Search

EP-4740000-A1 - PATHOGEN DETECTION IN BIOFLUID SAMPLES USING SPECTROSCOPY TECHNIQUES COUPLED WITH ANALYTICAL MODELS

EP4740000A1EP 4740000 A1EP4740000 A1EP 4740000A1EP-4740000-A1

Abstract

A system and method of performing pathogen detection from a collected biofluid sample is disclosed. The biofluid is subjected to one or more spectroscopy processes, which generates spectral data that may include specific biomarkers. The spectral data is then analyzed utilizing machine learning models trained on detection of specific pathogens to rapidly obtain result with a high degree of accuracy. The system may be configured as a partable assembly allowing for point-of-contact presentation of results to a patient in a matter of minutes. The sample preparation requires no follow-on processing of the sample (which could introduce contamination); rather, the collected biofluid is directly used to perform the spectroscopic procedure. Machine learning algorithms are trained on labelled data, which allows for an analysis of the submitted sample to look for the pathogen of concern. The machine learning models may be embedded within the assembly, or accessed via a secure communication link.

Inventors

  • PATHAK, SOHAM
  • SHASTRI, Ankita
  • HIRSCH, Pierre Martin
  • Rehman, Ihtesham Ur
  • DALAL, Sharvari Harsh
  • KUMAR, NEHA
  • PRENDIVILLE, WALTER

Assignees

  • NSV, Inc.

Dates

Publication Date
20260513
Application Date
20240705

Claims (20)

  1. 1. A diagnostic system for analyzing spectral data of biofluid samples to detect specific pathogens, comprising sample acquisition and preparation apparatus for collecting a biofluid sample; spectroscopy apparatus for illuminating the acquired biofluid sample and creating spectral data from the illumination; and a machine learning model processing unit responsive to the acquired biofluid sample, wherein the machine learning model processing unit includes one or more trained models to detect the presence or absence of one or more selected pathogens in the acquired spectral data.
  2. 2. The diagnostic system as defined in claim 1 wherein the collected biofluid is selected from the set comprising: blood, urine, saliva, serum, and plasma.
  3. 3. The diagnostic system as defined in claim 1 wherein the biofluid is collected using a non-invasive process.
  4. 4. The diagnostic system as defined in claim 1 wherein the sample acquisition and preparation apparatus utilizes a hygienic flow configuration to provide sample movement from collection to preparation to minimize introduction of any contaminant.
  5. 5. The diagnostic system as defined in claim 1 wherein the sample acquisition and preparation apparatus includes a desiccation chamber for drying a slide-mounted biofluid sample.
  6. 6. The diagnostic system as defined in claim 5 wherein the spectroscopy apparatus is utilized to scan for water peaks in the slide-mounted biofluid sample to determine if additional drying is required.
  7. 7. The diagnostic system as defined in claim 1 wherein the sample acquisition and preparation apparatus is configured to prepare a plurality of samples from the collected biofluid for independent analysis.
  8. 8. The diagnostic system as defined in claim 7 wherein at least two different preparation methods are used on individual samples from within the plurality of samples.
  9. 9. The diagnostic system as defined in claim 1 wherein the spectroscopy apparatus includes a Fourier Transform Infrared Spectrometer.
  10. 10. The diagnostic system as defined in claim 1 wherein the spectroscopy apparatus includes a Raman spectrometer.
  11. 11. The diagnostic system as defined in claim 1 wherein the machine learning model processing unit comprises a non-transitory, computer-readable storage medium for storing instructions that, when executed by circuitry included within the processing unit, cause the circuitry to perform analysis of presented spectral data utilizing one or more machine learning models stored in a memory module of the processing unit.
  12. 12. The diagnostic system as defined in claim 11 wherein the machine learning models stored in the memory unit include one or more of models based on SVM and logistic regression-based models, together with PCA, ANNs and CNNs, where one or a combination of these different models are developed for a given set of pathogens under study.
  13. 13. The diagnostic system as defined in claim 11 wherein the machine learning models provide the capability to detect different strains of a selected pathogen.
  14. 14. The diagnostic system as defined in claim 1 wherein at least the sample acquisition and preparation apparatus and the spectroscopy apparatus are formed as a single assembly, providing a portable diagnostic system.
  15. 15. The diagnostic system as defined in claim 14 wherein the single assembly further comprises the machine learning model processing unit, which embodies a plurality of downloaded machine learning models useful for pathogen detection.
  16. 16. The diagnostic system as defined in claim 14 wherein the machine learning model processing unit comprises a separate assembly configured to communicate with the single assembly.
  17. 17. A method of performing analysis of spectral data from presented biofluid samples to detect specific pathogens, the method including the steps of: a) performing a non-invasive sample collection process to acquire a selected biofluid from a patient; b) preparing a controlled amount of the selected biofluid from a cartridge for further processing; c) utilizing a spectrometer to illuminate the prepared sample and generate spectral data therefrom; and d) applying one or more machine learning models trained on detection of pathogens in biofluid to analyze the spectral data for detecting the presence or absence of a specific pathogen in the generated spectral data.
  18. 18. The method as defined in claim 17, further comprising the step of: repeating steps a) - d) over a period of time to monitor a response to a treatment of a detected pathogen.
  19. 19. The method as defined in claim 17, wherein the selected biofluid comprises urine and the selected pathogen is one of hr-HPV and Ir-HPV.
  20. 20. The method as defined in claim 17, wherein the selected biofluid comprises saliva and the selected pathogen is one of core pre-cancer and oral cancer.

Description

PATHOGEN DETECTION IN BIOFLUID SAMPLES USING SPECTROSCOPY TECHNIQUES COUPLED WITH ANALTYICAL MODELS Cross-Reference to Related Applications This application claims priority from U.S. Provisional Application 63/525,697 filed July 9, 2023, and U.S. Provisional Application 63/553,862 filed February 15, 2024, both of which are herein incorporated by reference. Technical Field The disclosed principles are directed to the detection of pathogens in biofluid samples and, more particularly, to the use of spectroscopic imaging of samples obtained from non-invasive biofluids (such as urine, saliva, blood, etc.) in combination with machine learning techniques to rapidly and accurately identify the presence of a specific pathogen. Background of the Invention Cervical cancer is the fourth most common cancer in women across the world. Cervical cancer is caused by high-risk human papillomavirus (hr-HPV). The World Health Organization (WHO) asserts that cervical cancer can be eliminated through its three pillar strategy: HPV vaccination; screening and treatment of pre-cancerous disease; and treatment of invasive disease. HPV vaccination prevents infection, but not in those already exposed to the virus. Thus, screening will continue to play a vital role in eliminating cervical cancer for the foreseeable future. In the UK, cervical screening has helped to reduce deaths from cervical cancer by 70% since its introduction. Yet despite its success, participation in cervical screening has declined over the years. In 2021, just 70.2% of eligible individuals were screened in the UK, which is an all-time low. Barriers to cervical screening include embarrassment, inconvenience and discomfort. To address some of these barriers, vaginal self-sampling for HPV detection has been developed. This is now available as an alternative method for cervical screening in at least 17 countries, either for under-screened populations or as a primary screening option. The effectiveness of vaginal sampling for improving screening uptake varies in different studies, ranging from 6-30%. Urine self-testing procedures have been demonstrated to have diagnostic accuracy similar to vaginal self-testing procedures, approaching that of clinician acquired cervical sampling. Urine testing is non-invasive, does not require a health worker to acquire the specimen and does not require a health care facility to house clinic examination rooms to take cervical specimens. Urine testing is perceived by women to be less 'invasive' than vaginal self testing. The need to improve laboratory diagnostic capacity at national and community levels of the healthcare system is an essential part of management of viral and bacterial infections. As discussed in detail below, it is contemplated that spectroscopy techniques (including both Infrared (IR) and Raman spectroscopy) allow analysis of bacteria and viruses with good specificity and sensitivity. IR spectroscopy is a form of vibrational spectroscopy that relies on the absorbance, transmittance, or reflectance of infrared light when a sample is illuminated. A particular type of IR spectroscopy, Fourier transform IR (FTIR) spectroscopy, is a preferable process in that it is able to simultaneously collect data across the complete IR wavelength range of interest. FTIR is considered to be a powerful tool for chemical analysis because of its ability to provide detailed information on the chemical composition of the constituents at the molecular level, such as proteins, nucleic acids, carbohydrates and lipids. Essentially, this provides a bio-chemical fingerprint of chemical and biomolecular structures in a variety of environments, for example in biofluids. IR spectroscopy takes advantage of the fact that chemical bonds or functional groups within the molecule vibrate at characteristic frequencies. IR spectroscopy mainly deals with the IR region of the electromagnetic spectrum and most commonly focuses on absorption, i.e. when the frequency of the IR radiation is the same as the vibrational frequency of a bond, absorption occurs. The wavelengths that are absorbed by the sample, represented as bands in the spectrum, are characteristic of its molecular structure. The use of IR methods in clinical practice is increasing due to the technology's rapid, cost-effective, and accurate disease prediction. To date, IR spectroscopy of biofluids has shown high sensitivity and specificity for the diagnosis of dementia, brain cancer and endometrial cancer in large cross-sectional studies. Others have demonstrated the potential of spectroscopy of liquid-based cytology samples as a triage to stratify women who were HPV-positive. Early proof-of-concept studies demonstrated similar potential for urine in the detection of gynecological cancers. Raman spectroscopy is a somewhat different technique, where it is used to measure the relative frequencies at which a biofluid sample scatters radiation. Raman spectroscopy is able to be used directly with a bio