Method for determining antibiotic susceptibility using stimulated Raman metabolic imaging

Abstract

A method for determining antibiotic susceptibility using stimulated Raman scattering microscopy is disclosed. The method utilizes an imaging device configured to collect laser signals from a sample having bacteria to image the metabolism of the sample. The sample can be treated with an antibiotic for imaging to determine the susceptibility of the bacteria to the given antibiotic.

Description

[Technical Field]

[0001] CROSS-REFERENCE TO RELATED APPLICATIONS This application claims priority to U.S. Provisional Application No. 62 / 556,013, filed September 8, 2017, entitled "METHOD FOR THE DETERMINATION OF ANTIBIOTIC SUSCEPTIBILITY THROUGH STIMULATED RAMAN METABOLIC IMAGING," the disclosure of which is considered part of the disclosure of this application and is incorporated herein by reference in its entirety.

[0002] The present invention relates generally to a method for detecting metabolic activity of viable bacteria by hyperspectral stimulated Raman scattering (SRS) microscopy for antibiotic susceptibility testing (AST). [Background technology]

[0003] The widespread misuse and overuse of antibiotics to combat various bacterial infections has led to an increase in the number of resistant bacteria. Therefore, to treat specific infections, it is important to profile the antibiotic response of a given bacterium through antibiotic susceptibility testing (AST). Traditional methods for AST generally utilize agar plates and broth dilution assays to grow cultures, which typically require at least 16 to 24 hours to complete, depending on the bacterial species. Although newer technologies have been developed for more rapid testing, these techniques still have limitations due to their applicability to only specific bacteria or the need for time-consuming sample concentration.

[0004] Raman spectroscopy, a label-free technique that measures molecular vibrations, has been used for rapid bacterial AST. Raman peaks in the fingerprint region or metabolic uptake of heavy water (DO) have been used as biomarkers to characterize bacterial responses to antibiotic treatment. However, Raman spectroscopy produces very weak signals due to its small cross-sectional area, thus requiring long integration times for large samples or cells. The Raman signal can be enhanced by surface-enhanced Raman spectroscopy (SERS) using metal colloids or rough metal surfaces. SERS has been used for rapid AST based on changes in Raman peaks upon antibiotic treatment, but SERS requires a substrate and the signal fluctuations are too large to be practical.

[0005] Coherent Raman scattering microscopy, such as coherent anti-Stokes Raman scattering (CARS) and stimulated Raman scattering (SRS), significantly improves signal quality over spontaneous Raman scattering. Unlike CARS, SRS microscopy does not suffer from nonresonant background and has been used for metabolic imaging of cells, tissues, and model organisms. However, SRS is not completely background-free. Hyperspectral SRS, which records a spectrum at each pixel, has been developed to distinguish the SRS signal from background caused by nonlinear absorption and cross-phase modulation.

[0006] Imaging the metabolic activity of single bacteria remains extremely challenging. The size of bacteria (approximately 1 μm in diameter) is much smaller than that of mammalian cells (approximately 10 μm), approaching the spatial resolution of CARS or SRS microscopy. Furthermore, the CD-Raman signal is much weaker than the CH signal. Therefore, improved methods for AST are needed. Preferably, this method can be performed without time extension for sample concentration and is applicable to various bacteria at the single-cell level. Summary of the Invention

[0007] In one aspect, the present disclosure relates to a rapid antibiotic susceptibility testing (AST) method applicable to all bacteria. Although overlapping with other descriptions, the present invention and its preferred embodiments are described below, however, the present invention is not limited to the following. [1] 1. A method for determining antibiotic susceptibility of bacteria, comprising: Preparing a bacteria-immobilized gel; Cultivating the bacteria in a medium containing a nutrient source; adding a preselected antibiotic to said medium to form a sample; culturing the antibiotic and bacteria in the medium for a preselected period of time; centrifuging the sample; washing the sample; depositing the sample on the cell-immobilizing gel; collecting images of the sample on the bacteria-immobilized gel by coherent Raman microscopy; A method comprising: [2] The method of [1], wherein the image of the sample is collected by hyperspectral coherent Raman microscopy. [3] The method of [1], wherein the image of the sample is collected by single-frequency coherent Raman microscopy. [4] The method according to [1], wherein the immobilizing gel comprises an agarose gel. [5] The nutrient source is 0.1 to 10% glucose-d 7 The method according to [1], wherein the carbon source is composed solely of [6] The method according to [1], wherein the nutrient source is composed of a medium containing 0 to 100% deuterium dioxide. [7] The method according to [1], wherein the carbon source consists of a combination of both glucose-d7 and deuterium dioxide. [8] The method according to [1], wherein the Raman imaging can be stimulated Raman scattering imaging, coherent anti-Stokes Raman scattering imaging, or coherent Raman stimulated Kerr effect imaging. [9] 1. A hyperspectral stimulated Raman scattering imaging device for collecting images for antibiotic susceptibility testing, comprising: A dual output femtosecond pulse laser with a repetition rate of 80 MHz, a pump beam; Stokes beam and a path of the Stokes beam modulated by an acousto-optic modulator; a pump beam path having a translation stage for adjusting the delay between the pump beam and the Stokes beam; a combiner configured to combine the pump beam and the Stokes beam; a dual output femtosecond pulse layer including: a laser scanning microscope having an objective configured to focus the pump beam and the Stokes beam on a sample; and an oil condenser configured to collect the laser from the sample; a filter positioned after the condenser and configured to filter out the Stokes beam; a photodiode positioned after the filter to detect the pump beam; and a lock-in amplifier configured to extract the pump beam signal loss; and 1. An imaging device comprising:

[10] [9] The imaging apparatus of [9], further comprising a chirp device, the chirp device configured to form different pulse durations between the Stokes beam and the pump beam.

[11]

[10] The imaging device according to

[10] , wherein the chirp device is two SF57 glass rods, each 15 cm long.

[12] [9] The imaging apparatus according to [9], wherein a pair of filters is used.

[13] The imaging device according to [9], wherein the laser scanning microscope objective lens is a 60x water objective lens.

[14] [9] The imaging device according to [9], wherein the pulse duration of the pump beam is between 1.5 and 2 ps, and the pulse duration of the Stokes beam is between 1.0 and 1.4 ps.

[15] [9] The imaging apparatus of [9], wherein the chirp device is configured to form a pulse duration of 1.9 ps for the pump beam and 1.3 ps for the Stokes beam.

[16] [9] An imaging device according to [9], wherein the pump beam is a 120 fs wavelength tunable laser and has a pump beam path, and the Stokes beam is a 220 fs laser with a center wavelength of 1040 nm and has a Stokes beam path.

[17] 1. A method for rapid determination of bacterial antibiotic susceptibility using microscopy, comprising: Cultivating the bacteria in a medium containing a nutrient source; adding a preselected antibiotic having a preselected concentration to said medium to form a sample; Incubating the sample in the medium for a preselected period of time; centrifuging the sample; washing the sample; depositing the sample onto the bacteria-immobilizing gel; imaging the sample on the bacteria-immobilized gel using coherent Raman microscopy; A method comprising:

[18] The method of

[17] , wherein the coherent Raman microscopy comprises a stimulated Raman scattering system.

[19] The method of

[18] , wherein the stimulated Raman scattering system uses a pump beam and a Stokes beam, both beams being chirped with one or more glass rods, thereby creating a pulse duration of 1-2 ps for both the pump beam and the Stokes beam.

[20] 19. The method of claim 19, wherein the microscopy system further comprises a translation stage for adjusting the delay between the pump beam and the Stokes beam.

[0008] In another aspect, the present disclosure relates to a method for monitoring metabolic activity of glucose-d7 in bacteria at the single-cell level using hyperspectral stimulated Raman scattering (SRS) imaging.

[0009] In another aspect, the present disclosure relates to a rapid AST method generally applicable to all bacteria by using hyperspectral SRS imaging to monitor the metabolic activity of glucose-d7 in live bacteria at the single-cell level. Other nutrient sources, such as DO, can be used in this method as well. Using vancomycin-susceptible (VSE) and -resistant (VRE) enterococci as a model, we demonstrate that SRS imaging can detect bacteria at the single-cell level and quantitatively monitor their metabolic activity. By monitoring the metabolic response to antibiotic treatment, bacterial susceptibility and minimum inhibitory concentration (MIC) can be determined within 30 minutes. Similarly, the imaging method can be applied to other antibiotics, regardless of the mechanism by which they inhibit or kill the bacteria. Glucose is a common carbon source for bacterial growth, and therefore, it is reasonable to expect that our method will be generally applicable to many bacterial species.

[0010] In another aspect, the present disclosure relates to sample preparation strategies, improved bacterial culture media methods, and imaging settings configured to maximize signal-to-noise ratios for more consistent imaging.

[0011] In another aspect, the present disclosure relates to a Raman imaging system for rapid AST testing, the Raman imaging system including a dual-output femtosecond pulse laser. The dual-output femtosecond pulse layer can include a pump beam and a Stokes beam. The Stokes beam path is modulated by an acousto-optic modulator. The pump beam path has a translation stage for adjusting the delay between the pump beam and the Stokes beam. A combiner is configured to combine the pump beam and the Stokes beam. A chirping device is configured to create different pulse durations between the Stokes beam and the pump beam. The laser scanning microscope can include an objective lens configured to focus the pump beam and the Stokes beam on the sample. Further, the system can include an oil condenser configured to collect the laser from the sample. The system can use one or more filters positioned after the condenser and configured to remove the Stokes beam. A photodiode can be positioned after the filter to detect the pump beam. A lock-in amplifier can be configured to extract signal loss from the pump beam.

[0012] The features and advantages of the present disclosure, and the manner in which they are achieved, will become more apparent and better understood by reference to the following description of the disclosed systems and processes taken in conjunction with the accompanying drawings. [Brief explanation of the drawings]

[0013] [Figure 1A] FIG. 1 is a diagram of an exemplary hyperspectral SRS imaging setup in accordance with the present disclosure.

[0014] [Figure 1B] FIG. 10 is a diagram of the delay between the pump and Stokes after chirp according to the present disclosure.

[0015] [Figure 2A] 1 illustrates the process of an AST procedure according to the present disclosure.

[0016] [Figure 2B] 1 shows SRS imaging data in the CD vibration region (2178 cm −1 ) of vancomycin-susceptible and -resistant E. faecalis with and without 20 μg / ml vancomycin treatment, according to the present disclosure.

[0017] [Figure 2C] 1 shows SRS spectral data of susceptible bacteria according to the present disclosure.

[0018] [Figure 2D] 1 shows SRS spectral data of resistant bacteria according to the present disclosure.

[0019] 3A-3D show multivariate curve resolution (MCR) analysis data of hyperspectral SRS imaging data to separate bacterial CD components from the gel background, according to the present disclosure. [Figure 3A] This is an SRS image of bacteria deposited on an agarose gel in the CD vibration region. [Figure 3B] 10 is an MCR output spectrum of glucose-d7 and gel components. [Figure 3C] This is the gel component after MCR. [Figure 3D] This is the glucose-d7 component after MCR.

[0020] [Figure 4A] The results of cross-phase modulation noise in SRS imaging are shown. [Figure 4B] The results of cross-phase modulation noise in SRS imaging are shown.

[0021] [Figure 5A] The results of MIC determination using SRS metabolic imaging are shown. [Figure 5B]The results of MIC determination using SRS metabolic imaging are shown.

[0022] [Figure 6A] 1 shows the results of time-lapse photography of CD component changes in bacteria grown in glucose-d7 containing medium in the presence and absence of 20 μg / ml vancomycin. [Figure 6B] 1 shows the results of time-lapse photography of CD component changes in bacteria grown in glucose-d7 containing medium in the presence and absence of 20 μg / ml vancomycin. [Figure 6C] 1 shows the results of time-lapse photography of CD component changes in bacteria grown in glucose-d7 containing medium in the presence and absence of 20 μg / ml vancomycin. [Figure 6D] 1 shows the results of time-lapse photography of CD component changes in bacteria grown in glucose-d7 containing medium in the presence and absence of 20 μg / ml vancomycin.

[0023] [Figure 7A] Results of hyperspectral SRS imaging at 2150 cm-1 are shown.

[0024] [Figure 7B] FIG. 1 is a graphical representation of the corresponding spectrum of a 1 M glucose-d7 solution in the CD region.

[0025] [Figure 8] The results of spontaneous Raman spectra of VSE bacteria cultured in normal medium and in medium containing 2% glucose-d7 are shown.

[0026] [Figure 9A] FIG. 1 is a graphical representation of bacterial growth measurements of glucose-d7 against E. faecalis. [Figure 9B] FIG. 1 is a graphical representation of bacterial growth measurements of glucose-d7 against E. faecalis.

[0027] [Figure 10A] 1 shows SRS metabolic imaging results of antibiotic susceptibility testing of E. faecalis 31970 to various antibiotics. [Figure 10B] 1 shows SRS metabolic imaging results of antibiotic susceptibility testing of E. faecalis 31970 to various antibiotics. [Figure 10C] 1 shows SRS metabolic imaging results of antibiotic susceptibility testing of E. faecalis 31970 to various antibiotics. [Figure 10D] 1 shows SRS metabolic imaging results of antibiotic susceptibility testing of E. faecalis 31970 to various antibiotics. [Figure 10E] 1 shows SRS metabolic imaging results of antibiotic susceptibility testing of E. faecalis 31970 to various antibiotics. [Figure 10F] 1 shows SRS metabolic imaging results of antibiotic susceptibility testing of E. faecalis 31970 to various antibiotics.

[0028] [Figure 11A] 1 shows the results of differentiation of oxacillin-susceptible and resistant S. aureus at 0.5 hours. [Figure 11B] 1 shows the results of differentiation of oxacillin-susceptible and resistant S. aureus at 0.5 hours. [Figure 11C] 1 shows the results of differentiation of oxacillin-susceptible and resistant S. aureus at 0.5 hours. [Figure 11D] 1 shows the results of differentiation of oxacillin-susceptible and resistant S. aureus at 0.5 hours.

[0029] [Figure 12]SRS imaging results and corresponding spectral data are shown for bacteria grown in glucose-d7 containing medium for 1 hour.

[0030] [Figure 13A] The results of SRS imaging of bacteria cultured in glucose-d7 containing medium for 3 hours are shown. [Figure 13B] The corresponding spectral data for bacteria grown for 3 hours in glucose-d7 containing medium are shown. [Figure 13C] The corresponding spectral data for bacteria grown for 3 hours in glucose-d7 containing medium are shown.

[0031] [Figure 14] 1 illustrates a method for an AST procedure according to the present disclosure.

[0032] [Figure 15A] The results of spontaneous Raman spectra of large amounts of P. aeruginosa cultured for 2 hours in normal medium and LB medium containing 70% DO are shown.

[0033] [Figure 15B] CD SRS imaging results at 2162 cm-1 and corresponding transmission imaging results of P. aeruginosa cultured for 2 hours in normal medium and LB medium containing 70% DO are shown.

[0034] [Figure 15C] The corresponding SRS spectrum of the single bacterium in Figure 15B is shown.

[0035] [Figure 16A] Picosecond SRS imaging of P. aeruginosa grown in LB medium containing 70% DO for 30 min at 2162 cm-1 after chirping with two glass SF57 rods is shown.

[0036] [Figure 16B]The dashed line in the bacterial intensity plot of Figure 16A.

[0037] [Figure 16C] The results of chirp-free picosecond SRS imaging at 2162 cm-1 of P. aeruginosa grown in LB medium containing 70% DO for 30 min are shown.

[0038] [Figure 16D] The dashed line in the bacterial intensity plot of Figure 16C.

[0039] [Figure 17A] This shows the results of CD SRS imaging of P. aeruginosa cultured in LB medium containing 70% DO for 0 to 3 hours.

[0040] [Figure 17B] A magnified image of Figure 17A is shown.

[0041] [Figure 17C] The dynamics of the mean SRS intensity for each P. aeruginosa strain are shown in Figure 17A, with error bars indicating standard deviation (N>10).

[0042] [Figure 17D] The dashed line is the bacterial intensity plot from the 10 minute results in Figure 17B.

[0043] [Figure 17E] The dashed line is the bacterial intensity plot from the 30 minute results in Figure 17B.

[0044] [Figure 18A] This figure shows the results of time-lapse imaging of DO biosynthesis in P. aeruginosa using CD SRS imaging of P. aeruginosa grown in LB medium containing 70% DO from 0 to 3 hours.

[0045] [Figure 18B] 18B is a magnified image of each P. aeruginosa within the square.

[0046] [Figure 18C] FIG. 18A is a graphical representation of the dynamics of the mean SRS intensity for each P. aeruginosa, with error bars indicating standard deviation (N>10).

[0047] [Figure 18D] The dashed line is the bacterial intensity plot from the 10 minute results in Figure 18B.

[0048] [Figure 18E] The dashed line is the bacterial intensity plot from the 30 minute results in Figure 18B.

[0049] [Figure 19A] The results of CD SRS imaging and transmission imaging at 2162 cm-1 of P. aeruginosa cultured in LB medium containing 70% DO and supplemented with 20 μg / ml of gentamicin are shown.

[0050] [Figure 19B] The results of CD SRS imaging and transmission imaging at 2162 cm-1 of P. aeruginosa cultured in LB medium containing 70% DO with the addition of 20 μg / ml cefotaxime are shown.

[0051] [Figure 19C] Plot of the mean CD SRS intensity of P. aeruginosa grown in LB medium containing 70% DO without antibiotic treatment (control (Ctrl)) and after treatment with gentamicin (Gen) or cefotaxime (Cef) for various times. [Figure 19D]Plot of mean CD SRS intensity of P. aeruginosa grown in LB medium containing 70% DO without antibiotic treatment (control) and with treatment with gentamicin or cefotaxime for various times. [Figure 19E] Plot of mean CD SRS intensity of P. aeruginosa grown in LB medium containing 70% DO without antibiotic treatment (control) and with treatment with gentamicin or cefotaxime for various times. [Figure 19F] Plot of mean CD SRS intensity of P. aeruginosa grown in LB medium containing 70% DO without antibiotic treatment (control) and with treatment with gentamicin or cefotaxime for various times. [Figure 19G] Plot of mean CD SRS intensity of P. aeruginosa grown in LB medium containing 70% DO without antibiotic treatment (control) and with treatment with gentamicin or cefotaxime for various times. [Figure 19H] Plot of mean CD SRS intensity of P. aeruginosa grown in LB medium containing 70% DO without antibiotic treatment (control) and with treatment with gentamicin or cefotaxime for various times.

[0052] [Figure 20A] The results of CD SRS imaging and transmission imaging at 2162 cm-1 are shown for P. aeruginosa cultured in normal LB medium (control) or LB medium containing 70% DO supplemented with various concentrations of gentamicin.

[0053] [Figure 20B] 19A is a plot of the mean SRS intensity of each P. aeruginosa. DETAILED DESCRIPTION OF THE INVENTION

[0054] The detailed description of various embodiments herein refers to the accompanying drawings, which illustrate various embodiments and their implementation by way of example and best mode, not limitation. While these embodiments have been described in sufficient detail to enable those skilled in the art to practice the embodiments, it should be understood that other embodiments may be realized and that mechanical and other changes may be made without departing from the spirit and scope of the present disclosure. Furthermore, references to the singular include plural embodiments, and any reference to plural components may include singular embodiments. Similarly, the order of designations of devices or device components or portions of devices, e.g., "first" and "second," etc., is for convenience and clarity and should not be construed as limiting or indicating any distinction beyond its scope. Furthermore, the recitation of multiple embodiments having stated features is not intended to exclude other embodiments having additional features or other embodiments incorporating different combinations of the stated features.

[0055] Various embodiments of the present disclosure provide systems, methods, and computer program products. References to "various embodiments," "one embodiment," "embodiments," "example embodiments," and the like indicate that the described embodiments may include particular features, structures, or characteristics, but not all embodiments necessarily include the particular features, structures, or characteristics. Moreover, such phrases do not necessarily refer to the same embodiment. Furthermore, when a particular feature, structure, or characteristic is described in connection with an embodiment, it is submitted that the impact of such feature, structure, or characteristic on other embodiments is within the knowledge of one of ordinary skill in the art, whether or not explicitly described. After reading the description, it will be apparent to one of ordinary skill in the art how to implement the present disclosure in alternative embodiments.

[0056] Reference is now made to FIGS. 1A-1B of a method for determining antibiotic susceptibility by Raman metabolic imaging. The imaging system may include any type of Raman imaging, including, but not limited to, stimulated Raman scattering imaging, coherent anti-Stokes Raman scattering imaging, or coherent Raman induced Kerr effect imaging. In an exemplary embodiment, the present disclosure discloses a hyperspectral stimulated Raman scattering (SRS) imaging setup. Within the imaging system of the present disclosure, the system may include a light source 101. In an exemplary embodiment, the light source may be a dual-output femtosecond pulsed laser (InSight) that may have a repetition rate of approximately 80 MHz. The light source 101 may be a tunable laser (DeepSee, Spectra-Physics), although any suitable repetition rate may be used. The light source 101 may be used for hyperspectral SRS imaging. Similarly, the imaging system may use monochromatic SRS imaging. Imaging in the CD vibration region may be performed using the light source 101, such as a tunable laser 103. In an exemplary embodiment, the tunable laser may be a 120 fs tunable laser tuned to approximately 847 nm. Additionally, filters may be incorporated to sharpen the laser beam. This beam 103 may function as a pump beam. The center of a second light beam 105, such as a 220 fs laser, The wavelength is approximately 1040 nm. The second beam 105 can function as a Stokes beam, which can be modulated by an acousto-optic modulator 107 (AOM, 1205-C, Isomet) at 2.35 MHz. Each beam can pass through one or more filters or lenses. Similarly, each beam can pass through a polarizing beam splitter (PBS) 133. In an exemplary embodiment, a motorized translation stage 109 (T-LS28E, Zaber) can be installed in the pump path to adjust the delay between the pump beam 103 and the Stokes beam 105. A combiner 111 can then be used to combine the two beams. One or more glass rods 113 can then chirp both beams.In one exemplary embodiment, two glass rods of about 15 cm in length can be used. Similarly, the glass rods can be SF57 glass rods.

[0057] After chirping, the pump and Stokes pulse durations in some exemplary embodiments can be between approximately 1 and 2 ps, or between 1.3 and 1.9 ps. In one exemplary embodiment, the pump and Stokes beams can be approximately 1.9 ps and 1.3 ps, respectively. In other embodiments, the imaging system may not use beam chirp. The pump beam 103 and Stokes beam 105 may then be directed to a laser scanning microscope having a scanning device 115 and one or more lenses 117. In one exemplary embodiment, lens 117 can be an approximately 60x water objective (NA = 1.2, UPlanApo / IR, Olympus) that can be used to focus the laser onto the sample 119. An oil condenser 121 (NA = 1.4, U-AAC, Olympus) can be used to collect the combined laser beam 120 from the sample 123. One or more filters 125 (HQ825 / 150m, Chroma) can be used to remove the Stokes beam, the pump beam can be detected with a photodiode 127 (S3994-01, Hamamatsu), and the pump beam loss signal can be extracted with a lock-in amplifier (HF2LI, Zurich Instrument). The system can include one or more mirrors to direct each beam throughout various components of the apparatus. In some exemplary embodiments, one or more mirrors 129 can be dichroic mirrors 131. As shown in FIG. 1B, Δω1 and Δω2 indicate possible Raman peaks that can be measured by the exemplary imaging system of the present disclosure. Each delay between the pump and Stokes after chirp corresponds to a molecular vibrational mode detected by the imaging system.

[0058] The present disclosure also relates to exemplary sample preparation methods for use with exemplary imaging systems, including but not limited to SRS imaging systems. The sample preparation methods can utilize a bacterial immobilization pad. In one exemplary embodiment, the immobilization pad can include an agarose gel pad for placing bacteria in solution onto the gel pad for live imaging. In some exemplary embodiments, an agarose gel pad can be prepared by the following steps: (1) adding 1 wt% agarose powder to approximately 2 mL of purified water in a plastic tube; (2) heating the tube in a microwave for approximately 20 seconds until the agarose powder is completely dissolved; (3) adding approximately 10 μL of the heated agarose gel solution to a cover glass with a pipette; (4) immediately placing a second cover glass on top of the agarose gel solution and flattening it, typically sandwiching the gel solution between the first and second cover glasses; (5) after approximately 2 minutes, sliding the two cover glasses to remove one of the cover glasses from the agarose gel; and (6) forming a pad out of the solid agarose gel remaining on one of the cover glasses.

[0059] The disclosed AST procedure uses bacterial AST to measure bacterial metabolic activity within one cell cycle (approximately 30 minutes) by Raman imaging, such as stimulated Raman scattering (SRS) microscopy. The imaging can monitor the metabolic activity of deuterated glucose, such as glucose-d7, with picosecond SRS chirped at the CD vibration frequency, and the CD signal is used as a marker to perform AST.

[0060] Similarly, water can be used for bacterial biomolecular synthesis, and its metabolism can be monitored by Raman spectroscopy using heavy water (DO) at CD frequencies. Unlike glucose-d7, DO itself does not have CD bonds, and as a result, DO metabolic activity provides a better control for rapid AST than glucose-d7. DO metabolic activity has been used to image metabolically active bacteria with spontaneous Raman spectroscopy. SRS imaging has several orders of magnitude of signal enhancement, thereby enabling fast imaging. Furthermore, SRS imaging of DO metabolism is a noninvasive and accurate method for visualizing metabolic dynamics in mammalian cells and animals. SRS imaging based on the disclosed method can determine bacterial susceptibility to antibiotics within approximately 10 minutes. Bacterial solutions for SRS imaging and placement on agarose gel pads can be prepared by culturing bacteria in broth, such as lysogeny broth (LB) (LB broth, Sigma-Aldrich), for approximately 2 hours to reach logarithmic phase. The bacteria can then be added to 2 mL of various types of media at a ratio of approximately 0.1:100 to approximately 10:100. In an exemplary embodiment, the ratio can be approximately 1:100 to 2 mL of media, such as specialized media. Some exemplary embodiments can use DO media or other specialized media, referred to herein as M9 media, where approximately 2% glucose-d7 can be the only carbon source. M9 media was developed and selected to maximize the signal within the CD oscillation region. To measure the resistance of a given bacterium to an antibiotic, the antibiotic is added to the specialized media at a concentration of approximately 20 μg / mL. After at least 30 minutes of incubation, approximately 500 µL of sample is centrifuged, washed twice with purified water, and placed on an agarose gel pad for imaging. The imaging setup shown in Figure 1A can be used to process the sample.

[0061] Referring now to Figures 2A-2D, the disclosed method can be generally implemented in an exemplary manner to determine the effectiveness of antibiotics, such as vancomycin. Thus, this method can be used to examine the effect of antibiotics on the metabolic uptake of glucose-d7 by bacteria. Vancomycin-susceptible (VSE) and vancomycin-resistant (VRE) E. faecalis strains can be grown to logarithmic phase by first culturing them in LB medium for approximately 2 hours. Then, VSE and VRE strains can be further cultured in M9 medium containing approximately 2% glucose-d7 as the sole carbon source, supplemented with vancomycin to a final concentration of approximately 20 μg / mL (Figure 2A). Other embodiments can include various types of carbon sources, either as the sole carbon source or in combination with other carbon or nutrient sources. One exemplary embodiment can use a nutrient source of approximately 0-100% deuterium dioxide. Similarly, the carbon source can include a combination of deuterium dioxide and glucose-d7. This concentration was chosen because it is between the minimum inhibitory concentration (MIC) of vancomycin for VSE (MIC 1 μg / mL) and VRE (MIC >100 μg / mL). As a control, VSE and VRE were cultured in glucose-d7 medium without vancomycin (Figure 2a). After 30 min (0.5 h) to 3 h of incubation, each sample was centrifuged, washed twice with water, and then deposited on an agarose gel pad for SRS imaging.

[0062] Figure 2B shows the CD oscillation region (approximately 2160 cm) of VSE and VRE cultured in glucose d7-containing medium for 30 min with and without vancomycin. -1) shows SRS imaging of the control group. In the control group, both VSE and VRE had strong CD signals, indicating high metabolic uptake of glucose-d7 by the bacteria, and individual bacteria could be clearly identified. In the vancomycin-treated group, the CD signal intensity of individual VSE decreased to approximately one-third of that of individual VSE in the control group. Because the SRS signal intensity of glucose-d7 is proportional to its concentration, the decrease in signal indicates a decrease in glucose-d7 uptake. In comparison, the CD signal intensity of each VRE was similar to that of the control group.

[0063] Figures 2C and 2D show normalized SRS spectra corresponding to the VSE and VRE shown in Figure 2A. All spectra show peaks in the CD vibration region, but the signal intensity of the VSE is reduced in the vancomycin-treated specimens compared to the control, while the signal intensity of the VRE remains the same regardless of vancomycin treatment, which is consistent with the SRS imaging shown in Figure 2B. These results indicate that the metabolic activity of VSE and VRE responds differently to vancomycin treatment, and thus the susceptibility of VSE and VRE can be determined within 30 minutes, which is close to one cell cycle of E. faecalis. Similar effects were observed with vancomycin treatment for 1 and 3 hours. Therefore, time is not critical for AST of E. faecalis using the imaging method disclosed herein.

[0064] To quantify the CD content of E. faecalis cells and the bacterial response to antibiotic treatment, multivariate curve resolution (MCR) analysis was applied to extract CD components from the hyperspectral SRS imaging data shown in Figures 3A-3D. MCR is a bilinear model that decomposes the experimental data base, including the spectrum of each pixel, into a concentration map and spectra of the principal components. Using the initial estimated spectra of each component as input, the MCR is iteratively calculated and optimized using an alternating least-squares algorithm until convergence is reached. The MCR output includes the concentration map and spectra of each component. Referring now to Figures 3A-3D, the spectrum of glucose-d7, representing the CD components, and gel, representing the background, were used as input elements. After MCR analysis, the CD components were clearly separated from the background and could be used to quantify the content of the collected SRS imaging data.

[0065] While this method generally describes hyperspectral SRS microscopy for metabolic imaging-based AST, another exemplary embodiment can use single-frequency SRS microscopy with an improved sample preparation method. Previous single-frequency SRS results used bacteria dried on glass to prevent bacterial movement during measurement, resulting in strong noise in the SRS signal. Preparing an agarose gel sample and depositing bacteria in solution on an agarose gel pad preserves bacterial viability and eliminates noise. Thus, the bacterial SRS signal at the CD vibration frequency is mostly derived from CD components. Therefore, metabolic imaging-based AST methods can detect bacterial changes by monitoring CD components using single-frequency SRS imaging. Because it requires only one image rather than multiple images, it requires much less time than hyperspectral SRS imaging.

[0066] Further, another exemplary embodiment may use additional steps to prepare samples. In one exemplary embodiment, samples for SRS imaging were prepared by drying bacteria on glass. As shown in Figures 4A-4B, bacteria deposited on an agarose gel significantly reduced cross-phase modulation noise in SRS imaging. Figure 4A shows SRS imaging at 2178 cm-1 of bacteria grown in normal medium and dried on glass, while Figure 4B shows SRS imaging at 2178 cm-1 of bacteria grown in normal medium and deposited on an agarose gel pad. Imaging using dried bacteria on glass exhibited a strong background, likely due to cross-phase modulation. For live cell imaging, an agarose gel pad was used, and bacteria in solution were deposited on the gel pad. As shown in Figure 4B, this sample preparation significantly reduced background noise in SRS imaging. To maximize the CD signal level, the culture medium was changed from the normal glucose-containing LB medium to a custom-designed M9 medium in which glucose-d7 was the sole carbon source, increasing the glucose-d7 concentration to 2%. To ensure signal and spectral resolution, a two-rod setup using two 15 cm long SF57 glass rods was configured to chirp the pump and Stokes beams for hyperspectral SRS imaging. The SNR of a single bacterium in the CD vibration region was approximately 5, approximately 20 cm. -1 along with a spectral resolution of

[0067] To test the toxicity of glucose-d7 to bacteria, E. faecalis 31970 and 31972 were cultured in custom-designed M9 medium, one containing normal glucose as a control and the other containing glucose-d7. Bacterial growth was monitored for up to 22 hours by optical density (OD) measurements at 600 nm (Figure 9). Similar growth curves were observed for E. faecalis 31970 and E. faecalis 31972 cultured in control and glucose-d7-containing media, demonstrating the lack of toxicity of glucose-d7 to bacterial growth. Figure 9A shows OD measurements of vancomycin-susceptible enterococci cultured in M9 medium containing either 2% normal glucose or 2% glucose-d7, while Figure 9B shows OD measurements of vancomycin-resistant enterococci cultured in M9 medium containing either 2% normal glucose or 2% glucose-d7.

[0068] The MIC, defined as the lowest concentration of an antibiotic capable of inhibiting visible bacterial growth in vitro, can be determined by our metabolic imaging method. VSE were cultured in glucose-d7-containing M9 medium supplemented with different concentrations of vancomycin. After 1 hour of incubation, each sample was centrifuged, washed twice with PBS, and deposited on an agarose gel pad for SRS imaging. Figure 5A shows the CD components of each sample after MCR analysis. To quantitatively compare the changes in bacterial glucose-d7 uptake with vancomycin treatment, the intensity of the CD components of each bacterium within the field of view was statistically analyzed and plotted (Figure 5B). Compared to the control without vancomycin treatment, vancomycin treatment at concentrations of 2 μg / mL or higher significantly reduced the intensity of the CD components. In contrast, vancomycin treatment at 1 μg / mL did not reduce this intensity. Therefore, the MIC of VSE determined by our metabolic imaging method is 2 μg / mL, which is twice the MIC determined by conventional culture-based methods. This discrepancy is primarily due to the manual serial dilution of antibiotics performed in conventional AST, which is acceptable and within the accuracy range. Figure 5A shows the CD component concentration maps of bacteria cultured for 1 hour in glucose-d7-containing medium at different vancomycin concentrations. Figure 5B is a graph showing the quantification of CD component intensity. The CD component intensity of bacteria treated with 2 μg / ml or more of vancomycin is significantly reduced compared to the control without vancomycin treatment. Figure 5C shows a comparison of the MICs determined from the conventional culture-based method and those determined from our metabolic imaging method.

[0069] Evaluation of bacterial metabolic activity by SRS imaging was performed on E. faecalis bacteria cultured in normal glucose-d7-containing medium and glucose-d7-containing medium, subsequently washed with PBS to remove traces of the medium, and then deposited on an agarose gel pad for SRS imaging. As shown in Figure 7B, hyperspectral SRS imaging of a 1 M glucose-d7 solution shows a peak at around 2130 cm-1 from the CD vibrational mode. This peak is in the silent region of the Raman spectrum, providing excellent contrast for imaging the metabolic uptake of deuterium-labeled metabolites. SRS imaging of bacteria cultured in glucose-d7-containing medium showed a strong signal in the CD vibrational region, indicating successful uptake and utilization of glucose-d7. As a control, no signal was observed in bacteria cultured in normal glucose medium. These results were further substantiated by spontaneous Raman measurements, as shown in Figure 8, where a peak in the CD region was observed only in bacteria cultured in glucose-d7-containing medium. These data jointly demonstrate that the metabolic activity of glucose-d7 uptake by bacteria can be monitored by hyperspectral SRS imaging at the single-cell level. The grey bar indicates the peak at approximately 2150 cm -1 The nearby CD region is shown.

[0070] Figures 6A-6D show the changes in CD components of bacteria cultured in glucose-d7 medium in the presence and absence of 20 μg / ml vancomycin. Figures 6A and 6B show the temporal dynamics of CD components after MCR analysis of VSE and VRE, respectively. The average intensity of each bacterial CD component was quantified and plotted in Figures 6C and 6D for VSE and VRE, respectively. For control VSE without vancomycin treatment, the CD component intensities of bacteria cultured in glucose-d7 medium for approximately 0.5 and 1 h were similar, while the CD component intensity of bacteria cultured for 3 h was higher (Figure 6C), indicating a slight increase in glucose-d7 uptake over time. For control VRE, the CD component intensities of each bacterial species were similar from approximately 0.5 to 3 h (Figure 6D). Vancomycin treatment significantly reduced the CD component intensity of VSE. Student's t-test was used to analyze significant differences between the control and treated VSE groups. The p-values ​​were 2.2 × 10 for the results at approximately 0.5, 1, and 3 hours, respectively. -13 , 2.0×10 -11 , and 3.9 × 10 -15 Vancomycin treatment did not significantly reduce the intensity of the CD component of VRE. Interestingly, when VRE was treated with vancomycin for 3 hours, the intensity of the CD component was higher than that of the control group (p value 5 × 10 -4), which is likely due to a metabolic response of VRE to vancomycin treatment, in which cells increase glucose uptake to combat antibiotic stress. The results show that vancomycin treatment significantly reduces glucose-d7 uptake in VSE, while glucose-d7 uptake in VRE remains unchanged over a 3-hour period. Therefore, the susceptibility of VSE and VRE to vancomycin can be rapidly determined. Figures 12A-12C show the corresponding and imaging data for bacteria cultured in glucose-d7-containing medium for 1 hour. Figure 12A shows SRS imaging in the CD region of vancomycin-susceptible and -resistant E. faecalis cultured in glucose-d7-containing medium with and without 20 μg / ml vancomycin treatment. Figure 12B shows the corresponding spectra for vancomycin-susceptible E. faecalis with and without vancomycin treatment (1203 and 1201). Additionally, Figure 12C shows the corresponding spectra of vancomycin-treated and untreated vancomycin-susceptible E. faecalis, demonstrating no change in the imaging data. Similarly, Figures 13A-13C show the corresponding data and imaging data, respectively, for bacteria grown in glucose-d7-containing medium for 3 hours. Figure 13B shows the corresponding spectra of vancomycin-susceptible E. faecalis with vancomycin treatment 1303 and without vancomycin treatment 1301. Figure 13C shows the corresponding spectra of vancomycin-treated and untreated vancomycin-susceptible E. faecalis, demonstrating no change in the imaging data. Additional tests were performed to determine whether the metabolic imaging-based AST method worked for other antibiotics. VSE E. faecalis 31970 was cultured in glucose-d7-containing M9 medium with five different antibiotics: vancomycin, linezolid, daptomycin, gentamicin, and erythromycin. These antibiotics are commonly prescribed in clinics for the treatment of bacterial infections and have different mechanisms of action. Figures 10A–10F show SRS imaging of VSE after 1 hour of culture in glucose-d7-containing medium without antibiotics (control) and with each antibiotic added at a final concentration of 20 μg / mL. A significant decrease in CD signal was observed only in bacteria treated with vancomycin and linezolid. Therefore, based on our metabolic imaging method, we conclude that E. faecalis 31970 is sensitive to vancomycin and linezolid but resistant to daptomycin, gentamicin, and erythromycin. The MICs of these antibiotics against VSE were determined by conventional culture-based methods and are shown and confirmed in Table 1 below. The susceptibility of VSE to these five antibiotics is consistent with the results determined by our metabolic imaging method. [Table 1]

[0071] To confirm that this method also works with other bacterial species, we tested Staphylococcus aureus (S. aureus) using one oxacillin-susceptible (MIC 0.0625 μg / ml) and one oxacillin-resistant strain. The metabolic activity of glucose-d7 was confirmed by SRS imaging in single bacteria cultured in glucose-d7-containing medium for 0.5 h (Figure 11B). When bacteria were treated with 1.3 μg / ml oxacillin for 0.5 h, the CD peak of the susceptible strain 1103 was reduced compared to the untreated control 1101 (Figure 11C), while no significant change was observed in the resistant strain (Figure 11D). Therefore, the susceptible and resistant S. aureus strains can be determined in approximately 0.5 h.

[0072] The present disclosure further provides a metabolic imaging method that can determine the susceptibility and antibiotic MIC of live bacteria within approximately 30 minutes by monitoring bacterial metabolic activity at the single-cell level. This method reduces the time required for AST using traditional culture-based methods, which typically takes at least 16 to 24 hours, to 0.5 hours. Metabolic imaging-based AST at the single-bacterium level is particularly useful for non-culturable or fastidious bacteria because metabolic activity occurs faster than phenotypic growth, eliminating the need to replicate bacteria to detect metabolic activity responses to antibiotic treatment.

[0073] In the case of the DO method, SRS imaging-based methods demonstrated metabolic activity using heavy water (DO) using Pseudomonas aeruginosa. Given the broad CD vibrational spectrum, the bacterial signal could be improved by over five times using chirp-free femtosecond SRS compared to chirp-based SRS. Bacterial DO metabolism may respond differently to various antibiotics, depending on the bacterial sensitivity, in approximately 10 minutes. This allows for SRS microscopy to be used for metabolic activity studies of single bacteria and for fast AST, which works generally for different bacteria.

[0074] The toxicity of DO was first tested in bacteria by measuring their growth in DO-containing medium. Three types of bacteria (E. coli, Staphylococcus aureus, and P. aeruginosa) were cultured in LB medium containing different concentrations of DO, and their growth was monitored by measuring the optical density (OD) at approximately 600 nm. As shown by the growth curves in various DO concentrations, DO concentrations up to 100% did not significantly toxicate the growth of E. coli and S. aureus (Figures 15A-15B). Growth of P. aeruginosa initially slowed in LB medium containing DO concentrations above 70% but returned to normal after approximately 18 h at 70% and 80% DO concentrations and after approximately 22 h at 100% DO (Figure 15c). Therefore, DO in the medium does not induce significant toxicity to the bacteria.

[0075] We cultured bacteria in 70% DO-containing LB medium and used P. aeruginosa as a model to test whether DO metabolic activity in single bacteria could be monitored by SRS microscopy. P. aeruginosa was cultured separately in regular LB medium and 70% DO-containing LB medium for approximately 2 hours, then centrifuged and washed with water to remove the medium. (See Figure 16A.) The spontaneous Raman spectrum of the high-density bacteria showed a broad peak (1601) from approximately 2070 to approximately 2250 cm-1 in the CD vibrational region of the bacteria cultured in DO-containing medium, indicating successful utilization of DO in biomolecular synthesis. As a control, bacteria cultured in regular medium (1603) did not exhibit this peak in this region (Figure 16A.) To image single bacteria, the bacteria were further diluted and deposited on an agarose gel pad. By tuning the SRS frequency to the CD region around 2162 cm-1, a strong signal was observed for individual bacteria grown in DO-containing medium (Fig. 16B, right). As a control, no CD signal was observed for bacteria grown in normal medium (Fig. 16B, left). The results were confirmed by the SRS spectrum (Fig. 16C) obtained by time-tuning the chirped pump and Stokes femtosecond pulses.

[0076] To further improve the CD signal of a single bacterium, we tested unchirped femtosecond pulses to improve the signal-to-noise (SNR) ratio of chirped picosecond pulses. Because the CD vibrational band is relatively broad, approximately 180 cm-1 (Figure 16C), unchirped femtosecond SRS may further improve the SNR. To test this, we cultured P. aeruginosa in LB medium containing 70% DO for approximately 30 min and imaged them at 2162 cm-1 with chirped picosecond and unchirped femtosecond pulses, as shown in Figures 17A and 17C, respectively. The pump and Stokes powers were adjusted to ensure that the average pump and Stokes powers used on the sample were the same. The SNRs of individual bacteria for picosecond and femtosecond SRS were 1.43 and 7.81, respectively, as shown in Figures 17B and 17D, indicating an approximately 5.46x improvement in the SNR of femtosecond SRS over picosecond SRS. This improvement is due to two different aspects of the setup. First, the CD vibration band is wider, allowing femtosecond SRS to detect wider bandwidth signals than chirped picosecond SRS. Second, chirped pulses can reduce the peak power of the pulse. Although the same average power was used in the sample, the reduction in peak power significantly reduced the SNR due to nonlinear effects of SRS.

[0077] Time-lapse imaging of DO metabolic activity can be performed on single P. aeruginosa cells using femtosecond SRS imaging. P. aeruginosa cells were first cultured in LB medium containing approximately 70% DO for up to approximately 3 hours. At different time points, approximately 500 μl of bacteria were centrifuged, washed, and deposited on an agarose gel pad for imaging. Figure 18A shows SRS imaging of a single P. aeruginosa cell at approximately 2162 cm-1. The CD signal of individual P. aeruginosa cells can be observed after as little as 10 minutes. Statistical analysis showed that the average CD signal intensity of individual bacteria increased over time and saturated after approximately 1.5 hours, which corresponds to approximately three generations, since the generation time of P. aeruginosa cells cultured in LB medium is 24–27 minutes, as shown in Figure 18C. To more clearly view individual bacteria, the image in Figure 18A was further enlarged in Figure 18B. Interestingly, the 10-minute results shown in Figure 18B confirm that the signal is stronger at the periphery of the bacterial cell, as shown in the bacterial intensity plot shown in Figure 18D. In contrast, at 30 minutes and 30 minutes later, Figure 17B shows that the signal intensity is stronger at the center of the bacterial cell, as shown in the 30-minute bacterial intensity plot shown in Figure 18E. Taken together, these results suggest that water is initially used to synthesize the cell membrane and / or cell wall of P. aeruginosa.

[0078] P. aeruginosa was cultured in LB medium containing 70% DO and supplemented with 20 μg / ml gentamicin or cefotaxime to determine the effect of antibiotics on the bacterial DO metabolic activity. This is suitable for use in rapid AST with SRS imaging. The susceptibility of P. aeruginosa was previously determined to be sensitive to gentamicin and resistant to cefotaxime at this concentration using a conventional culture-based microdilution method. SRS imaging at approximately 2162 cm-1 showed a significant decrease in the CD signal after incubation with 20 μg / ml gentamicin, as shown in Figure 19A, indicating that the DO metabolic activity of P. aeruginosa was inhibited by gentamicin. In contrast, P. aeruginosa cultured in cefotaxime exhibited a peak at approximately 2162 cm-1 at all time points, indicating active DO metabolism in P. aeruginosa cultured in cefotaxime. P. aeruginosa cultured in cefotaxime tended to form long rod-shaped structures, as shown in Figure 19B. This may be due to the fact that cefotaxime is a β-lactam antibiotic that inhibits bacterial cell wall synthesis, so P. aeruginosa cultured in cefotaxime can still grow but cannot divide. This filamentous formation can also be observed when P. aeruginosa is treated with other β-lactam antibiotics.

[0079] Using the bacterial DO metabolic activity, we were able to rapidly identify bacterial antibiotic susceptibility using the mean CD signal intensity of bacteria compared among three groups shown in Figures 19C-19H: a control without antibiotic treatment (Figure 18A), gentamicin treatment (Figure 19A), and cefotaxime treatment (Figure 19B). To distinguish between susceptible and resistant groups, a threshold line was determined at approximately 65%, the mean CD intensity of control bacteria, for all plots of results from approximately 10 minutes to approximately 3 hours, shown in Figures 19C-19H. This threshold clearly separated susceptible and resistant groups, with the gentamicin-treated group always below the threshold and the cefotaxime-treated group always above the threshold. Therefore, the susceptibility of P. aeruginosa to gentamicin and cefotaxime can be determined in as little as approximately 10 minutes.

[0080] SRS metabolic imaging, which can determine the minimum inhibitory concentration (MIC) of antibiotics against bacteria, was performed by culturing P. aeruginosa in LB medium containing approximately 70% DO for approximately 1 hour with serially varying dilutions of gentamicin. As shown in Figure 20A, SRS imaging at 2162 cm-1 indicated that DO metabolic activity of P. aeruginosa was inhibited by gentamicin at concentrations above 8 μg / ml. As a control, no CD signal was observed in P. aeruginosa cultured in standard LB medium. As shown in Figure 20B, the average intensity of the P. aeruginosa CD signal was plotted for comparison. At a 65% intensity threshold, the MIC was determined to be approximately 8 μg / ml using the SRS-based DO metabolic imaging method. This is consistent with the results determined using the conventional culture-based method.

[0081] Those skilled in the art can readily devise numerous and varied other configurations without departing from the spirit and scope of the present invention. Moreover, all statements herein reciting principles, aspects, and embodiments of the present invention, as well as specific examples thereof, are intended to encompass both structural and functional equivalents thereof. Additionally, such equivalents are intended to include both currently known equivalents as well as equivalents developed in the future, i.e., elements developed that perform the same function, regardless of structure. All references cited herein are incorporated by reference. Additional disclosure is provided in Appendix A, which is incorporated herein by reference in its entirety.

Claims

1. 1. A method for determining antibiotic susceptibility of bacteria, comprising: Preparing a bacteria-immobilized gel; Cultivating the bacteria in a medium containing a nutrient source; adding a preselected antibiotic to D 2 O medium and / or M9 medium containing glucose-d 7 as the sole carbon source to form a sample; culturing the bacteria in the medium supplemented with antibiotics for a preselected period of time; centrifuging the sample; washing the sample; depositing the sample onto the bacteria-immobilizing gel; collecting images of the sample on the bacteria-immobilized gel by coherent Raman microscopy; A method comprising:

2. The method of claim 1 , wherein the image of the sample is collected by hyperspectral coherent Raman microscopy.

3. The method of claim 1 , wherein the image of the sample is collected by single-frequency coherent Raman microscopy.

4. The method of claim 1 , wherein the immobilizing gel comprises an agarose gel.

5. The M9 medium contains 0.1 to 10% glucose-d ７ The method of claim 1 , consisting solely of

6. The above D ２ O medium is heavy water (D ２ 0). The method of claim 1 , comprising:

7. The above D ２ O medium and glucose-d ７ M9 medium containing as the sole carbon source was supplemented with glucose-d7 and deuterium oxide (D ２ 0) a combination of both.

8. The method of claim 1 , wherein the Raman imaging by coherent Raman microscopy can be stimulated Raman scattering imaging, coherent anti-Stokes Raman scattering imaging, or coherent Raman stimulated Kerr effect imaging.

9. 10. A hyperspectral stimulated Raman scattering imaging apparatus for the method of claim 1, comprising: A dual output femtosecond pulse laser with a repetition rate of 80 MHz, a pump beam; Stokes beam and a path of the Stokes beam modulated by an acousto-optic modulator; a pump beam path having a translation stage for adjusting the delay between the pump beam and the Stokes beam; a combiner configured to combine the pump beam and the Stokes beam; a dual output femtosecond pulse layer including: a laser scanning microscope having an objective configured to focus the pump beam and the Stokes beam on a sample; and an oil condenser configured to collect the laser from the sample; a filter positioned after the condenser and configured to filter out the Stokes beam; a photodiode positioned after the filter to detect the pump beam; and a lock-in amplifier configured to extract the pump beam signal loss; and 1. An imaging device comprising:

10. The imaging device of claim 9 , further comprising a chirp device, the chirp device configured to create different pulse durations between the Stokes beam and the pump beam.

11. 11. The imaging device of claim 10, wherein the chirp device is two SF57 glass rods, each 15 cm long.

12. 10. The imaging device of claim 9, wherein a pair of filters is used.

13. 10. The imaging device of claim 9, wherein the objective of the laser scanning microscope is a 60x water objective.

14. 10. The imaging apparatus of claim 9, wherein the pulse duration of the pump beam is between 1.5 and 2 ps and the pulse duration of the Stokes beam is between 1.0 and 1.4 ps.

15. The imaging device of claim 10 , wherein the chirp device is configured to produce a pulse duration of 1.9 ps for the pump beam and 1.3 ps for the Stokes beam.

16. 10. The imaging apparatus of claim 9, wherein the pump beam is a 120 fs tunable laser and has a pump beam path, and the Stokes beam is a 220 fs laser centered at 1040 nm and has a Stokes beam path.

17. 1. A method for rapid determination of bacterial antibiotic susceptibility using microscopy, comprising: Cultivating the bacteria in a medium containing a nutrient source; adding a preselected antibiotic having a preselected concentration to D 2 O medium and / or M9 medium containing glucose-d 7 as the sole carbon source to form a sample; Incubating the sample in the medium for a preselected period of time; centrifuging the sample; washing the sample; depositing the sample onto a bacteria-immobilizing gel; imaging the sample on the bacteria-immobilized gel using coherent Raman microscopy; A method comprising:

18. The method of claim 17 , wherein the coherent Raman microscopy comprises a stimulated Raman scattering system.

19. 20. The method of claim 18, wherein the stimulated Raman scattering system uses a pump beam and a Stokes beam, both beams chirped with one or more glass rods, thereby creating a pulse duration of 1-2 ps for both the pump beam and the Stokes beam.

20. 20. The method of claim 19, wherein the stimulated Raman scattering system further comprises a translation stage that adjusts the delay between the pump beam and the Stokes beam.

Images