[go: up one dir, main page]

WO2016036314A1 - Sonde à fibre optique sensible à la pression pour spectroscopie optique de tissu in vivo en temps réel, système l'incorporant, et son procédé d'utilisation - Google Patents

Sonde à fibre optique sensible à la pression pour spectroscopie optique de tissu in vivo en temps réel, système l'incorporant, et son procédé d'utilisation Download PDF

Info

Publication number
WO2016036314A1
WO2016036314A1 PCT/SG2015/050289 SG2015050289W WO2016036314A1 WO 2016036314 A1 WO2016036314 A1 WO 2016036314A1 SG 2015050289 W SG2015050289 W SG 2015050289W WO 2016036314 A1 WO2016036314 A1 WO 2016036314A1
Authority
WO
WIPO (PCT)
Prior art keywords
pressure
spectra
probe
spectrograph
tissue
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Ceased
Application number
PCT/SG2015/050289
Other languages
English (en)
Inventor
Zhiwei Huang
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
National University of Singapore
Original Assignee
National University of Singapore
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by National University of Singapore filed Critical National University of Singapore
Publication of WO2016036314A1 publication Critical patent/WO2016036314A1/fr
Anticipated expiration legal-status Critical
Ceased legal-status Critical Current

Links

Classifications

    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
    • G01N21/62—Systems in which the material investigated is excited whereby it emits light or causes a change in wavelength of the incident light
    • G01N21/63—Systems in which the material investigated is excited whereby it emits light or causes a change in wavelength of the incident light optically excited
    • G01N21/65—Raman scattering
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00—Measuring for diagnostic purposes; Identification of persons
    • A61B5/0059—Measuring for diagnostic purposes; Identification of persons using light, e.g. diagnosis by transillumination, diascopy, fluorescence
    • A61B5/0075—Measuring for diagnostic purposes; Identification of persons using light, e.g. diagnosis by transillumination, diascopy, fluorescence by spectroscopy, i.e. measuring spectra, e.g. Raman spectroscopy, infrared absorption spectroscopy
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00—Measuring for diagnostic purposes; Identification of persons
    • A61B5/68—Arrangements of detecting, measuring or recording means, e.g. sensors, in relation to patient
    • A61B5/6801—Arrangements of detecting, measuring or recording means, e.g. sensors, in relation to patient specially adapted to be attached to or worn on the body surface
    • A61B5/6843—Monitoring or controlling sensor contact pressure
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
    • G01N21/17—Systems in which incident light is modified in accordance with the properties of the material investigated
    • G01N21/47—Scattering, i.e. diffuse reflection
    • G01N21/4738—Diffuse reflection, e.g. also for testing fluids, fibrous materials
    • G01N21/474—Details of optical heads therefor, e.g. using optical fibres
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
    • G01N21/17—Systems in which incident light is modified in accordance with the properties of the material investigated
    • G01N21/47—Scattering, i.e. diffuse reflection
    • G01N21/49—Scattering, i.e. diffuse reflection within a body or fluid
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
    • G01N21/62—Systems in which the material investigated is excited whereby it emits light or causes a change in wavelength of the incident light
    • G01N21/63—Systems in which the material investigated is excited whereby it emits light or causes a change in wavelength of the incident light optically excited
    • G01N21/64—Fluorescence; Phosphorescence
    • G01N21/645—Specially adapted constructive features of fluorimeters
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B2562/00—Details of sensors; Constructional details of sensor housings or probes; Accessories for sensors
    • A61B2562/22—Arrangements of medical sensors with cables or leads; Connectors or couplings specifically adapted for medical sensors
    • A61B2562/221—Arrangements of sensors with cables or leads, e.g. cable harnesses
    • A61B2562/223—Optical cables therefor
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
    • G01N21/62—Systems in which the material investigated is excited whereby it emits light or causes a change in wavelength of the incident light
    • G01N21/63—Systems in which the material investigated is excited whereby it emits light or causes a change in wavelength of the incident light optically excited
    • G01N21/64—Fluorescence; Phosphorescence
    • G01N2021/6417—Spectrofluorimetric devices
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
    • G01N21/62—Systems in which the material investigated is excited whereby it emits light or causes a change in wavelength of the incident light
    • G01N21/63—Systems in which the material investigated is excited whereby it emits light or causes a change in wavelength of the incident light optically excited
    • G01N21/64—Fluorescence; Phosphorescence
    • G01N21/645—Specially adapted constructive features of fluorimeters
    • G01N2021/6484—Optical fibres

Definitions

  • a pressure-sensitive fiber optic probe for real-time in vivo tissue optical spectroscopy a system incorporating the same and a method for using the same FIELD OF INVENTION
  • the present invention relates to a pressure-sensitive fiber optic probe for realtime in vivo tissue optical spectroscopy, a system incorporating the same and a method of using such a probe.
  • the probe is placed against the target tissue during laser irradiation, and remains in substantially the same place during in vivo spectroscopy measurements. This ensures probe-tissue contact, avoids inconsistent illumination-detection geometry, reduces refractive index mismatch, and increases light penetration to enhance the diagnostic capability of spectroscopy instruments.
  • high probe-tissue contact pressures generally produce the following effects on tissue under compression: decreased thickness, reduced blood volume and oxygen saturation, vasodilation, microcirculation impairment, densely packed scatter effects and occlusion under the probe tip, which can alter the optical properties of underlying tissue.
  • the applied probe pressure is especially hard to control in case of clinical scenario such as in vivo diagnosis of internal organs, for example endoscopy, due to problems of access.
  • tissue optical spectroscopy may introduce operator-induced variations in the magnitude of pressure exerted onto the tissue. This leads to potential concerns such as changes in tissue spectral profiles, disparities in data collection and calibration and potential inaccuracies in diagnosis. Hence, acquisition of high- quality in vivo tissue spectra with a short integration time of say ⁇ lsec and instantaneous quantitative probe-tissue contact pressure feedback to the operators is necessary to mitigate the user-induced tissue spectral profile variations.
  • one object of the present invention is to overcome at least some of the problems associated with the prior art and the present techniques.
  • a further object of the present invention is to provide a pressure-sensitive fiber optic probe for real-time in vivo tissue optical spectroscopy and a method of using such a probe.
  • a spectroscopy system for obtaining biomedical data for diagnosis purposes, the system comprising: a source of radiation for irradiating a tissue sample; a spectrograph for receiving a spectra collected from the tissue sample as a result of irradiation of the tissue sample; a pressure sensitive probe for delivering radiation to the sample and collecting and returning the spectra to the spectrograph; a pressure sensor for measuring the pressure exerted by the probe on the tissue sample; and a pressure adjustment module for adjusting the pressure to be at a predetermined value when the spectra is collected.
  • the pressure sensitive probe includes a tip which is a thin film piezo-resistive pressure sensor.
  • the pressure sensitive probe is pre-calibrated.
  • the pressure sensor is a pressure to voltage converter.
  • the pressure adjustment module automatically adjust the exerted pressure.
  • system may further comprise a user interface for controlling the analysis and data processing of the spectra to enable a diagnosis from the spectra.
  • user interface presents spectra and pressure readings.
  • the user interface enables statistical analysis of the spectra to enable diagnosis from the spectra.
  • the spectrograph is a Raman spectrograph. In an implementation the spectrograph is a diffuse reflectance spectrograph.
  • the spectrograph is a near infrared (NIR) auto fluorescence (AF) spectrograph.
  • NIR near infrared
  • AF auto fluorescence
  • a method for obtaining biomedical data for diagnosis purposes comprising: irradiating a tissue sample with a source of radiation; receiving a spectra collected from the tissue sample as a result of irradiation of the tissue sample from a spectrograph; delivering radiation to the sample and collecting and returning the spectra to the spectrograph via a pressure sensitive probe; measuring the pressure exerted by the probe on the tissue sample using a pressure sensor; adjusting the pressure to be at a predetermined value when the spectra is collected using a pressure adjustment module.
  • a pressure sensitive probe for use in the method and system of previous aspects.
  • a pressure sensitive probe for use in a spectroscope comprising a source of radiation for irradiating a tissue sample; a spectrograph for receiving a spectra collected from the tissue sample as a result of irradiation of the tissue sample; wherein the pressure sensitive probe delivers radiation to the sample and collects and returns the spectra to the spectrograph; and is coupled to a pressure sensor for measuring the pressure exerted by the probe on the tissue sample; and a pressure adjustment module for adjusting the pressure to be at a predetermined value when the spectra is collected.
  • a fifth aspect of the invention there is provided a method of using the apparatus of the first aspect to diagnose an abnormal growth.
  • Figure 1 is a schematic diagram of a pressure-sensitive fiber-optic autofluorescence (AF) spectroscopy system, in accordance with an embodiment of the invention
  • Figure 2 is a graph of a typical response function ⁇ 1 standard deviation (SD) of a thin film piezo-resistive pressure sensor calibrated for a range of pressures of between about 0 - 200KPa, in accordance with an embodiment of the invention
  • Figure 4 is a histogram displaying the peak AF intensity [I 4 67nm] of the acquired fingertip and volar forearm AF spectra among three pressure levels (LP ⁇ 10KPa, MP ⁇ 50KPa and HP ⁇ 130KPa), in accordance with an embodiment of the invention;
  • Figure 5 shows the number of latent variables (LVs) against cross-validation error for identifying an ideal number of LVs to be utilized for classifying spectra measured at different probe-tissue contact pressures LP ⁇ 10KPa, MP ⁇ 50KPa and HP ⁇ 130KPa, in accordance with an embodiment of the invention
  • Figure 6 shows the diagnostically significant LVs calculated from the autofluorescence (AF) spectra of (a) fingertip skin and (b) volar forearm skin measured at three different probe pressure levels LP ⁇ 10KPa, MP ⁇ 50KPa and HP ⁇ 130KPa, in accordance with an embodiment of the invention
  • Figure 7 shows a two-dimensional ternary plot of the posterior probabilities belonging to low pressure (LP), medium pressure (MP), and high pressure (HP) categories calculated from the PLS-DA models together with leave-one patient- out, cross validation, generated from in vivo autofluorescence (AF) spectra of (a) fingertip skin and (b) volar forearm skin, in accordance with an embodiment of the invention;
  • LP low pressure
  • MP medium pressure
  • HP high pressure
  • LP low pressure
  • MP medium pressure
  • HP high pressure
  • Figure 9 is a diagram showing number of latent variables (LVs) vs. cross- validation error for identifying optimum number of LVs to be utilized for developing PLS-DA diagnostic models for differentiating spectra acquired at different reference pressure levels (LP ⁇ 10 kPa, MP ⁇ 50 kPa and HP ⁇ 130 kPa), in accordance with an embodiment of the invention;
  • Figure 10 is a diagram showing diagnostically significant latent variables (LVs) calculated from the NIR AF spectra of (a) fingertip, (b) palm, and (c) volar forearm measured at three different probe pressure levels (LP ⁇ 10 kPa, MP ⁇ 50 kPa and HP ⁇ 130 kPa), in accordance with an embodiment of the invention;
  • LVs diagnostically significant latent variables
  • Figure 11 is a two-dimensional ternary plot of calculated posterior probabilities belonging to in vivo (a) fingertip, (b) palm, and (c) volar forearm NIR AF spectra measured under low pressure (LP ⁇ 10 kPa), medium pressure (MP ⁇ 50 kPa), and high pressure (HP ⁇ 130 kPa) using PLS-DA models together with leave-one patient-out, cross validation), in accordance with an embodiment of the invention;
  • LP low pressure
  • MP ⁇ 50 kPa medium pressure
  • HP high pressure
  • FIG 12 is a schematic drawing of the pressure sensitive fiber-optic diffuse reflectance (DR) spectroscopy system, in accordance with an embodiment of the invention.
  • DR diffuse reflectance
  • LP low pressure
  • MP medium pressure
  • HP high pressure
  • Figure 14 is a graph of the number of latent variables (LVs) against cross- validation error for identifying ideal number of LVs to be utilized for classifying spectra measured at different probe-tissue contact pressures (LP ⁇ 10 kPa, MP ⁇ 50 kPa and HP ⁇ 130 kPa) , in accordance with an embodiment of the invention;
  • Figure 15 is a graph of the diagnostically significant LVs calculated from the diffuse reflectance (DR) spectra of (a) fingertip (b) volar forearm and (c) palm measured at three different probe pressure levels (LP ⁇ 10 kPa, MP ⁇ 50 kPa and HP ⁇ 130 kPa), in accordance with an embodiment of the invention;
  • DR diffuse reflectance
  • Figure 16 is a two-dimensional ternary plot of the posterior probabilities belonging to low pressure (LP), medium pressure (MP), and high pressure (HP) categories calculated from the PLS-DA models together with leave-one patient- out, cross validation, generated from in vivo diffuse reflectance (DR) spectra of (a) fingertip (b) volar forearm and (c) palm, in accordance with an embodiment of the invention;
  • LP low pressure
  • MP medium pressure
  • HP high pressure
  • SD standard deviation
  • FIG. 19 is a graph of the diagnostically significant latent variables (LVs) calculated from the Raman spectra of (a) fingertip, (b) palm, and (c) volar forearm measured at three different probe pressure levels (LP ⁇ 10 kPa, MP ⁇ 50 kPa and HP ⁇ 130 kPa). The loading values of volar forearm have been shifted vertically for better visualization, in accordance with an embodiment of the invention.
  • Figure 20 is a two-dimensional ternary plot of calculated posterior probabilities belonging to in vivo (a) fingertip, (b) palm, and (c) volar forearm Raman spectra measured under low pressure (LP ⁇ 10 kPa), medium pressure (MP ⁇ 50 kPa), and high pressure (HP ⁇ 130 kPa) using PLS-DA models together with leave-one patient-out, cross validation, in accordance with an embodiment of the invention.
  • LP low pressure
  • MP ⁇ 50 kPa medium pressure
  • HP high pressure
  • the present invention provides a novel pressure sensitive fiber optic probe applicable to various biomedical spectroscopy platforms, for example Raman, fluorescence, and reflectance, for real-time quantitative monitoring of in vivo (and ex vivo) probe-tissue contact pressure.
  • the pressure sensitive fiber optic probe is used to evaluate the probe pressure effects on in vivo skin tissue autofluorescence (AF) measurements.
  • Skin tissue such as the fingertip and volar forearm skin, were purposefully selected as these act as highly scattering media in the visible to near-infrared region. Skin tissue represents one of the most challenging organs with complex and inhomogeneous morphological structures.
  • Multivariate analysis including partial least squares-discriminant analysis (PLS-DA) was carried out on the measured tissue AF spectra to determine the spectral information associated with probe pressure variability.
  • PLS-DA partial least squares-discriminant analysis
  • probe-tissue contact pressure is one of the major problems that can induce variations on fiber-based in vivo spectroscopic measurements.
  • a novel miniaturized pressure sensitive fiber optic probe for real-time, quantitative monitoring of probe-tissue contact pressure and investigate its effect on in vivo skin autofluorescence (AF) measurements has been developed in accordance with the present invention.
  • the miniaturized pressure sensitive fiber probe couples a customized piezo-resistive pressure sensor that has high sensitivity of about 20.6 mV/KPa and short response time of about ⁇ 5 ⁇ 5, facilitating the precise, real-time measurement of clinically relevant probe-tissue contact pressures of between about 0 and 200KPa in vivo.
  • the emission peak intensity (U67nm) of the skin AF spectra demonstrates that MP has high implications on the fingertip AF (p ⁇ 5xl0 -4 ) compared to the volar forearm skin AF (p > 0.05).
  • Multiclass partial least squares-discriminant analysis provides accuracies of 66.72% and 34.04%, for differentiating the fingertip skin and volar forearm skin AF spectra measured under MP vs. LP and HP, respectively, confirming the site-specific effect of probe pressures exerted.
  • MP affects the fingertip AF, but not the volar forearm AF.
  • the fingertip AF and the volar forearm AF measured at LP vs. MP and HP yield accuracies of 78.97% and 89.89%, respectively, suggesting that the probe pressure has significant effect on in vivo skin AF.
  • the pressure sensitive fiber-optic probe according to the present invention has the potential to mitigate probe pressure induced alterations in tissue AF in real-time by providing instant feedback to the operators with quantitative probe-tissue contact pressure values during in vivo spectroscopic measurements.
  • FIG. 1 shows the schematic diagram of pressure sensitive in vivo tissue AF spectroscopy platform 100 developed for real-time, quantitative probe-tissue contact pressure measurements.
  • the spectroscopy system consists of a spectrum-stabilized 405-nm diode laser 102 having a maximum output of about 100 mW, such as that supplied by Power Technology Inc., Alexander, AR, USA; a transmissive imaging spectrograph 104, such as a QE65000, supplied by Ocean Optics Inc., Dunedin, FL, USA, which is equipped with a back-thinned charge-coupled device (CCD) detector 106, such as a S7031-1006, 1024X58with pixel sizes of 24.6mm, QE > 90%, supplied by Hamamatsu, Shizuoka, Japan; and a customized piezo-resistive sensor coupled bifurcated fiber optic probe 108.
  • CCD charge-coupled device
  • the hand-held bifurcated fiber probe is about 2 m in length; about 6.33 mm in outer diameter.
  • An example may be the R200-7- VIS/NIR fiber supplied by Ocean Optics Inc., Dunedin, FL, USA.
  • a narrow band-pass filter 116 such as a FBH 405 as supplied by Thorlabs Inc., Newton, NJ, USA which is used to suppress laser noise for tissue excitation
  • the collection fiber bundle is integrated with a long-pass filter 118, such as the HQ 430 as supplied by Thorlabs Inc., Newton, NJ, USA which reduces scattered laser light, while
  • the round-to-line fiber adapter is used for matching with the CCD height for maximizing the signal detection with improved signal-to-noise ratios (SNR).
  • the round-to-line fiber bundle adapter further improves the signal-to-noise ratio of the measured signal of up to 7.6-fold ( 58) by vertical binning of the entire CCD for maximizing in vivo tissue NIR AF detection.
  • the hand-held fiber-optic probe was customized by coupling a customized piezo-resistive flexi-force pressure sensor 122 to the probe tip to facilitate realtime, quantitative monitoring of probe pressure exerted on the tissue.
  • the pressure sensor having for example a 6.33mm sensing area diameter, 51 mm length and 0.203 mm thickness, such as that from Tekscan Inc., South Boston, MA, USA.
  • the standard thin film pressure sensor was made of a piezo-resistive polyester substrate whose resistance varies with application of pressure on its sensing area.
  • the pressure sensor has a large force working range of between about 0-111 N, a high sensitivity of about 20.6 mV/KPa, a short response time of about ⁇ 5 ⁇ and an operating temperature of about -9°C to 60°C. This greatly facilitates the adoption of piezo-resistive sensor coupled pressure sensitive fiber-optic probe for real-time quantitative monitoring of probe pressure and its effects on in vivo tissue optical spectroscopy.
  • the pressure sensor was connected to a force/pressure-to-voltage circuit and analog to digital converter 124 (ADC) to accurately quantify the magnitude of exerted probe pressure onto the tissue.
  • the force-to-voltage circuit uses an inverting operational amplifier and a fixed reference resistance of about 1.5 ⁇ to produce an analog output based on the sensor resistance (i.e., a voltage divider circuit followed by an inverting amplifier with an adjustable gain and DC bias (0 to 5 V)).
  • the ADC further converted the analog output of the sensor to a digital voltage value.
  • the sensitivity of the sensor was adjusted either by changing the reference resistance or the drive voltage. Lower reference resistance or drive voltage allows the sensor to be less sensitive with a large active force range.
  • GUI graphical user interface
  • Matlab environment such as that supplied by Mathworks Inc., Natick, MA, USA.
  • the user interface triggers data acquisition and analysis including laser power control, CCD shutter and camera readout synchronization, CCD dark- noise subtraction, outlier detection, wavelength calibration, system spectral response calibration, normalization, and real-time display of in vivo skin tissue spectra as well as quantified probe-tissue contact pressure.
  • the force/pressure-to-voltage circuit is just one example of a pressure sensor to be used to measure the pressure exerted by the sensor tip on the tissue sample. Any other appropriate pressure sensor could be used to provide the necessary pressure readings.
  • Precise calibration of the fiber-optic pressure sensor is essential for accurate real-time quantification of applied probe pressure during in vivo tissue spectroscopic measurements.
  • the sensor was pre-calibrated using standard weights in the range of about 0-2 kg with small increments of about 50g and the response of the sensor is noted.
  • the force (mass X acceleration due to gravity) and pressure (force / sensing area) values corresponding to the applied weights were calculated.
  • the resistance of the sensor gradually decreases with applied pressure and causes an increase in the output voltage, as shown in Figure 2. This calibration procedure was repeated about three times and a reproducible calibration curve was obtained as shown in Figure 2, depicting the relationship between applied pressure and sensor response (i.e., output voltage).
  • the pre-calibrated piezo-resistive pressure sensor was coupled to the fiberoptic probe tip and this pressure sensor coupled probe was mounted horizontally on a translation stage.
  • the tissue site i.e., index fingertip or volar forearm
  • the translation stage was promptly displaced to apply desirable probe pressures upon in vivo target tissue site. These pressures were about low pressure (LP) -lOKPa, moderate pressure (MP) - 50KPa and high pressure (HP) - 130KPa.
  • the probe pressure was instantaneously displayed in the real-time diagnostic software, guiding the operator to acquire spectra at the required pressure levels.
  • Multiple spectra ( ⁇ 3-4) were obtained from each site for all the three pressure levels to include inter- and/or intra-tissue variability for data analysis. Pronounced spectral changes occurred mainly due to different skin prototypes and anatomical areas associated with variation in melanin and hemoglobin content.
  • the volunteers were restricted to Chinese people with an age limit of 18 to 35 and no history of skin cancer, to reduce the significant spectral variability associated with differences in skin optical properties across individuals (i.e., race, age, and sun exposure).
  • Displaying the pressure reading of the pressure applied by the tip to the tissue sample means that the pressure can be adjusted either manually or automatically to be a specific required value or within a specific range. If the adjustment is manual the operator can adjust the pressure. If the adjustment is automatic the movement of the translation stage may be varied. If the probe is brought into contact with the tissue sample by any other form of device appropriate adjustment can be made.
  • the adjustment of pressure is controlled by a pressure adjustment module, which may be a simple display in the manual adjustment scenario, and may include control circuitry for automatic adjustment.
  • the system may operate at a number or predetermined pressures (e.g. LP, MP and HP) with a spectra being measured at each pressure.
  • the system may alternatively operate at a predetermined pressure which is determined to be the optimal pressure for a specific tissue sample.
  • the system may further operate in a manner in which the pressure is increased in a step wise manner and spectra obtained at each step to give a representation of the effect of the pressure on the spectra and to provide different diagnostic information.
  • Multi-class probabilistic PLS-DA was performed to realize in vivo discrimination among skin tissue spectra acquired at different pressure levels.
  • PLS-DA employs the fundamental principle of principal component analysis (PCA), but further rotates the components such as latent variables (LVs), by maximizing the covariance between the spectral variations and group affinity, so that the LVs explain diagnostically relevant variations rather than the most prominent variations in the spectral dataset.
  • PCA principal component analysis
  • LVs latent variables
  • the performance of the PLS-DA model was validated using a leave-one patient-out, cross validation methodology. In this validation procedure, the spectral data of one subject was left out and the PLS-DA model was redeveloped from the remaining spectra.
  • the redeveloped model was used to classify the withheld spectra and this process was repeated iteratively until all the withheld spectra were classified.
  • the complexity of the developed PLS-DA model i.e., the number of retained model LVs
  • the one-against-all multi-class terminology was further employed to classify the tissue spectra measured under different pressure levels.
  • the multivariate statistical analysis was performed using a PLS toolbox as supplied by Eigenvector Research, Wenatchee, WA in the Matlab programming environment.
  • the tissue AF spectra emerges due to the superposition of light re-emission from endogenous skin tissue fluorophores (e.g., nicotinamide adenine dinucleotide (NADH), flavin adenine dinucleotide (FAD), and collagen) that were excited, and distorted by re- absorption of intrinsic tissue pigments (e.g., blood in dermis).
  • endogenous skin tissue fluorophores e.g., nicotinamide adenine dinucleotide (NADH), flavin adenine dinucleotide (FAD), and collagen
  • NADH nicotinamide adenine dinucleotide
  • FAD flavin adenine dinucleotide
  • collagen collagen
  • the blue shift of the primary tissue emission peak ( ⁇ 467 nm) and narrowing of fluorescence emission spectra were also noticeable with increased probe-tissue contact pressure.
  • the significant changes (p ⁇ 0.0001, one-way AN OVA with Bonferroni correction at 5%) with the application of pressure were especially observed around 440-460 nm (p ⁇ 1X10- 8 ), 518-526 nm (p ⁇ 1X10" 4 ) and 540-570 nm (p ⁇ 1X10" 7 ), confirming that the observed AF variation may be primarily caused due to change in concentration of NADH (A em ⁇ 460nm), FAD (A em ⁇ 520nm), dermal collagen (A em ⁇ 460nm) and blood (i.e., oxy/deoxy-hemoglobin in blood, Aem ⁇ 540, 557, 577 nm).
  • the fingertip skin and volar forearm skin AF spectra measured under HP exhibited high spectral variance ( ⁇ 10 times) compared to LP, signifying that LP applied against the tissue causes considerably less changes in tissue optical properties, thereby trivial changes in measured AF spectra, in agreement with the reported literatures.
  • SE standard error
  • the leave-one patient-out, cross- validated PLS-DA multi-class diagnostic models were developed for the fingertip and volar forearm datasets using an optimum number of components, estimated based on the local minimum of cross-validation classification error values.
  • the optimum number of components was found to be 3 LVs for the fingertip and 1 LV for the volar forearm dataset as is seen in Figure 5, accounting for 99.47% and 83.39% of total AF spectral variations, respectively, for differentiating pressure induced changes in the measured skin tissue AF spectra.
  • the posterior probability values were further calculated for the fingertip and volar forearm datasets, using the developed PLS-DA multi-class models and shown as a 2-D ternary scatter plot of Figure 7.
  • the generated models for fingertip and volar forearm dataset classify the AF spectra measured under LP from MP + HP with accuracies of 78.97% and 89.89%; the spectra acquired at MP from LP + HP with accuracies of 66.72% and 34.04%; the spectra obtained at HP from LP + MP with accuracies of 57.76% and 61.67%, respectively.
  • Fiber-optic spectroscopy techniques such as reflectance, fluorescence and Raman spectroscopy have emerged as a compelling tool for disease diagnosis.
  • the recent technical advances including high-performance spectroscopy instrumentation, fiber-optic probe design, and chemometric techniques have enabled non-invasive real-time in vivo tissue diagnosis ( ⁇ 1 sec), translating the spectroscopy diagnostic methods into clinical healthcare.
  • These potential clinical spectroscopy modalities especially require hand-held fiber-optic probes to non-invasively acquire in vivo tissue spectra from remote internal organs.
  • the probe-tissue contact measurements provide relatively less spectral variability compared to non-contact measurement due to reduced refractive index mismatch.
  • the precise control of applied probe pressure is arduous in clinical scenario, particularly during endoscopic procedures.
  • the use of fiber-optic probes may induce spectral variations that can dilute the diagnostic variations present across tissue spectra and further lead to reduced diagnostic accuracy, raising potential concerns in incorporating spectroscopy modalities for the diagnosis of diseases in clinics
  • the present invention provides for the first time a novel pressure sensitive fiber-optic probe for quantitative monitoring of the magnitude of exerted probe-tissue contact pressure in real-time during in vivo tissue spectroscopic measurements.
  • the pressure sensitive fiber probe has been tested on in vivo skin tissue AF spectroscopy to characterize the AF spectral variations related to underlying physiological changes induced by varied probe pressure.
  • skin tissue blanches when pressure is applied. This is mitigated by use of the developed novel pressure sensitive spectroscopy platform.
  • the AF spectra measured from the fingertip and volar forearm seen in Figure 3 exhibits increased intensity for high probe pressure over the range 450-480nm with peak maximum at ⁇ 467 nm. Pressure-induced spectral variability was also observed around ⁇ 440-460, 518-526, and 540-570nm (p ⁇ 0.0001, one-way ANOVA with Bonferroni correction at 5%).
  • the pressure-induced enhancement in skin AF may be primarily associated with the increase of epithelial fluorescence (i.e., NADH fluorescence, A em ⁇ 460nm), because the penetration depth of 405 nm excited laser on skin tissue is ⁇ 50-100 ⁇ .
  • epithelial fluorophores such as free/bound NADH (A em ⁇ 460nm), and FADH (A em ⁇ 520nm) have also been reported due to reduction in metabolic activity caused by local ischemia produced at the tissue site under pressure.
  • the blood is squeezed out of the dermis while compressing the underlying tissue, leading to decreased blood absorption and increased AF signals that may arise from dermal collagen (A em ⁇ 460nm).
  • the AF changes that occur at 540-570 nm correspond to re-absorption of blood oxy/deoxy- hemoglobin and transition of oxy-hemoglobin to deoxy-hemoglobin with applied probe pressure onto the tissue.
  • interesting features such as minor blue- shift of AF emission peak (467 to 464 nm) due to pressure induced increase in rigidity of chromophore environment and narrowing of skin tissue AF spectra can also be noted with the increase of probe pressure.
  • the multivariate analysis based on PLS-DA diagnostic modeling further rendered accuracies of 66.72% and 34.04% for the finger and volar forearm, respectively, for discriminating AF spectra measured at MP from those measured at LP and HP, reconfirming the site-specific effect of probe-tissue contact pressure.
  • This site-specific effect of probe tissue-contact pressure may be due to increased concentration of tissue absorber melanin in the volar forearm, requiring HP to cause changes in its AF.
  • the AF spectra of the fingertip and volar forearm measured under LP can be discriminated with accuracies of 78.97% and 89.89%, respectively, from the spectra acquired with MP and HP, demonstrating that the native skin tissue AF changes significantly with the increase of exerted probe pressure.
  • pressure sensitive spectroscopy platform enables quantitative assessment of probe-tissue contact pressure in real-time during in vivo spectroscopy measurements and can be useful in realizing dynamic probe pressure control.
  • the real-time pressure sensitive fiber-optic probe further automates probe handling training (e.g., ensuring probe-tissue contact, maintaining constant LP during spectral acquisition) for clinicians for effective disease diagnosis and clinical decision making during spectral measurements.
  • probe handling training e.g., ensuring probe-tissue contact, maintaining constant LP during spectral acquisition
  • the developed pressure-sensitive fiber probe for real-time quantitative probe pressure monitoring can be easily adapted to other in vivo tissue optical spectroscopic (e.g., diffuse reflectance, Raman spectroscopy) applications, and is particularly appealing for challenging endoscopic applications in clinical settings.
  • the diagnostic efficacy of spectroscopy techniques can further be enhanced.
  • the present invention provides for the first time a novel pressure-sensitive fiber-optic probe applicable to various biomedical diagnostic spectroscopy platforms for realizing real-time ( ⁇ 1 sec) quantitative monitoring of in vivo probe-tissue contact pressure.
  • the successful utilization of the probe for characterizing probe pressure induced changes on in vivo AF demonstrates that the probe pressure effect is significant and tissue site-specific.
  • the integration of pressure sensitive fiber-optic probe with the rapid spectroscopy diagnostic technique greatly reduces the confounding sources of spectral variations, opening a new pathway for improving real-time in vivo tissue optical spectroscopy measurements for better tissue diagnosis and characterization.
  • the pressure sensitive fiber-optic probe has the potential to mitigate probe pressure induced alterations in tissue AF in real-time by providing instant feedback to the operators with quantitative probe-tissue contact pressure values during in vivo spectroscopic measurements.
  • This invention will greatly improve the real-time, in vivo optical spectroscopy diagnosis and characterization of disease tissue in clinical settings
  • This invention has no limitations in in vivo tissue diagnosis and characterizations.
  • the invention with miniaturized pressure sensors is not confined to the applications in the skin, but can also be extended to any other organs (e.g., gastric, esophagus, colorectal, lung, bladder, liver, breast, skin, cervix, etc.) in humans and other species.
  • organs e.g., gastric, esophagus, colorectal, lung, bladder, liver, breast, skin, cervix, etc.
  • the pressure sensitive in vivo spectroscopy platform utilized for real-time, quantitative monitoring of probe-tissue contact pressure as described above is adapted as follows.
  • the confocal probe is specially designed to collect signals particularly from the epithelial layer ( ⁇ 300 Um).
  • the hand-held fiber-optic probe is coupled to a customized piezo-resistive pressure sensor at the probe tip to realize real-time, dynamic probe pressure control and to facilitate quantitative monitoring of probe-tissue contact pressure.
  • the entire control of the system was implemented by a personal computer using in-house developed graphical user interface (GUI) under Matlab environment (Mathworks Inc., Natick, MA, USA).
  • GUI graphical user interface
  • Matlab environment Matlab environment
  • the acquired spectra were over the range of 800-1800 cm 1 with a spectral resolution of 9 cm 1 .
  • Each spectrum in this study was measured with an integration time of 1 sec under the 785 nm laser excitation.
  • the in vivo Raman spectra were acquired from index fingertip, palm and volar forearm of 20 healthy subjects using the pre-calibrated piezo-resistive pressure sensor coupled confocal fiber-optic probe.
  • LP low pressure
  • MP moderate pressure
  • HP high pressure
  • Probe pressure variations due to orientation of the probe and movement of target tissue site were avoided by fixing the target tissue perpendicular to the probe tip during spectral data acquisition.
  • Different probe pressures (LP, MP and HP) applied on to the target tissue site were displayed instantaneously in the developed real-time Raman diagnostic platform during spectroscopic measurements.
  • Spectral data can be acquired at constant probe pressures by making use of the real-time instantaneous probe pressure display during spectral data acquisition.
  • the acquired raw spectrum is a composite signal that was a combination of Raman signals, intense AF and noise.
  • the noise was suppressed by the use of first-order Savitsky-Golay smoothing filter (window width of 5 pixels).
  • a fifth - order polynomial fitting was found to be optimal for extracting the auto- fluorescence signal from the raw spectrum.
  • the extracted AF datasets from the fingertip, palm and volar forearm were subjected to multi-class probabilistic PLS-DA for differentiation among skin tissue spectra acquired at different pressure levels.
  • PLS is a two-block regression method that correlates the variations in the dataset with the response variable to explain the diagnostically relevant spectral variations associated with pressure variability in the first few latent variables (LVs).
  • _ENREF_27 The number of components (LVs) corresponding to minimum leave-one patient-out cross validation error were selected to build the optimal PLS-DA model. All the multivariate statistical analysis were performed in the Matlab (Mathworks Inc., Natick, MA, USA) scripting environment using the PLS toolbox (Eigenvector Research, Wenatchee, WA).
  • SE standard error
  • the MP (50 KPa) level caused an AF signal increase of 10.63%, 9.67% and 1.34% while the HP (130 KPa) level caused an increase of 25.43%, 13.88% and 9.21% for the fingertip, palm and volar forearm spectra respectively; suggesting that the extent of AF signal variation is dependent on the skin tissue type.
  • the differences in spectral signal intensities also indicate that variations in the exerted probe pressure affect the acquired AF spectra to a certain extent.
  • Multivariate statistical analysis i.e., PLS-DA was performed on the fingertip, palm and forearm spectral datasets to further explore spectral changes associated with operator-induced probe pressure variability.
  • Outlier analysis based on PCA coupled with Hotelling's T2 and Q-residual statistics was further utilized to remove the spectra with unusual variations (e.g., light interference).
  • PLS-DA together with leave-one patient-out, cross validation, was employed to realize in vivo discrimination of AF spectra acquired at the different reference pressure levels.
  • the optimum number of components i.e. latent variables- LVs
  • the optimum number of components i.e. latent variables- LVs
  • the optimum number of components was found to be 2 LV (LV1 - 99.72%; LV2 - 0.27%) for fingertip, 1 LV (LV1 - 99.73%) for palm and 3 LVs (LV1 - 99.98%; LV2 - 0.11%; LV3 - 0.01%) for volar forearm dataset, representing the variations in the AF spectral profiles which may be due to the changes in concentrations of the endogenous fluorophores with increase in applied probe pressure.
  • the first latent variable (LV1) which closely resembles the AF signals in the fingerprint region (800 - 1800 cm 4 ); contribute to the greatest variations in the AF signal (i.e., 99.72% - fingertip; 99.73% - palm and 99.88% - volar forearm) and is shown in Figure 10.
  • the generated PLS-DA models from the AF spectra of fingertip, palm and forearm provided correct classification rate of 52.69%, 62.19% and 53.70% for separating LP spectra from MP + HP; accuracies of 21.18%, 35.97% and 50.95% for differentiating MP spectra from LP + HP and accuracies of 50.93%, 50.95% and 34.19% for differentiating HP spectra from LP + MP respectively.
  • These classification results were plotted in a 2-D ternary scatter plot shown in Figure 11 for visualization of the classification results. From the results, it can be observed that the applied probe pressures produces notable variations ( ⁇ 50-60% discrimination between LP vs. MP + HP) in the AF spectroscopic properties of skin tissue.
  • the observed probe pressure variations on skin tissue AF spectra can be minimized by applying short-term ( ⁇ lsec) constant LP against the tissue which can be achieved by real-time quantitative monitoring of probe-tissue contact pressure using the developed pressure sensitive probe.
  • ⁇ lsec short-term
  • FIG 12 shows the schematic diagram of the pressure sensitive in vivo tissue diffuse reflectance (DR) spectroscopy platform developed for real-time, quantitative probe-tissue contact pressure measurement in accordance with an embodiment of the invention.
  • the spectroscopy system consists of a tungsten halogen lamp (LS1LL, Ocean Optics Inc., Dunedin, FL, USA), a transmissive imaging spectrograph (QE65000, Ocean Optics Inc., Dunedin, FL, USA) equipped with a back-thinned charge-coupled device (CCD) detector (S7031- 1006, 1024X58with pixel sizes of 24.6mm, QE > 90%, Hamamatsu, Shizuoka, Japan), and a customized piezo-resistive sensor coupled bifurcated fiber-optic probe.
  • CCD charge-coupled device
  • the hand-held fiber-optic probe is customized by coupling a customized piezo- resistive flexiforce pressure sensor (6.33 mm in sensing area diameter, 51 mm in length, 0.203 mm in thickness, Tekscan Inc., South Boston, MA, USA) to the probe tip to facilitate real-time, quantitative monitoring of probe pressure exerted on the tissue.
  • the DR spectra were acquired within the spectral bandwidth 375-1150 nm with an integration time of 10 ms and tungsten light incident power of O.lmW on tissue surface.
  • DR spectra measurements were carried out on the skin surface of ten healthy volunteers at three different measurement sites (the index fingertip, volar forearm and palm skin on abductor pollicis muscle). Desirable probe pressures (i.e., low pressure (LP) -lOKPa, moderate pressure (MP) - 50KPa and high pressure (HP) - 130KPa) were applied upon in vivo target tissue site by mounting the sensor coupled probe on a translation stage and controlling its movement appropriately [refj.
  • LP low pressure
  • MP moderate pressure
  • HP high pressure
  • the probe pressure was instantaneously displayed in the real-time diagnostic software, guiding the operator to acquire spectra at required pressure levels.
  • Multiple spectra ( ⁇ 4-5) were obtained from each site for all the three pressure levels to include inter- and/or intra-tissue variability for data analysis.
  • Multi-class probabilistic PLS-DA modeling was performed to realize in vivo discrimination among skin tissue spectra acquired at different pressure levels [refj.
  • the performance of PLS-DA model was validated using leave-one patient- out, cross validation methodology in which the spectral data of one subject is left out and the PLS-DA model is redeveloped from the remaining spectra.
  • the complexity of the developed PLS-DA model i.e., number of retained model LVs
  • the one-against- all multi-class terminology was further employed to classify the tissue spectra measured under different pressure levels.
  • the multivariate statistical analysis was performed using the PLS toolbox (Eigenvector Research, Wenatchee, WA) in the Matlab (Mathworks Inc., Natick, MA, USA) programming environment.
  • the custom designed pressure sensitive fiber optic probe was used to acquire high quality in vivo DR spectra in the spectral range 375-1150 nm from three measurement sites on the human hand (fingertip, volar forearm and palm).
  • SE standard error
  • the reference pressure levels were chosen such that it included clinically relevant probe pressures (LP and MP) as well as an extreme, intense high pressure level (HP) to study the pressure induced spectral distortions over a wide pressure range (0-150KPa).
  • the acquired DR spectra contained distinct characteristic minima (e.g. 420 nm (Soret Band), 540, 578, 750, 815, 920, 980, 1090 nm) that correspond to major endogenous chromophores like hemoglobin, fats, lipids, water and melanin, in agreement with existing literature.
  • the DR spectra exhibited an overall decreasing trend with the increase of applied probe pressure.
  • the dual peak of oxy-hemoglobin decreases in its height with applied probe pressure, indicating a gradual transition from oxy-hemoglobin to deoxy-hemoglobin.
  • the excitation peaks of porphyrins (400 nm), flavins (460 nm), lipo-pigments (435 nm) and melanin (1090 nm) also overlap with the peaks in the acquired DR spectra.
  • DR signal may be primarily due to pressure induced changes in the concentration of blood (i.e., oxy/deoxy-hemoglobin - A a b ⁇ 540, 560, 578 nm, fat - A a b ⁇ 750, 815, 900 nm, lipids - A a b ⁇ 920 nm and water - A a b ⁇ 980 nm) with minor contributions from other endogenous chromophores.
  • the fingertip, volar forearm and palm DR spectra measured under MP and HP exhibited considerably higher spectral variance compared to LP, signifying that LP applied against the tissue causes trivial changes in tissue optical properties, in agreement with the reported literatures.
  • the DR spectra were further analyzed using multivariate statistical algorithms for in-depth understanding of pressure induced DR spectral distortions and to differentiate DR spectra acquired at different pressure levels.
  • PLS-DA together with leave-one patient-out, cross validation, was employed for classification of the DR spectra based on the applied probe pressure during spectral data acquisition.
  • Outlier analysis based on PCA coupled with Hotelling's T2 and Q- residual statistics was further utilized to remove the spectra with unusual variations (e.g., light interference).
  • the leave-one patient-out, cross-validated PLS-DA multi-class diagnostic modeling was performed on the fingertip, volar forearm and palm DR datasets using optimum number of components, estimated based on the local minimum of cross-validation classification error values.
  • the optimal number of components differed for each measurement site and was found to be 1 (fingertip), 4 (volar forearm) and 1 (palm) as shown in Figure 14.
  • the principal components accounted for 92.95% (fingertip), 99.32% (volar forearm) and 91.27% (palm) of total DR spectral variations for differentiating pressure induced changes in the measured skin tissue DR spectra.
  • the loading vectors (see Figure 15) clearly show spectral changes that occur at prominent wavelengths located around 420, 540, 578, 750, 815, 920, 980 and 1090 nm of the acquired reflectance signal which are associated with blood (oxy/deoxy-hemoglobin), water, fats and lipid absorption spectra.
  • the posterior probability values were further calculated for each dataset, using the developed PLS-DA multi-class models and are shown as a 2-D ternary scatter plot ( Figure 16).
  • the generated models for fingertip, volar forearm and palm dataset classify the DR spectra measured under LP from MP + HP with accuracies of 82.11%, 82.40% and 76.01%; the spectra acquired at MP from LP + HP with accuracies of 49.48%, 64.90% and 52.00%; the spectra obtained at HP from LP + MP with accuracies of 76.39%, 81.95% and 67.34%, respectively.
  • the above results show that there is a distinct separation between LP and HP spectra at the three measurement sites.
  • the variation in the accuracy for the differentiation of MP spectra among the measurement sites show the site- specific nature of the pressure induced spectral distortions. Thus, it is highly imperative to monitor the applied probe pressure in real-time to reduce the pressure induced distortions to a great extent.
  • the pressure sensitive in vivo tissue confocal Raman spectroscopy platform utilized for real-time, quantitative monitoring of probe-tissue contact pressure has been reported in detail elsewhere.
  • NIR-coated sapphire ball lens 5 mm in diameter, refr
  • the probe was integrated with optical filtering modules for suppressing laser noise, Rayleigh scattered light, while allowing only the frequency-shifted tissue Raman signal to pass towards the spectrograph.
  • the hand-held fiber-optic probe is coupled to a customized piezo-resistive pressure sensor at the probe tip to facilitate real-time, quantitative monitoring of probe pressure exerted onto the tissue.
  • the entire control of the system was implemented by a personal computer using in-house developed graphical user interface (GUI) under Matlab environment (Mathworks Inc., Natick, MA, USA).
  • GUI graphical user interface
  • the system acquires Raman spectra over the range of 800-1800 cm " -*- with a spectral resolution of 9 cm " -'-.
  • Each Raman spectrum in this study was measured with an integration time of 1 sec under the 785 nm laser excitation.
  • the in vivo Raman spectra were acquired from index fingertip, palm and volar forearm of healthy volunteers using the pre-calibrated piezo-resistive pressure sensor coupled confocal fiber-optic probe.
  • the tissue sites to be measured were fixed perpendicular to probe tip to avoid movement of tissue site and variations in exerted probe pressure due to orientation of probe during Raman spectral acquisition.
  • probe pressures i.e., low pressure (LP) -lOKPa, moderate pressure (MP) - 50KPa and high pressure (HP) - 130KPa
  • LP low pressure
  • MP moderate pressure
  • HP high pressure
  • the real-time display of probe-tissue contact pressure can guide the operator to maintain exactly the same magnitude of pressure (i.e., LP, MP or HP) on to the target tissue during in vivo tissue Raman spectral acquisition.
  • the in-house developed GUI triggers data acquisition and analysis including laser power control, spectrometer, CCD shutter and camera readout synchronization, CCD dark-noise subtraction, probe background subtraction, outlier detection, wavelength calibration, system spectral response calibration, auto fluorescence subtraction, normalization, and real-time display of in vivo skin tissue Raman spectra together with the magnitude of applied probe-tissue contact pressure.
  • Multi-class probabilistic PLS-DA was further performed on the dataset to realize in vivo discrimination among skin tissue spectra acquired at different pressure levels.
  • PLS is a two-block regression method that correlates the variations in the dataset with the response variable to explain the diagnostically relevant spectral variations associated with pressure variability in the first few components (i.e., latent variables (LVs)).
  • _ENREF_27 The number of components (LVs) corresponding to minimum leave- one patient-out cross validation error were selected to build optimal PLS-DA models. All the multivariate statistical analysis were performed in the Matlab (Mathworks Inc., Natick, MA, USA) scripting environment using the PLS toolbox (Eigenvector Research, Wenatchee, WA).
  • SD standard deviation
  • Raman spectra acquired on various sites of skin tissue indicated differences in spectral properties associated with variability in applied probe pressure ( Figure 17 (b), 1(d), and 1(e)).
  • Raman spectra acquired from fingertip expressed significant changes only for HP ((p ⁇ 0.005, one-way AN OVA with Bonferroni correction at 5%). Palm showed substantial changes (p ⁇ 0.05, one-way ANOVA with Bonferroni correction at 5%) with probe exerted pressure (i.e., MP and HP) around the major Raman peaks particularly related to proteins -1244-1247, 1275-1295, 1335-1360, 1400-1420, 1617-1630 and 1650-1680 cm -1 .
  • the spectral changes associated with probe pressure variability were found to be negligible (p>0.05) for volar forearm.
  • the above pressure-related changes observed in Raman active components of skin tissue reveals that the probe pressure effects are site-specific.
  • Multivariate PLS-DA models were further generated from fingertip, palm and forearm spectral dataset to further explore these site-specific spectral changes associated with operator-induced probe pressure variability.
  • the measured Raman spectra were first mean-centered and spectrum quality was assured with outlier analysis (i.e., principal component analysis (PCA) coupled with Hotelling's T2 and Q-residual statistics), where Hotelling's T2 represents the major variations in the data and Q-residuals represents the random noise or variations that are not present in the dataset (e.g., new variations).
  • PCA principal component analysis
  • Hotelling's T2 represents the major variations in the data
  • Q-residuals represents the random noise or variations that are not present in the dataset (e.g., new variations).
  • the spectra with unusual variations were successfully removed from the spectral database of three skin tissue sites.
  • the PLS-DA multi-class models together with leave-one patient-out, cross-validation were further developed for fingertip, palm and volar forearm dataset using optimum number of components.
  • the optimum number of components was found to be 1 LV (14.23%) for fingertip, 5 LVs (LV1 - 25.60%; LV2 - 13.64%; LV3 - 3.25%; LV4 - 3.93%; LV5 - 2.52%) for palm and 2 LVs (LV1 - 25.13%; LV2 - 7.43%) for volar forearm dataset (Figure 18), representing the variations around Raman peaks (-853, 936, 1004, 1012, 1260, 1313, 1345, 1451, 1468, 1618, 1642, 1656, and 1665 cm -1 ) corresponding to major skin tissue Raman-active biomolecules (Figure 19).
  • the observed small probe pressure variations on skin tissue Raman spectra can further be eliminated by applying short-term ( ⁇ lsec) LP against the tissue through monitoring the probe-tissue contact pressure in real-time using the developed pressure sensitive Raman probe, moving the capability of novel pressure sensitive Raman spectroscopy one step forward in disease diagnosis by preserving only the diagnostically relevant variations.

Landscapes

  • Health & Medical Sciences (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Physics & Mathematics (AREA)
  • General Health & Medical Sciences (AREA)
  • Pathology (AREA)
  • Chemical & Material Sciences (AREA)
  • Analytical Chemistry (AREA)
  • Biochemistry (AREA)
  • General Physics & Mathematics (AREA)
  • Immunology (AREA)
  • Biophysics (AREA)
  • Engineering & Computer Science (AREA)
  • Biomedical Technology (AREA)
  • Heart & Thoracic Surgery (AREA)
  • Medical Informatics (AREA)
  • Molecular Biology (AREA)
  • Surgery (AREA)
  • Animal Behavior & Ethology (AREA)
  • Public Health (AREA)
  • Veterinary Medicine (AREA)
  • Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
  • Spectroscopy & Molecular Physics (AREA)
  • Investigating, Analyzing Materials By Fluorescence Or Luminescence (AREA)

Abstract

L'invention concerne un procédé et un système pour obtenir des données biomédicales à des fins de diagnostic, le procédé consistant à : irradier un échantillon de tissu avec une source de rayonnement ; recevoir un spectre collecté à partir de l'échantillon de tissu en conséquence de l'irradiation de l'échantillon de tissu à partir d'un spectrographe ; délivrer un rayonnement à l'échantillon, et collecter et renvoyer le spectre au spectrographe par l'intermédiaire d'une sonde sensible à la pression ; mesurer la pression exercée par la sonde sur l'échantillon de tissu à l'aide d'un capteur de pression ; ajuster la pression pour être à une valeur prédéterminée lorsque le spectre est collecté à l'aide d'un module d'ajustement de pression.
PCT/SG2015/050289 2014-09-01 2015-08-31 Sonde à fibre optique sensible à la pression pour spectroscopie optique de tissu in vivo en temps réel, système l'incorporant, et son procédé d'utilisation Ceased WO2016036314A1 (fr)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
GBGB1415439.7A GB201415439D0 (en) 2014-09-01 2014-09-01 A pressure-sensitive fiber optic probe for real-time in vivo tissue optical spectroscopy,a system incorporating the same and a method for using the same
GB1415439.7 2014-09-01

Publications (1)

Publication Number Publication Date
WO2016036314A1 true WO2016036314A1 (fr) 2016-03-10

Family

ID=51752430

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/SG2015/050289 Ceased WO2016036314A1 (fr) 2014-09-01 2015-08-31 Sonde à fibre optique sensible à la pression pour spectroscopie optique de tissu in vivo en temps réel, système l'incorporant, et son procédé d'utilisation

Country Status (2)

Country Link
GB (1) GB201415439D0 (fr)
WO (1) WO2016036314A1 (fr)

Cited By (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108872142A (zh) * 2018-06-19 2018-11-23 温州大学 一种波长选择算法中多参数的选择优化方法
CN109901279A (zh) * 2019-02-25 2019-06-18 桂林电子科技大学 基于同轴三波导光纤的微球自组装激光器
US11147453B2 (en) 2017-10-03 2021-10-19 Canon U.S.A., Inc. Calibration for OCT-NIRAF multimodality probe
CN114795123A (zh) * 2022-04-25 2022-07-29 四川省肿瘤医院 放疗患者皮肤辐射损伤预测系统及方法
WO2024228012A1 (fr) * 2023-05-03 2024-11-07 The University Of Warwick Système de balayage térahertz
US12336844B2 (en) 2018-09-05 2025-06-24 The University Of Nottingham Monitoring physiological parameters
WO2025252689A1 (fr) 2024-06-05 2025-12-11 Trinamix Gmbh Dispositif spectroscopique
CN114795123B (zh) * 2022-04-25 2026-01-30 四川省肿瘤医院 放疗患者皮肤辐射损伤预测系统及方法

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20040254457A1 (en) * 2003-06-02 2004-12-16 Van Der Weide Daniel Warren Apparatus and method for near-field imaging of tissue
US20090204009A1 (en) * 2008-02-07 2009-08-13 Los Alamos National Security Medical device system and related methods for diagnosing abnormal medical conditions based on in-vivo optical properties of tissue
US20090234206A1 (en) * 2007-11-27 2009-09-17 Sanna Gaspard Medical device for diagnosing pressure ulcers
WO2012039679A2 (fr) * 2010-09-22 2012-03-29 National University Of Singapore Système de mesure d'autofluorescence dans le proche infrarouge chez un sujet, et procédé associé
WO2013049677A1 (fr) * 2011-09-30 2013-04-04 The Trustees Of Columbia University In The City Of New York Procédés, et systèmes et dispositifs optiques compacts d'imagerie
US20130100439A1 (en) * 2009-12-04 2013-04-25 Duke University Smart fiber optic sensors systems and methods for quantitative optical spectroscopy

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20040254457A1 (en) * 2003-06-02 2004-12-16 Van Der Weide Daniel Warren Apparatus and method for near-field imaging of tissue
US20090234206A1 (en) * 2007-11-27 2009-09-17 Sanna Gaspard Medical device for diagnosing pressure ulcers
US20090204009A1 (en) * 2008-02-07 2009-08-13 Los Alamos National Security Medical device system and related methods for diagnosing abnormal medical conditions based on in-vivo optical properties of tissue
US20130100439A1 (en) * 2009-12-04 2013-04-25 Duke University Smart fiber optic sensors systems and methods for quantitative optical spectroscopy
WO2012039679A2 (fr) * 2010-09-22 2012-03-29 National University Of Singapore Système de mesure d'autofluorescence dans le proche infrarouge chez un sujet, et procédé associé
WO2013049677A1 (fr) * 2011-09-30 2013-04-04 The Trustees Of Columbia University In The City Of New York Procédés, et systèmes et dispositifs optiques compacts d'imagerie

Cited By (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US11147453B2 (en) 2017-10-03 2021-10-19 Canon U.S.A., Inc. Calibration for OCT-NIRAF multimodality probe
CN108872142A (zh) * 2018-06-19 2018-11-23 温州大学 一种波长选择算法中多参数的选择优化方法
CN108872142B (zh) * 2018-06-19 2020-12-22 温州大学 一种波长选择算法中多参数的选择优化方法
US12336844B2 (en) 2018-09-05 2025-06-24 The University Of Nottingham Monitoring physiological parameters
CN109901279A (zh) * 2019-02-25 2019-06-18 桂林电子科技大学 基于同轴三波导光纤的微球自组装激光器
CN114795123A (zh) * 2022-04-25 2022-07-29 四川省肿瘤医院 放疗患者皮肤辐射损伤预测系统及方法
CN114795123B (zh) * 2022-04-25 2026-01-30 四川省肿瘤医院 放疗患者皮肤辐射损伤预测系统及方法
WO2024228012A1 (fr) * 2023-05-03 2024-11-07 The University Of Warwick Système de balayage térahertz
WO2025252689A1 (fr) 2024-06-05 2025-12-11 Trinamix Gmbh Dispositif spectroscopique

Also Published As

Publication number Publication date
GB201415439D0 (en) 2014-10-15

Similar Documents

Publication Publication Date Title
Lim et al. Probe pressure effects on human skin diffuse reflectance and fluorescence spectroscopy measurements
Pence et al. Clinical instrumentation and applications of Raman spectroscopy
JP5574246B2 (ja) 生物組織中のカロテノイドの非侵襲的測定
JP6230077B2 (ja) 光ファイバーラマン分光法を利用する内視鏡検査でのリアルタイム癌診断に関連する方法
WO2016036314A1 (fr) Sonde à fibre optique sensible à la pression pour spectroscopie optique de tissu in vivo en temps réel, système l'incorporant, et son procédé d'utilisation
US20140330098A1 (en) Reflectance calibration of fluorescence-based glucose measurements
US20060247532A1 (en) Method for extraction of optical properties from diffuse reflectance spectra
CN102088906A (zh) 利用组织荧光确定糖基化终产物或疾病状态的测量的改进的方法和装置
JP2013514520A (ja) ラマン分光法によるインビボでの組織の特徴付けのための装置および方法
Zhao et al. Real-time Raman spectroscopy for noninvasive in vivo skin analysis and diagnosis
WO2013073270A1 (fr) Dispositif de mesure, procédé de mesure, programme, et support d'enregistrement
CN104603601A (zh) 与拉曼光谱学有关的诊断仪器和方法
EP4175538A1 (fr) Dispositif non invasif de mesure de la glycémie
Zhang et al. Subcutaneous depth-selective spectral imaging with mμSORS enables noninvasive glucose monitoring
Saeed et al. Hyperspectral retinal imaging biomarkers of ocular and systemic diseases
WO1999043255A1 (fr) Spectroscopie de transmission a infrarouge proche sur tissu de la langue
US20090198113A1 (en) Dedicated spectral illumination spectroscopy
JP2019219375A (ja) スペクトルイメージ補正装置及び方法、ならびに対象体の成分分析装置
US20180000390A9 (en) System and method for non-invasive glucose monitoring using near infrared spectroscopy
US20060063991A1 (en) Method and apparatus for non-invasive measurement of blood analytes with dynamic spectral calibration
CN109984725B (zh) 漫反射测量中接触压力干扰抑制方法、装置及测量方法
CN113974618B (zh) 基于水峰血糖修正的无创血糖测试方法
Deng et al. Highly sensitive imaging spectrometer system based on areal array to linear array optical fiber probe for biological spectral detection
Zhou et al. Influence of water on noninvasive hemoglobin measurement by Dynamic Spectrum
Leizaola et al. Objective assessment of clinical debridement and healing status of diabetic foot ulcers using a smartphone-based NIRS device

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 15838108

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 15838108

Country of ref document: EP

Kind code of ref document: A1