with a processor via a button, switch, trigger, or any combination thereof configured to initiate illumination of the biological sample and detection of the multi-dimensional dataset of the surface. In some embodiments, the system further comprises a rigid guide member mechanically coupled to the compressible body configured to position the compressible body relative to the biological sample. In some embodiments, the compressible body is housed in a locking mechanical member configured to fix the three-dimensional position of the compressible body. In some embodiments, the multi-dimensional dataset comprises at least one dimension.

[0006] Aspects of the disclosure provided herein include a method of training a predictive model for outputting a phenotypic characterization of a topographical dataset of one or more subjects, the method comprising: (a) receiving one or more biological samples and phenotypic characterizations from a first set of subjects; (b) determining a first topographical dataset from one or more biological samples of the first set of subjects; (c) calculating a first set of features of the first topographical dataset; and (d) training a predictive model with the first set of features and the phenotypic characterizations of the first set of subjects to generate the trained predictive model configured to output a phenotypic characterization of a second set of one or more subjects when the trained predictive model is input with a second set of features of a second topographical dataset. In some embodiments, the one or more biological samples comprise teeth. In some embodiments, the first set of features of the topography dataset includes the following features of the topography dataset: a number of peaks of the topography dataset detected, a slope of one or more of the peaks, a two-dimensional gradient across the topography data, or any combination thereof. In some embodiments, the predictive model includes a machine learning algorithm. In some embodiments, the machine learning algorithm includes a support vector machine (SVM), a naive Bayes classification, a random forest, a neural network, a deep neural network (DNN), a recurrent neural network (RNN), a deep RNN, a long short-term memory (LSTM) recurrent neural network (RNN), a gated recurrent unit (GRU), a supervised learning algorithm, an unsupervised machine learning algorithm, or any combination thereof. In some embodiments, the phenotypic characterization comprises the presence or absence of a disease or disorder, wherein the disease or disorder comprises autism spectrum disorder (ASD), attention deficit hyperactivity disorder (ADHD), amyotrophic lateral sclerosis (ALS), schizophrenia, irritable bowel disease (IBD), pediatric kidney disease, kidney transplant rejection, pediatric cancer, or any combination thereof.

[0007] Aspects of the disclosure provided herein include a method of training a predictive model to output a phenotypic characterization of phenotypic data of one or more subjects, the method comprising: (a) generating a first topographical dataset by imaging one or more biological samples of a first set of subjects; (b) calculating a first set of features of the first topographical dataset; and (c) training a predictive model with the first set of features and the phenotypic characterization of the first set of subjects to generate the trained predictive model configured to output a phenotypic characterization of a second set of one or more subjects when the trained predictive model is input with a second set of features of a second topographical dataset. In some embodiments, the first and second sets of subjects are the same or different. In some embodiments, the one or more biological samples include teeth. In some embodiments, the first or second set of features of the topography dataset comprises the following features of the topography dataset: a number of peaks of the topography dataset detected, a slope of one or more of the peaks, a two-dimensional gradient across the topography data, or any combination thereof. In some embodiments, the first or second set of phenotypic characterizations comprises the presence or absence of a disease or disorder, the disease or disorder comprising autism spectrum disorder (ASD), attention deficit hyperactivity disorder (ADHD), amyotrophic lateral sclerosis (ALS), schizophrenia, irritable bowel disease (IBD), pediatric kidney disease, kidney transplant rejection, pediatric cancer, or any combination thereof. In some embodiments, the predictive model comprises a machine learning algorithm. In some embodiments, the machine learning algorithm comprises a support vector machine (SVM), a naive Bayes classification, a random forest, a neural network, a deep neural network (DNN), a recurrent neural network (RNN), a deep RNN, a long short-term memory (LSTM) recurrent neural network (RNN), a gated recurrent unit (GRU), a supervised learning algorithm, an unsupervised machine learning algorithm, or any combination thereof.

[0008] Aspects of the disclosure provided herein include a method of predicting a phenotypic classification of a topography dataset of one or more subjects, the method comprising: (a) generating a topography dataset by imaging one or more biological samples of one or more subjects; (b) calculating a set of features of the topography dataset; and (c) predicting a phenotypic characterization as an output of a trained predictive model when the trained predictive model is provided with the set of features of the one or more subjects as input. In some embodiments, the one or more biological samples comprise teeth. In some embodiments, the set of features of the topography dataset comprises a number of peaks of the topography dataset detected, a slope of one or more of the peaks, a two-dimensional gradient across the topography data, or any combination thereof. In some embodiments, the phenotypic characterization comprises the presence or absence of a disease or disorder, the disease or disorder comprising autism spectrum disorder (ASD), attention deficit hyperactivity disorder (ADHD), amyotrophic lateral sclerosis (ALS), schizophrenia, irritable bowel disease (IBD), pediatric kidney disease, kidney transplant rejection, pediatric cancer, or any combination thereof. In some embodiments, the predictive model comprises a machine learning algorithm. In some embodiments, the machine learning algorithm comprises a support vector machine (SVM), a naive Bayes classification, a random forest, a neural network, a deep neural network (DNN), a recurrent neural network (RNN), a deep RNN, a long short-term memory (LSTM) recurrent neural network (RNN), a gated recurrent unit (GRU), a supervised learning algorithm, an unsupervised machine learning algorithm, or any combination thereof.

[0009] Aspects of the disclosure provided herein include a method of predicting a phenotypic characterization of a topographical dataset of one or more subjects, the method comprising: (a) receiving one or more biological samples and phenotypic data of one or more subjects; (b) determining a topographical dataset from one or more biological samples of the one or more subjects; (c) calculating a set of features of a first topographical dataset; and (d) predicting a phenotypic classification as an output of a trained predictive model when the trained predictive model is provided with the set of features of the one or more subjects as input. In some embodiments, the one or more biological samples comprise teeth. In some embodiments, the set of features of the topographical dataset comprises a number of peaks of the topographical dataset detected, a slope of one or more of the peaks, a two-dimensional gradient across the topographical data, or any combination thereof. In some embodiments, the phenotypic characterization comprises the presence or absence of a disease or disorder, the disease or disorder comprising autism spectrum disorder (ASD), attention deficit hyperactivity disorder (ADHD), amyotrophic lateral sclerosis (ALS), schizophrenia, irritable bowel disease (IBD), pediatric kidney disease, kidney transplant rejection, pediatric cancer, or any combination thereof. In some embodiments, the trained predictive model comprises a machine learning algorithm. In some embodiments, the machine learning algorithm comprises a support vector machine (SVM), a naive Bayes classification, a random forest, a neural network, a deep neural network (DNN), a recurrent neural network (RNN), a deep RNN, a long short-term memory (LSTM) recurrent neural network (RNN), a gated recurrent unit (GRU), a supervised learning algorithm, an unsupervised machine learning algorithm, or any combination thereof.

[0010] Aspects of the disclosure provided herein include a method for outputting a feature importance score of one or more features of a subject that correlate with a phenotypic characterization, the method comprising: (a) receiving a biological sample of the subject; (b) contacting a compressible body optically coupled to at least one light source and at least one sensor with a surface of the biological sample, the at least one light source generating a multidimensional dataset of the surface of the biological sample detected by the at least one sensor; (c) determining one or more features of the multidimensional dataset of the surface of the biological sample by analyzing the multidimensional dataset; and (d) outputting a feature importance score of one or more features of a topographical dataset that correlate with a phenotypic characterization. In some embodiments, the biological sample comprises a tooth. In some embodiments, one or more of the features of the multidimensional dataset include a feature of the topographical dataset of the number of peaks of the topographical dataset detected, a slope of one or more of the peaks, a two-dimensional gradient across the topographical data, or any combination thereof. In some embodiments, the phenotypic characterization comprises the presence or absence of a disease or disorder, the disease or disorder comprising autism spectrum disorder (ASD), attention deficit hyperactivity disorder (ADHD), amyotrophic lateral sclerosis (ALS), schizophrenia, irritable bowel disease (IBD), pediatric kidney disease, kidney transplant rejection, pediatric cancer, or any combination thereof. In some embodiments, the feature importance score is output by a trained predictive model, the trained predictive model comprising a machine learning algorithm. In some embodiments, the machine learning algorithm comprises a support vector machine (SVM), a naive Bayes classification, a random forest, a neural network, a deep neural network (DNN), a recurrent neural network (RNN), a deep RNN, a long short-term memory (LSTM) recurrent neural network (RNN), a gated recurrent unit (GRU), a supervised learning algorithm, an unsupervised machine learning algorithm, or any combination thereof.

[0011] Incorporation by Reference All publications, patents, and patent applications mentioned in this specification are herein incorporated 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. [Brief description of the drawings]

[0012] 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 present invention will be obtained by reference to the following detailed description that sets forth illustrative embodiments, in which the principles of the invention are utilized, and the accompanying drawings of which: [Figure 1] FIG. 1 illustrates an example apparatus configuration for collecting ex-vivo biological sample surface topography data as described in some embodiments herein. [Diagram 2] FIG. 1 is a flow diagram of a method for collecting topographical data of a biological surface of an ex-vivo biological sample as described in some embodiments herein. [Figure 3A] FIG. 1 shows an example dataset generated by application of the device, method and system for predicting anthropomorphic and clinical phenotypic traits (height and / or weight) in a population of children from a dataset of biological surface topography data (Study 1). The plot demonstrates device performance by showing the correlation between predicted height and height measured in a clinical evaluation. [Figure 3B] FIG. 1 shows an example dataset generated by application of the device, method and system for predicting anthropomorphic and clinical phenotypic traits (height and / or weight) in a population of children from a dataset of biological surface topography data (Study 1). The plot demonstrates device performance by showing the correlation between predicted weight and weight measured in a clinical evaluation. [Figure 4A]FIG. 1 shows an example dataset generated by application of the device, method and system for predicting anthropomorphic and clinical phenotypic traits (height and / or body mass index (BMI)) in a population of children from a dataset of biological surface topography data (Study 2). The plot demonstrates device performance by showing the correlation between predicted height and height measured in a clinical evaluation. [Figure 4B] FIG. 1 shows an example dataset generated by application of the device, method and system for predicting anthropomorphic and clinical phenotypic traits (height and / or body mass index (BMI)) in a population of children from a dataset of biological surface topography data (Study 2). The plot demonstrates device performance by showing the correlation between predicted BMI and BMI measured in a clinical evaluation. [Figure 5A] 1 shows an example data set generated by application of the device, method and system for predicting intelligence quotient (IQ) from a data set of biological surface topography data. The plots demonstrate device performance by showing correlations between predicted scores for these measures based on analysis of surface topography data and scores measured by clinical evaluation. [Figure 5B] 1 shows an example data set generated by application of the device, method and system for predicting a pediatric cognitive composite score from a data set of biological surface topography data. The plots demonstrate device performance by showing correlations between predicted scores for these measures based on analysis of surface topography data and scores measured by clinical assessment. [Figure 6] FIG. 1 is a diagram of a system configured to process, execute, and / or implement the methods of the disclosure provided herein. [Figure 7A]1 shows an example data set generated by application of the device, method and system for predicting the concentration of essential and non-essential elements in dental samples from a data set of biological surface topography data. The plots demonstrate device performance for certain elements by showing the correlation between predicted element concentrations based on analysis of surface topography data compared to detection of essential and non-essential elements using standard mass spectrometry methods, as described in some embodiments herein. [Figure 7B] 1 shows an example data set generated by application of the device, method and system for predicting the concentration of essential and non-essential elements in dental samples from a data set of biological surface topography data. The plots demonstrate device performance for certain elements by showing the correlation between predicted element concentrations based on analysis of surface topography data compared to detection of essential and non-essential elements using standard mass spectrometry methods, as described in some embodiments herein. [Figure 7C] 1 shows an example data set generated by application of the device, method and system for predicting the concentration of essential and non-essential elements in dental samples from a data set of biological surface topography data. The plots demonstrate device performance for certain elements by showing the correlation between predicted element concentrations based on analysis of surface topography data compared to detection of essential and non-essential elements using standard mass spectrometry methods, as described in some embodiments herein. [Figure 7D] 1 shows an example data set generated by application of the device, method and system for predicting the concentration of essential and non-essential elements in dental samples from a data set of biological surface topography data. The plots demonstrate device performance for certain elements by showing the correlation between predicted element concentrations based on analysis of surface topography data compared to detection of essential and non-essential elements using standard mass spectrometry methods, as described in some embodiments herein. [Figure 7E]1 shows an example data set generated by application of the device, method and system for predicting the concentration of essential and non-essential elements in dental samples from a data set of biological surface topography data. The plots demonstrate device performance for certain elements by showing the correlation between predicted element concentrations based on analysis of surface topography data compared to detection of essential and non-essential elements using standard mass spectrometry methods, as described in some embodiments herein. [Figure 7F] 1 shows an example data set generated by application of the device, method and system for predicting the concentration of essential and non-essential elements in dental samples from a data set of biological surface topography data. The plots demonstrate device performance for certain elements by showing the correlation between predicted element concentrations based on analysis of surface topography data compared to detection of essential and non-essential elements using standard mass spectrometry methods, as described in some embodiments herein. [Figure 7G] 1 shows an example data set generated by application of the device, method and system for predicting the concentration of essential and non-essential elements in dental samples from a data set of biological surface topography data. The plots demonstrate device performance for certain elements by showing the correlation between predicted element concentrations based on analysis of surface topography data compared to detection of essential and non-essential elements using standard mass spectrometry methods, as described in some embodiments herein. [Figure 7H] 1 shows an example data set generated by application of the device, method and system for predicting the concentration of essential and non-essential elements in dental samples from a data set of biological surface topography data. The plots demonstrate device performance for certain elements by showing the correlation between predicted element concentrations based on analysis of surface topography data compared to detection of essential and non-essential elements using standard mass spectrometry methods, as described in some embodiments herein. [Figure 7I]1 shows an example data set generated by application of the device, method and system for predicting the concentration of essential and non-essential elements in dental samples from a data set of biological surface topography data. The plots demonstrate device performance for certain elements by showing the correlation between predicted element concentrations based on analysis of surface topography data compared to detection of essential and non-essential elements using standard mass spectrometry methods, as described in some embodiments herein. [Figure 8A] FIG. 1 illustrates the use of a handheld device to measure the topography of a biological surface in-vivo as described in some embodiments herein. [Figure 8B] FIG. 1 illustrates the use of a handheld device to measure the topography of a biological surface in-vivo as described in some embodiments herein. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0013] The disclosure provided herein describes devices, systems, and methods for measuring, capturing, and analyzing the topography of a biological surface. In some cases, the topography of the biological surface may be the in-vivo or ex-vivo biological surface of a subject. The devices described herein may include handheld or tabletop devices. The disclosure provided herein may also describe an analysis method applied to the measured topography of the biological surface. In some examples, the analysis method may include training a predictive model with one or more features of the measured topography of the biological surface and corresponding phenotypic characteristics associated with the measured biological surface of the subject. In some cases, the phenotypic characteristics may include the subject's intelligence quotient (IQ), body mass index (BMI), height, weight, etc.

[0014] Apparatus and system The disclosure provided herein describes devices and systems that may be utilized to determine the topography of a biological surface. The device 100 may include at least one light source, at least one sensor, an optical assembly, and optionally a compressible body 106 optically coupled to the at least one light source, the optical assembly, and / or the at least one sensor.

[0015] In some cases, the device may be a handheld device 100 as seen in FIGS. 8A-8B configured to be minimally or non-invasively placed over or in contact with a biological surface 806 (FIG. 8B) of a subject 800. In some cases, the biological surface 806 may be contacted by the device via a contact surface of a compressible body 106 as described elsewhere herein. In some cases, the handheld device may provide access to the biological surface in-vivo within the oral cavity of the subject. In some cases, the biological surface may include the surface of one or more of the subject's teeth 807. As shown in FIG. 8A, the device 100 used in a handheld configuration may be in electrical communication with a processor 114 via wired communication 116 or wireless communication 802. In some cases, the handheld device may include magnetic, gyroscopic, accelerometer, or any combination thereof, sensors to determine the three-dimensional position of the device in space. In some examples, the three-dimensional position of the device in space may be used to reconstruct collected data of multiple continuous and / or non-contiguous biological surfaces.

[0016] Alternatively, the device may be a benchtop system, as seen in FIG. 1. In some examples, the device 100 may be docked to a benchtop system (112, 120, 102) as shown in FIG. 1 and used as a benchtop system. In some examples, the device 100 may be in electrical communication with the processor 114 via a wired 116 or wireless 113 communication platform. In some cases, the processor may include a personal desktop computer, a server, a cloud-based server, a laptop computer, a tablet computer, a smartphone, or any combination thereof. In some examples, the processor may initiate collection of or analyze topographical data of a biological surface of the biological sample 108 collected by the device 100. In some cases, the topographical data of the biological surface may include a multidimensional dataset. In some cases, the multidimensional dataset includes at least one dimensional data, at least two dimensional data, or at least three dimensional data. In some cases, the data may be converted into one or more color and / or grayscale images.

[0017] In some examples, device 100 may include a housing that houses electrical, optical, and / or mechanical components. In some cases, the housing may include a grip or ergonomic handheld mechanism 804 to allow a user to hold and position the device over or in contact with a subject's biological surface 806 for compressive purposes, as seen in Figures 8A-8B. In some cases, the device may include a battery, charging circuitry, battery indicator, or any combination thereof, configured to allow the device to be used in a handheld manner.

[0018] In some cases, the at least one light source may include a coherent laser, an incoherent laser, a pulsed laser, a confocal laser, a light emitting diode (LED), an organic light emitting diode (OLED), a superluminescent diode, or any combination thereof.

[0019] In some examples, the at least one light source emits light within a wavelength range of about 350 nm to about 1,000 nm.In some examples, the at least one light source has a wavelength of about 350 nm to about 400 nm, about 350 nm to about 440 nm, about 350 nm to about 480 nm, about 350 nm to about 520 nm, about 350 nm to about 560 nm, about 350 nm to about 600 nm, about 350 nm to about 640 nm, about 350 nm to about 670 nm, about 350 nm to about 600 nm, about 350 nm to about 800 nm, about 350 nm to about 1,000 nm, about 400 nm to about 440 nm, about 400 nm to about 480 nm, about 400 nm to about 520 nm, about 400 nm to about 560 nm, about 400 nm to about 60 ...600 nm, about 400 nm to about 600 nm, about 400 nm to about 600 nm, about 400 nm to about 600 nm, about 400 nm to about 600 nm, about 400 nm to about 600 nm, about 400 nm to about 600 nm, about 400 nm to about 600 nm, about 400 nm to about 600 nm, about 400 nm to about 600 nm, about 4 nm, about 400nm to about 640nm, about 400nm to about 670nm, about 400nm to about 600nm, about 400nm to about 800nm, about 4 00nm ~ approx. 1,000nm, approx. 440nm ~ approx. 480nm, approx. 440nm ~ approx. 520nm, approx. 440nm ~ approx. 560nm, approx. 440nm ~about 600nm, about 440nm to about 640nm, about 440nm to about 670nm, about 440nm to about 600nm, about 440nm to about 800 nm, approximately 440 nm to approximately 1,000 nm, approximately 480 nm to approximately 520 nm, approximately 480 nm to approximately 560 nm, approximately 480 nm to approximately 600 nm, approximately 480nm to about 640nm, about 480nm to about 670nm, about 480nm to about 600nm, about 480nm to about 800nm, about 480nm ~approx. 1,000nm, approx. 520nm ~ approx. 560nm, approx. 520nm ~ approx. 600nm, approx. 520nm ~ approx. 640nm, approx. 520nm ~ approx. 67 0nm, about 520nm to about 600nm, about 520nm to about 800nm, about 520nm to about 1,000nm, about 560nm to about 600nm , about 560nm to about 640nm, about 560nm to about 670nm, about 560nm to about 600nm, about 560nm to about 800nm, about 560n The optical fiber emits light within a wavelength range of about 600 nm to about 1,000 nm, about 600 nm to about 640 nm, about 600 nm to about 670 nm, about 600 nm to about 600 nm, about 600 nm to about 800 nm, about 600 nm to about 1,000 nm, about 640 nm to about 670 nm, about 640 nm to about 600 nm, about 640 nm to about 800 nm, about 640 nm to about 1,000 nm, about 670 nm to about 600 nm, about 670 nm to about 800 nm, about 670 nm to about 1,000 nm, about 600 nm to about 800 nm, about 600 nm to about 1,000 nm, or about 800 nm to about 1,000 nm.In some examples, the at least one light source emits light within a wavelength range of about 350 nm, about 400 nm, about 440 nm, about 480 nm, about 520 nm, about 560 nm, about 600 nm, about 640 nm, about 670 nm, about 600 nm, about 800 nm, or about 1,000 nm. In some examples, the at least one light source emits light within a wavelength range of at least about 350 nm, about 400 nm, about 440 nm, about 480 nm, about 520 nm, about 560 nm, about 600 nm, about 640 nm, about 670 nm, about 600 nm, or about 800 nm. In some examples, at least one light source emits light within a wavelength range of at most about 400 nm, about 440 nm, about 480 nm, about 520 nm, about 560 nm, about 600 nm, about 640 nm, about 670 nm, about 600 nm, about 800 nm, or about 1,000 nm.

[0020] In some cases, the at least one sensor includes an elastomeric gel membrane, a camera, a photomultiplier tube, a charge-coupled device (CCD), a complementary metal oxide semiconductor (CMOS), or any combination thereof. In some cases, the at least one sensor may be configured to capture a three-dimensional spatial dataset of topographical data of the biological surface. In some cases, the sensor may be configured to capture at least a one-dimensional, two-dimensional, or three-dimensional spatial dataset of topographical data of the biological surface.

[0021] In some cases, the at least one sensor may be configured to generate a data set having a spatial resolution of about 0.01 μm to about 2 μm.In some cases, the at least one sensor has a diameter of about 0.01 μm to about 0.03 μm, about 0.01 μm to about 0.05 μm, about 0.01 μm to about 0.08 μm, about 0.01 μm to about 0.1 μm, about 0.01 μm to about 0.5 μm, about 0.01 μm to about 0.8 μm, about 0.01 μm to about 1 μm, about 0.01 μm to about 1.2 μm, about 0.01 μm to about 1.5 μm, about 0.01 μm to about 1.7 μm, about 0.01 μm to about 2 μm, about 0.03 μm to about 0.05 μm, about 0.03 μm to about 0.08 μm, about 0.03 μm to about 0.1 μm, about 0.03 μm to about 0.5 μm. m, about 0.03 μm to about 0.8 μm, about 0.03 μm to about 1 μm, about 0.03 μm to about 1.2 μm, about 0.03 μm to about 1.5 μm, about 0.03μm to about 1.7μm, about 0.03μm to about 2μm, about 0.05μm to about 0.08μm, about 0.05μm to about 0.1μm, about 0. 05μm~about 0.5μm, about 0.05μm~about 0.8μm, about 0.05μm~about 1μm, about 0.05μm~about 1.2μm, about 0.05μm m ~ approx. 1.5 μm, approx. 0.05 μm ~ approx. 1.7 μm, approx. 0.05 μm ~ approx. 2 μm, approx. 0.08 μm ~ approx. 0.1 μm, approx. 0.08 μm ~ approx. 0 .5μm, about 0.08μm to about 0.8μm, about 0.08μm to about 1μm, about 0.08μm to about 1.2μm, about 0.08μm to about 1.5μ m, about 0.08 μm to about 1.7 μm, about 0.08 μm to about 2 μm, about 0.1 μm to about 0.5 μm, about 0.1 μm to about 0.8 μm, about 0. 1μm to about 1μm, about 0.1μm to about 1.2μm, about 0.1μm to about 1.5μm, about 0.1μm to about 1.7μm, about 0.1μm to about 2μ m, about 0.5 μm to about 0.8 μm, about 0.5 μm to about 1 μm, about 0.5 μm to about 1.2 μm, about 0.5 μm to about 1.5 μm, about 0.5 μm The system may be configured to generate a data set having a spatial resolution of about 1.7 μm, about 0.5 μm to about 2 μm, about 0.8 μm to about 1 μm, about 0.8 μm to about 1.2 μm, about 0.8 μm to about 1.5 μm, about 0.8 μm to about 1.7 μm, about 0.8 μm to about 2 μm, about 1 μm to about 1.2 μm, about 1 μm to about 1.5 μm, about 1 μm to about 1.7 μm, about 1 μm to about 2 μm, about 1.2 μm to about 1.5 μm, about 1.2 μm to about 1.7 μm, about 1.2 μm to about 2 μm, about 1.5 μm to about 1.7 μm, about 1.5 μm to about 2 μm, or about 1.7 μm to about 2 μm.In some cases, the at least one sensor may be configured to generate a data set having a spatial resolution of about 0.01 μm, about 0.03 μm, about 0.05 μm, about 0.08 μm, about 0.1 μm, about 0.5 μm, about 0.8 μm, about 1 μm, about 1.2 μm, about 1.5 μm, about 1.7 μm, or about 2 μm. In some cases, the at least one sensor may be configured to generate a data set having a spatial resolution of at least about 0.01 μm, about 0.03 μm, about 0.05 μm, about 0.08 μm, about 0.1 μm, about 0.5 μm, about 0.8 μm, about 1 μm, about 1.2 μm, about 1.5 μm, or about 1.7 μm. In some cases, the at least one sensor may be configured to generate a data set having a spatial resolution of at most about 0.03 μm, about 0.05 μm, about 0.08 μm, about 0.1 μm, about 0.5 μm, about 0.8 μm, about 1 μm, about 1.2 μm, about 1.5 μm, about 1.7 μm, or about 2 μm.

[0022] In some cases, the compressible body 106 may include a compressible gel cartridge including a deformable elastomeric gel matrix and membrane configured to increase the topography of the biological sample 108 surface upon contact with the biological sample 108. In some examples, the deformable plastic may include a thermoplastic elastomer (TPE) or silicone material. In some cases, the device may operate without the compressible body 106 by removing the compressible cartridge 106 from the optical path of the device 100. In some cases, the device may include a switch or an actionable button 104 configured to start, pause, or end the collection of topographical data of the biological surface. In some cases, the switch or actionable button 104 may include a physical push button, a force sensor, a capacitive touch button, or any combination thereof. In some examples, by pressing and holding or activating the switch or actionable button 104 for an extended period of time, a user may initiate various functions of the device and / or system, such as a particular light source illumination pattern or averaging of collected data.

[0023] In some cases, the device may include an optical assembly including one or more lenses or lens elements. In some cases, the one or more lenses may include a concave lens, a convex lens, a biconcave lens, a biconvex lens, a plano-concave lens, a plano-convex lens, a spherical lens, or any combination thereof. In some examples, the one or more lenses may be configured to direct, magnify, collimate, focus, or any combination thereof, beam manipulation of the light emitted by the at least one light source or reflected from the biological sample 108 collected by the at least one sensor. In some cases, the one or more lenses may include an anti-reflective coating configured to reflect and / or transmit a particular band of light. In some cases, the optical assembly may further include a dichroic filter, a hot mirror, a cold mirror, or any combination thereof.

[0024] system The disclosure provided herein describes a system configured to determine a topography of a biological surface. In some examples, the system may include a handheld system configured to image the topography of the biological surface in-vivo. Alternatively, the system may include a tabletop system configured to image the topography of the biological surface ex-vivo. In some cases, the system may include (a) a fixture mechanically configured to constrain a three-dimensional position of the biological sample; (b) a compressible body optically coupled to at least one light source and at least one sensor, the at least one light source generating a multidimensional dataset of the surface of the biological sample detected by the at least one sensor; and (c) a processor electrically coupled to the at least one light source and the at least one sensor, the processor including a set of programmatic instructions stored in a non-transitory storage medium configured to cause the processor to determine a topography of the surface of the biological sample from the multidimensional dataset. In some cases, the biological sample may be a tooth. In some cases, the compressible body may include a gel cartridge. In some cases, the compressible body is housed in a locking mechanical member configured to fix the three-dimensional position of the compressible body.

[0025] In some examples, the fixture 110 may include putty, mechanical fasteners, adhesives, or any combination thereof. In some examples, the device 100 may be mechanically coupled to an attachment mechanism 102 configured to releasably lock the position of the device relative to the biological sample 108. In some examples, the spatial position of the compressible body of the device may be releasably lockable by a locking mechanism of the attachment mechanism 102. In some cases, the locking mechanism may include a removable or latching locking mechanism. In some examples, the locking mechanism may fix the three-dimensional position of the compressible body relative to the biological sample. In some examples, the attachment mechanism is mechanically coupled to a rigid guide member 112, which may allow the attachment mechanism to move along the rigid guide member 112 toward and away from the attached biological sample 108 within limited dimensions. In some cases, the rigid guide member 112 may include tracks, rails, or posts 112. In some examples, the rigid guide member may be mechanically coupled to a base 120 configured to reduce vibrations when collecting biological surface topography data using an apparatus mounted on a tabletop system as shown in FIG.

[0026] In some cases, the at least one light source may include at least one coherent laser, incoherent laser, pulsed laser, light emitting diode, superluminescent diode, or any combination thereof. In some cases, the at least one light source and the at least one sensor are in electrical communication with a button, switch, trigger (104), or any combination thereof, configured to initiate illumination of the biological sample and detection of a multidimensional dataset of the biological sample surface. In some cases, the multidimensional dataset includes at least one dimension.

[0027] Computer Systems 6 illustrates a computer system 601 suitable for implementing and / or training the models and / or predictive models described herein. The computer system 601 may process various aspects of the data and / or information of the present disclosure, such as, for example, subject biological sample topography raw data, topographically generated images, processed topography data parameters or features, corresponding subject phenotypic characterization, clinical metadata, or any combination thereof. The computer system 601 may be an electronic device. The electronic device may be a portable electronic device.

[0028] The computer system 601 may include a central processing unit (CPU, also referred to herein as a "processor" and a "computer processor") 605, which may be a single-core or multi-core processor, or multiple processors for parallel processing. The computer system 601 may further include memory or memory locations 604 (e.g., random access memory, read-only memory, flash memory), electronic storage 606 (e.g., hard disk), communication interface 608 (e.g., network adapter) for communicating with one or more other devices, and peripheral devices 607, such as cache, other memory, data storage, and / or electronic display adapters. The memory 604, storage 606, interface 608, and peripheral devices 607 may be in communication with the CPU 605 via a communication bus (solid lines), such as a motherboard. The storage 606 may be a data storage device (or data repository) for storing data. The computer system 601 may be operatively coupled to a computer network ("network") 600 with the aid of the communication interface 608. Network 600 may be the Internet, an Internet and / or an extranet, or an intranet and / or an extranet in communication with the Internet. Network 600 may in some cases be a telecommunications and / or data network. Network 600 may include one or more computer servers, which may enable distributed computing, such as cloud computing. Network 600 may in some cases implement a peer-to-peer network with the aid of computer system 601, which may enable devices coupled to computer system 601 to operate as clients or servers.

[0029] CPU 605 may execute sequences of machine-readable instructions, which may be embodied in a program or software. The instructions may be directed to CPU 605, which may then be programmed or otherwise configured to implement methods of the present disclosure. Examples of operations performed by CPU 605 include fetch, decode, execute, and writeback.

[0030] The CPU 605 may be part of a circuit, such as an integrated circuit. One or more other components of the system 601 may be included in the circuit. In some cases, the circuit is an application specific integrated circuit (ASIC).

[0031] The storage device 606 may store files such as drivers, libraries, and saved programs. The storage device 606 may store subject biological sample topography raw data, topographically generated images, processed topography data parameters or features, corresponding phenotypic characterizations, clinical metadata, or any combination thereof. The computer system 601 may include one or more additional data storage devices external to the computer system 601, such as in some cases located on a remote server in communication with the computer system 601 via an intranet or the Internet.

[0032] The methods described herein may be implemented by machine (e.g., a computer processor) executable code stored in an electronic storage location of computing device 601, such as memory 604 or electronic storage 606. The machine executable or machine readable code may be provided in the form of software. During use, the code may be executed by processor 605. In some examples, the code may be retrieved from storage 606 and stored in memory 604 so as to be immediately accessible by processor 605. In some examples, electronic storage 606 may be eliminated and machine executable instructions stored in memory 604.

[0033] The code may be precompiled and configured for use on a machine having a processor adapted to execute the code, or it may be compiled at run time. The code may be provided in a programming language that may be selected to enable the code to be executed in a precompiled or compiled-at-once manner.

[0034] Aspects of the systems and methods provided herein, such as computer system 601, may be implemented in programming. Various aspects of the technology may be considered as a "product" or "article of manufacture" typically in the form of machine (or processor) executable code and / or associated data transmitted on or embodied in some type of machine-readable medium. The machine executable code may be stored in electronic storage such as memory (e.g., read-only memory, random access memory, flash memory) or hard disk. A "storage" type medium may include any or all of the tangible memory of a computer, processor, etc., or its associated modules, such as various semiconductor memories, tape drives, disk drives, etc., which may provide non-transitory storage at any time for software programming. All or parts of the software may be communicated from time to time over the Internet or various other remote communication networks. Such communication may enable, for example, loading of the software from one computer or processor to another, for example, from a management server or host computer into the computer platform of an application server. Thus, other types of media that may bear software elements include light waves, radio waves, and electromagnetic waves, such as those used across physical interfaces between local devices, through wired and optical networks, and across various air links. Physical elements that transmit such waves, such as wired or wireless lines, optical lines, or the like, may also be considered media bearing the software. As used herein, unless limited to non-transitory tangible "storage" media, terms such as computer or machine "readable medium" refer to any medium that participates in the issue of instructions to a processor for execution.

[0035] Thus, a machine-readable medium such as a computer-executable code may take many forms, including, but not limited to, a tangible storage medium, a carrier wave medium, or a physical transmission medium. Non-volatile storage media may include, for example, optical or magnetic disks, such as any of the storage devices in any computer, such as may be used to implement a database, etc. Volatile storage media include dynamic memory, such as the main memory of such a computer platform. Tangible transmission media include coaxial cables, copper wire and fiber optics, including the wires that comprise a bus within a computing device. Carrier wave transmission media may take the form of electric or electromagnetic signals, or acoustic or light waves, such as those generated during radio frequency (RF) and infrared (IR) data communications. Common forms of computer readable media for this purpose include, for example, a floppy disk, a flexible disk, a hard disk, a magnetic tape, any other magnetic medium, a CD-ROM, a DVD or DVD-ROM, any other optical medium, punch cards, paper tape, any other physical storage medium having a pattern of holes, RAM, ROM, PROM and EPROM, FLASH-EPROM, any other memory chip or cartridge, a carrier wave transmitting data or instructions, a cable or line transmitting such a carrier wave, or any other medium from which a computer may read programming 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.

[0036] The computer system may include or be in communication with an electronic display 602 that includes a user interface (UI) 603 for viewing a subject's phenotypic characterization prediction based on the topographical data of the subject's biological sample. Examples of UIs include, but are not limited to, graphical user interfaces (GUIs) and web-based user interfaces.

[0037] The methods and systems of the present disclosure may be implemented by one or more algorithms and / or predictive models having instructions provided with one or more processors as disclosed herein. The algorithms and / or predictive models may be implemented by software upon execution by the central processing unit 905. In some cases, the predictive models may include machine learning predictive models. In some cases, the machine learning predictive models may include one or more statistical, machine learning, or artificial intelligence algorithms. Examples of algorithms utilized may include support vector machines (SVMs), 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)), supervised learning, unsupervised machine learning, statistical deep learning algorithms for classification and regression. Machine learning predictive models may also include estimation of ensemble models consisting of multiple predictive models, and may utilize techniques such as gradient boosting in the construction of gradient boosting decision trees. Machine learning predictive models may be trained using one or more training datasets corresponding to the data of interest. In some embodiments, the one or more training datasets may include exposome biochemical signatures, dynamic exposome biochemical signatures, clinical metadata, clinical trial information, exposome biochemical signature information of pharmaceutical and nutritional supplement treatments, or any combination thereof.

[0038] method The disclosure provided herein describes a method for collecting or determining topographic data of a biological surface from a biological sample of a subject. In some examples, the method of the disclosure provided herein may include a method for generating, calculating, or determining features or characteristics of the topographic data of a biological surface of a subject that can be used in combination with the phenotypic characterization of the subject to train a predictive model. In some cases, the predictive model may include a machine learning algorithm described elsewhere herein. In some examples, the phenotypic characterization of the subject may include the presence or absence of a disease state (e.g., autism, ADHD, cancer, etc.), physiological parameters (e.g., weight, height, body mass index), or any combination thereof.

[0039] In some cases, the disclosure provided herein describes a method 121 for determining the topography of a biological sample, as shown in FIG. 2. In some cases, the method may include (a) receiving 122 a biological sample from a subject, (b) mounting 126 the biological sample in a fixture, (c) contacting 128 a compressible body optically coupled to at least one light source and at least one sensor with the biological sample, (d) generating 130 a multidimensional dataset of a surface of the biological sample, (e) deriving 132 quantitative measures (i.e., features) representative of the multidimensional dataset, (f) generating 134 a predictive model based on correlation data between the subject's health metadata and the quantitative measures of the multidimensional dataset, and (g) using the predictive model to provide a subject's score on the dental topography measures to predict features and / or phenotypic characterizations correlated with health outcomes 135. In some cases, steps (a)-(c) may be replaced with imaging the biological sample to generate a multidimensional dataset of a surface of the biological sample. In some cases, the method may further include (b) cleaning 124 the surface of the biological sample to mount the biological sample in the fixture. In some examples, cleaning the surface of the biological sample may remove dirt, debris, or other particles that may adversely affect the measured topography of the biological sample surface. In some examples, cleaning the surface of the biological sample may be done with a moist lint-free cloth or fabric that leaves no residue after cleaning. In some cases, deionized water may be used with a Kimwipe to clean the surface of the biological sample. In some examples, the biological sample may include an ex-vivo or in-vivo biological sample (e.g., an extracted tooth of a subject compared to a tooth of the subject in the mouth of the subject). In some cases, the method may further include calibrating the compressible body on a standard glass, a ball grid array (BGA), a groove target, a validation plate, or any combination thereof, prior to contacting the compressible body with the biological sample. In some examples, the spatial position of the compressible body may be releasably lockable.

[0040] In some cases, the fixture 110 may include putty, a mechanical attachment, an adhesive, or any combination thereof. In some examples, the fixture may limit or restrict movement of the biological sample when determining a topographical dataset of a surface of the biological sample using an apparatus described elsewhere herein. In some examples, the at least one light source and the at least one sensor may be in electrical communication with a button, switch, trigger, or any combination thereof configured to perform the following actions: start illuminating the biological sample, pause illuminating the biological sample, stop illuminating the biological sample, start collecting a topographical dataset with the at least one sensor, pause collecting a topographical dataset with the at least one sensor, stop collecting a topographical dataset with the at least one sensor, or any combination thereof. In some cases, the multidimensional dataset includes at least one dimension. In some cases, the method may include generating a trained predictive model by training the model with the topographical dataset and the corresponding subject health metadata.

[0041] In some cases, the disclosure provided herein describes a method for outputting a feature importance score of one or more features that correlate with a phenotypic characterization of a subject. In some cases, the method may include (a) receiving a biological sample from a subject; (b) mounting the biological sample in a fixture; (c) contacting a compressible body optically coupled to at least one light source and at least one sensor with a surface of the biological sample, the at least one light source being detected by the at least one sensor to generate a multidimensional dataset of the surface of the biological sample; (d) determining one or more features of the multidimensional dataset of the surface of the biological sample by analyzing the multidimensional dataset; and (e) outputting a feature importance score of one or more features of the topography dataset. In some cases, steps (a)-(c) may be replaced with imaging one or more in-vivo biological samples of one or more subjects to generate a multidimensional dataset of the surface of the one or more in-vivo biological samples. In some cases, the method may further include (b) cleaning the surface of the biological sample to mount the biological sample in the fixture. In some examples, the surface of the biological sample may be washed to remove dirt, debris, or other particles that may adversely affect the measured topography of the biological sample surface. In some examples, washing the surface of the biological sample may be done with a moist lint-free cloth or fabric that leaves no residue after washing. In some cases, deionized water may be used with a Kimwipe to clean the surface of the biological sample. In some examples, the biological sample may include an ex-vivo or in-vivo biological sample (e.g., a subject's extracted teeth compared to the subject's teeth in the subject's mouth). In some cases, the method may further include calibrating the compressible body on a standard glass, a ball grid array (BGA), a groove target, a validation plate, or any combination thereof, prior to contacting the compressible body with the biological sample. In some examples, the spatial position of the compressible body may be releasably lockable.

[0042] In some cases, one or more features of the topographical data set may be correlated or associated with one or more exposome signatures. In some examples, the one or more exposome signatures may include metal ion concentrations. In some cases, the metal ions may include ions of chemical elements including zinc (Zn), lead (Pb), copper (Cu), arsenic (As), manganese (Mn), cadmium (Cd), magnesium (Mg), calcium (Ca), and chromium (Cr). In some examples, the temporal metal ion concentrations are determined from the subject's biological sample by standard mass spectrometry.

[0043] In some cases, the fixture may include putty, a mechanical attachment, an adhesive, or any combination thereof. In some examples, the fixture may limit or restrict movement of the biological sample when determining a topographical dataset of the surface of the biological sample using an apparatus described elsewhere herein. In some examples, the at least one light source and the at least one sensor may be in electrical communication with a button, switch, trigger, or any combination thereof configured to perform the following actions: start illuminating the biological sample, stop illuminating the biological sample, stop illuminating the biological sample, start collecting a topographical dataset with the at least one sensor, stop collecting a topographical dataset with the at least one sensor, stop collecting a topographical dataset with the at least one sensor, or any combination thereof. In some cases, the multidimensional dataset includes at least one dimension. In some cases, the method may include generating a trained predictive model by training the model with the topographical dataset and the corresponding subject health metadata.

[0044] In some cases, the disclosure provided herein describes a method of training a predictive model to output a phenotypic characterization of a phenotypic trait of one or more subjects. In some cases, the method of the present disclosure may include: (a) receiving one or more ex-vivo biological samples and phenotypic characterization data from a first set of subjects; (b) determining a first topographical dataset of the surface of one or more biological samples of the first set of subjects; (c) calculating a first set of features of the first topographical dataset; and (d) training a predictive model with the first set of features and the phenotypic data of the first set of subjects to generate a trained predictive model configured to output a phenotypic characterization of a second set of one or more subjects when the trained predictive model is input with a second set of features of a second set of topographical data. In some cases, steps (a) and (b) may be replaced by imaging one or more in-vivo biological samples of one or more subjects. In some examples, the first set and the second set of subjects are the same or different.

[0045] In some examples, the first or second set of features may include features of the topography data, such as the number of peaks detected in a two-dimensional surface profile, the number of peaks detected in a one-dimensional surface profile, the slope of one or more peaks, the two-dimensional gradient across the topography data, or any combination thereof. In some examples, the first or second set of features may include measurements obtained from the application of a recursive quantification analysis to the surface topography data. In some examples, the first or second set of features may include measures obtained from the surface topography data, particularly measurements obtained from the application of a recursive quantification analysis to peak spacing. In some examples, the first or second set of features may involve measures of the surface topography data that relate to aspects of information theory, particularly the derivation of Shannon entropy in the surface topography waveform, or Shannon entropy in the peak spacing. In some cases, the first or second features may include a linear profile of the topography data. In some examples, the profile may be drawn by a user or operator on a user interface as described elsewhere herein. In some examples, the linear profile may be taken from an anatomically co-registered region of the topographical data set. In some cases, the anatomically co-registered region may include a linear segment between the incisal edge and cervical edge of the crown of the tooth ex-vivo or in-vivo. In some cases, the first or second set of features may include parameters of the concavity and convexity of the linear profile, the average height, or any combination thereof.

[0046] In some cases, the phenotypic characterization of the first set of subjects may include the presence or absence of a disease state (e.g., autism, ADHD, cancer, etc.), physiological parameters (e.g., weight, height, body mass index), aspects of psychological development including intelligence quotient and / or cognitive measures, or any combination thereof. In some cases, determining the first topographical data set from one or more biological samples of the first set of subjects may be accomplished by a tabletop system as described elsewhere herein.

[0047] In some cases, the disclosure provided herein describes a method for predicting a phenotypic characterization of a topographical data set of one or more subjects. In some examples, the method may include (a) receiving one or more biological samples and phenotypic data from one or more subjects, (b) determining topographical data from one or more biological samples of the one or more subjects, (c) calculating a set of features of the topographical data, and (d) predicting a phenotypic characterization as an output of the trained predictive model when the trained predictive model is provided with the set of features of the one or more subjects as input. In some cases, steps (a) and (b) may be replaced by imaging one or more in-vivo biological samples of the one or more subjects.

[0048] In some examples, the set of features may include features of the topography data: a number of peaks detected, a slope of one or more peaks, a two-dimensional gradient across the topography data, or any combination thereof. In some examples, the first or second set of features may include measurements obtained from application of recursive quantification analysis (RQA) to the surface topography data.

[0049] The RQA may include the construction of recurrence plots to visualize and analyze features derived from the topographical surface data. Such recurrence plots may illustrate phasic processes in the spatial measurements. From the spatial topography measured from the surface of a biological sample, e.g., a tooth sample, additional dimensions may be computationally derived to embed the spatial measurements in a higher dimensional space called a phase diagram, where x refers to the dimension of the topographical data set and dimensions (x+τ) and (x+2τ) may be derived from the original spatial dimensions shifted by an interval τ. The following analysis may then be performed on the embedded phase diagram to construct a recurrence plot and a recursive quantification analysis. A recurrence quantification plot may be obtained from the phase diagram by application of a threshold function to each point in the phase diagram. On the corresponding recurrence plot, which consists of a square binary matrix, typically represented as a black and white space, a given point is assigned a value of 1 in each spatial interval where another point in the phase diagram shares the spatial limit of the boundary of the assigned threshold. The RQA method may be applied to a recurrence plot to examine the intervals between states in a given system, with the black dots reflecting the spatial intervals when the system revisits the same state. A periodic process, where a system continuously repeats a given pattern of states, may appear in a recurrence plot as a black diagonal line, while periods of stability may appear as a square structure, with pseudorepetitions as black dots and unique events as white space.

[0050] In some embodiments, recurrence plots may be constructed for one spatial dimension or a combination of two or more dimensions of the topographic data (e.g., to visualize interactive periodic patterns in two or more dimensions of a topographic dataset), which may be referred to as cross-recurrence quantification analysis, or joint recurrence quantification analysis.

[0051] In some embodiments, the data analysis may include analyzing the recurrence plot to obtain a set of features associated with the recurrence plot. The features, interchangeably referred to as "rhythmic features" or "dynamic features", provide quantitative measures describing the periodicity, predictability, and transitivity present in one or more dimensions of the topographical data set. The features are selected from a set including recurrence rate, determinism, average diagonal length, maximum diagonal length, divergence, Shannon entropy in diagonal length, trend in recurrences, laminarity, trapping time, maximum vertical line length, Shannon entropy in vertical line length, average recurrence time, Shannon entropy in recurrence time, number of the most probable recurrences, and / or any combination thereof.

[0052] In some examples, the first or second set of features may include measures derived from the surface topography data, particularly measurements derived from the application of Recursive Quantification Analysis (RQA) to peak spacing. In some examples, the first or second set of features may involve measures of the surface topography data that involve aspects of information theory, particularly the derivation of Shannon entropy in the surface topography waveform, or Shannon entropy in the peak spacing. In some cases, the set of features may include a linear profile of the topography data. In some examples, the profile may be drawn by a user or operator on a user interface described elsewhere herein. In some examples, the linear profile may be taken from an anatomically co-registered region of the topography data set. In some cases, the anatomically co-registered region may include a linear segment between the incisal edge and cervical edge of the crown of the tooth ex-vivo or in-vivo. In some cases, the set of features may include parameters of the concavity and convexity of the linear profile, the average height, or any combination thereof.

[0053] In some examples, applying recurrence quantification analysis to surface topography data or derived features of the surface topography data, such as detected peaks, peak spacing, or measured irregularities of the surface topography, may yield measures including signal recurrence rate, determinism, average diagonal length, maximum diagonal length, divergence, Shannon entropy in diagonal length, trend in recurrences, laminarity, dwell time, maximum vertical line length, Shannon entropy in vertical line length, average recurrence time, Shannon entropy in recurrence time, number of most likely recurrences, or any combination thereof.

[0054] In some examples, derivation of features from the surface topography data or from derived features of the surface topography data, such as detected peaks, peak spacing, etc., may include estimation of Lyapunov exponents, entropy estimation, convergent intersection mapping, nonlinear modeling and parameter estimation, change-point estimation, frequency domain representations of the surface topography data or features derived from the surface topography data, power spectral domain representations of the surface topography data, features derived from the surface topography data, or any combination thereof.

[0055] In some examples, recursive matrices obtained from analysis of surface topography data or derived features of the surface topography data, such as detected peaks, peak spacing, or measured irregularities of the surface topography, may be extended into a network-based model, where features representing network connectivity, efficiency, feature importance, path importance, and related graph theory based metrics may be obtained.

[0056] Methods and features of RQA are described, for example, in Webber et al., "Simpler Methods Do It Better: Success of Recurrence Quantification Analysis as a General Purpose Data Analysis Tool," Physics Letters A 373, 3753-3756 (2009), and Marwan et al., "Recurrence Plots for the Analysis of Complex Systems," Physics Reports 438, 237-239 (2007), the contents of each of which are incorporated herein by reference in their entirety. In some embodiments, the spatial dimensions of the topographical data are analyzed by using other analytical methods, such as Fourier transforms, wavelet analysis, and cosinor analysis. Such techniques can be applied to obtain similar metrics, including spectral analysis of frequency content and their associated power. These metrics and associated derived measures may be used in place of features obtained from RQA to analyze one or more spatial dimensions of topographical data obtained from the surface of a biological sample for predictive classification purposes.

[0057] In some examples, the machine learning predictive model utilizing features obtained from the analysis of the surface topography data may include one or more statistical, machine learning, or artificial intelligence algorithms. Examples of algorithms utilized may include gradient boosting ensemble learners, support vector machines (SVMs), 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 or unsupervised machine learning, statistical, or deep learning algorithms for classification and regression. Machine learning classifiers may also include estimation of ensemble models consisting of multiple predictive models, and may utilize techniques such as gradient boosting in the construction of gradient boosting decision trees. Machine learning classifiers may be trained using one or more training datasets corresponding to the data of interest. In some embodiments, the one or more training datasets may include exposome biochemical signatures, dynamic exposome biochemical signatures, clinical metadata, clinical trial information, exposome biochemical signature information of pharmaceutical and nutritional supplement treatments, or any combination thereof.

[0058] In some embodiments, the classifier 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, 3371-3408; Larochelle et al., 2009, "Exploring strategies for training deep neural networks," J Mach Learn Res 10, 1-40; and Hassoun, 1995, Fundamentals of Artificial Neural Networks, Massachusetts Institute of Technology, each of which is incorporated herein by reference.

[0059] 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", Proceedings of the 5th 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 (Fundamentals of Statistical Learning), Springer, New York, and Furey et al., 2000, Bioinformatics 16, 906-914, each of which is incorporated herein by reference in its entirety. When used for classification, SVM separates a given set of binary labeled data by a hyperplane that is the maximum distance away from the labeled data. For cases where linear separation is not possible, SVM can work in combination with the technique of "kernels" that automatically realize a nonlinear mapping to the feature space. The hyperplane discovered by SVM in the feature space corresponds to a nonlinear decision boundary in the input space.

[0060] Decision trees are reviewed by Duda, 2001, Pattern Classification, John Wiley & Sons, New York, pp. 395-396, which is incorporated herein by reference. Tree-based methods divide the feature space into a set of rectangles and then fit a model (as a constant) to each. In some embodiments, the decision tree is a random forest regression. One specific algorithm that can be used is Classification and Regression Trees (CART). Other specific decision tree algorithms include, but are not limited to, ID3, C4.5, MART, and Random Forest. CART, ID3, and C4.5 are described in Duda, 2001, Pattern Classification, John Wiley & Sons, New York, pp. 396-408 and 411-412, which is incorporated herein 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 incorporated herein by reference in its entirety. Random forests are described in Breiman, 1999, "Random Forests - Random Features," Technical Report 567, Department of Statistics, University of California, Berkeley, September 1999, which is incorporated herein by reference in its entirety.

[0061] Clustering (e.g., unsupervised and supervised clustering model algorithms) is described in Duda and Hart, Pattern Classification and Scene Analysis, 1973, John Wiley & Sons, Inc., New York, pp. 211-256 (hereinafter "Duda 1973"), which is incorporated herein by reference in its entirety. As described in section 6.7 of Duda 1973, the clustering problem is described as finding natural groupings in a data set. To identify natural groupings, two problems are addressed. First, a method for measuring the similarity (or dissimilarity) between two samples is determined. This metric (similarity measure) is used to ensure that samples in one cluster are more similar to samples in the other cluster. Second, a mechanism for dividing the data into clusters using the similarity measure is determined. Similarity measures are covered in Duda 1973, section 6.7, where it is stated that one way to begin a clustering search is to define a distance function and compute a matrix of distances between all pairs of samples in the training set. If distance were a good measure of similarity, the distance between reference entities in the same cluster would be significantly smaller than the distance between reference entities in different clusters. However, as stated in Duda 1973, p. 215, clustering does not require the use of a distance metric. For example, a non-metric similarity function s(x,x') can be used to compare two vectors x and x'. Traditionally, s(x,x') is a symmetric function whose value is large when x and x' are somehow "similar". An example of a non-metric similarity function s(x,x') is given in Duda 1973, p. 218. Once a method for measuring "similarity" or "dissimilarity" between points in a data set has been selected, clustering requires a criterion function that measures the clustering quality of any partition of the data.The data are clustered using a partition of the data set that extremizes a criterion function. See Duda 1973, p. 217. Criterion functions are discussed in section 6.8 of Duda 1973. More recently, Duda et al., Pattern Classification, 2nd ed., John Wiley & Sons, New York, pp. 537-563 discusses clustering in detail. Further information on clustering techniques is available in Kaufman and Rousseeuw, 1990, Finding Groups in Data: An Introduction to Cluster Analysis, Wiley, New York, NY; Everitt, 1993, Cluster analysis (3rd ed.), Wiley, New York, NY; and Backer, 1995, Computer-Assisted Reasoning in Cluster Analysis, Prentice Hall, Upper Saddle River, NJ, each of which is incorporated herein by reference. Particular exemplary clustering techniques that can be used in the present disclosure include, but are not limited to, hierarchical clustering (agglomerative clustering using nearest neighbor, furthest neighbor, average linkage, centroid, or sum of squares algorithms), k-means clustering, fuzzy k-means clustering algorithm, and Jarvis-Patrick clustering. In some embodiments, the clustering includes unsupervised clustering, where when the training set is clustered, no preconceived notions are assumed about what the clusters should form.

[0062] Regression models, such as those of the multicategory logit model, are described in Agresti, An Introduction to Categorical Data Analysis, 1996, John Wiley & Sons, New York, Chapter 8, which is incorporated herein by reference in its entirety. In some embodiments, the classifier utilizes the regression model disclosed in Hastie et al., The Elements of Statistical Learning, 2001, Springer-Verlag, New York, which is incorporated herein by reference in its entirety. In some embodiments, gradient boosting models are used, for example, for the classification algorithms described herein, and these gradient boosting models are "Gradient Boosting" by Bradley Boehmke, Brandon Greenwell (2019). Hands-On Machine Learning with R. Chapman & Hall. Pages 221-245. ISBN978-1-138-49568-5, which is incorporated herein by reference in its entirety. In some embodiments, ensemble modeling techniques are used for the classification algorithms described herein, for example, and these ensemble modeling techniques are described in the implementation of classification models herein and in Ensemble Methods: Foundations and Algorithms by Zhou Zhihua (2012). Chapman and Hall / CRC. ISBN978-1-439-83003-1, which is incorporated herein by reference in its entirety.

[0063] In some embodiments, the machine learning analysis may be performed by a device executing one or more programs (e.g., one or more programs stored in non-persistent memory (i.e., RAM or ROM) 604 or storage 606 (i.e., hard disk) shown in FIG. 6) that include instructions to perform the data analysis. In some embodiments, the data analysis is performed by a system that includes at least one processor (e.g., processing core 605) and memory (e.g., one or more programs stored in non-persistent memory 604 or storage 606) that include instructions to perform the data analysis.

[0064] In some cases, the phenotypic characterization of one or more subjects may include the presence or absence of a disease state (e.g., autism, ADHD, cancer, etc.), physiological parameters (e.g., weight, height, body mass index), aspects of psychological development including intelligence quotient and / or cognitive measures, or any combination thereof. In some cases, determining the topographical dataset from one or more biological samples of one or more subjects may be accomplished by benchtop and / or handheld systems as described elsewhere herein.

[0065] In some cases, the accuracy of one or more features of the calculated feature set in predicting a phenotypic trait, as described elsewhere herein, may be analyzed as seen from the experimental data shown in Figures 3A-3B, 4A-4B, and 5A-5B. In some cases, one or more features of the feature set may predict a phenotypic characterization of intelligence quotient (IQ), as described in Figure 5A. Alternatively, or in addition, one or more features of the feature set may predict a phenotypic characterization of a disease state, e.g., autism, ADHD, the presence or absence of cancer, a physiological parameter, e.g., weight (Figure 3B), height (Figures 3A and 4A), body mass index (Figure 4B), cognitive composite score (Figure 5B), or any combination thereof. In some examples, the device may utilize one or more features of the feature set to predict the concentration of a chemical biomarker, such as an essential or non-essential element, as described elsewhere herein, that may be collected in a biological tissue or tissue sample of a subject, e.g., dentate tissue, as shown in Figures 7A-7I.

[0066] Although the above steps illustrate each of the methods or sets of operations according to the embodiments, one of ordinary skill in the art will recognize numerous variations based on the teachings described herein. Steps may be completed in different orders. Steps may be added or omitted. Some of the steps may include sub-steps. Many of the steps may be repeated as often as is beneficial.

[0067] One or more of the steps of each of the methods or sets of operations may be performed by one or more of the circuitry described herein, e.g., a processor or logic circuitry such as programmable array logic for a field programmable gate array. The circuitry may be programmed to provide one or more of the steps of each of the methods or sets of operations, and the program may include program instructions stored on a computer readable memory or programmed steps of logic circuitry, e.g., programmable array logic or a field programmable gate array.

[0068] definition Unless otherwise defined, all technical terms, notations and other scientific or terminological terms used herein are intended to have the same meaning as commonly understood by one of ordinary skill in the art to which the claimed subject matter belongs. In some cases, terms having commonly understood meanings are defined herein for clarity and / or ready reference, and the inclusion of such definitions herein should not necessarily be construed as representing a substantial departure from what is commonly understood in the art.

[0069] Throughout the application, various embodiments may be presented in a range format. It should be noted that the range format is merely for convenience and brevity and should not be construed as an inflexible limitation on the scope of the disclosure. Thus, the description of a range should be considered to have specifically disclosed all possible subranges and individual numerical values ​​within that range. For example, the description of a range such as 1-6 should be considered to have specifically disclosed subranges such as 1-3, 1-4, 1-5, 2-4, 2-6, 3-6, etc., as well as individual numerical values ​​within that range, such as 1, 2, 3, 4, 5, and 6. This is true regardless of the breadth of the range.

[0070] As used in this specification and claims, the singular forms "a," "an," and "the" include plural referents unless the context clearly dictates otherwise. For example, the term "a sample" includes a plurality of samples, including mixtures thereof.

[0071] The terms "determining," "measuring," "evaluating," "assessing," "assaying," and "analyzing" are often used interchangeably herein to refer to forms of measurement. The terms include determining whether an element is present or not (e.g., detecting). These words can include quantitative, qualitative, or both quantitative and qualitative determinations. Assessment can be relative or absolute. "Detecting the presence of" can include determining the amount of something present in addition to determining whether something is present or absent, depending on the context.

[0072] The terms "subject", "individual", or "patient" are often used interchangeably herein. A "subject" may be a biological entity that contains expressed genetic material. The biological entity may be a plant, an animal, or a microorganism, including, for example, bacteria, viruses, fungi, and protozoa. A subject may be tissues, cells, and progeny of a biological entity obtained in vivo or cultured in vitro. A subject may be a mammal. A mammal may be a human. A subject may have been diagnosed with or be suspected of being at risk for a disease. In some cases, a subject may not necessarily have been diagnosed with or be suspected of being at risk for the disease.

[0073] The term "in vivo" is used to describe events that take place within the body of a subject.

[0074] The term "ex vivo" is used to describe events that occur outside of a subject's body. Ex vivo assays are not performed on a subject. Rather, they are performed on a sample that is separate from the subject. An example of an ex vivo assay that is performed on a sample is an "in vitro" assay.

[0075] The term "in vitro" is used to describe events that occur in a container that holds laboratory reagents separate from the biological source from which the material is obtained. In vitro assays can include cell-based assays in which living or dead cells are used. In vitro assays can also include cell-free assays in which no intact cells are used.

[0076] As used herein, the term "about" and a number refers to the number plus or minus 10%. The term "about" and a range refers to a range from the lower limit minus 10% to the upper limit plus 10%.

[0077] As used herein, the term "treatment" or "treating" is used in reference to a pharmaceutical or other intervention regimen to obtain a beneficial or desired result in a recipient. Beneficial or desired results include, but are not limited to, therapeutic benefit and / or prophylactic benefit. Therapeutic benefit may refer to the eradication or amelioration of symptoms or the underlying disease being treated. Therapeutic benefit may also be achieved by eradication or amelioration of one or more of the physiological symptoms associated with the underlying disease, such that an improvement is observed in the subject, although the subject may still suffer from the underlying disease. Prophylactic effects include delaying, preventing or eliminating the appearance of a disease or condition, delaying or eliminating the manifestation of a symptom of a disease or condition, slowing, stopping or reversing the progression of a disease or condition, or any combination thereof. For prophylactic benefit, subjects at risk of developing a particular disease or reporting one or more of the physiological symptoms of a disease may be treated even if the disease has not been diagnosed.

[0078] The section headings used herein are for organizational purposes only and are not to be construed as limiting the subject matter described. EXAMPLES

[0079] The following examples are included for illustrative purposes only and are not intended to limit the scope of the invention.

[0080] Example 1: Dental topography data as predictors for measured height and weight in children In a sample set of 52 subjects collected in a population-based cohort study, ex vivo teeth were analyzed with a tabletop system with a TPE gel cartridge as described elsewhere herein. Ex vivo tooth samples were obtained between 1 and 5 years from the time of extraction. Each tooth was then scanned with a tabletop system as described elsewhere herein to generate a 3D topography data set of the upper surface of the tooth. The 3D topography data set was then analyzed as described elsewhere herein to generate one or more features of the 3D topography data set. One such feature calculated from the 3D topography data set included a linear profile of the 3D topography data set measured from the incisal edge of the tooth to the cervical edge of the labial / buccal crown of the tooth. Other features representing the tooth topography were obtained as described and used in constructing a model to predict anthropometric characteristics at age 6 years, including height and weight measured at the clinical visit. FIG. 3A shows, for each sample, the measured height of the child associated with the sample and the predicted height of the child based on a model trained on the data obtained for the dental topography. The predicted height was highly correlated with the measured height (r=0.98), indicating high accuracy of the prediction model. FIG. 3B shows, for each sample, the measured weight of the child associated with the sample and the predicted weight of the child based on a model trained on the features obtained from the dental topography of the sample. The predicted BMI child weight was highly correlated with the measured weight (r=0.92), indicating high accuracy of the prediction model.

[0081] Example 2: Dental topography data as predictors for measured height and weight in children In a sample set of 150 subjects collected in another population-based cohort study, ex vivo teeth were analyzed by a tabletop system with a TPE gel cartridge. Ex vivo tooth samples were obtained between 1 and 5 years from the time of extraction. Each tooth was then scanned by a tabletop system as described elsewhere herein to generate a 3D topography data set of the top surface of the tooth. The 3D topography data set was then analyzed as described elsewhere herein to generate one or more features of the 3D topography data set. One such feature calculated from the 3D topography data set included a linear profile of the 3D topography data set measured from the incisal edge of the tooth to the cervical edge of the labial / buccal crown of the tooth. Further features representing the tooth topography were obtained as described and used in constructing a model predicting anthropometric characteristics at age 6 years, including height and body mass index (BMI) measured at the clinical visit. Figure 4A shows, for each sample, the measured height of the child associated with the sample and the predicted height of the child based on a model trained on the data obtained for the dental topography. The predicted height was highly correlated with the measured height (r=0.97), indicating high accuracy of the prediction model. Figure 4B shows, for each sample, the measured BMI of the child associated with the sample and the predicted BMI based on a model trained on the features obtained from the dental topography of the sample. The predicted BMI was highly correlated with the measured height (r=0.99), indicating high accuracy of the prediction model.

[0082] Example 3: Dental Topography Data as Predictors for IQ and Child Cognitive Composite Score In a sample set of 150 subjects collected in population-based cohort study 2 (as seen in Example 2), ex vivo teeth were analyzed by a tabletop system with a TPE gel cartridge. Ex vivo tooth samples were obtained between 1 and 5 years from the time of extraction. Each tooth was then scanned by a tabletop system as described elsewhere herein to generate a 3D topography data set of the top surface of the tooth. The 3D topography data set was then analyzed as described elsewhere herein to generate one or more features of the 3D topography data set. One such feature calculated from the 3D topography data set included a linear profile of the 3D topography data set measured from the incisal edge of the tooth to the cervical edge of the labial / buccal crown of the tooth. Another feature describing the tooth topography was also obtained as described and used in constructing a model predicting IQ and cognitive development. These phenotypic characteristics were measured at age 6 using standard psychological assays, specifically the Wechsler Intelligence Scale for Children (WISC) and the Behavioral Assessment System for Children (BASC), respectively. Figure 5A shows, for each sample, the measured IQ of the child associated with the sample and the predicted IQ of the child based on the model trained on the data obtained for dental topography. The predicted IQ score and the measured IQ score were highly correlated (r=0.99), indicating the high accuracy of the prediction model. Figure 5B shows, for each child, the measured cognitive total score and the predicted cognitive total score based on the prediction model. The measured cognitive total score and the predicted cognitive total score were highly correlated (r=0.98), indicating the high accuracy of the model.

[0083] Example 4: Dental Topography Data as Predictors for IQ and Child Cognitive Composite Score In a sample set of 127 subjects collected in a population-based cohort (a subset of subjects in Example 2), parallel analysis was performed by standard mass spectrometry to determine the concentrations of various chemical elements, including zinc (Zn), lead (Pb), copper (Cu), arsenic (As), manganese (Mn), cadmium (Cd), magnesium (Mg), calcium (Ca), and chromium (Cr). Features representative of the tooth topography were then obtained as described (prior to chemical analysis) and used in constructing models predicting elemental biomarker concentrations. For each element, the metal concentration measured in each tooth by mass spectrometry was plotted against the predicted concentration for that tooth based on the model trained on features obtained from the analysis of the tooth surface topography (Figures 7A-7I). The model performance was accurate, with correlations between measured and predicted element concentrations ranging from 0.87 to 0.96. More specifically, for Zn (Figure 7A), the model performance was r=0.96, with a correlation between measured and predicted values. For Pb (Figure 7B), the model performance was r = 0.91 for the measured and predicted values. For Cu (Figure 7C), the model performance was r = 0.96 for the measured and predicted values. For As (Figure 7D), the model performance was r = 0.87 for the measured and predicted values. For Mn (Figure 7E), the model performance was r = 0.91 for the measured and predicted values. For Cd (Figure 7F), the model performance was r = 0.90 for the measured and predicted values. For Mg (Figure 7G), the model performance was r = 0.91 for the measured and predicted values. For Ca (Figure 7H), the model performance was r = 0.91 for the measured and predicted values. For Cr (Figure 7I), the model performance was r = 0.92 for the measured and predicted values.

[0084] While preferred 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 by way of example only. Numerous variations, changes, and substitutions will now occur to those skilled in the art without departing from the invention. It is understood that various alternatives to the embodiments of the invention described herein may be used in practicing the invention. It is intended that the following claims define the scope of the invention and that methods and structures within the scope of these claims and their equivalents be covered therein.

Claims

1. 1. A method for determining the topography of a surface of a biological sample, comprising: (a) receiving a biological sample; (b) mounting the biological sample in a fixture; (c) contacting the biological sample with a compressible body optically coupled to at least one light source and at least one sensor, the at least one light source generating a multidimensional dataset of a surface of the biological sample detected by the at least one sensor; and (d) determining a surface profile of the biological sample from the multidimensional data set. The method includes:

2. The method of claim 1 , wherein the biological sample comprises an ex-vivo or in-vivo tooth.

3. The method of claim 1 , wherein the compressible body comprises a gel cartridge.

4. The method of claim 1 , wherein the fastener comprises a putty, a mechanical attachment, an adhesive, or any combination thereof.

5. The method of claim 1 , further comprising calibrating the compressible body on a standard glass, a ball grid array, a groove target, or any combination thereof.

6. The method of claim 1 , wherein the at least one light source comprises at least one coherent laser, an incoherent laser, a pulsed laser, a light emitting diode, a superluminescent diode, or any combination thereof.

7. The method of claim 1 , further comprising washing the biological sample with water.

8. The method of claim 1 , wherein the spatial position of the compressible body is releasably lockable.

9. 10. The method of claim 1, wherein the at least one light source and the at least one sensor are in electrical communication with a processor via a button, switch, trigger, or any combination thereof configured to initiate illumination of the biological sample and detection of the multidimensional dataset of the surface.

10. The method of claim 1 , wherein the multidimensional data set includes at least one dimension.

11. The method of claim 2, wherein the surface profile is measured from the incisal edge to the cervical edge of the crown of the ex-vivo or in-vivo tooth.

12. 1. A method for determining the topography of a surface of a biological sample, comprising: (a) receiving a biological sample; (b) mounting the biological sample in a fixture; (c) illuminating the biological specimen with at least one light source and at least one sensor, the at least one light source generating a multi-dimensional dataset of a surface of the biological specimen detected by the at least one sensor; and (d) determining a surface profile of the biological sample from the multidimensional data set. The method includes:

13. The method of claim 12, wherein the biological sample comprises an ex-vivo or in-vivo tooth.

14. The method of claim 12 , wherein the fastener comprises a putty, a mechanical attachment, a stabilizer, an adhesive, or any combination thereof.

15. 13. The method of claim 12, wherein the at least one light source comprises at least one coherent laser, an incoherent laser, a pulsed laser, a light emitting diode, a superluminescent diode, or any combination thereof.

16. 13. The method of claim 12, further comprising washing the biological sample with water.

17. 13. The method of claim 12, wherein the at least one light source and the at least one sensor are in electrical communication with a processor via a button, switch, trigger, or any combination thereof configured to initiate illumination of the biological sample and detection of the multidimensional dataset of the surface.

18. The method of claim 12 , wherein the multidimensional data set includes at least one dimension.

19. The method of claim 13, wherein the surface profile is measured from the incisal edge to the cervical edge of the crown of the ex-vivo or in-vivo tooth.

20. 1. A system for determining the topography of a surface of a biological specimen, comprising: (a) a fixture mechanically configured to constrain a three-dimensional position of a biological specimen; (b) a compressible body optically coupled to at least one light source and at least one sensor, the at least one light source generating a multidimensional data set of a surface of the biological sample detected by the at least one sensor; and (c) a processor electrically coupled to the at least one light source and the at least one sensor, the processor including a set of programmatic instructions stored on a non-transitory storage medium configured to cause the processor to determine a topography of the surface of the biological specimen from the multidimensional dataset; and A system including:

21. The system of claim 20 , wherein the biological sample comprises a tooth.

22. The system of claim 20 , wherein the compressible body comprises a gel cartridge.

23. 21. The system of claim 20, wherein the fastener comprises a putty, a mechanical attachment, an adhesive, or any combination thereof.

24. 21. The system of claim 20, wherein the at least one light source comprises at least one coherent laser, an incoherent laser, a pulsed laser, a light emitting diode, a superluminescent diode, or any combination thereof.

25. The system of claim 20 , wherein the spatial position of the compressible body is releasably lockable.

26. 21. The system of claim 20, wherein the at least one light source and the at least one sensor are in electrical communication with a processor via a button, switch, trigger, or any combination thereof configured to initiate illumination of the biological sample and detection of the multidimensional dataset of the surface.

27. 21. The system of claim 20, further comprising a rigid guide member mechanically coupled to the compressible body configured to position the compressible body relative to the biological sample.

28. 28. The system of claim 27, wherein the compressible body is contained in a locking mechanical member configured to fix the three-dimensional position of the compressible body.

29. The system of claim 20 , wherein the multidimensional data set includes at least one dimension.

30. 1. A method of training a predictive model for outputting a phenotypic characterization of one or more subject topographical datasets, comprising: (a) receiving one or more biological samples and phenotypic characterization from a first set of subjects; (b) determining a first topographical data set from one or more biological samples of a first set of subjects; (c) calculating a first set of features of the first topographical data set; and (d) training a predictive model with the first set of features and the phenotypic characterization of the first set of subjects to generate the trained predictive model configured to output a phenotypic characterization of a second set of one or more subjects when the trained predictive model is input with a second set of features of a second topographical dataset. The method includes:

31. 31. The method of claim 30, wherein the one or more biological samples comprise a tooth.

32. 31. The method of claim 30, wherein the first set of features of the topography dataset comprises the following features of the topography dataset: a number of peaks of the topography dataset detected, a slope of one or more of the peaks, a two-dimensional gradient across the topography data, or any combination thereof.

33. The method of claim 30 , wherein the predictive model comprises a machine learning algorithm.

34. 34. The method of claim 33, wherein the machine learning algorithm comprises a support vector machine (SVM), a naive Bayes classification, a random forest, a neural network, a deep neural network (DNN), a recurrent neural network (RNN), a deep RNN, a long short-term memory (LSTM) recurrent neural network (RNN), a gated recurrent unit (GRU), a supervised learning algorithm, an unsupervised machine learning algorithm, or any combination thereof.

35. 31. The method of claim 30, wherein the phenotypic characterization comprises the presence or absence of a disease or disorder, wherein the disease or disorder comprises autism spectrum disorder (ASD), attention deficit hyperactivity disorder (ADHD), amyotrophic lateral sclerosis (ALS), schizophrenia, irritable bowel disease (IBD), pediatric kidney disease, kidney transplant rejection, pediatric cancer, or any combination thereof.

36. 1. A method for training a predictive model to output a phenotypic characterization of phenotypic data of one or more subjects, comprising: (a) generating a first topographical data set by imaging one or more biological samples of a first set of subjects; (b) calculating a first set of features of the first topographical data set; and (c) training a predictive model with the first set of features and the phenotypic characterizations of the first set of subjects to generate the trained predictive model configured to output a phenotypic characterization of a second set of one or more subjects when the trained predictive model is input with a second set of features of a second topographical dataset. The method includes:

37. 37. The method of claim 36, wherein the first and second sets of subjects are the same or different.

38. 37. The method of claim 36, wherein the one or more biological samples comprise a tooth.

39. 37. The method of claim 36, wherein the first or second set of features of the topography dataset comprises the following features of the topography dataset: a number of peaks of the topography dataset detected, a slope of one or more of the peaks, a two-dimensional gradient across the topography data, or any combination thereof.

40. 37. The method of claim 36, wherein the first or second set of phenotypic characterizations comprises the presence or absence of a disease or disorder, wherein the disease or disorder comprises autism spectrum disorder (ASD), attention deficit hyperactivity disorder (ADHD), amyotrophic lateral sclerosis (ALS), schizophrenia, irritable bowel disease (IBD), pediatric kidney disease, kidney transplant rejection, pediatric cancer, or any combination thereof.

41. 37. The method of claim 36, wherein the predictive model comprises a machine learning algorithm.

42. 42. The method of claim 41, wherein the machine learning algorithm comprises a support vector machine (SVM), a naive Bayes classification, a random forest, a neural network, a deep neural network (DNN), a recurrent neural network (RNN), a deep RNN, a long short-term memory (LSTM) recurrent neural network (RNN), a gated recurrent unit (GRU), a supervised learning algorithm, an unsupervised machine learning algorithm, or any combination thereof.

43. 1. A method of predicting a phenotype classification of a topographical dataset of one or more subjects, comprising: (a) generating a topographical data set by imaging one or more biological samples of one or more subjects; (b) calculating a set of features of the topographical dataset; and (c) predicting a phenotypic characterization as an output of the trained predictive model when the trained predictive model is provided with the set of one or more subject features as input. The method includes:

44. 44. The method of claim 43, wherein the one or more biological samples comprise a tooth.

45. 44. The method of claim 43, wherein the set of features of the topography dataset comprises the following features of the topography dataset: a number of peaks of the topography dataset detected, a slope of one or more of the peaks, a two-dimensional gradient across the topography data, or any combination thereof.

46. 44. The method of claim 43, wherein the phenotypic characterization comprises the presence or absence of a disease or disorder, wherein the disease or disorder comprises autism spectrum disorder (ASD), attention deficit hyperactivity disorder (ADHD), amyotrophic lateral sclerosis (ALS), schizophrenia, irritable bowel disease (IBD), pediatric kidney disease, kidney transplant rejection, pediatric cancer, or any combination thereof.

47. 44. The method of claim 43, wherein the predictive model comprises a machine learning algorithm.

48. 48. The method of claim 47, wherein the machine learning algorithm comprises a support vector machine (SVM), a naive Bayes classification, a random forest, a neural network, a deep neural network (DNN), a recurrent neural network (RNN), a deep RNN, a long short-term memory (LSTM) recurrent neural network (RNN), a gated recurrent unit (GRU), a supervised learning algorithm, an unsupervised machine learning algorithm, or any combination thereof.

49. 1. A method for predicting a phenotypic characterization of one or more subjects' topographical datasets, comprising: (a) receiving one or more biological samples and phenotypic data of one or more subjects; (b) determining a topographical dataset from one or more biological samples of said one or more subjects; (c) calculating a set of features of the first topographical data set; and (d) predicting a phenotype classification as an output of the trained predictive model when the trained predictive model is provided with the set of one or more subject features as an input. The method includes:

50. 50. The method of claim 49, wherein the one or more biological samples comprise a tooth.

51. 50. The method of claim 49, wherein the set of features of the topography dataset comprises the following features of the topography dataset: a number of peaks of the topography dataset detected, a slope of one or more of the peaks, a two-dimensional gradient across the topography data, or any combination thereof.

52. 50. The method of claim 49, wherein the phenotypic characterization comprises the presence or absence of a disease or disorder, wherein the disease or disorder comprises autism spectrum disorder (ASD), attention deficit hyperactivity disorder (ADHD), amyotrophic lateral sclerosis (ALS), schizophrenia, irritable bowel disease (IBD), pediatric kidney disease, kidney transplant rejection, pediatric cancer, or any combination thereof.

53. 50. The method of claim 49, wherein the trained predictive model comprises a machine learning algorithm.

54. 54. The method of claim 53, wherein the machine learning algorithm comprises a support vector machine (SVM), a naive Bayes classification, a random forest, a neural network, a deep neural network (DNN), a recurrent neural network (RNN), a deep RNN, a long short-term memory (LSTM) recurrent neural network (RNN), a gated recurrent unit (GRU), a supervised learning algorithm, an unsupervised machine learning algorithm, or any combination thereof.

55. 1. A method for outputting a feature importance score for one or more features of a subject that correlates with a phenotypic characterization, comprising: (a) receiving a biological sample from a subject; (b) contacting a compressible body optically coupled to at least one light source and at least one sensor with a surface of the biological sample, the at least one light source generating a multidimensional dataset of the surface of the biological sample detected by the at least one sensor; (c) determining one or more features of the multidimensional dataset of the surface of the biological sample by analyzing the multidimensional dataset; and (d) outputting a feature importance score for one or more features of the topographical dataset that correlates with the phenotypic characterization. The method includes:

56. 56. The method of claim 55, wherein the biological sample comprises a tooth.

57. 56. The method of claim 55, wherein the one or more of the features of the multidimensional dataset comprise the following features of the topography dataset: a number of peaks of the topography dataset detected, a slope of one or more of the peaks, a two-dimensional gradient across the topography data, or any combination thereof.

58. 56. The method of claim 55, wherein the phenotypic characterization comprises the presence or absence of a disease or disorder, wherein the disease or disorder comprises autism spectrum disorder (ASD), attention deficit hyperactivity disorder (ADHD), amyotrophic lateral sclerosis (ALS), schizophrenia, irritable bowel disease (IBD), pediatric kidney disease, kidney transplant rejection, pediatric cancer, or any combination thereof.

59. 56. The method of claim 55, wherein the feature importance scores are output by a trained predictive model, the trained predictive model comprising a machine learning algorithm.

60. 60. The method of claim 59, wherein the machine learning algorithm comprises a support vector machine (SVM), a naive Bayes classification, a random forest, a neural network, a deep neural network (DNN), a recurrent neural network (RNN), a deep RNN, a long short-term memory (LSTM) recurrent neural network (RNN), a gated recurrent unit (GRU), a supervised learning algorithm, an unsupervised machine learning algorithm, or any combination thereof.