Methods and systems for identifying sepsis

Abstract

Methods and systems for identifying sepsis are provided. The method includes obtaining two-dimensional tandem mass spectrometry (2DMS) data of a sample from a subject. The method includes determining a concentration of at least three biomarkers from the 2DMS data. Each biomarker is individually selected from the group consisting of a peptide, an amino acid, a lipid, a metabolite, and a pathogen. The method includes determining, using a machine learning model stored in a memory of a computing device, a presence of sepsis in the subject based on the concentrations of the at least three biomarkers. The method includes providing an indicia indicating the presence of sepsis in the subject based on the machine learning model.

Description

Reference No. P240021-711 -WO01 / 242278 PCTTITLEMETHODS AND SYSTEMS FOR IDENTIFYING SEPSISPRIORITY

[0001] The present application claims priority under 35 U.S.C. Section 119(e) from United States Provisional Patent Application No. 63 / 714,391, entitled “METHODS AND SYSTEMS FOR IDENTIFYING SEPSIS,” filed October 31, 2024, which is incorporated herein by reference in its entirety.FIELD

[0002] This disclosure relates generally to methods and systems for identifying sepsis.BACKGROUND

[0003] Sepsis is a life-threatening medical emergency that occurs when the body's response to infection spirals out of control, leading to widespread inflammation and potential organ failure. Identifying sepsis early is crucial for improving patient outcomes and reducing mortality rates. The condition can progress rapidly, with symptoms such as fever, rapid heart rate, difficulty breathing, and confusion often developing quickly. Prompt recognition allows for timely intervention, including administering antibiotics, managing fluid levels, and providing organ support if necessary. Healthcare providers use various criteria and screening tools to assess patients for sepsis, as early detection can significantly increase the chances of survival. There is a need for improved methods of determining the presence of sepsis in a sample obtained from a subject.SUMMARY

[0004] In one general aspect, the present disclosure is directed to a method for determining presence of sepsis in a subject. The method comprises obtaining two- dimensional tandem mass spectrometry (2DMS) data of a sample from the subject. The method comprises determining a concentration of at least three biomarkers from the 2DMS data. Each biomarker is individually selected from the group consisting of a peptide, an amino acid, a lipid, a metabolite, and a pathogen. The method comprises determining, using a machine learning model stored in a memory of a computing device, a presence of sepsis in the subject based on the concentrations of the at least three biomarkers. The method comprises providing an indicia indicating the presence of sepsis in the subject based on the machine learning model.11602919598.1Reference No. P240021-711 -WO01 / 242278 PCT

[0005] Another general aspect of the present disclosure is directed to a system for determining presence of sepsis in a subject, the system comprising a control circuit and a memory. The memory stores a plurality of instructions and a pre-trained machine learning model executable by the control circuit, and the instructions are configured to cause the control circuit to perform a method according to the present disclosure for determining presence of sepsis in a subject.

[0006] Although the present disclosure relates to different aspects and embodiments, it is understood that the different aspects and embodiments disclosed herein can be integrated, combined, or used together as a combination system, or in part, as separate components, devices, and systems, as appropriate. Thus, each embodiment disclosed herein can be incorporated in each of the aspects to varying degrees as appropriate for a given implementation.

[0007] These and other features of the applicant’s teachings are set forth herein.BRIEF DESCRIPTION OF THE FIGURES

[0008] Unless specified otherwise, the accompanying drawings illustrate aspects of the innovations described herein. Referring to the drawings, wherein like numerals refer to like parts throughout the several views and this specification, several embodiments of presently disclosed principles are illustrated by way of example, and not by way of limitation. The drawings are not intended to be to scale. A more complete understanding of the disclosure may be realized by reference to the accompanying drawings in which:

[0009] FIG. 1 is a flow diagram illustrating a non-limiting embodiment of a method for identifying sepsis in a sample;

[0010] FIG. 2 is a flow diagram illustrating a non-limiting embodiment of a method for training a machine learning model;

[0011] FIG. 3 is a schematic diagram of a non-limiting embodiment of a system for identifying sepsis in a sample;

[0012] FIG. 4 is a schematic diagram of a non-limiting embodiment of the mass spectrometer of FIG. 3;

[0013] FIGs. 5A-5D are spectra data of samples for various degrees of sepsis; and

[0014] FIGs. 6A-6D are 2DMS data of samples for various degrees of sepsis.21602919598.1Reference No. P240021-711 -WO01 / 242278 PCTDETAILED DESCRIPTION

[0015] The fast and accurate diagnosis of sepsis can be a clinical problem which can result in high mortality rates and costs. For example, sepsis costs the U.S. healthcare system an estimated $38 billion annually and is the cause of greater than 19% of all annual global deaths. It is also the number one cost of hospitalization and rate of patient readmission in the U.S. Detection of sepsis can enable antibiotic administration, which decreases the likelihood of death by approximately 8% per hour. Enabling early and rapid sepsis screening in emergency rooms and ambulatory care units can enhance diagnosis and facilitate rapid antibiotic treatments, thereby enhancing patient outcomes and reducing cost burdens on the U.S. healthcare system.

[0016] The present inventors have observed that the ability to diagnose sepsis is a challenge as symptoms and clinical observations may be confounded with other diagnostic criteria such as pulmonary embolism, anaphylaxis, and toxin overdose. Current approaches in sepsis diagnosis include the confirmation of an infection and clinical observations such as serum lactate or procalcitonin concentration and patient vitals. Confirmation of a sepsis infection currently involves a blood culture requiring 12 to 72 hours to analyze. In addition, there is a low concentration of bacteria at the onset of sepsis (1-100 CFU / mL), which can increase the risk of a false negative diagnosis as the concentration may be below detection limits.

[0017] Typically, only 30-40% of patients who are ultimately diagnosed with sepsis test positive in a blood culture. While recent improvements in advanced technologies have enabled earlier sepsis detection, such technologies require laborious sample preparation and expert technicians to perform techniques such as polymerase chain reaction (PCR) and matrix-assisted laser desorption ionization time-of-flight (MALDI-TOF). Moreover, these technologies may require long analysis times and still may not be able to detect sepsis in its earliest stages. The present inventors believe there is an unmet need in sepsis detection to enable efficient sample preparation, rapid analysis, and the ability to detect the earliest stages of sepsis.

[0018] The present disclosure relates to the application of mass spectrometry to identify presence of sepsis. Two-dimensional tandem mass spectrometry (“2D MS / MS” or “2DMS”) enables the collection of the fragmentation patterns and intact masses of all components in a sample in a matter of seconds using a single scan of a linear ion trap mass spectrometer and a single ion injection. A method according to the present disclosure for determining presence of sepsis in a subject comprises obtaining 2DMS data of a sample from the31602919598.1Reference No. P240021-711 -WO01 / 242278 PCT subject. The method comprises determining a concentration of at least three biomarkers from the 2DMS data. Each biomarker is individually selected from the group consisting of a peptide, an amino acid, a lipid, a metabolite, and a pathogen. The method comprises determining, using a machine learning model stored in a memory of a computing device, a presence of sepsis in the subject based on the concentrations of the at least three biomarkers. The method comprises providing an indicia indicating the presence of sepsis in the subject based on the machine learning model. The method according to the present disclosure can enable efficient sample preparation, rapid analysis, and / or the ability to detect the earliest stages of sepsis in a sample and the subject from which it originates.

[0019] Referring to FIG. 1, a method for determining sepsis in a subject is provided. The method optionally comprises obtaining a sample from a subject at step 102. The subject can be a human patient or an animal patient. The sample can be obtained at, for example, a point of care (e.g., a clinic), a bedside, and / or in an ambulatory environment. The sample may have been previously obtained or can be obtained from the subject near the time at which it is analyzed for sepsis. The sample can be a single sample or at least two samples from the subject. In various non-limiting embodiments, the sample can be taken over a period of time from the subject.

[0020] The sample can comprise a homogenous or heterogeneous mixture of chemical compounds and can be a fluid sample and / or a solid sample. For example, the sample can comprise blood, urine, saliva, breath condensate, lung condensate (e.g., from a ventilator), or a combination thereof. The sample can comprise blood, such as, for example, blood serum, blood plasma, or whole blood. The blood may be untreated or treated. For example, the blood may be provided as a liquid or the blood may be dried onto a substrate (e.g., a blood spot card). The fluid sample may be in a collection vial with or without anticoagulants, for example.

[0021] Referring to FIG. 1, the method comprises obtaining 2DMS data of the sample at step 104. Obtaining the 2DMS data can comprise receiving 2DMS data of the sample from a data store (e.g., database, cloud) and / or receiving the 2DMS data from a mass spectrometer, such as, for example, a mass spectrometer 310 as described with respect to FIGs. 3 and 4 below; a mass spectrometer as described in PCT App. Pub. No.PCT / US2023 / 079983, filed November 16, 2023, which is hereby incorporated by reference; and / or other mass spectrometer. For example, the sample can be provided to a mass spectrometer, and the sample can be analyzed using the mass spectrometer.41602919598.1Reference No. P240021-711 -WO01 / 242278 PCT

[0022] To generate the 2DMS data, the sample can be provided to a mass spectrometer and ionized, and a mass-to-charge ratio (m / z) of the ions generated from the sample can be measured. The ionization can comprise spray ionization, electron ionization, photo ionization, chemical ionization, collisionally activated disassociation, electrospray ionization (e.g., nanoelectrospray ionization), desorption electrospray ionization, and / or paper spray ionization. The sample can be introduced to the mass spectrometer, for example, by injection, or the mass spectrometer can actively sample the surrounding environment and / or otherwise receive a sample provided to it or that it encounters.

[0023] The 2DMS data may include mass-to-charge (m / z) ratios of precursor ions, m / z ratios of product ions, and intensities. The 2DMS data can be at least a two-dimensional data set with respect to mass (e.g., at least two dimensions of m / z). For example, the precursor ions can be the individual molecules in the sample modified by addition of an electrical charge. The product ions may be formed through fragmentation of the precursor ions. The m / z ratios of the ions generated may be the same or different. The 2DMS data can comprise a m / z ratio and abundance for the precursor ion, and the m / z ratios and abundance for any product ions generated from the precursor ion, whether first-generation, second-generation, or other generation.

[0024] The 2DMS data can comprises a m / z ratio of no greater than 10,000, such as, for example, no greater than 5,000, no greater than 4,000, no greater than 3,000, no greater than 2,000, or no greater than 1 ,000. The 2DMS data can comprises a m / z ratio of in a range of 1 to 10,000, such as, for example, 1 to 5,000 or 1 to 3,000.

[0025] 2DMS can enable the collection of the fragmentation patterns and intact masses of components in a sample in less than a minute (e.g., less than 10 seconds) using a single scan of a linear ion trap mass spectrometer and a single ion injection. 2DMS can increase both molecular specificity and analytical signal-to-noise ratios that can be beneficial for detecting difficult-to-identify diseases and / or preventing false positive identifications.

[0026] In certain non-limiting embodiments, precursor ions can be produced by ionizing the sample, and the sample can be unfragmented. Ionization electrically charges a molecule and thereby generates an ion from the molecule through gain or loss of one or more electrons and / or charged particles (e.g., proton, sodium ion, chloride ion) from the molecule. For example, the precursor ions can be the individual molecules in the sample modified by addition of an electrical charge.51602919598.1Reference No. P240021-711 -WO01 / 242278 PCT

[0027] The method can optionally comprise fragmenting at least a portion of the sample and / or ions. For example, fragmenting precursor ions forms first-generation product ions, fragmenting first-generation product ions forms second-generation product ions, and fragmenting second-generation product ions forms third-generation product ions. The product ions can be fragmented a number of times based on the desired application and may be fragmented to third-generation product ions or further. Fragmenting is a chemical disassociation caused by, for example, the removal of at least one electron from an ion, collision with a gas molecule and / or solid surface, electron capture or transfer, and / or ultraviolet and / or infrared photon absorption. The m / z ratios of the ions generated may be the same or different. In various non-limiting embodiments, fragmenting at least the portion of the sample can comprise at least one of in-source collision induced dissociation, beamtype collision induced dissociation, collision induced dissociation by resonance excitation, surface-induced dissociation, infrared multiphoton dissociation, ultraviolet photodissociation, electron capture dissociation, electron transfer dissociation, and electron impact dissociation.

[0028] In various non-limiting embodiments, fragmenting at least a portion of the sample and / or ions can occur prior to introducing the ions and / or sample to an ion trap of the mass spectrometer. In various other non-limiting embodiments, fragmenting at least a portion of the sample and / or ions can occur within the ion trap. In various non-limiting embodiments, fragmenting at least a portion of the sample and / or ions can occur before the ion trap in a collision device. For example, fragmenting at least the portion of the sample and / or ions can comprise beam type collision induced dissociation within the collision cell prior to the ion trap. In various non-limiting embodiments, fragmenting at least the portion of the sample and / or ions can comprise at least one of beam type collision induced dissociation within the ion trap and collision induced dissociation by resonance excitation within the ion trap.

[0029] In certain non-limiting embodiments, fragmenting at least the portion of the sample and / or ions is performed over a range of different collision energies (e.g., by varying the ion’s acceleration prior to the collision device) within the collision cell. In various nonlimiting embodiments, fragmenting at least a portion of the sample and / or ions can occur both in a collision device and within the ion trap. The ions can be fragmented based on their m / z ratio and applied electric fields such that an m / z ratio of the ion that was fragmented can be determined based on the applied electric field and, thus, based on a time of detection.

[0030] The analysis of the sample can comprise storing the ions in an ion trap of the mass spectrometer. For example, at least one of precursor ions produced from the sample, first-generation product ions produced from precursor ions, second-generation product ions61602919598.1Reference No. P240021-711 -WO01 / 242278 PCT produced from the first-generation product ions, third-generation product ions produced from the second-generation product ions, and further generation product ions can be stored in the ion trap.

[0031] The analysis can comprise applying a scan function to the ion trap. Applying the scan function can comprise exciting at least a portion of the ions stored in the ion trap selectively over time to fragment ions into product ions and / or ejecting ions from the ion trap. In various non-limiting embodiments, at least one of a precursor ion can be fragmented into first-generation product ions, a first-generation product ion can be fragmented into second- generation product ions, a second-generation product ion can be fragmented into third- generation product ions, or further generation product ion can be fragmented by the scan function. In various non-limiting embodiments, fragmenting is performed on ions with relatively low m / z ratio values first, and fragmenting is next performed on ions with successively greater m / z ratio values. In certain non-limiting embodiments, at least one of a precursor ion, a first-generation product ion, a second-generation product ion, a third- generation product ion, or a further generation product ion can be ejected from the ion trap by the scan function.

[0032] The scan function can produce one further generation of ions from ions and / or sample stored in the ion trap and eject ions from the ion trap. For example, a single scan function may fragment a first-generation product ion into a first set of second-generation product ions, but the single scan function may not further fragment the second-generation product ions. In various non-limiting embodiments, the second-generation product ions can be stored in the ion trap and fragmented into third-generation product ions. Certain scan functions can fragment ions within the ion trap and may not eject ions from the ion trap. Thus, the scan functions can be applied until a desired generation of product ions is achieved.

[0033] The scan function may comprise a radio frequency (RF) voltage (e.g., a trapping voltage), an excitation frequency, and an ejection frequency. The RF voltage, the excitation frequency, and the ejection frequency can be applied to electrodes in the ion trap to form a dynamic electric field within the ion trap configured to control (e.g., store, eject) the ions as desired. For example, the scan function can store ions based on an m / z ratio of the ions and / or eject ions based on an m / z ratio of the ions such that the ions can be sorted based on the m / z ratio of the respective ion.

[0034] The analysis of the sample can comprise detecting ions ejected from the ion trap and generating spectrum data based on the detected ions. The ions can be detected with a71602919598.1Reference No. P240021-711 -WO01 / 242278 PCT single detector or two or more detectors. For example, at least one of a precursor ion, a first-generation product ion, a second-generation product ion, a third-generation product ion, or a further generation product ion can be detected.

[0035] The detected data can be analyzed. The analysis can comprise correlating and / or combining spectrum data for precursor ions with the spectrum data for respective product ions. For example, based on the scan function, the time of detection can be correlated to an m / z ratio value and / or a generation of product (e.g., precursor, first- generation, etc.). For example, spectrum data can be generated for a precursor ion of a sample, the sample data can comprise an m / z ratio and abundance for the precursor ion, and the m / z ratio and abundance for any product ion generated from the precursor ion, whether first-generation, second-generation, or other generation.

[0036] Spectrum data can be combined from a first spectrum data, a second spectrum data, and optionally other spectrum data to form the 2DMS data. For example, first spectrum data can be generated from detecting a first composition comprising at least a portion of a first set of precursor ions produced from a sample, a first set of first-generation product ions produced from the first set of precursor ions, and a first set second-generation product ions produced from the first set of first-generation product ions. Second spectrum data can be generated from detecting a second set of precursor ions produced from the sample and a second set of first-generation product ions produced from the second set of precursor ions. The first and second spectrum data can be combined to correlate the precursor ions to the first-generation product ions and the second-generation product ions. For example, the second spectrum data may provide the identity of the precursor ions, the first spectrum data may provide the identity of the second-generation product ions, and the precursor ions can be matched with the second-generation product ions by the overlap of the first-generation product ions. The second spectrum data can be generated prior to, at least partially concurrently with, or after generating the first spectrum data.

[0037] An example of spectra data for a blood sample from a healthy patient is illustrated in FIG. 5A, and 2DMS data created from the spectra data of FIG. 5A is illustrated in FIG. 6A. An example of spectra data for a blood sample from a patient with a low degree of sepsis severity is illustrated in FIG. 5B, and 2DMS data created from the spectra data in FIG. 5B is illustrated in FIG. 6B. An example of spectra data for a blood sample from a patient with a severe degree of sepsis severity is illustrated in FIG. 5C, and 2DMS data created from the spectra data in FIG. 5C is illustrated in FIG. 6C. An example of spectra data for a blood sample from a patient with a multiple organ dysfunction syndrome (MODS)81602919598.1Reference No. P240021-711 -WO01 / 242278 PCT degree of sepsis severity is illustrated in FIG. 5D, and 2DMS data created from the spectra data in FIG. 5D is illustrated in FIG. 6D.

[0038] An example of biomarker concentrations for various degrees / levels of sepsis severity is shown in Table 1 below:91602919598.1Reference No. P240021-711 -WO01 / 242278 PCT101602919598.1Reference No. P240021-711 -WO01 / 242278 PCT

[0039] The 2DMS data can enable efficient and rapid characterization of complex mixtures on a molecular level by providing direct measurements of m / z ratios (e.g., correlating to molecular weight) of the intact ionized molecules (e.g., precursor ions). The detected product ions can be used to deduce structural information (e.g., molecular substructure) based on the fragmentation patterns. In various non-limiting embodiments, the 2DMS data can be a two-dimensional plot (e.g., an image).

[0040] Optionally, the 2DMS data can be pre-processed and / or have features extracted. For example, the method can comprise topological data analysis of the 2DMS data.Topological data analysis can analyze the data in a way that is agnostic / blind to the biomarkers that are measured. This form of analysis can enable the use of biomarkers without the exact identity of the biomarker. In various non-limiting embodiments, the data can be analyzed as an image, as an array of m / z intensities, or a combination thereof to build a model.

[0041] In certain non-limiting embodiments, the 2DMS data may be calibrated along the precursor and product m / z ratio axes. The 2DMS data may be normalized to calculate the intensity between 0 and 1. The data may also be subject to digital processing and / or smoothing, for example using a 1D or 2D Gaussian filter or a Savitzky-Golay filter. The 2DMS data may also be subjected to processing to remove noise.

[0042] Mass spectrometry can enable rapid analysis and detection of sepsis. For example, the analysis can be performed on a mass spectrometer in a time period ranging from 10 ms to 10 seconds, such as, for example, 100 milliseconds (ms) to 5 seconds, 200 ms to 2 seconds, 200 ms to 1 second, or 300 ms to 900 ms. In various non-limiting embodiments, the method can be performed with a single ion injection, and the entire ion population can be characterized (by measuring precursor m / z ratio and product m / z ratio simultaneously) in a single analysis scan. In various non-limiting embodiments, the analysis can comprise two ion injections and multiple spectrum data can be combined.

[0043] Again referring to FIG. 1, the method can comprise determining a concentration of at least three biomarkers from the 2DMS data at step 106, such as, for example, three biomarkers, at least four biomarkers, at least five biomarkers, or at least six biomarkers. 2DMS data can enable the determination of the concentration of the at least three biomarkers simultaneously. As used herein, a “concentrations” of a biomarker can be an actual concentration, a relative concentration, and / or an intensity unit in the 2DMS data.111602919598.1Reference No. P240021-711 -WO01 / 242278 PCT

[0044] Certain difficult-to-detect conditions, such as, for example, sepsis, may only have a low concentration of a particular pathogen at the outset, which may lead to false positives if only testing for that single biomarker. Monitoring the effect of sepsis on additional biomarkers (e.g., ones that are not pathogens) can enable more rapid and accurate detection. Furthermore, by utilizing 2DMS, all of the at least three biomarkers can be determined in a single sample, thereby enabling more rapid detection and enhancing efficiency.

[0045] Each biomarker can be individually selected from the group consisting of a peptide, an amino acid, a lipid (e.g., phospholipids), a metabolite, and a pathogen (e.g., bacteria, virus, microorganism). In various non-limiting embodiments, at least three biomarkers can be individually selected from the group consisting of a peptide, an amino acid, a lipid (e.g., phospholipids), and a metabolite. The amino acid can comprise Arginine, Phenylalanine, Glutamic Acid, Glycine, Histidine, 4-Hydroxyproline, Leucine, Isoleucine, Lysine, Ornithine, Threonine, Tyrosine, Asparagine, Glutamine, Homocitrulline, or a combination thereof. The lipid can comprise lipopolysaccharide, acylcarnitine, lysophosphatidylcholine, phosphatidylcholine, or a combination thereof. The metabolite can comprise ethanolamine, lactate, creatinine, glucose, cholesterol, or a combination thereof. In various non-limiting embodiments, each biomarker can be individually selected from the group consisting of Arginine, Phenylalanine, Glutamic Acid, Glycine, Histidine, 4- Hydroxyproline, Leucine, Isoleucine, Lysine, Ornithine, Threonine, Tyrosine, Asparagine, Glutamine, Ethanolamine, Homocitrulline, Lactate, Creatinine, Glucose, cholesterol, lipopolysaccharide, acylcarnitine, lysophosphatidylcholine, and phosphatidylcholine. In various non-limiting embodiments, each biomarker can be individually selected from the group consisting of Arginine, Phenylalanine, Glycine, 4-Hydroxyproline, Isoleucine, Ornithine, Asparagine, Ethanolamine, Homocitrulline, Lactate, Glucose, and lipopolysaccharide.

[0046] The biomarkers can be produced by the subject, can be produced by a pathogen, can be a pathogen itself, or a can be a combination thereof. For example, the biomarkers can be produced by the subject in response to the pathogen, an endotoxin from the pathogen, the pathogen itself, or a combination thereof.

[0047] With reference to FIG. 1, the method can comprise, at step 108, determining, using a machine learning model stored in a memory of a computing device, a presence of sepsis in the subject based on the concentrations of the at least three biomarkers. The presence of sepsis can be a binary determination of yes / no, a degree of severity of sepsis in121602919598.1Reference No. P240021-711 -WO01 / 242278 PCT the subject, or a combination of both. The degree of severity of sepsis can be, for example, non-present (e.g., healthy), present, severe, multiple organ dysfunction syndrome (MODS), or other degree. Steps 106 and 108 can be performed simultaneously and can both be performed using a machine learning model.

[0048] The machine learning model can be trained to determine the presence of sepsis. For example, the machine learning model can be selected from the group consisting of a random forest classifier, XGboost, lightGBM, K-nearest neighbors, k-d-tree, a recurrent neural network, convolutional neural network, transformer-based neural network, and spiking neural network. For example, the machine learning model can be a convolutional neural network.

[0049] The 2DMS data can be a dense collection of the fragmentation patterns and intact masses of biomarkers from the sample. Utilizing machine learning, the various biomarkers can be rapidly analyzed and used to determine the presence and / or severity of sepsis where a single biomarker may not be effective in quickly diagnosing sepsis, let alone in its early stages. Utilizing 2DMS and machine learning can enable the determination of sepsis based on less samples and can enable providing the results in less time.

[0050] Embodiments of the method can comprise providing an indicia indicating the presence and / or severity of sepsis in the subject based on the machine learning model at step 108. In various non-limiting embodiments, a predication score, a quality score, a degree of severity of sepsis, or a combination thereof can be provided as the indicia. The indicia can be presented to the user on, for example, a display device and / or stored in memory. The indicia can comprise, for example, alphanumeric text, graphics, audio, or other indicia.

[0051] In various embodiments of the method, the machine learning model can be trained. For example, referring to FIG. 2, a method of training (e.g., developing) a machine learning model for determining presence of sepsis according to a non-limiting embodiment of this disclosure is provided. The method can obtain data from a database comprising a plurality of sets of data at step 202. Each set of data may include a determine for presence of sepsis and 2DMS data corresponding to the determination. For example, each set of data comprises a determination for the presence of sepsis and concentrations of at least three biomarkers for the determination, such as, for example, at least four, at least five, or at least six biomarkers for the determination. The database may comprise a plurality of sets of data for each of the degrees of severity of sepsis. The database may be formed from the data received by, for example and without limitation, any of the mass spectrometers herein.131602919598.1Reference No. P240021-711 -WO01 / 242278 PCT

[0052] Optionally, the method comprises forming the database with 2DMS data collected for known substances using mass spectrometers. The method may optionally comprise augmenting the sets of 2DMS data in the database. For example, augmenting occurs by making copies of each 2DMS data with random intensity and mass shifts. Mass shifts are assumed to affect both the precursor and product axes, though to different extents since the scan rate for the latter typically is much higher than for the former. The augmented sets of data are added to the database. The database may comprise both augmented 2DMS data and 2DMS data from a mass spectrometer. The database may comprise, alone or in combination with the augmented 2DMS data and 2DMS from a mass spectrometer, artificial 2DMS data such as, for example, 2DMS data created using spectral libraries of known compounds. The database may include 2DMS data.

[0053] Embodiments of the method may include calibrating the 2DMS data in the database at step 204. Calibration may involve calibrating the 2DMS data along the precursor and product m / z ratio axes. For example, the method may include normalizing the 2DMS data in the database. The intensity of the precursor and product ions are between, for example, 0 and 1 after normalization. Normalization may be performed before the machine learning model is trained. Optionally, the data may be rectangularized by interpolation and resampling.

[0054] The method can comprise inputting a first portion of the data from the database into the machine learning model at step 206. The first portion of data may comprise a portion of the sets of data in the database. The first portion may comprise, for example, more than half of the sets of data, such as, for example, at least 70% of the data sets. For example, the first portion may comprise 70% of the data sets. The database, for example, may be split into a first portion and a second portion, and the first portion and the second portion may or may not be the same size.

[0055] The method can comprise training the machine learning model with the first portion of the data at step 208. The machine learning model can be trained to identify the presence of sepsis based on the concentrations of the at least three biomarkers.

[0056] The method can comprise inputting a second portion of data from the database into the machine learning model at step 210. The second portion of the data may comprise, for example, the remaining portion of data in the database. The first and second portion of the data can comprise different data sets of the data from the database.141602919598.1Reference No. P240021-711 -WO01 / 242278 PCT

[0057] The method can comprise evaluating the machine learning model using the second portion of the data at step 212. Evaluating the machine learning model may include determining, using the trained machine learning model, the identity of each sample in the second portion of the data based on the 2DMS data. The method can determine whether the determined identity of each sample matches the identity of each sample stored in the database for the corresponding 2DMS data. Based on the determined identity of the sample matching the identity of the sample in the database, the accuracy of the machine learning model can be determined. If the identity determined for the sample does not match the identity of the sample in the database, the machine learning model may need to be retrained with a different data set or with additional data sets and / or using a different machine learning model type.

[0058] In various non-limiting embodiments, the training of the machine learning model may be repeated. For example, the machine learning model may be re-trained with additional data from any of the mass spectrometers described herein. The machine learning model may be able to “learn” (i.e. , be re-trained) as new 2DMS data comes in from a mass spectrometer. The model can be re-trained, for example, onboard the mass spectrometer, or the data can be sent back to a computing device at a central location where the model can be re-trained. The re-trained model, for example, can be utilized through the cloud at the central location or the re-trained model can be pushed to the instrument (e.g., mass spectrometer) in the field.

[0059] In certain non-limiting embodiments, developing a neural network trained on 2DMS data for identifying samples may involve utilizing a plurality of 2DMS spectra per degree of severity of sepsis, including biological replicates, replicates at varying concentrations, and under varying growth conditions. The 2DMS data may also include replicates from different mass spectrometers such as the different types of mass spectrometers. Depending on the machine learning model, the data may then be calibrated and processed.

[0060] In the example of a convolutional neural network, the data can be reformed into an image, smoothed, and normalized (precursor m / z, product m / z, and intensities) between 0 and 1 prior to being sent to the machine learning model. The model can be trained on the dataset, the accuracy and loss can be calculated, and, if the accuracy is high (>95%), the model can be saved and can be used to identify sepsis from 2DMS data not yet seen by the machine learning model.151602919598.1Reference No. P240021-711 -WO01 / 242278 PCT

[0061] In various non-limiting embodiments, a convolutional neural network can be constructed using several layers of a 2D convolutional layer (Conv2d) and rectified linear activation unit (ReLU) activation functions as well as 2D max pooling over an input signal composed of several input planes (MaxPool2D) for pooling the convolution results. The final activation function can be a sigmoid, and a binary cross entropy loss function can be chosen to enable multilabel classification. The model can be trained, for example, for 30 epochs using a batch size of 20 files and a 70 / 30 data split (i.e. , 70% of the data is used to train the model and 30% of the data is used to validate the model). The model may be trained several times on different training sets to achieve different accuracies. For example, the system may vary the first portion of data and second portion of data to achieve better accuracies.

[0062] One considering the present disclosure will understand that the example convolutional neural network discussed herein is only a single example of applying machine learning to 2DMS data and that other models and methods of processing the data may be appropriate or preferred.

[0063] Referring to FIG. 3, a system 300 for determining a presence of sepsis in a subject is provided. The system 300 can comprises a control circuit 302 and a non- transitory memory 304. The control circuit 302 can comprise a microprocessor, microcontroller, or other basic computing device that incorporates the functions of a computer’s central processing unit (CPU) on an integrated circuit (e.g., a GPU).

[0064] The memory 304 can comprise primary storage (e.g., main memory that is directly accessible by a processor, such as RAM, ROM processor registers or processor cache); secondary storage (e.g., SSDs or HDDs that are not directly accessible by a processor); and / or off-line storage. The memory 304 can be onboard and / or offboard memory. For example, the memory 304 may be present in one physical device, available through the cloud / internet, or a combination thereof. In certain non-limiting embodiments, the processes described herein can be executed across multiple computer systems that are communicably connected together in a network, a computer system communicably connected to a cloud computing system configured to execute one or more of the described steps, and so on.

[0065] The memory 304 can comprise a plurality of instructions 306 and a pre-trained machine learning model 308 executable by the control circuit 302. The instructions can be configured to cause the control circuit 302 to perform the method of FIG. 1 and / or FIG. 2 as described herein.161602919598.1Reference No. P240021-711 -WO01 / 242278 PCT

[0066] In various non-limiting embodiments, the system 300 can utilize at least two machine learning models. For example, a first machine learning model 308 may be stored on the system 300, and a second machine learning model can be stored off-device. The second machine learning model can be accessible through the cloud / internet and the system can choose which model to use based on connectivity to the internet and / or quality of 2DMS data received. For example, the second machine learning model can be larger than the first machine learning model and the second machine learning model may not be able to be stored on the system 300. The second machine learning model can be based on general patient data, which may be secured, anonymized, and / or otherwise protected from unauthorized disclosure of patient data.

[0067] Optionally, the system 300 can comprise a mass spectrometer 310. The mass spectrometer 310 can be configured to measure the m / z ratios of various ions generated from a sample and provide 2DMS data to the control circuit 302.

[0068] Referring to FIG. 4, a detailed schematic view of the mass spectrometer 310 is shown. For example, the mass spectrometer 310 can comprise an ionizer 420, an ion trap 422 in ion communication with the ionizer 420, and a detector 424 in ion communication with the ion trap 422. In various non-limiting embodiments, the mass spectrometer comprises a single ion trap 422.

[0069] As used herein, “ion communication” means that the recited elements are configured with features such that ions, gas molecules, and / or the like can be transmitted between the two elements. In various non-limiting embodiments, the ion communication can be generation by control of ions between the recited elements by an electric field or by a physical structure (e.g., a tube, walls defining a bore).

[0070] The ionizer 420 can be in ion communication with the ion trap 422 via an ion conduit 428 suitable to transfer the sample and / or ions from the ionizer 420 to the ion trap 422 such that gases, vapors, particles entrained in a gas, and / or ions can be transferred from the ionizer 420 to the ion trap 422 through the ion conduit 428. The ion trap 422 can be in ion communication with the detector 424 by an ion conduit 430 suitable to transfer ions ejected from the ion trap 422 to the detector 424 and so that gases, vapors, particles entrained in a gas, and / or ions can flow from the ion trap 422 to the detector 424 through the ion conduit 430. The ion conduits 428 and 430 can be a suitable ion pathway, which may be generated by an electric field and / or comprise a physical structure.171602919598.1Reference No. P240021-711 -WO01 / 242278 PCT

[0071] The ionizer 420 can be capable to ionize a sample, thereby generating precursor ions or other generation product ions from the sample. The ionizer 420 can be capable to perform at least one of spray ionization, electron ionization, photo ionization, chemical ionization, collisionally activated disassociation, electrospray ionization (e.g., nanoelectrospray ionization), desorption electrospray ionization, paper spray ionization, and other suitable methods of ionization on the sample.

[0072] The control circuit 302 can be in signal communication with various components of the mass spectrometer 310 or the mass spectrometer 310 can comprise its own control circuit. Regardless, the control circuit 302 can be configured to control the functionality of the mass spectrometer 310. The control circuit 302 can be in signal communication with the ionizer 420, the ion trap 422, and the detector 424. For example, the control circuit 302 can communicate with the ionizer 420, the ion trap 422, and the detector 424 through physical wires and / or wireless signals. The control circuit 302 can comprise, for example, a processor operatively coupled to non-transitory memory, a DC voltage source, an AC voltage source, and a rectifier, and may comprise other hardware components. For example, the DC voltage source, AC voltage source, and rectifier may be used to generate the scan function.

[0073] The machine executable instructions 306 can include instructions for controlling the ionizer 420, ion trap 422, and detector 424 and / or for analyzing the data received from the detector 424.

[0074] The ion trap 422 can be capable to receive the sample and / or ions, ionize the sample, store the ions in the ion trap 422, excite at least a portion of the ions in the ion trap 422 selectively over time to fragment the ions, and / or eject ions from the ion trap 422. For example, the control circuit 302 can be configured to apply a scan function to the ion trap 422 to control ions within the ion trap 422. The control circuit 302 can generate and apply to the ion trap 422 an RF voltage, an excitation frequency, and / or an ejection frequency to create an electric field, such as, for example, an oscillating potential well within the ion trap 422 that can selectively store, excite, and / or eject ions. As the RF voltage, the excitation frequency, and / or the ejection frequency changes, the electric field within the ion trap 422 can change, which can affect the ions stored, excited, and / or ejected by the ion trap 422. In various non-limiting embodiments, the ion trap 422 can be capable to ionize the sample, thereby generating precursor ions from the sample within the ion trap 422.181602919598.1Reference No. P240021-711 -WO01 / 242278 PCT

[0075] In various embodiments, the ion trap 422 can be a quadrupole ion trap, such as, for example, a 3D quadrupole ion trap, a linear quadrupole ion trap, a toroidal ion trap, a cylindrical ion trap, or a rectilinear ion trap.

[0076] Referring back to FIG. 4, in various non-limiting embodiments, the ion trap 422 can be capable to fragment ions within the ion trap 422. For example, the control circuit 302 can apply a scan function to the ion trap 422 and introduce a gas that causes the ions to collide with gas molecules within the ion trap 422, thereby creating product ions. For example, the ion trap 422, as supported by the control circuit 302, can fragment a precursor ion into first-generation product ions, a first-generation product ion into second-generation product ions, a second-generation product ion into a third-generation product ions, and / or another product ion.

[0077] Referring yet again to FIG. 4, the detector 424 can be capable to detect ions ejected from the ion trap 422. For example, the detector 424 can generate data (e.g., a mass spectrum data) comprising the m / z value of the detected ions and an abundance (e.g., intensity) of the detected ions at the m / z value and / or 2DMS data. In various non-limiting embodiments, the detector 424 can comprise at least one of an electron multiplier, a Faraday cup collector, a photographic and stimulation-type detector, and other detector type.

[0078] Referring back to FIG. 4, the detector 424 can be capable to produce data based on the ions received and detected from the ion trap 422. The detector 424 can send the data to the control circuit 302. In various non-limiting embodiments, the data can be processed separately and may be combined to form 2DMS data or the data may be directly generated as 2DMS data.

[0079] The mass spectrometer 310 can optionally comprise a collision cell in fluid communication with and intermediate the ion trap 422 and the ionizer 420. The collision cell can be configured for fragmenting at least a portion of a sample and / or ions, thereby forming a first mixture of ions prior to the ion trap 422. The first mixture can comprise one or more precursor ions and first-generation product ions formed from fragmentation of the precursor ions. In various non-limiting embodiments, the collision cell can be configured for beam-type collision fragmentation.

[0080] The following numbered clauses are directed to various non-limiting embodiments according to the present disclosure:191602919598.1Reference No. P240021-711 -WO01 / 242278 PCT

[0081] Clause 1. A method for determining sepsis in a subject, the method comprising: obtaining two-dimensional tandem mass spectrometry (2DMS) data of a sample from the subject; determining a concentration of at least three biomarkers from the 2DMS data, wherein each biomarker is individually selected from the group consisting of a peptide, an amino acid, a lipid, a metabolite, and a pathogen; determining, using a machine learning model stored in a memory of a computing device, a presence of sepsis in the subject based on the concentrations of the at least three biomarkers; and providing an indicia indicating the presence of sepsis in the subject based on the machine learning model.

[0082] Clause 2. The method of clause 1 , wherein each biomarker is individually selected from the group consisting of Arginine, Phenylalanine, Glutamic Acid, Glycine, Histidine, 4-Hydroxyproline, Leucine, Isoleucine, Lysine, Ornithine, Threonine, Tyrosine, Asparagine, Glutamine, Ethanolamine, Homocitrulline, Lactate, Creatinine, Glucose, cholesterol, lipopolysaccharide, acylcarnitine, lysophosphatidylcholine, and phosphatidylcholine.

[0083] Clause 3. The method of any of clauses 1-2, wherein each biomarker is individually selected from the group consisting of Arginine, Phenylalanine, Glycine, 4- Hydroxyproline, Isoleucine, Ornithine, Asparagine, Ethanolamine, Homocitrulline, Lactate, Glucose, and lipopolysaccharide.

[0084] Clause 4. The method of any of clauses 1-3, wherein a concentration of at least four biomarkers is determined from the 2DMS data, and determining a presence of sepsis in the subject is based on the concentration of the at least four biomarkers, wherein each biomarker is individually selected from the group consisting of a peptide, an amino acid, a lipid, and a metabolite.

[0085] Clause 5. The method of any of clauses 1-4, wherein a concentration of at least five biomarkers is determined from the 2DMS data, and determining a presence of sepsis in the subject is based on the concentration of the at least five biomarkers, wherein each biomarker is individually selected from the group consisting of a peptide, an amino acid, a lipid, and a metabolite.

[0086] Clause 6. The method of any of clauses 1-5, wherein the 2DMS data comprises a mass to charge ratio of no greater than 10,000.

[0087] Clause 7. The method of any of clauses 1-6, wherein the 2DMS data comprises a mass to charge ratio of no greater than 1 ,000.201602919598.1Reference No. P240021-711 -WO01 / 242278 PCT

[0088] Clause 8. The method of any of clauses 1-7, wherein the 2DMS data comprises a precursor mass to charge ratio, a product mass to charge ratio, and an intensity.

[0089] Clause 9. The method of any of clauses 1-8, wherein the sample comprises blood, urine, saliva, breath condensate, lung condensate, or a combination thereof.

[0090] Clause 10. The method of any of clauses 1-9, wherein the sample comprises blood.

[0091] Clause 11. The method of any of clauses 1-10, wherein the indicia further comprises a predication score, a quality score, or a combination thereof.

[0092] Clause 12. The method of any of clauses 1-11 , further comprising obtaining the sample from a subject, and providing the sample to a mass spectrometer, wherein obtaining the 2DMS data of the sample comprises receiving data from the mass spectrometer.

[0093] Clause 13. The method of clause 12, wherein the indicia indicating the presence of sepsis in the subject based on the machine learning model is based on a single sample from the subject.

[0094] Clause 14. The method of any of clauses 12-13, wherein the sample is obtained from the subject at a point of care, a bedside, or in an ambulatory environment.

[0095] Clause 15. The method of any of clauses 12-14, wherein providing the sample to the mass spectrometer comprises electron ionization, photo ionization, chemical ionization, collisionally activated disassociation, electrospray ionization, desorption electrospray ionization, or paper spray ionization.

[0096] Clause 16. The method of any of clauses 1-15, wherein the machine learning model is selected from the group consisting of a random forest classifier, XGboost, lightGBM, K-nearest neighbors, k-d-tree, a recurrent neural network, convolutional neural network, transformer-based neural network, and spiking neural network.

[0097] Clause 17. The method of any of clauses 1-16, wherein the machine learning model is a convolutional neural network.

[0098] Clause 18. The method of any of clauses 1-17, further comprising: determining, using the machine learning model stored in a memory of a computing device, a degree of severity of sepsis in the subject based on the concentrations of the at least three biomarkers;211602919598.1Reference No. P240021-711 -WO01 / 242278 PCT and providing an indicia indicating the degree of severity of sepsis in the subject based on the machine learning model.

[0099] Clause 19. The method of any of clauses 1-18, further comprising training the machine learning model, wherein the training comprises: obtaining data from a database comprising a plurality of sets of data, wherein each set of data comprises a presence of sepsis and concentrations for at least three biomarkers for the presence of sepsis; inputting a first portion of the data from the database into the machine learning model; training the machine learning model with the first portion of the data, wherein the machine learning model is trained to identify the presence of sepsis based on the concentrations of the at least three biomarkers; inputting a second portion of data from the database into the machine learning model, wherein the second portion and the first portion comprise different sets of data from the database; and evaluating the machine learning model using the second portion of the data.

[0100] Clause 20. A system for determining a presence of sepsis in a subject, the system comprising: a control circuit; and a memory storing a plurality of instructions and a pre-trained machine learning model executable by the control circuit, wherein the instructions are configured to cause the control circuit to perform the method of any of clauses 1-19.

[0101] Having thus described several aspects and embodiments of the technology of this application, it is to be appreciated that various alterations, modifications, and improvements will readily occur to those of ordinary skill in the art. Such alterations, modifications, and improvements are intended to be within the scope of the technology described in the application. It is, therefore, to be understood that the foregoing embodiments are presented by way of example only and that, within the scope of the appended claims and equivalents thereto, inventive embodiments may be practiced otherwise than as specifically described. In addition, any combinations of two or more features, systems, articles, materials, and / or methods described herein, if such features, systems, articles, materials, and / or methods are not mutually inconsistent, are included within the scope of the present disclosure.

[0102] In certain embodiments, a processor may be a physical or virtual processor. In other embodiments, a virtual processor may be spread across one or more portions of one or more physical processors. In certain embodiments, one or more of the embodiments described herein may be embodied in hardware such as a Digital Signal Processor (DSP). In certain embodiments, one or more of the embodiments herein may be executed on a221602919598.1Reference No. P240021-711 -WO01 / 242278 PCTDSP. One or more of the embodiments herein may be programmed into a DSP. In some embodiments, a DSP may have one or more processors and one or more memories. In certain embodiments, a DSP may have one or more computer readable storages. In many embodiments, a DSP may be a custom designed ASIC chip. In other embodiments, one or more of the embodiments stored on a computer readable medium may be loaded into a processor and executed.

[0103] Also, as described, some aspects may be embodied as one or more methods. The acts performed as part of the method may be ordered in any suitable way. Accordingly, embodiments may be constructed in which acts are performed in an order different than illustrated, which may include performing some acts simultaneously, even though shown as sequential acts in illustrative embodiments.

[0104] The phrase “and / or”, as used herein in the specification and in the claims, should be understood to mean “either or both” of the elements so conjoined, i.e., elements that are conjunctively present in some cases and disjunctively present in other cases.

[0105] As used herein in the specification and in the claims, the phrase “at least one”, in reference to a list of one or more elements, should be understood to mean at least one element selected from any one or more of the elements in the list of elements, but not necessarily including at least one of each and every element specifically listed within the list of elements and not excluding any combinations of elements in the list of elements. This definition also allows that elements may optionally be present other than the elements specifically identified within the list of elements to which the phrase “at least one” refers, whether related or unrelated to those elements specifically identified.

[0106] In the claims, as well as in the specification above, all transitional phrases such as “comprising,” “including,” “carrying,” “having,” “containing,” “involving,” “holding,” “composed of,” and the like are to be understood to be open-ended, i.e., to mean including but not limited to. The transitional phrases “consisting of” and “consisting essentially of” shall be closed or semi-closed transitional phrases, respectively.

[0107] Where a range or list of values is provided, each intervening value between the upper and lower limits of that range or list of values is individually contemplated and is encompassed within the disclosure as if each value were specifically enumerated herein. In addition, smaller ranges between and including the upper and lower limits of a given range are contemplated and encompassed within the disclosure. The listing of exemplary values231602919598.1Reference No. P240021-711 -WO01 / 242278 PCT or ranges is not a disclaimer of other values or ranges between and including the upper and lower limits of a given range.

[0108] Embodiments disclosed herein may be embodied as a system, method, or computer program product. Accordingly, embodiments may take the form of an entirely hardware embodiment, an entirely software embodiment (including firmware, resident software, micro-code, etc.), or an embodiment combining software and hardware aspects that may all generally be referred to herein as a "circuit", "module", or "system".Furthermore, embodiments may take the form of a computer program product embodied in one or more computer readable medium(s) having computer readable program code embodied thereon.241602919598.1

Claims

1. Reference No. P240021-711 -WO01 / 242278 PCTCLAIMSWhat is claimed is:

1. A method for determining presence of sepsis in a subject, the method comprising: obtaining two-dimensional tandem mass spectrometry (2DMS) data of a sample from the subject; determining a concentration of at least three biomarkers from the 2DMS data, wherein each biomarker is individually selected from the group consisting of a peptide, an amino acid, a lipid, a metabolite, and a pathogen; determining, using a machine learning model stored in a memory of a computing device, a presence of sepsis in the subject based on the concentrations of the at least three biomarkers; and providing an indicia indicating the presence of sepsis in the subject based on the machine learning model.

2. The method of claim 1, wherein each biomarker is individually selected from the group consisting of Arginine, Phenylalanine, Glutamic Acid, Glycine, Histidine, 4- Hydroxyproline, Leucine, Isoleucine, Lysine, Ornithine, Threonine, Tyrosine, Asparagine, Glutamine, Ethanolamine, Homocitrulline, Lactate, Creatinine, Glucose, cholesterol, lipopolysaccharide, acylcarnitine, lysophosphatidylcholine, and phosphatidylcholine.

3. The method of claim 1, wherein each biomarker is individually selected from the group consisting of Arginine, Phenylalanine, Glycine, 4-Hydroxproline, Isoleucine, Ornithine, Asparagine, Ethanolamine, Homocitrulline, Lactate, Glucose, and lipopolysaccharide.

4. The method of claim 1 , wherein a concentration of at least four biomarkers is determined from the 2DMS data and determining a presence of sepsis in the subject is based on the concentration of the at least four biomarkers, wherein each biomarker is individually selected from the group consisting of a peptide, an amino acid, a lipid, and a metabolite.

5. The method of claim 1, wherein a concentration of at least five biomarkers is determined from the 2DMS data and determining a presence of sepsis in the subject is based on the concentration of the at least five biomarkers, wherein each biomarker is251602919598.1Reference No. P240021-711 -WO01 / 242278 PCT individually selected from the group consisting of a peptide, an amino acid, a lipid, and a metabolite.

6. The method of claim 1, wherein the 2DMS data comprises a mass to charge ratio of no greater than 10,000.

7. The method of claim 1, wherein the 2DMS data comprises a mass to charge ratio of no greater than 1,000.

8. The method of claim 1, wherein the 2DMS data comprises: a precursor mass to charge ratio; a product mass to charge ratio; and an intensity.

9. The method of claim 1, wherein the sample comprises a material selected from the group consisting of blood, urine, saliva, breath condensate, lung condensate, and a combination thereof.

10. The method of claim 1, wherein the sample comprises blood.

11. The method of claim 1 , wherein the indicia further comprises a value selected from the group consisting of a predication score, a quality score, and a combination thereof.

12. The method of claim 1 , further comprising: obtaining the sample from a subject; and providing the sample to a mass spectrometer, wherein obtaining the 2DMS data of the sample comprises receiving data from the mass spectrometer.

13. The method of claim 12, wherein the indicia indicating the presence of sepsis in the subject based on the machine learning model is based on a single sample from the subject.

14. The method of claim 12, wherein the sample is obtained from the subject at a point of care, a bedside, or in an ambulatory environment.261602919598.1Reference No. P240021-711 -WO01 / 242278 PCT15. The method of claim 12, wherein providing the sample to the mass spectrometer comprises a technique selected from the group consisting of electron ionization, photo ionization, chemical ionization, collisionally activated disassociation, electrospray ionization, desorption electrospray ionization, and paper spray ionization.

16. The method of claim 1, wherein the machine learning model is selected from the group consisting of a random forest classifier, XGboost, lightGBM, K-nearest neighbors, k-d- tree, a recurrent neural network, convolutional neural network, transformer-based neural network, and spiking neural network.

17. The method of claim 1 , wherein the machine learning model is a convolutional neural network.

18. The method of claim 1, further comprising: determining, using the machine learning model stored in a memory of a computing device, a degree of severity of sepsis in the subject based on the concentrations of the at least three biomarkers; and providing an indicia indicating the degree of severity of sepsis in the subject based on the machine learning model.

19. The method of claim 1, further comprising: training the machine learning model, wherein the training comprises: obtaining data from a database comprising a plurality of sets of data, wherein each set of data comprises a presence of sepsis and concentrations for at least three biomarkers for the presence of sepsis; inputting a first portion of the data from the database into the machine learning model; training the machine learning model with the first portion of the data, wherein the machine learning model is trained to identify the presence of sepsis based on the concentrations of the at least three biomarkers; inputting a second portion of data from the database into the machine learning model, wherein the second portion and the first portion comprise different sets of data from the database; and evaluating the machine learning model using the second portion of the data.271602919598.1Reference No. P240021-711 -WO01 / 242278 PCT20. An system for determining a presence of sepsis in a subject, the system comprising: a control circuit; and a memory storing a plurality of instructions and a pre-trained machine learning model executable by the control circuit, wherein the instructions are configured to cause the control circuit to perform the method of claim 1.281602919598.1

Images