WO2023275527A1 - Method and apparatus using absorption spectroscopy for discrimination of tissue or cells - Google Patents

Abstract

A method of discriminating between first cell types or first tissue types and second cell types or second tissue types is described. At S601, the method comprises providing a cell sample or a tissue sample including a first spatial region, comprising a first cell type or a first tissue type, and a second spatial region. At S602, the method comprises obtaining a first absorption response comprising: illuminating the first spatial region using emitted infrared, IR, electromagnetic radiation, EMR, incident thereupon, wherein the IR EMR comprises a set of pairs of discrete wavelengths, including a first pair of discrete wavelengths consisting of a first wavelength and a second wavelength, wherein the first pair of discrete wavelengths is characteristic of a discriminating biomarker; and detecting respective absorbances of the incident first wavelength and the incident second wavelength of the first pair of discrete wavelengths by the first spatial region. At S603, the method comprises obtaining a second absorption response comprising: illuminating the second spatial region using the IR EMR; and detecting respective absorbances of the incident first wavelength and the incident second wavelength of the first pair of discrete wavelengths by the second spatial region. At S604, the method comprises discriminating between the first spatial region and the second spatial region based on a result of comparing the obtained first absorption response and the obtained second absorption response.

Description

METHOD AND APPARATUS USING ABSORPTION SPECTROSCOPY FOR DISCRIMINATION OF TISSUE OR CELLS

Field The present invention relates to a method of and an apparatus for of discriminating between first cell types or first tissue types and second cell types or second tissue types.

Background to the invention Absorption spectroscopy is an important tool in the diagnosis of disease. For example, Fourier transform infrared (FTIR) spectroscopy and Raman spectroscopy are routinely applied to studies of cancer. Known methods of analysing the acquired spectral data use techniques such as principal component analysis and random forest analysis. These techniques involve the identification of radiative absorption ‘fingerprints’ for characterizing different samples. When used for identifying the presence or absence of a given cell type or tissue type (e.g. a cancerous cell type or tissue type) in a sample from a patient, these techniques are known to result in false positives and/or fail to identify the presence of the cell type or tissue type. Such failures can be very damaging to the patient and may result in unnecessary and potentially harmful treatment being performed and/or a lack of treatment when treatment is needed. Furthermore, FTIR spectroscopy and Raman spectroscopy are not amenable to point of care (POC) applications, due to a cost, complexity and size of the instruments as well as the requirement for specifically-trained operators. Hence, early stage disease detection and monitoring of progression of disease and/or treatment is generally not possible. Hence, there is a need to improve methods of and apparatuses for diagnosis of disease.

Summary of the Invention

It is one aim of the present invention, amongst others, to provide a method and an apparatus which at least partially obviate or mitigate at least some of the disadvantages of the prior art, whether identified herein or elsewhere. For instance, it is an aim of embodiments of the invention to provide a method of discriminating between first cell types or first tissue types and second cell types or second tissue types that reduces a false positive rate and/or false negative rate of discrimination therebetween. For instance, it is an aim of embodiments of the invention to provide an apparatus for discriminating between first cell types or first tissue types and second cell types or second tissue types that may be used point of care, POC, applications. A first aspect provides a method of discriminating between first cell types or first tissue types and second cell types or second tissue types, the method comprising: providing a cell sample or a tissue sample including a first spatial region, comprising a first cell type or a first tissue type, and a second spatial region; obtaining a first absorption response comprising: illuminating the first spatial region using emitted infrared, IR, electromagnetic radiation, EMR, incident thereupon, wherein the IR EMR comprises a set of pairs of discrete wavelengths, including a first pair of discrete wavelengths consisting of a first wavelength and a second wavelength, wherein the first pair of discrete wavelengths is characteristic of a discriminating biomarker; and detecting respective absorbances of the incident first wavelength and the incident second wavelength of the first pair of discrete wavelengths by the first spatial region; obtaining a second absorption response comprising: illuminating the second spatial region using the IR EMR; and detecting respective absorbances of the incident first wavelength and the incident second wavelength of the first pair of discrete wavelengths by the second spatial region; and discriminating between the first spatial region and the second spatial region based on a result of comparing the obtained first absorption response and the obtained second absorption response.

A second aspect provides an apparatus for discriminating between first cell types or first tissue types and second cell types or second tissue types, the apparatus comprising: a set of lasers, including a first laser tuned to emit a first wavelength of infrared, IR, electromagnetic radiation, EMR, of a first pair of discrete wavelengths of a set of pairs of discrete wavelengths and a second laser tuned to emit a second wavelength of IR EMR of the first pair of discrete wavelengths, wherein the first pair of discrete wavelengths is characteristic of a discriminating biomarker; a probe, optically coupled to the set of lasers, for illuminating a cell sample or a tissue sample including a first spatial region, comprising a first cell type or first tissue type, and a second spatial region, using the first wavelength and the second wavelength of the first pair of discrete wavelengths incident thereupon; a detector, optically coupled to the probe, for detecting respective absorbances of the incident first wavelength and the incident second wavelength of the first pair of discrete wavelengths by the first spatial region and by the second spatial region; and a controller configured to control the detector to obtain a first absorption response from the first spatial region and a second absorption response from the second spatial region; and to discriminate between the first spatial region and the second spatial region based on a result of comparing the obtained first absorption response and the obtained second absorption response. A third aspect provides a computer comprising a processor and a memory configured to implement a method according to the first aspect.

A fourth aspect provides a computer program comprising instructions which, when executed by a computer comprising a processor and a memory, cause the computer to perform a method according to the first aspect.

A fifth aspect provides a non-transient computer-readable storage medium comprising instructions which, when executed by a computer comprising a processor and a memory, cause the computer to perform a method according to the first aspect.

Detailed Description of the Invention

According to the present invention there is provided a method, as set forth in the appended claims. Also provided is an apparatus. Other features of the invention will be apparent from the dependent claims, and the description that follows.

Method The first aspect provides a method of discriminating between first cell types or first tissue types and second cell types or second tissue types, the method comprising: providing a cell sample or a tissue sample including a first spatial region, comprising a first cell type or a first tissue type, and a second spatial region; obtaining a first absorption response comprising: illuminating the first spatial region using emitted infrared, IR, electromagnetic radiation, EMR, incident thereupon, wherein the IR EMR comprises a set of pairs of discrete wavelengths, including a first pair of discrete wavelengths consisting of a first wavelength and a second wavelength, wherein the first pair of discrete wavelengths is characteristic of a discriminating biomarker; and detecting respective absorbances of the incident first wavelength and the incident second wavelength of the first pair of discrete wavelengths by the first spatial region; obtaining a second absorption response comprising: illuminating the second spatial region using the IR EMR; and detecting respective absorbances of the incident first wavelength and the incident second wavelength of the first pair of discrete wavelengths by the second spatial region; and discriminating between the first spatial region and the second spatial region based on a result of comparing the obtained first absorption response and the obtained second absorption response. In this way, presence or absence of the second cell type or the second tissue type, for example a diseased cell type or a diseased tissue type such as a cancerous cell type or a cancerous tissue type, in the second spatial region of the cell sample or the tissue sample may be identified, for example for POC applications. Particularly, the cell type or tissue type of the second spatial region (i.e. a sample region) is compared with the first cell type or the first tissue type, for example a healthy cell type or a healthy tissue type, of the first spatial region (i.e. a control or reference region) of the cell sample or the tissue sample by the discriminating, which is based on the result of comparing the obtained first absorption response and the obtained second absorption response. Notably, the first absorption response and the second absorption response are respectively obtained for the first pair of discrete wavelengths of the IR EMR, thereby providing respective pairwise correlations of the first wavelength and the second wavelength for the first spatial region and the second spatial region i.e. based on respective absorbances of the same pair of discrete wavelengths for each region. The first pair of discrete wavelengths is characteristic of the discriminating biomarker. It should be understood that the biomarker discriminates between the first cell type or the first tissue type and the second cell type or the second tissue type. For example, the biomarker may be upregulated, downregulated, present or absent in the second cell type or the second tissue type relative to the first cell type or the first tissue type, thereby enabling the first cell type or the first tissue type and the second cell type or the second tissue type to be mutually discriminated (i.e. distinguished, differentiated, classified). Since the first wavelength and the second wavelength are discrete, rather than of a continuum of wavelengths or scanned wavelengths of spectral analysis, a false positive rate and/or false negative rate of discrimination between the first cell type or first tissue type and the second cell type or second tissue type is reduced because an accuracy and/or precision of the first wavelength and the second wavelength is increased, compared with spectral analysis. Conversely, a true positive rate and/or a true negative rate of discrimination between the first cell type or first tissue type and the second cell type or second tissue type is improved, mutatis mutandis. The inventors have observed that successful discrimination requires detection of absorbance of reproducible pairs of discrete wavelengths of the IR EMR. While conventional spectral analysis, for example using broadband IR EMR or by wavelength scanning of IR EMR, may provide pairs of continuum wavelengths of the IR EMR, reproducibility of such pairs of continuum wavelengths and/or reproducibility of detection thereof may be compromised by random and/or systematic errors of the associated conventional apparatus, such as IR EMR source and/or detector, resulting in linear and/or non-linear offsets in the first wavelength and/or the second wavelength and hence erroneous detective absorbances. This is particularly important since the first pair of discrete wavelengths may not correspond with spectral maxima or minima and may instead correspond with spectral shoulders. While intensities and/or wavelengths of spectral maxima or minima may be readily estimated by conventional peak fitting, for example, and hence correction for offsets applied, such correction may not be accurately applied for shoulders, which may also be associated with relatively rapid rates of change of intensity as a function of wavelength. Furthermore, such a conventional apparatus typically also requires periodic and/or frequent calibration which may nevertheless be susceptible to drift, for example. Hence, offsets in the first wavelength and/or the second wavelength may significantly affect detecting respective absorbances thereof, thereby adversely affecting a false positive rate and/or false negative rate of discrimination between the first cell type or first tissue type and the second cell type or second tissue type. That is, confidence in determining a presence or absence of the second cell type or the second tissue type, for example a diseased cell type or a diseased tissue type, in the second spatial region of the cell sample or the tissue sample using conventional methods is relatively lower. In contrast, since the first pair of wavelengths is a pair of discrete wavelengths, reproducibility thereof and/or reproducibility of detection thereof is improved, thereby enhancing confidence in determining a presence or absence of the second cell type or the second tissue type, for example a diseased cell type or a diseased tissue type, in the second spatial region of the cell sample or the tissue sample. Additionally and/or alternatively, since the first pair of wavelengths is a pair of discrete wavelengths, a cost, a size and/or a complexity of a suitable apparatus, for example an apparatus according to the second aspect, is lowered while a physical robustness raised, compared with conventional spectral analysis, thereby facilitating deployment for POC applications. Particularly, spectrometers are essential for Raman and FTIR techniques. Generally, spectrometers are relatively large, costly, complex and/or delicate optical instruments. In contrast, a spectrometer is not required for the method according to the first aspect, nor for the apparatus according to the second aspect. Additionally and/or alternatively, since the first pair of wavelengths is a pair of discrete wavelengths, data acquisition (i.e. detecting respective absorbances of the incident first wavelength and the incident second wavelength of the first pair of discrete wavelengths) is accelerated compared with relatively long scan times (typically 30 s to 300 s) associated with spectral techniques. Additionally and/or alternatively, since the first pair of wavelengths is a pair of discrete wavelengths, signal to noise (i.e. sensitivity) is improved compared with spectral techniques since the respective absorbances of the incident first wavelength and the incident second wavelength of the first pair of discrete wavelengths are detected for relatively longer, even though the total detection time is reduced, thereby improving discrimination and/or limits of detection. Additionally and/or alternatively, since the first pair of wavelengths is a pair of discrete wavelengths, calibration may be eliminated without degrading the discriminating, thereby lessening operational burden of a suitable apparatus, for example an apparatus according to the second aspect. Additionally and/or alternatively, since the first pair of wavelengths is a pair of discrete wavelengths, comparing the obtained first absorption response and the obtained second absorption response is simplified since spectral analysis is not required, thereby accelerating data processing and enabling real-time discrimination. For example, results of analysis of a mouth lesion may be provided while the mouth lesion is being illuminated. Since the first pair of discrete wavelengths is characteristic of the discriminating biomarker, the method provides accurate real-time diagnosis, for example of oral cancers. Additionally and/or alternatively, the relative simplicity and robustness of the method and apparatus provide suitability for in vivo (e.g. in the mouth or on the skin of a patient i.e. non- invasive) POC applications and ex vivo (e.g. biopsies or cell scrapes) applications. Such realtime diagnosis enables early stage disease detection and monitoring of progression of diseases and/or treatment.

The method is of discriminating between the first cell types or the first tissue types and the second cell types or the second tissue types.

In one example, at least one of the first cell type or the first tissue type and/or the second cell type or the second tissue type comprises and/or is a diseased cell or a diseased tissue. In one example, the disease is cancer i.e. the diseased tissue is cancerous. In such an example, at least one of the other respective first cell type or first tissue type and/or the second cell type or the second tissue type comprises and/or is a diseased cell or a diseased tissue e.g. wherein the method is able to discriminate between different types of diseased cells or diseased tissues in a patient sample. The at least one of the other respective first cell type or first tissue type and/or the second cell type or the second tissue type comprises and /or is a healthy cell or a healthy tissue e.g. wherein the method is able to discriminate between diseased and healthy cells or tissue in a patient sample.

In one example, the first cell type or the first tissue type and the second cell type or the second tissue type is a diseased cell or a diseased tissue. In one example, the first cell type or the first tissue type and/or the second cell type or the second tissue type comprises and/or is a cell or a tissue associated with different diseases, or different stages of the same disease. In one example, at least one of the first cell type or the first tissue type and/or the second cell type or the second tissue type is in a pre-diseased state, a state that is known to be at risk of developing into diseased state and/or a state causing disease (e.g. a cell type or tissue type in a pre-cancerous state).

Esophageal cancer is the sixth most common cause of cancer mortality and is the cancer with the fastest rise in incidence in the western world. There are two main forms of esophageal cancer. One is squamous cell carcinoma, which is most common in Asia and is associated with smoking and poor diet. The other is adenocarcinoma, which is more common in the west and is associated with the gastro-esophageal reflux of acid and bile salts and the preneoplastic condition of Barrett’s metaplasia of the esophagus. Both cancers consist of malignant epithelial cells and stroma and the latter is important for facilitating cancer progression. One of the most important cell types in the stroma is a specialized fibroblast called the myofibroblast that produces growth factors and cytokines that promote cancer growth and metastasis. The diagnosis of esophageal cancer follows the standard approach of examining images of excised tissue, obtained by endoscopy, after staining with Haematoxylin and Eosin (H&E). This highlights the nucleic acid and protein content of the specimen at blue and red visible wavelengths respectively. Typically, the interobserver discordance for the diagnosis of the low- grade dysplasia, which is characteristic of the earliest preneoplastic stage of disease is greater than 50%. Although this discordance is reduced to ~15% for the diagnosis of the more serious condition of high-grade dysplasia, there is a need to improve the accuracy of diagnosis since false positives can give rise to unnecessary procedures and false negatives can be fatal. As with all cancers, early detection is critical for the best patient outcome and there is a need for cheaper, more accurate and ideally automated methods for cancer diagnosis and for identification of those patients with Barrett’s esophagus at most risk of progressing to dysplasia and cancer.

In more detail, head and neck squamous cell carcinoma (HNSCC) is the sixth most common cancer world-wide. Over 90% of these cancers are oral squamous cell carcinomas (OSCC), and this is a growing problem with 300,000 new cases per annum globally and 12,200 new cases in the UK in 2017. Oral cancer has an overall survival rate of only 50%, due mainly to late stage presentation. The immediate aim of this proposal is to improve early diagnosis of this disease in order to enhance survival and reduce morbidity associated with unnecessary treatment. The current gold standard for early detection of oral cancer is the longitudinal pathological assessment and clinical observation of premalignant lesions such as oral epithelial dysplasia (OED), but it is acknowledged that other factors must be considered to improve prediction of malignant transformation. The method according to the first aspect and the apparatus according to the second aspect may be used to interrogate fixed tissue specimens using novel biomarkers that show high specificity and sensitivity and augment the current gold standard.

At present, the best prospect of improving survival in OSCC patients is early detection. A recognised pre-neoplastic lesion, OED, is present at a population rate of 2.5-5 per 1000 in the UK, with a rate of malignant transformation to oral cancer of around 12% and estimated monitoring costs of £600 per patient per annum. Current best practice is clinical observation with pathological assessment of biopsy material. However, transformation may not occur for several years after diagnosis of OED yet lesions with only mild histological changes cannot be safely discharged from monitoring, as 10-20% are at risk of malignant change over a 5-10 year window. Conversely, surgical intervention for even the more advanced grades of dysplasia is probably overtreatment, as 60-80% do not transform into cancer. Clinical and histopathological attributes alone are not sufficient to distinguish those oral dysplastic lesions that will undergo transformation into oral cancer from those that will not. Furthermore, attempts to identify a molecular biomarker have failed to enter clinical practice. This uncertainty impacts on both individual treatment planning (e.g. early surgical excision or chemoprevention) and on service provision, which necessarily includes a large proportion of individuals who are not at risk of developing cancer. What is required are biomarkers to augment current risk models that can be easily incorporated into current clinical practice.

While the description is focussed on esophageal cancer, particularly HNSCC, the method according to the first aspect and the apparatus according to the second aspect may be used for improved early diagnosis and staging of a wide range of cancers through analysis of fixed biopsies or surrogate tissue (for example: lung, esophageal, cervical and/or prostate cancers) and for improving the in situ diagnostic accuracy of clinical observation (for example: oral, cervical and/or esophageal cancers).

In one example, the first cell type, the first tissue type, the second cell type, the second tissue type, the cell sample and/or the tissue sample is diseased or pre-diseased, for example wherein the disease is cancer or at risk of becoming cancerous or gastresophageal reflux disease e.g. Barrett’s esophagus. In one example, the first cell type, the first tissue type, the second cell type, the second tissue type, the cell sample and/or the tissue sample is cancerous (for example comprising cancer associated myofibroblast cells) or at risk of becoming cancerous, such as cancer associated with esophageal cancer cell line OE19 and / or OE21 .

The method comprises providing the cell sample or the tissue sample including the first spatial region, comprising the first cell type or the first tissue type, and the second spatial region. In one example, providing the cell sample or a tissue sample is in vivo, for example non-invasive or invasive in situ examination of a patient. For example, oral lesions may be examined non- invasively in a patient’s mouth and skin cancer (melanoma) may be examined on a patient’s body. In one example, providing the cell sample or a tissue sample is ex vivo, for example by providing a biopsy or a cell scrape. It should be understood that the first spatial region and the second spatial region are mutually different, for example mutually adjacent spatial regions such as in a patient’s mouth, on a patient’s skin, on a biopsy or on a cell scrape. Additionally and/or alternatively, the first spatial region may be provided by a reference sample or a calibration standard, comprising a known first cell type or a known first tissue type. It should be understood that the first absorption response and the second absorption response are obtained from surfaces and/or sub-surfaces of the first spatial region and the second spatial region, respectively, depending, at least in part, on a depth of penetration of the IR EMR therein. It should be understood that the first spatial region, comprising the first cell type or the first tissue type, provides a control sample or a reference sample, against which the second spatial region is compared. For example, the first spatial region may comprise a healthy cell type or a healthy tissue type. It should be understood that the second spatial region may or may not comprise the second cell type or the second tissue type. For example, the second spatial region may comprise a healthy cell type or a healthy tissue type, such as similar or identical to the first spatial region, or a diseased cell type or a diseased tissue type, such as distinguished from the first spatial region. In one example, the first spatial region comprises a plurality of spatial regions, for example mutually separated. In this way, the first absorption response is obtained from the plurality of spatial regions, for example by averaging (e.g. mean absorption response). In one example, the second spatial region comprises a plurality of spatial regions, for example mutually separated. In this way, the second absorption response is obtained from the plurality of spatial regions, for example by averaging (e.g. mean absorption response).

The method comprises obtaining the first absorption response comprising: illuminating the first spatial region using the IR EMR, incident thereupon. That is, the first spatial region of the cell sample or the tissue sample is illuminated or irradiated using the IR EMR, for example by emitting the IR EMR theretowards from a probe (as described below with respect to the second aspect), directly or indirectly (such as via a loop). In one example, illuminating the first spatial region using the IR EMR comprises illuminating the first spatial region using an evanescent wave formed from the IR EMR. Generally, evanescent wave fibre optic sensors are a subset of fibre optic sensors that perform sensing along the fibre's cylindrical length. In all optical fibres, light propagates by means of total internal reflection, wherein the propagating light is launched into waveguide at angles such that upon reaching the cladding-core interface, the energy is reflected and remains in the core of the fire. However, for light reflecting at angles near the critical angle, a significant portion of the power extends into the cladding or medium which surrounds the core. This phenomenon, known as the evanescent wave, extends only to a short distance from the interface, with power dropping exponentially with distance. The evanescent wave has been exploited to allow for real-time interrogation of surface-specific recognition events. If the discriminating biomarker is present, for example in the first spatial region and/or the second spatial region, the biomarker absorbs the evanescent wave, causing a change in intensity of the detected wavelengths.

In one example, illuminating the first spatial region using the IR EMR incident thereupon comprises illuminating the first spatial region simultaneously, for example intermittently or continuously, using the first wavelength and the second wavelength of the first pair of discrete wavelengths. In this way, the respective absorbances are obtained simultaneously and throughout the illuminating. In one example, illuminating the first spatial region using the IR EMR incident thereupon comprises illuminating the first spatial region alternately using the first wavelength and the second wavelength of the first pair of discrete wavelengths, for example for the first time period and a second period respectively. That is, the first spatial region is repeatedly illuminated successively by the first wavelength and the second wavelength, for the same or different time periods. In one example, illuminating the first spatial region using the IR EMR incident thereupon comprises illuminating the first spatial region using a single wavelength during any particular time period. In one example, the first time period and/or the second time period is in a range from 1 ps to 10 s, preferably in a range from 1 ms to 1 s. In one example, the first time period and/or the second time period is pre-determined, for example according to expected respective absorbances of the incident first wavelength and the incident second wavelength of the first pair of discrete wavelengths, so as to improve sensitivity of the result. In one example, the first time period and/or the second time period is dynamically determined, for example according to obtained respective absorbances of the incident first wavelength and the incident second wavelength of the first pair of discrete wavelengths, so as to improve sensitivity of the result. In one example, illuminating the first spatial region using the IR EMR incident thereupon comprises illuminating the first spatial region alternately using the respective pairs of discrete wavelengths of the set thereof, for example as described with respect to the first wavelength and the second wavelength of the first pair of discrete wavelengths mutatis mutandis.

In one example, illuminating the first spatial region using the emitted infrared IR EMR incident thereupon comprises contacting the first spatial region with a probe, for example as described with respect to the second aspect.

In one example, the method comprises determining non-contact of the first spatial region with the probe, for example as described with respect to the second aspect. In one example, the probe comprises and/or is a loop probe, for example as described with respect to the second aspect.

The IR EMR comprises the set of pairs of discrete wavelengths, including the first pair of discrete wavelengths consisting of the first wavelength and the second wavelength. It should be understood that the set of pairs of discrete wavelengths is predetermined, for example fixed. In this way, complexity is reduced while providing selectivity for the second cell type or second tissue type. It should be understood that the first wavelength and the second wavelength are discrete wavelengths i.e. each monochromatic or quasi-monochromatic, for example each having a wavelength range of at most 20 nm (i.e. ±10 nm), at most 10 nm (i.e. ±5 nm) or , at most 5 nm (i.e. ±2.5 nm) or having a corresponding wavenumber range of at most 4 cnr¹ (i.e. ±2 cm ¹), at most 2 cnr¹ (i.e. ±1 cm^-1) or at most 1 cnr¹ (i.e. ±0.5 cm^-1). For example, the wavelength range may be the line width of a corresponding laser transition. In one example, the set of pairs of discrete wavelengths is not provided by and/or does not form a continuum of wavelengths. In one example, the set of pairs of discrete wavelengths includes only the first pair of discrete wavelengths i.e. a single pair of discrete wavelengths. In this way, complexity is reduced further. In one example, the set of pairs of discrete wavelengths includes N pairs of discrete wavelengths, including the first pair of discrete wavelengths, wherein N is a natural number greater than or equal to 1 , for example 1 , 2, 3, 4, 5, 6, 7, 8, 9, 10 or more. By increasing the number of pairs of discrete wavelengths, a false positive rate and/or a false negative rate may be improved. Additionally and/or alternatively, by increasing the number of pairs of discrete wavelengths, additional cell types or tissue types may be discriminated using the same method and/or the same apparatus. In one example, the first wavelength of the first pair of discrete wavelengths is common with (i.e. the same as) the first wavelength of a second pair of discrete wavelengths. In this way, a false positive rate and/or a false negative rate may be improved while attenuating complexity. In one example, the set of pairs of discrete wavelengths includes a second pair of discrete wavelengths consisting of a first wavelength and a second wavelength, wherein the second pair of discrete wavelengths is characteristic of the discriminating biomarker.

The first pair of discrete wavelengths is characteristic of the discriminating biomarker. If the discriminating biomarker is present, for example in the first spatial region and/or the second spatial region, the biomarker absorbs the first wavelength or the second wavelength of the first pair of discrete wavelengths, causing a change in intensity of the detected wavelengths.

In one example, the first pair of discrete wavelengths corresponds with a pair of wavenumbers (cm ¹) of radiation and/or the set of pairs of discrete wavelengths corresponds with respective pairs of wavenumbers (cm ¹) of radiation selected using pre-existing knowledge of the first cell type or first tissue type’s radiation absorption behaviour. In one example, the first pair of discrete wavelengths corresponds with a pair of wavenumbers (cm ¹) of radiation and/or the set of pairs of discrete wavelengths corresponds with respective pairs of wavenumbers (cm ¹) of radiation selected as described in WO 2019/197806 A1 and/or A novel FTIR analysis method for rapid high-confidence discrimination of esophageal cancer (2019) James Ingham, Michael J. Pilling, David S. Martin, Caroline I. Smith, Barnaby G. Ellis, Conor A. Whitley, Michele R.F. Siggel-King, Paul Harrison, Timothy Craig, Andrea Varro, D. Mark Pritchard, Akos Varga, Peter Gardner, Peter Weightman, Steve Barrett; Infrared Physics and Technology 102, 103007, https://doi.orq/10.1016/1. infrared.2019.103007, which are incorporated by reference herein in entirety.

In one example, the first pair of discrete wavelengths corresponds with a pair of wavenumbers (cm ¹) of radiation and/or the set of pairs of discrete wavelengths corresponds with respective pairs of wavenumbers (cm ¹) of radiation selected from the range of about (i.e. within 10 cm^-1) 900 cnr¹ to about 4000 cm ¹, preferably selected from the range of about 1000 cnr¹ to about 3000 cm^-1, more preferably selected from within the range of about 1000 cnr¹ to about 1800 cm^-1.

In one example, the first wavelength and/or the second wavelength of the first pair of discrete wavelengths corresponds with a spectral shoulder of the discriminating biomarker. As described previously, the inventors have observed that successful discrimination requires detection of absorbance of reproducible pairs of discrete wavelengths of the IR EMR. While conventional spectral analysis, for example using broadband IR EMR or by wavelength scanning of IR EMR, may provide pairs of continuum wavelengths of the IR EMR, reproducibility of such pairs of continuum wavelengths and/or reproducibility of detection thereof may be compromised by random and/or systematic errors of the associated conventional apparatus, such as IR EMR source and/or detector, resulting in linear and/or nonlinear offsets in the first wavelength and/or the second wavelength and hence erroneous detective absorbances. This is particularly important since the first pair of discrete wavelengths may not correspond with spectral maxima or minima and may instead correspond with spectral shoulders. While intensities and/or wavelengths of spectral maxima or minima may be readily estimated by conventional peak fitting, for example, and hence correction for offsets applied, such correction may not be accurately applied for shoulders, which may also be associated with relatively rapid rates of change of intensity as a function of wavelength. Furthermore, such a conventional apparatus typically also requires periodic and/or frequent calibration while may nevertheless be susceptible to drift, for example. Hence, offsets in the first wavelength and/or the second wavelength may significantly affect detecting respective absorbances thereof, thereby adversely affecting a false positive rate and/or false negative rate of discrimination between the first cell type or first tissue type and the second cell type or second tissue type. That is, confidence in determining a presence or absence of the second cell type or the second tissue type, for example a diseased cell type or a diseased tissue type, in the second spatial region of the cell sample or the tissue sample using conventional methods is relatively lower. In contrast, since the first pair of wavelengths is a pair of discrete wavelengths, reproducibility thereof and/or reproducibility of detection thereof is improved, thereby enhancing confidence in determining a presence or absence of the second cell type or the second tissue type, for example a diseased cell type or a diseased tissue type, in the second spatial region of the cell sample or the tissue sample.

In one example, the first cell type or first tissue type is oral squamous cell carcinoma (OSCC) nodal metastases. In one example, the first cell type or first tissue type is oral squamous cell carcinoma (OSCC) nodal metastases, wherein the first pair of discrete wavelengths corresponds with a pair of wavenumbers (cnr¹) of radiation and/or wherein the set of pairs of discrete wavelengths corresponds with respective pairs of wavenumbers (cnr¹) of radiation selected from a group consisting of: 1751 , 1650, 1369, 1285, 1252 cnr¹. In one example, the first cell type or first tissue type is oral squamous cell carcinoma (OSCC) nodal metastases, wherein the first pair of discrete wavelengths corresponds with a pair of wavenumbers (cm ¹) of radiation: 1285, 1252.

In one example, the first cell type or first tissue type is esophageal cancer cell line OE19. In one example, the first cell type or first tissue type is esophageal cancer cell line OE21. In one example, the first cell type or first tissue type is cancer associated myofibroblast cells. In one example, the first cell type or first tissue type is adjacent tissue myofibroblast cells. In one example, the first cell type or first tissue type is cancerous tissue. In one example, the first cell type or first tissue type is cancer associated stroma. In one example, the first cell type or first tissue type is Barrett’s tissue. In one example, the first cell type or first tissue type is Barrett’s associated stroma.

In one example, the first cell type is esophageal cancer cell line OE19 and the cell sample or the tissue sample comprises one or more of esophageal cancer cell line OE21 , cancer associated myofibroblast cells and adjacent tissue myofibroblast cells (optionally wherein the sample comprises all of these), wherein the first pair of discrete wavelengths corresponds with a pair of wavenumbers (cm ¹) of radiation and/or wherein the set of pairs of discrete wavelengths corresponds with respective pairs of wavenumbers (cm ¹) of radiation selected from a group consisting of: 1375, 1381 , 1400, 1406, 1418, 1692, 1697.

In one example, the first cell type is esophageal cancer cell line OE21 and the cell sample or the tissue sample comprises one or more of esophageal cancer cell line OE19, cancer associated myofibroblast cells and adjacent tissue myofibroblast cells (optionally wherein the sample comprises all of these), wherein the first pair of discrete wavelengths corresponds with a pair of wavenumbers (cm ¹) of radiation and/or wherein the set of pairs of discrete wavelengths corresponds with respective pairs of wavenumbers (cm ¹) of radiation selected from a group consisting of: 1443, 1449, 1466, 1472, 1539, 1545, 1551.

In one example, the first cell type is cancer associated myofibroblast cells and the cell sample or the tissue sample comprises one or more of esophageal cancer cell line OE19, esophageal cancer cell line OE21 , and adjacent tissue myofibroblast cells (optionally wherein the sample comprises all of these), wherein the first pair of discrete wavelengths corresponds with a pair of wavenumbers (cm ¹) of radiation and/or wherein the set of pairs of discrete wavelengths corresponds with respective pairs of wavenumbers (cm ¹) of radiation selected from a group consisting of: 1443, 1508, 1522, 1678, 1684, 1692.

In one example, the first cell type is adjacent tissue myofibroblast cells and the cell sample or the tissue sample comprises one or more of esophageal cancer cell line OE19, esophageal cancer cell line OE21 , and cancer associated myofibroblast cells (optionally wherein the sample comprises all of these), wherein the first pair of discrete wavelengths corresponds with a pair of wavenumbers (cm ¹) of radiation and/or wherein the set of pairs of discrete wavelengths corresponds with respective pairs of wavenumbers (cm ¹) of radiation selected from a group consisting of: 1049, 1103, 1146, 1200, 1206, 1400, 1424, 1466, 1472.

In one example, the first tissue type is esophageal cancerous tissue and the tissue sample comprises one or more of cancer associated stroma, Barrett’s tissue and Barrett’s associated stroma (optionally wherein the sample comprises all of these), wherein the first pair of discrete wavelengths corresponds with a pair of wavenumbers (cm ¹) of radiation and/or wherein the set of pairs of discrete wavelengths corresponds with respective pairs of wavenumbers (cm ¹) of radiation selected from a group consisting of: 1460, 1466, 1472, 1480, 1485.

In one example, the first tissue type is cancer associated stroma and the tissue sample comprises one or more of esophageal cancerous tissue, Barrett’s tissue and Barrett’s associated stroma (optionally wherein the sample comprises all of these), wherein the first pair of discrete wavelengths corresponds with a pair of wavenumbers (cm ¹) of radiation and/or wherein the set of pairs of discrete wavelengths corresponds with respective pairs of wavenumbers (cm ¹) of radiation selected from a group consisting of: 999, 1007, 1018, 1061 , 1067, 1073.

In one example, the first tissue type is Barrett’s tissue and the tissue sample comprises one or more of esophageal cancerous tissue, cancer associated stroma and Barrett’s associated stroma (optionally wherein the sample comprises all of these), wherein the first pair of discrete wavelengths corresponds with a pair of wavenumbers (cm ¹) of radiation and/or wherein the set of pairs of discrete wavelengths corresponds with respective pairs of wavenumbers (cm ¹) of radiation selected from a group consisting of: 1375, 1406, 1412, 1418, 1443, 1449, 1466.

In one example, the first tissue type is Barrett’s associated stroma and the tissue sample comprises one or more of esophageal cancerous tissue, cancer associated stroma, and Barrett’s tissue (optionally wherein the sample comprises all of these), wherein the first pair of discrete wavelengths corresponds with a pair of wavenumbers (cm ¹) of radiation and/or wherein the set of pairs of discrete wavelengths corresponds with respective pairs of wavenumbers (cm ¹) of radiation selected from a group consisting of: 1375, 1406, 1412, 1418, 1443, 1449, 1466.

The method comprises obtaining the first absorption response comprising: detecting respective absorbances of the incident first wavelength and the incident second wavelength of the first pair of discrete wavelengths by the first spatial region. In one example, detecting respective absorbances of the incident first wavelength and the incident second wavelength of the first pair of discrete wavelengths by the first spatial region comprises detecting the respective absorbances of the incident first wavelength and the incident second wavelength of the first pair of discrete wavelengths by the first spatial region using a Mercury Cadmium Telluride (MCT) detector, for example a MCT amplified photodetector, optionally thermoelectric cooled (TEC). Suitable MCT detectors are known.

In one example, detecting respective absorbances of the incident first wavelength and the incident second wavelength of the first pair of discrete wavelengths by the first spatial region comprises phase-sensitive detection thereof, for example as described with respect to the second aspect.

In one example, obtaining the first absorption response comprises averaging the detected respective absorbances of the incident first wavelength and the incident second wavelength of the first pair of discrete wavelengths by the first spatial region. In this way, spatially and/or temporally local changes in the detective respective absorbances, for example such as due to relative motion of the cell sample or tissue sample, may be attenuated.

It should be understood that the method according to the first aspect does not comprise and/or is not a spectroscopic method, for example using a spectrometer for illuminating the respective spatial regions and/or detecting respective absorbances of the IR EMR, such as used in FTIR spectroscopy and Raman spectroscopy, respectively. In this way, cost, complexity and/or size is lowered while physical robustness is raised.

The method comprises obtaining the second absorption response comprising: illuminating the second spatial region using the IR EMR, for example as described with respect to obtaining the first absorption response mutatis mutandis.

The method comprises obtaining the second absorption response comprising: detecting respective absorbances of the incident first wavelength and the incident second wavelength of the first pair of discrete wavelengths by the second spatial region, for example as described with respect to obtaining the first absorption response mutatis mutandis.

It should be understood that the first absorption response and the second absorption response are obtained successively. In one example, obtaining the first absorption response comprises inputting, for example by a user via a human computer interface or a graphical user interface, a first input, such as indicating that a probe is positioned over or in contact with the first spatial region. In this way, the first absorption response is correlated with the first spatial region, for example for calibration with respect thereto. In one example, illuminating the first spatial region using the IR EMR and/or detecting respective absorbances of the incident first wavelength and the incident second wavelength of the first pair of discrete wavelengths by the first spatial region is in response to receiving the first input. In one example, obtaining the second absorption response comprises inputting, for example by a user via a human computer interface or a graphical user interface, a second input, such as indicating that a probe is positioned over or in contact with the second spatial region. In one example, illuminating the second spatial region using the IR EMR and/or detecting respective absorbances of the incident first wavelength and the incident second wavelength of the first pair of discrete wavelengths by the second spatial region is in response to receiving the second input. In this way, the second absorption response is correlated with the second spatial region, for example for discriminating the second spatial region with respect to the first spatial region.

The method comprises discriminating (i.e. distinguishing or differentiating) between the first spatial region and the second spatial region based on the result of comparing the obtained first absorption response and the obtained second absorption response. In this way, presence or absence of the second cell type or second tissue type in the second spatial region may be identified, as described previously. In one example, comparing the obtained first absorption response AR1 and the obtained second absorption response AR2 comprises calculating a difference (AR1 - AR2) therebetween (i.e. the result), for example a magnitude |AR1 - AR2|. In one example, comparing the obtained first absorption response AR1 and the obtained second absorption response AR2 comprises calculating a ratio (AR1 / AR2) (i.e. the result) thereof. Determining a ratio between the obtained first absorption response and the obtained second absorption response advantageously negates measurement variables such as a thickness of the sample of the cell sample or the tissue sample and/or the particular cell type or tissue type. This in turn improves a confidence of the discriminating because fewer variables influence the result. In one example, comparing the obtained first absorption response AR1 and the obtained second absorption response AR2 comprises calculating a ratio (AR1 - AR2) / (AR1 + AR2) (i.e. the result) of a difference (AR1 - AR2) therebetween divided by a total (AR1 + AR2) thereof. Such a ratio calculation may be useful if one of the absorption responses is very small, for example.

In one example, the method comprises balancing the respective absorbances of the incident first wavelength and the incident second wavelength of the first pair of discrete wavelengths by the first spatial region by adjusting respective intensities of the first pair of discrete wavelengths of the IR EMR incident thereupon. In this way, similar signals may be detected for the first wavelength and the second wavelength, thereby improving a sensitivity of the result. In one example, the respective intensities are pre-determined, for example according to expected respective absorbances of the incident first wavelength and the incident second wavelength of the first pair of discrete wavelengths, so as to improve sensitivity of the result. In one example, the respective intensities are dynamically determined, for example according to obtained respective absorbances of the incident first wavelength and the incident second wavelength of the first pair of discrete wavelengths, so as to improve sensitivity of the result.

Where at least one of the first cell type or first tissue type and the different cell type or tissue type is identified as being a diseased tissue, the method may further comprise treating the identified disease in the patient, e.g. by therapy or surgery, such as by administering to the patient one or more therapeutic agents having efficacy in treating the disease, i.e. wherein the one or more therapeutic agents are administered in a therapeutically effective amounts. In other words, the present disclosure also provides a method of treating a disease in a patient comprising performing the method of identifying the presence or absence of a first cell type or first tissue type in a cell type or tissue sample obtained from a patient comprising multiple different cell type or tissue types according to the fourth aspect (or any embodiments thereof), and in the event that at least one of the first cell type or first tissue type and the different cell type or tissue type is identified as being in a diseased or pre-diseased state, treating the respective disease in the patient (e.g. by therapy or surgery, such as by administering one or more therapeutic agents having efficacy in treating the disease to the patient, i.e. in therapeutic amounts). Where the identified cell type or tissue type is attributed with a cell type or tissue type that is in a pre-diseased state, or a state that is known to be at risk of developing into diseased state, or causing disease (e.g. a cell type or tissue type in a pre-cancerous state), the method may comprise prophylactically treating the patient (i.e. in an attempt to prevent development or further development of the disease). In some embodiments, the method does not further comprise treating the identified disease in the patient, e.g. by surgery or therapy. In embodiments, where at least one of the first cell type or first tissue type and the different cell type or tissue type is a diseased tissue, the method may further comprise contacting the diseased cell type or tissue type with an active agent (e.g. therapeutic agent) in vitro, such as with a therapeutic agent known to have efficacy in treating the disease, or an agent that is determined to be a therapeutic candidate for treating the disease.

Apparatus

The second aspect provides an apparatus for discriminating between first cell types or first tissue types and second cell types or second tissue types, the apparatus comprising: a set of lasers, including a first laser tuned to emit a first wavelength of infrared, IR, electromagnetic radiation, EMR, of a first pair of discrete wavelengths of a set of pairs of discrete wavelengths and a second laser tuned to emit a second wavelength of IR EMR of the first pair of discrete wavelengths, wherein the first pair of discrete wavelengths is characteristic of a discriminating biomarker; a probe, optically coupled to the set of lasers, for illuminating a cell sample or a tissue sample including a first spatial region, comprising a first cell type or first tissue type, and a second spatial region, using the first wavelength and the second wavelength of the first pair of discrete wavelengths incident thereupon; a detector, optically coupled to the probe, for detecting respective absorbances of the incident first wavelength and the incident second wavelength of the first pair of discrete wavelengths by the first spatial region and by the second spatial region; and a controller configured to control the detector to obtain a first absorption response from the first spatial region and a second absorption response from the second spatial region; and to discriminate between the first spatial region and the second spatial region based on a result of comparing the obtained first absorption response and the obtained second absorption response.

The first cell types, the first tissue types, the second cell types, the second tissue types, the first wavelength of IR EMR of the first pair of discrete wavelengths, the first pair of discrete wavelengths, the set of pairs of discrete wavelengths, the second wavelength of IR EMR of the first pair of discrete wavelengths, the biomarker, the second cell type, the second tissue type, the illuminating, the cell sample, the tissue sample, the first spatial region, the first cell type, the first tissue type, the second spatial region, the detecting, the respective absorbances, the first absorption response, the second absorption response, the discriminating, the result, and/or the comparing may be as described with respect to the first aspect mutatis mutandis.

In other words, the apparatus may be used to discriminate between different types of biological cells or tissues. An example of the tissue types might be healthy tissue comprising normal cells and diseased tissue comprising cancerous or pre-cancerous cells, as described with respect to the first aspect. The apparatus may be used by pathologists, for example, for early diagnosis of a disease, by clinicians, for example, as a tool to grade the progression of a disease and hence make a more accurate prognosis, and/or by surgeons, for example, to indicate when resection has removed all of the intended tissue.

The set of lasers includes the first laser tuned to emit the first wavelength of the IR EMR of the first pair of discrete wavelengths and the second laser tuned to emit the second wavelength of the IR EMR of the first pair of discrete wavelengths. It should be understood that the first laser is specifically tuned, rather than tunable, to emit the first wavelength of the IR EMR of the first pair of discrete wavelengths. That is, the first laser is specifically adapted to emit this first wavelength, for example using optics. In this way, drift of the first wavelength is reduced compared with a tunable laser. In one example, the first laser is tuned to emit only the first wavelength of the IR EMR of the first pair of discrete wavelengths (i.e. single line tuning). In one example, the first laser is tuned to emit a plurality of wavelengths including the first wavelength of the IR EMR of the first pair of discrete wavelengths (i.e. multi line tuning). If the other wavelengths are not of interest, these may be filtered. The second laser may be as described with respect to the first laser. In one example, the set of lasers includes L lasers, wherein L is a natural number greater than or equal to 1 , for example 1 , 2, 3, 4, 5, 6, 7, 8, 9, 10 or more, for example to correspond with wherein the set of pairs of discrete wavelengths includes N pairs of discrete wavelengths, including the first pair of discrete wavelengths, as described previously. In one example, the set of lasers includes a third laser tuned to emit a third wavelength of infrared, IR, electromagnetic radiation, EMR, of a second pair of discrete wavelengths of a set of pairs of discrete wavelengths and a optionally a fourth laser tuned to emit a fourth wavelength of IR EMR of the second pair of discrete wavelengths, wherein the second pair of discrete wavelengths is characteristic of the biomarker of the second cell type or a second tissue type, for example as described with respect to the first aspect. Suitable lasers include quantum cascade laser (QCL) modules that can be tuned to operate at a wavelength selected in the range 3-11 microns (corresponding to wavenumbers from above 3000 cnr¹ to below 1000 cm^-1).

In one example, the controller is configured to control the set of lasers to alternately to emit the first wavelength of IR EMR and the second wavelength of IR EMR of the first pair of discrete wavelengths, for example as described with respect to the first aspect.

In one example, intensities of the emitted first pair of discrete wavelengths of the IR EMR are adjusted to balance the respective absorbances of the incident first wavelength and the incident second wavelength of the first pair of discrete wavelengths by the first spatial region, for example as described with respect to the first aspect.

The apparatus comprises the probe, optically coupled, for example using an optical fibre, to the set of lasers. In one example, the probe comprises and/or is a contact probe for contacting the first spatial region. In one example, the probe comprises and/or is a loop probe. In one example, the probe comprises and/or is an attenuated total reflection (ATR) probe, for example comprising a small crystal (pyramidal or conical) made of an infrared-transparent material (such as diamond, silicon, germanium, ZnSe or ZrC>2). Suitable ATR probes include fibre optic spectroscopy probes that terminate in an ATR crystal and can be sterilised for use with biological tissue. In one example, the probe comprises and/or is an ATR loop probe, for example having a single, a double or a triple loop. Suitable ATR loop probes include fibre optic spectroscopy probes that terminate in a detachable loop that can be removed/replaced for each application to mitigate cross contamination between different tissue samples. In one example, the apparatus comprises a laser beam combiner and the probe is optically coupled to the set of lasers via the laser beam combiner. Suitable laser beam combiners include semi- silvered mirrors or dichroic mirrors. In one example, the apparatus comprises a probe housing, arranged to house the probe therein, for example a pen, such as ergonomically adapted to be held in a user’s hand while protecting the probe and/or optical coupling(s), such as optical fibres, therein. In this way, the probe and/or optical coupling(s), such as optical fibres, are protected while the user may examine the cell sample or the tissue sample conveniently. In one example, the probe housing comprises and/or is an optically transparent housing. In this way, the user can see the probe and the first spatial region and/or the second spatial region through the optically transparent housing, improving localisation accuracy of the probe on the first spatial region and/or the second spatial region. In one example, the probe housing comprises an aperture and the probe protrudes therethrough, for example by a predetermined protrusion, for example to just touch the first spatial region and/or the second spatial region. In this way, damage to the probe is prevented and/or unintentional contacting with spatial regions avoided while pressure applied by the user on the first spatial region and/or the second spatial region controlled. In one example, a shape of the probe housing, for example of a distal end thereof such as a cell sample or tissue sample contacting surface thereof, is adapted to control the presentation of the probe to the first spatial region and/or the second spatial region, for example by controlling an angle of the probe with respect to the probe housing and hence an angle of the probe respect to the first spatial region and/or the second spatial region when in contact therewith. This way, angular presentation of the probe with respect to the first spatial region and/or the second spatial region by the user is controlled. In one example, the probe housing comprises one or more buttons, as described below. In this way, the user may initiate the controller to control the detector to obtain the first absorption response from the first spatial region and the second absorption response from the second spatial region, for example including illuminating the first spatial region and/or the second spatial region and/or obtaining the first absorption response and/or the second absorption response, for example as described with respect to the first aspect. In one example, individual buttons are provided corresponding to the first wavelength to the second wavelength. In one example, the probe housing is interchangeable, enabling different probe housings to be used and hence different presentations of the probe to be provided. In one example, the probe is interchangeable, for example between a single, a double or a triple loop.

In one example, the detector comprises and/or is a phase-sensitive detector. If the phase- sensitive detection is synchronised ("locked-in") with the modulation/alternation of the lasers then the detector is able to separate low-intensity signals even in the presence of significant noise levels. In one example, the controller is configured to determined non-contact of the first spatial region and/or the second spatial region with the probe. In this way, respective absorbances detected when non-contact is determined may be excluded from the comparing.

In one example, the controller is configured to average the detected respective absorbances of the incident first wavelength and the incident second wavelength of the first pair of discrete wavelengths by the first spatial region, for example as described with respect to the first aspect.

In one example, the controller is configured to control the set of lasers, for example wherein the controller is configured to control the first laser to emit the first wavelength of IR EMR and the second laser to emit the second wavelength of IR EMR, for example as described with respect to the first aspect regarding illuminating the first spatial region and/or illuminating the second spatial region mutatis mutandis.

In one example, the apparatus comprises an output device, for example an audio output device and/or a visual output device, and the controller is configured to control the output device according to whether the first spatial region and the second spatial region are mutually discriminated. For example, a particular audio signal and/or a particular visual signal may be output to indicate presence or absence of the second cell type or second tissue type.

In one example, the apparatus does not comprise a spectrometer.

In one example, the apparatus comprises a human computer interface and/or a graphical user interface for inputting inputs, such as indicating that the probe probe is positioned over or in contact with the first spatial region and/or the second spatial region. In one example, the human computer interface or a part of thereof is provided in and/or on the probe housing, for example as a button. In this way, input is facilitated.

In one example, the controller comprises a processor and a memory.

Computer, computer program and non-transient computer-readable storage medium

The third aspect provides a computer comprising a processor and a memory configured to implement a method according to the first aspect.

The fourth aspect provides a computer program comprising instructions which, when executed by a computer comprising a processor and a memory, cause the computer to perform a method according to the first aspect. The fifth aspect provides a non-transient computer-readable storage medium comprising instructions which, when executed by a computer comprising a processor and a memory, cause the computer to perform a method according to the first aspect.

Definitions

Throughout this specification, the term “comprising” or “comprises” means including the component(s) specified but not to the exclusion of the presence of other components. The term “consisting essentially of or “consists essentially of means including the components specified but excluding other components except for materials present as impurities, unavoidable materials present as a result of processes used to provide the components, and components added for a purpose other than achieving the technical effect of the invention, such as colourants, and the like.

The term “consisting of or “consists of means including the components specified but excluding other components.

Whenever appropriate, depending upon the context, the use of the term “comprises” or “comprising” may also be taken to include the meaning “consists essentially of or “consisting essentially of, and also may also be taken to include the meaning “consists of or “consisting of.

The optional features set out herein may be used either individually or in combination with each other where appropriate and particularly in the combinations as set out in the accompanying claims. The optional features for each aspect or exemplary embodiment of the invention, as set out herein are also applicable to all other aspects or exemplary embodiments of the invention, where appropriate. In otherwords, the skilled person reading this specification should consider the optional features for each aspect or exemplary embodiment of the invention as interchangeable and combinable between different aspects and exemplary embodiments.

Brief description of the drawings For a better understanding of the invention, and to show how exemplary embodiments of the same may be brought into effect, reference will be made, by way of example only, to the accompanying diagrammatic Figures, in which: Figure 1 shows a graph of normal distributions (blue lines) fitted to the training data (not shown) for OSCC and lymphoid tissue. Histograms of OSCC (black) and lymphoid tissue (grey) testing spectra are also shown.

Figure 2 shows Average FTIR profiles for (a) lymphoid tissue (grey) spectra and (b) OSCC (black). The shaded grey rectangles show the regions of 1250-1254 cnr¹ and 1285-1289 cm^-1.

Figure 3 shows images of a tissue core containing OSCC and lymphoid tissue: (a) IHC image stained for pan-cytokeratins (dark brown); (b) FTIR image at 1285 cnr¹; (c) FTIR image at 1252 cnr¹; and (d) FTIR ratio image 1252 cnr¹/1285 cnr¹. Black arrows indicate the periphery of the tumour; white arrows identify highly keratinised areas of the tumour. Each FTIR image is plotted with a colour table covering the 5th to 95th percentiles of the image intensity range. Image (a) was obtained from a section adjacent to that used to obtain images (b), (c) and (d).

Figure 4 shows (a) H&E stained image; (b) IHC image stained for pan-cytokeratins (dark brown); (c) Topography, IR SNOM images at (d) 1751 cnr¹; (e) 1650 cnr¹; (f) 1369 cnr¹; (g) 1285 cnr¹; (h) 1252 cnr¹; and (i) ratio of 1252 cnr¹/1285 cnr¹ [i.e. (h)/(g)]. All images are 300 pm x 300 pm. Each SNOM IR image is plotted with a colour table covering the 5th to 95th percentiles of the image intensity range. Image (a) was obtained from a section adjacent to that used to obtain image (b), which was in turn adjacent to that used to obtain images (c) to

(i)·

Figure 5 shows H&E stained image (top) and line profiles (bottom) taken through the core at the white line showing (a) Topography; (b) 1751 cm ¹; (c) 1650 cm ¹; (d) 1369 cnr¹; (e) 1285 cnr¹; (f) 1252 cnr¹; and (g) ratio of 1252 cm Vl285 cnr¹ [i.e. (f)/(e)j. H&E image (top) was obtained from a section adjacent to that used to obtain SNOM line profiles. Each line profile has been normalised to its min/max values.

Figure 6 schematically depicts a method according to an exemplary embodiment; and Figure 7 schematically depicts an apparatus according to an exemplary embodiment.

Detailed Description of the Drawings

Summary

A novel machine learning algorithm is shown to accurately discriminate between oral squamous cell carcinoma (OSCC) nodal metastases and surrounding lymphoid tissue on the basis of a single metric, the ratio of FTIR absorption intensities at 1252 cnr¹ and 1285 cm ¹. The metric yields discriminating sensitivities, specificities and precision of 99 ± 0.14%, 100 ± 0.01% and 100 ± 0.02% respectively, and an area under receiver operator characteristic (AUC) of 0.99 ± 0.0007. The delineation of the OSCC and lymphoid tissue revealed by the image formed from the metric is in better agreement with an immunohistochemistry (IHC) stained image than are either of the FTIR images obtained at the individual wavenumbers. Scanning near-field optical microscopy (SNOM) images of the tissue obtained at a number of key wavenumbers, with high spatial resolution, show variations in the chemical structure of the tissue with a feature size down to ~4 pm. The image formed from the ratio of the SNOM images obtained at 1252 cnr¹ and 1285 cnr¹ shows more contrast than the SNOM images obtained at these or a number of other individual wavenumbers. The discrimination between the two tissue types is dominated by the contribution from the 1252 cnr¹ signal, which is representative of nucleic acids, and this shows the OSCC tissue to be accompanied by two wide arcs of tissue which are particularly low in nucleic acids. Haematoxylin and Eosin (H&E) staining shows the tumour core in this specimen to be ~40 pm wide and the SNOM topography shows that the core centre is raised by ~1 pm compared to the surrounding tissue. Line profiles of the SNOM signal intensity taken through the highly keratinised core show that the increase in height correlates with an increase in the protein signal. SNOM line profiles show that the nucleic acids signal decreases at the centre of the tumour core between two peaks of higher intensity. All these nucleic acid features are ~25 pm wide, roughly the width of two cancer cells.

Introduction

There is considerable interest in the detection of cancer by applying machine learning algorithms to the analysis of the extensive datasets obtained by the application of infrared (IR) imaging spectroscopies to fixed human tissue. Considerable improvement in sensitivity and specificity has been demonstrated in the Gleason grading of prostate cancer when applying principal component discriminant function analysis (PC-DFA) to a Fourier transform infrared (FTIR) imaging dataset. Similarly, the application of convolutional neural networks to a combination of results obtained from FTIR spectral imaging and associated spatial information obtained from tissue microarrays was able to identify six major cellular and acellular constituents associated with breast cancer. There have been several reviews of advances in the instrumentation and application of the FTIR technique to cancer and the application of techniques for obtaining chemical information from FTIR. The inventors recently applied a novel machine learning multivariate metrics analysis (MA) technique to the analysis of FTIR images obtained from four cell lines associated with esophageal cancer. This was able to discriminate with accuracies in the range of 81% to 97% between OE19 and OE21 cell lines, associated respectively with adenocarcinoma and squamous carcinoma, and more importantly between cancer associated myofibroblasts (CAM) and adjacent tissue myofibroblasts (ATM) obtained from the same patient. In addition to discriminating between these cell lines, the MA yielded a number of key spectral biomarkers that had not been identified in previous FTIR studies of esophageal cancer. FTIR and Raman imaging has previously been applied to the discrimination of oral cancer from histologically normal or benign tissue in a number of studies. For example, PCA and cluster analysis has been used to produce pseudo-colour images of OSCC tissue microarrays and showed correspondence between FTIR and routine histology, suggesting that tissue types are separable by their IR spectra when appropriate methods are used to analyse the dataset. A multivariate analysis technique has been developed that combined principal component analysis (PCA) followed by linear discriminant analysis (LDA) to results obtained by Raman spectroscopy. This was able to discriminate between lymph nodes with benign pathology from those harbouring lymphoma or metastases of head and neck cancer with sensitivities and specificities of 81% and 89% respectively. Another study used a framework of feature selection and classification algorithms to identify spectral features which distinguished normal mucosa, precancerous tissue and cancer of the oral cavity. Particular wavenumbers, previously correlated with chemical moieties such as glycogen and proteins, were discriminatory which suggests that relevant information comparable to that previously obtained via other methodologies is achievable from such data. A comprehensive review of Raman and FTIR studies of oral cancers has recently been published. The present investigation examines the value of the MA technique in discriminating between lymph nodal metastasis of oral cancer and indigenous lymphoid tissue. High spatial resolution measurements using an aperture scanning near-field optical microscope (SNOM) provide additional insight into the chemical biology of the metastatic tissue.

Experimental

Preparation of samples for analysis

Archival formalin-fixed, paraffin-embedded (FFPE) tissue from cervical lymph node metastases were obtained from a single patient with oral squamous cell carcinoma (OSCC) following informed consent and under ethical approval (REC number EC 47.01). Regions of interest (ROIs) (n=2) containing both metastatic OSCC and surrounding lymphoid tissue were identified by light microscopy on sections routinely prepared and stained with Haematoxylin and Eosin (H&E). Cores of 1 mm diameter corresponding to the ROIs were then obtained from the FFPE blocks using a Beecher MTA-1 tissue microarrayer for constructing a tissue microarray block. Serial, 5 pm thick, sections were cut from the tissue microarray block and floated onto charged glass slides for histopathology and immunohistochemistry (IHC) and onto calcium fluoride (CaF2) disks for FTIR imaging. While sections for IHC were eventually subjected to deparaffin isation, sections for FTIR remained in paraffin wax to minimise further changes in hydration and structure of the samples. Six serial sections were utilised and comprised two sections for FTIR imaging sandwiched between two sections stained with H&E and two with IHC for pancytokeratins using the AE1AE3 antibody (Agilent DAKO, Stockport, UK) and a Bond RXTM autostainer (Leica Biosystems, Milton Keynes, UK). The H&E and IHC stained sections were scanned using an Aperio CS2scanner (Leica Biosystems) to facilitate co-registration with IR images.

FTIR experiments

Mid-IR spectroscopic images were acquired from each ROI using an Agilent Cary 620 FTIR microscope coupled to an Agilent Cary 670 FTIR spectrometer (Agilent, Stockport, UK) as described previously. Poor quality spectra, defined as having an Amide I absorbance (peak centre 1650 cm ¹) less than 0.1 or greater than 2, were removed from the dataset. This range was chosen so that outlier spectra arising from sub-optimal sample thickness would be discarded whilst retaining the vast majority of data. The spectra were then truncated to the fingerprint region (900 cnr¹ to 1800 cm ¹) and the region dominated by paraffin contributions (1350 cnr¹ to 1500 cnr¹) was omitted from the analysis. Each spectrum in the truncated dataset was then subject to a rubber-band baseline correction, followed by vector normalisation. Corrections for Mie scattering are unnecessary for FFPE tissue due to the refractive index matching between the tissue and paraffin, thus significantly reducing scattering artefacts. The histopathological and FTIR images were cross-referenced and spectra from the ROIs were identified and labelled as OSCC or lymphoid tissue as appropriate. Labelled FTIR data were used to train a discriminatory model using the MA technique. An equal number of spectra were randomly sampled from each image so as to mitigate the risk of overfitting to image-specific features. The MA model was trained using a three-fold cross validation regime, whereby the data is divided into three partitions, selecting two for training and holding out the third for testing. This process is repeated three times so that all data appears in both the training and testing sets.

SNOM experiments

Experiments were also performed using an aperture SNOM described previously. The infrared source was a quantum cascade laser (QCL) instrument (Daylight Solutions, San Diego, USA), equipped with three modules enabling an effective spectral range of 1965 cnr¹ to 1145 cnr¹ and pulsing at a rate of 80 kHz with pulse widths of 200-500 ns. The x-y piezo-stage was configured to scan a region of 300 ^c 300 pm with a step of 2 pm. The SNOM imaging tip was a specially prepared IR transmitting chalcogenide fibre (CorActive, Quebec, Canada) of core diameter 100 pm, sharpened by etching, to create a small aperture through which the SNOM images were collected concurrently with shear-force topography. The images were corrected for non-linearity of the piezo stage, and other common processing techniques such as streak removal and line levelling were applied. The images were co-registered and then a Gaussian smoothing of 2 pixels (4 pm), full-width half maximum (FWHM) was applied.

Results

Discrimination of OSCC metastases from lymph node tissue

The MA algorithm produces a ranked list of metrics, an ensemble of which produces the optimum discrimination. The trained MA model was able to discriminate between metastatic OSCC and the surrounding lymphoid tissue with a high sensitivity and specificity by utilising only the highest-ranking metric, specifically the ratio of intensities at 1252 cnr¹ and 1285 cnr¹ (Table 1 , Figure 1). The success of this metric is shown in Figure 1 in which the histograms of the ratio of the intensities of the discriminating wavenumbers obtained at each pixel in the areas of the FTIR images identified with each tissue type in the images used to train the algorithm are plotted. This shows that the test spectra conform very well with the decision boundaries formed by this metric, explaining the high AUC. These two wavenumbers and those contained in the next four metrics in rank order, 1254 cnr ¹/1285 cnr¹, 1250 cnr¹/1289 cnr¹, 1252 cnr¹/1287 cnr¹ and 1252 cnrVl289 cnr¹, draw attention to a very narrow region of the FTIR spectrum, wherein the average spectra of different types of tissue show differences (Figure 2). This highest-ranking metric discriminates between OSCC and lymphoid tissue better than the individual wavenumbers (Figure 3). Thus, although a correspondence between tumour cells stained by IHC [Figure 3(a)] and the low absorbance at 1252 cnr¹ [Figure 3(c)] is observed, a greater co-registration is seen between the IHC and the ratio of 1252 cnr¹/1285 cnr¹ [Figure 3(d)] However, topographically different areas of the metastasis (e.g. periphery versus the more heavily keratinised centre as appreciated on H&E sections) are not discriminated by the metric (Figure 3).

Table 1 : Measures of discrimination between metastatic OSCC and lymphoid nodal tissue for the highest-ranking metric. The mean and standard deviation are taken from across three cross validation partitions. *Area under the receiver operating characteristic (ROC) curve.

SNOM analysis of OSCC nodal metastases To further investigate the biological changes that underly the discriminatory metric, a second core from a different region in the same lymph node metastasis specimen shown in Figure 3 was dewaxed using the protocol described recently and topography and SNOM images were obtained of a small region of this tissue that contained the OSCC-lymphoid tissue interface. The results obtained in these experiments are shown in Figure 4. Figures 4(a) and 4(b) show the H&E and IHC stained images of this region of the core, respectively, and Figure 4(c) shows the topography obtained during the collection of SNOM images. SNOM images were collected at a number of wavenumbers that have been shown to be important in discriminating esophageal cancer cells and in the development of a dewaxing protocol for SNOM experiments: 1751 cnr¹, 1650 cnr¹, 1369 cnr¹ (shown in Figure 4(d), (e) and (f), respectively). The SNOM images obtained at the discriminating wavenumbers 1285 cnr¹ and 1252 cm ¹, defined above, are shown in Figure 4(g) and 4(h), respectively, and the ratio of the intensity of these two images is shown in Figure 4(i). The images indicate presence of tumour mass in the bottom right corner of the image, while the heterogeneity of the images indicates that additional, higher resolution, differences might also be identified in the tissue by this method. In order to bring out in more detail the information captured in the images obtained with high spatial resolution using the SNOM (Figure 4), the smaller region of the tumour in the bottom right-hand corner of the H&E image [Figure 4(a)] was used to create line profiles of the topography and the SNOM intensities at each wavenumber. Each profile was obtained along a 1 -pixel wide line close to the centre of the OSCC nodal metastasis (Figure 5). The noise levels in the SNOM images (and hence the profiles) were quantified by comparing raw images with de-noised images, and the noise-to-signal ratios were found to be <5% for all wavenumbers. Line profiles taken within 8 pm of those shown in Figure 5 show only very small differences from those shown in the Figure. The topography [Figure 5(a)] of the centre of the tumour can be seen to be higher than the surrounding tissue. This increase in height correlates with an increase in the protein signal [Figure 5(c)] in this region of the image. The line profiles obtained at other wavenumbers show more marked variations in intensity across smaller distances, indicating that there are many subtle changes in the chemistry of the metastasis. The SNOM images were taken in IR transmission mode and so the SNOM intensity profiles in Figure 5 have been inverted to present a more intuitive interpretation - peaks (valleys) in the profiles correspond to more (less) absorption. The profiles are presented on vertical scales that have been corrected for image acquisition parameters such as detector sensitivity. Comparison between profiles at different wavenumbers should not be taken as providing values for relative molecular concentrations, as the SNOM fibre transmission varies with wavenumber and each molecular vibration has a different transition dipole strength.

Discussion

The MA algorithm applied to FTIR data is able to discriminate between OSCC nodal metastases and surrounding lymphoid tissue on the basis of a single metric, the ratio of intensities at 1252 cnr¹ and 1285 cnr¹ (Table 1). It is apparent that, with sufficient spatial resolution and control of the signal-to-noise ratio, a direct inspection of the region of the spectrum between 1250 cnr¹ and 1289 cnr¹ might be used to identify these two tissue types in similar specimens. However, this observation is not necessarily true of all cancer types and it is often not possible to identify tissue types by direct inspection of spectra. Although there is a correspondence between the FTIR image obtained at 1252 cnr¹ [Figure 3(c)] and IHC [Figure 3(a)], the image formed from the ratio of the intensities of the images obtained at 1252 cnr¹ and 1285 cnr¹ [Figure 3(d)] seems in better agreement with the IHC than are either of the images obtained at the individual wavenumbers. This is to be expected given the high specificity, sensitivity and precision attributed to this metric during the MA of the FTIR spectra (Table 1 , Figure 1). FTIR absorbance at 1252 cnr¹ would be expected to be related to nucleic acid content. However, absorbance at this wavenumber was observed to be lower in OSCC metastasis compared with the surrounding lymphoid tissue [Fig, 3(c)] and this is reflected in the ratio of 1252 cnr¹ and 1285 cnr¹. This is surprising since it is known that OSCC, like many solid tumours, often shows changes in DNA ploidy and, indeed, that such changes may be an early event. This might be explained by the fact that the nuclei in lymphoid tissue are more closely packed than in the tumour, with its typically larger cells, and hence the IR absorbance at 1252 cnr¹ would be higher for lymphoid tissue. The inability of FTIR to discriminate between the periphery and highly keratinised centre of the metastasis [Figure 3 (a), (c) and (d)] was overcome in higher resolution studies utilising SNOM. The high spatial resolution of the SNOM images have the potential to provide some chemical information, although over a smaller region of the specimen and at a limited number of wavenumbers. This makes the choice of wavenumbers particularly important since biological macromolecules give complex IR absorbance spectra. Nevertheless, with a careful choice of wavenumbers, the SNOM images and line profile data, obtained with higher intrinsic spatial resolution than FTIR, can be used to infer on basic chemistry of individual tissues. The wavenumbers 1751 cnr¹, 1650 cnr¹ and 1369 cnr¹ are commonly attributed to lipids, the Amide I peak of proteins and the C-N stretch vibrations of the cytosine and guanine components of nucleic acids respectively29 and have been employed in previous SNOM studies, whereas the 1285 cnr¹ signal is characteristic of collagen. The image obtained at 1252 cnr¹ can be attributed to the (PO2 ) nucleic acids and/or RNA signal, since this wavenumber is within a broad range of absorption from these molecules. As expected, the SNOM images of the small region of the tissue microarray core shown in Figure 4 show detail on a finer length scale than is obtained in the diffraction-limited FTIR images of Figure 3. These images show variations in the chemical structure of the tissue with a feature size down to ~4 pm. All the images indicate differences in spectral intensities in the region of the OSCC nodal metastatic core and this region of the image is also clearly delineated in the topographic image [Figure 4(c)] The image formed from the ratio of the intensities of the SNOM images obtained at 1252 cnr¹ and 1285 cnr¹ [Figure 4(i)] shows more contrast between different areas of the tissue than the images obtained at any of the individual wavenumbers. In particular it shows that the centre of the tumour in the bottom right of Fig. 4(a) is bounded by two broad arcs of tissue in which the ratio of the intensity of the discriminating wavenumbers is particularly low. Thus, the SNOM images are able to provide more detail than the FTIR images and highlight differences between the centre and periphery of the metastasis. The line profiles obtained in the small region of the tumour core shown in Figure 5 provide more detail of the chemical differences therein. As regards topography [Figure 5(a)] the centre of the tumour was higher than the periphery. Although it is not possible to quantify this difference precisely due to the difficulty in calibrating the vertical scale of the topographic image, it was found to be ~1 pm. The increase in height correlates with an increase in the protein signal [Figure 5(c)] in this region of the image. The centre of the metastasis appeared highly keratinised and this is mirrored in the 1650 cnr¹ Amide I line profile which can be attributed to the a-helical structure of cytokeratins. Furthermore, changes in spatial arrangement and subpopulations of cytokeratins and the molecules related to keratinisation (involucrin, etc) are expected between the often heavily keratinised centre of tumour cells aggregates and the less keratinised periphery, the latter corresponding to the advancing front of the primary, which could be reflected in the line profile at 1650 cnr¹. In contrast to the smooth increase and decrease of both the height and the protein intensity in this region of the image, the line profiles obtained at other wavenumbers show more marked variations in intensity over smaller distances, indicating that there are subtle changes in the chemistry of the tissue. The attribution of the 1252 cnr¹ signal to the (PO2 ) vibration of nucleic acids is supported by the very close correspondence between the line profiles obtained at 1252 cnr¹ [Figure 5(f)] and 1369 cnr¹ [Figure 5(d)], since the latter is attributed to the C-N stretch vibrations of the cytosine and guanine components of nucleic acids. A similar correspondence between the line profiles of these two wavenumbers was found in all regions of the images examined. As previously mentioned, the line profile obtained at 1285 cnr¹ [Figure 5(e)] is attributable to collagen. However, given the relative paucity of collagen in lymph nodes, the discriminating metric of Table 1 possibly arises from variations in the levels of nucleic acids and collagen in the tissue, with the signal from the nucleic acids dominating the discrimination. This would be consistent with the relative discrimination between OSCC and lymphoid tissue obtained from FTIR data [Figure 3(c) compared to Figure 3(b)] Taking the peak in the line profile of the topography as a reference for the centre of the tumour, the nucleic acid line profile shows a small central reduction in intensity in the centre of the metastasis with two peaks in intensity ~25 pm on either side, which is consistent with the increased keratinisation at that sub-site. Two further reductions in intensity are observed at ~50 pm from the centre and correlate with the periphery of the metastasis, with each of these features ~25 pm in width, roughly corresponding to 2-3 layers of cancer cells. If this signal were based solely on absorbance by nucleic acids, this would appear counter-intuitive because the more differentiated, keratinised core of the tumour most likely contains fewer, mitotically inactive nuclei compared with the tumour periphery. However, if we use 1252 cnr¹ as a wavenumber characteristically absorbed by the phosphate groups in all nucleic acids and in the phosphate groups of phospholipids, we hypothesise that this increase in absorbance reflects a change in the RNA signature and/or an increase in endoplasmic reticulum commensurate with an increased proteinosynthetic events in this sub-site. The 1285 cnr¹ line profile represents a complex pattern of relative absorbance across the whole section, but notably indicates an increase immediately to the right of the tumour centre. The amount and distribution of collagen, including fibre alignment, density, width length and straightness, appear to differ between cancer types and at different sites within a tumour. These attributes have an effect on invasion, metastasis and apoptosis as well as being a prognostic factor correlated with cancer differentiation, invasion, lymph node metastasis, and clinical stage. Collagen concentration is also influenced by the hypoxic microenvironment and affects intensity of immune cell response. It is thus plausible that the differences observed in the 1285 cnr¹ SNOM line profiles are due to more subtle changes in collagen fibre structure than in concentration and require further investigation.

Conclusions

A novel machine learning algorithm, MA, has been shown to accurately discriminate between OSCC nodal metastasis and surrounding lymphoid tissue on the basis of a single metric, the ratio of FTIR intensities at 1252 cnr¹ and 1285 cnr¹. This metric yields discriminating sensitivities, specificities and precision of 99 ± 0.14%, 100 ± 0.01% and 100 ± 0.02%, respectively, and an AUC of 0.99 ± 0.0007. However, the topographically different periphery and highly keratinised centre of the metastasis are not discriminated by the metric in the diffraction-limited FTIR images. SNOM images of the tissues obtained at a number of key wavenumbers, with a higher spatial resolution, show variations in chemistry with a feature size down to ~4 pm. The image obtained from the ratio of the intensities of the SNOM images obtained at the discriminating wavenumbers supports the finding from the FTIR images that the discrimination between the two tissue types is dominated by the contribution from the 1252 cnr¹ signal which is representative of nucleic acids. Additional insight into the chemistry is revealed by line profiles of the SNOM intensity obtained at specific wavenumbers, representative of particular chemical moieties, in the region of the OSCC-lymphoid tissue interface. The differences between the periphery and the centre of the metastasis reflect our current biological knowledge, but also raise additional, more subtle, questions at the cellular level. This study demonstrates that a combination of the MA technique applied to labelled FTIR spectra together with SNOM images obtained at key wavenumbers identified by MA provides insight into the chemistry of tissues.

Method

Figure 6 schematically depicts a method according to an exemplary embodiment. The method is of discriminating between first cell types or first tissue types and second cell types or second tissue types.

At S601 , the method comprises providing a cell sample or a tissue sample including a first spatial region, comprising a first cell type or a first tissue type, and a second spatial region.

At S602, the method comprises obtaining a first absorption response comprising: illuminating the first spatial region using emitted infrared, IR, electromagnetic radiation, EMR, incident thereupon, wherein the IR EMR comprises a set of pairs of discrete wavelengths, including a first pair of discrete wavelengths consisting of a first wavelength and a second wavelength, wherein the first pair of discrete wavelengths is characteristic of a discriminating biomarker; and detecting respective absorbances of the incident first wavelength and the incident second wavelength of the first pair of discrete wavelengths by the first spatial region.

At S603, the method comprises obtaining a second absorption response comprising: illuminating the second spatial region using the IR EMR; and detecting respective absorbances of the incident first wavelength and the incident second wavelength of the first pair of discrete wavelengths by the second spatial region.

At S604, the method comprises discriminating between the first spatial region and the second spatial region based on a result of comparing the obtained first absorption response and the obtained second absorption response.

The method may include any of the steps described with respect to the first aspect.

Apparatus

Figure 7 schematically depicts an apparatus 1 according to an exemplary embodiment.

The apparatus 1 is for discriminating between first cell types or first tissue types and second cell types or second tissue types.

The apparatus 1 comprises a set of lasers 11 , including a first laser 11A tuned to emit a first wavelength of infrared, IR, electromagnetic radiation, EMR, of a first pair of discrete wavelengths of a set of pairs of discrete wavelengths and a second laser 11 B tuned to emit a second wavelength of IR EMR of the first pair of discrete wavelengths, wherein the first pair of discrete wavelengths is characteristic of a discriminating biomarker.

The apparatus 1 comprises a probe 12 (art photonics GmbH AP10100 ATR probe for detachable loops and art photonics GmbH AP10109 Detachable single loops for detachable loop probe or art photonics GmbH AP10191 Detachable triple loops for detachable loop probe), optically coupled to the set of lasers 11 (Thorlabs Inc. QD7500CM1 Distributed Feedback (DFB) quantum cascade laser (QCL) module (part number depends on wavelength specified) together with Thorlabs Inc. LDMC20/M Thermoelectrically cooled mounting for C- mount lasers and Thorlabs Inc. ITC4002QCL Benchtop laser diode/TEC controller for QCLs), for illuminating a cell sample or a tissue sample including a first spatial region, comprising a first cell type or first tissue type, and a second spatial region, using the first wavelength and the second wavelength of the first pair of discrete wavelengths incident thereupon. Suitable probes and lasers are also available from other manufacturers. The apparatus 1 comprises a detector 13 (Thorlabs Inc. PDAVJ10 HgCdTe (MCT) photodetector), optically coupled to the probe 12, for detecting respective absorbances of the incident first wavelength and the incident second wavelength of the first pair of discrete wavelengths by the first spatial region and by the second spatial region. Suitable detectors are also available from other manufacturers.

The apparatus 1 comprises a controller 14 configured to control the detector 13 to obtain a first absorption response from the first spatial region and a second absorption response from the second spatial region; and to discriminate between the first spatial region and the second spatial region based on a result of comparing the obtained first absorption response and the obtained second absorption response.

In this example, the first laser 11A is tuned to emit the first wavelength having a wavenumber 1285 cnr¹ and the second laser 11 B is tuned to emit the second wavelength having the wavenumber 1252 cnr¹, as described with respect to Figures 1 to 5. In this way, the apparatus may be used to accurately discriminate between oral squamous cell carcinoma (OSCC) nodal metastases and surrounding lymphoid tissue on the basis of a single metric 1252 cnr ¹/1285 cnr¹.

In more detail, the apparatus provides a relatively cheap, robust hand-held IR probe that can be used to discriminate between potentially malignant lesions, such as oral precancerous lesions, that look similar but have different outcomes. The apparatus is relatively simple and provides an immediate assessment of fixed, sectioned biopsies from any cancer for which the key biomarkers have been previously determined by applying the MLA to well-characterised tissue with known pathology and/or outcome. The apparatus is based on a closed loop of IR transmitting fibre illuminated by narrow band laser sources operating at the wavelengths corresponding to the IR biomarkers identified by the MLA. It is important to note that the IR biomarkers revealed by the MLA are actually ratios of the intensity of different IR wavelengths absorbed by the tissue. If one of the biomarker wavelengths is present in the tissue it will absorb the IR evanescent wave causing a change in the intensity of the corresponding ratio which can be monitored by phase-sensitive detector electronics and result in an immediate response. By detecting a change in the ratio of the intensities of two balanced signals using phase-sensitive detection, the sensitivity of the apparatus is significantly improved over conventional FTIR and Raman techniques.

This apparatus uses two laser sources and the design has been optimised by analysing unstained, formalin-fixed, paraffin embedded (FFPE) specimens of lymph node containing metastatic oral cancer tissue, as described previously. Although this is not a clinically important issue, the inventors have demonstrated that these two tissue types can be discriminated by just one ratio with each IR wavelength preferentially localised in just one tissue type, thus allowing the optimisation of signal delivery and detection, in agreement with the gold standard of cytokeratin stained sections.

By using more laser sources, early diagnosis of oral cancer may be optimised. The MLA has already been used to determine potential IR biomarkers that provide this discrimination, as described previously.

Although a preferred embodiment has been shown and described, it will be appreciated by those skilled in the art that various changes and modifications might be made without departing from the scope of the invention, as defined in the appended claims and as described above.

Attention is directed to all papers and documents which are filed concurrently with or previous to this specification in connection with this application and which are open to public inspection with this specification, and the contents of all such papers and documents are incorporated herein by reference.

All of the features disclosed in this specification (including any accompanying claims and drawings), and/or all of the steps of any method or process so disclosed, may be combined in any combination, except combinations where at most some of such features and/or steps are mutually exclusive.

Each feature disclosed in this specification (including any accompanying claims, and drawings) may be replaced by alternative features serving the same, equivalent or similar purpose, unless expressly stated otherwise. Thus, unless expressly stated otherwise, each feature disclosed is one example only of a generic series of equivalent or similar features.

The invention is not restricted to the details of the foregoing embodiment(s). The invention extends to any novel one, or any novel combination, of the features disclosed in this specification (including any accompanying claims and drawings), or to any novel one, or any novel combination, of the steps of any method or process so disclosed.

Claims

1 . A method of discriminating between first cell types or first tissue types and second cell types or second tissue types, the method comprising: providing a cell sample or a tissue sample including a first spatial region, comprising a first cell type or a first tissue type, and a second spatial region; obtaining a first absorption response comprising: illuminating the first spatial region using emitted infrared, IR, electromagnetic radiation, EMR, incident thereupon, wherein the IR EMR comprises a set of pairs of discrete wavelengths, including a first pair of discrete wavelengths consisting of a first wavelength and a second wavelength, wherein the first pair of discrete wavelengths is characteristic of a discriminating biomarker; and detecting respective absorbances of the incident first wavelength and the incident second wavelength of the first pair of discrete wavelengths by the first spatial region; obtaining a second absorption response comprising: illuminating the second spatial region using the IR EMR; and detecting respective absorbances of the incident first wavelength and the incident second wavelength of the first pair of discrete wavelengths by the second spatial region; and discriminating between the first spatial region and the second spatial region based on a result of comparing the obtained first absorption response and the obtained second absorption response.

2. The method according to claim 1 , wherein illuminating the first spatial region using the IR EMR incident thereupon comprises illuminating the first spatial region alternately using the first wavelength and the second wavelength of the first pair of discrete wavelengths.

3. The method according to any previous claim, comprising balancing the respective absorbances of the incident first wavelength and the incident second wavelength of the first pair of discrete wavelengths by the first spatial region by adjusting respective intensities of the first pair of discrete wavelengths of the IR EMR incident thereupon.

4. The method according to any previous claim, wherein detecting respective absorbances of the incident first wavelength and the incident second wavelength of the first pair of discrete wavelengths by the first spatial region comprises phase-sensitive detection thereof.

5. The method according to any previous claim, wherein illuminating the first spatial region using the emitted infrared IR EMR incident thereupon comprises contacting the first spatial region with a probe.

6. The method according to claim 5, comprising determining non-contact of the first spatial region with the probe.

7. The method according to any of claims 5 to 6, wherein the probe comprises and/or is a loop probe.

8. The method according to any previous claim, wherein obtaining the first absorption response comprises averaging the detected respective absorbances of the incident first wavelength and the incident second wavelength of the first pair of discrete wavelengths by the first spatial region.

9. The method according to any previous claim, wherein the set of pairs of discrete wavelengths includes a second pair of discrete wavelengths consisting of a first wavelength and a second wavelength, wherein the second pair of discrete wavelengths is characteristic of a discriminating biomarker.

10. The method according to any previous claim, wherein the first wavelength and/or the second wavelength of the first pair of discrete wavelengths corresponds with a spectral shoulder of the discriminating biomarker.

11 . An apparatus for discriminating between first cell types or first tissue types and second cell types or second tissue types, the apparatus comprising: a set of lasers, including a first laser tuned to emit a first wavelength of infrared, IR, electromagnetic radiation, EMR, of a first pair of discrete wavelengths of a set of pairs of discrete wavelengths and a second laser tuned to emit a second wavelength of IR EMR of the first pair of discrete wavelengths, wherein the first pair of discrete wavelengths is characteristic of a discriminating biomarker; a probe, optically coupled to the set of lasers, for illuminating a cell sample or a tissue sample including a first spatial region, comprising a first cell type or first tissue type, and a second spatial region, using the first wavelength and the second wavelength of the first pair of discrete wavelengths incident thereupon; a detector, optically coupled to the probe, for detecting respective absorbances of the incident first wavelength and the incident second wavelength of the first pair of discrete wavelengths by the first spatial region and by the second spatial region; and a controller configured to control the detector to obtain a first absorption response from the first spatial region and a second absorption response from the second spatial region; and to discriminate between the first spatial region and the second spatial region based on a result of comparing the obtained first absorption response and the obtained second absorption response.

12. The apparatus according to claim 11 , wherein the controller is configured to control the set of lasers to alternately to emit the first wavelength of IR EMR and the second wavelength of IR EMR of the first pair of discrete wavelengths.

13. The apparatus according to any of claims 11 to 12, wherein intensities of the emitted first pair of discrete wavelengths of the IR EMR are adjusted to balance the respective absorbances of the incident first wavelength and the incident second wavelength of the first pair of discrete wavelengths by the first spatial region.

14. The apparatus according to any of claims 11 to 13, wherein the detector comprises and/or is a phase-sensitive detector.

15. The apparatus according to any of claims 11 to 14, wherein the probe comprises and/or is a contact probe for contacting the first spatial region.

16. The apparatus according to any of claims 11 to 15, wherein the controller is configured to determined non-contact of the first spatial region with the probe.

17. The apparatus according to any of claims 15 to 16, wherein the probe comprises and/or is a loop probe.

18. The apparatus according to any of claims 11 to 17, wherein controller is configured to average the detected respective absorbances of the incident first wavelength and the incident second wavelength of the first pair of discrete wavelengths by the first spatial region.

19. The apparatus according to any of claims 11 to 18, wherein set of lasers includes a third laser tuned to emit a third wavelength of infrared, IR, electromagnetic radiation, EMR, of a second pair of discrete wavelengths of a set of pairs of discrete wavelengths and a optionally a fourth laser tuned to emit a fourth wavelength of IR EMR of the second pair of discrete wavelengths, wherein the second pair of discrete wavelengths is characteristic of a discriminating biomarker.

20. The apparatus according to any of claims 11 to 19, comprising an output device and wherein the controller is configured to control the output device according to whether the first spatial region and the second spatial region are mutually discriminated.

21. A computer comprising a processor and a memory configured to implement a method according to any of claims 1 to 10, a computer program comprising instructions which, when executed by a computer comprising a processor and a memory, cause the computer to perform a method according to any of claims 1 to 10 or a non-transient computer-readable storage medium comprising instructions which, when executed by a computer comprising a processor and a memory, cause the computer to perform a method according to any of claims 1 to 10.