DETAILED ACTION
Notice of Pre-AIA or AIA Status
The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
Priority
Foreign priority is not claimed.
Information Disclosure Statement
The information disclosure statement (IDS) submitted on 02/12/2025 is in compliance with the provisions of 37 CFR 1.97. Accordingly, the information disclosure statement is being considered by the examiner.
Claim Rejections - 35 USC § 103
In the event the determination of the status of the application as subject to AIA 35 U.S.C. 102 and 103 (or as subject to pre-AIA 35 U.S.C. 102 and 103) is incorrect, any correction of the statutory basis for the rejection will not be considered a new ground of rejection if the prior art relied upon, and the rationale supporting the rejection, would be the same under either status.
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102 of this title, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
Claims 1-2, 5-11, 15-19, 22-25 and 29-30 are rejected under 35 U.S.C. 103 as being unpatentable over Papagiannakis et al. (US Pub No. 20210249118 A1) in view of Struble et al. (US Pub No. 20220157400 A1).
Regarding Claim 1,
Papagiannakis A method for performing sex determination from a histological image, the method comprising: receiving, at one or more processors, a histological image of a thin section tissue sample from a subject; (Papagiannakis, [0008], [0029], Fig. 1, discloses a schematic diagram of a computing system in accordance with the disclosure. The computing system comprises: an image collection module 10; a local processor 15; a main server 20; an identity database 30; an imaging database 40. A local interface 12 couples the image collection module 10 with the local processor 15. A processing interface 22 couples the local processor 15 with the main server 20. A first identity interface 32 couples the identity database 30 with the local processor 15 and a second identity interface 34 couples the identity database 30 with the main server 20. A first image data interface 42 couples the imaging database 40 with the local processor 15 and a second image data interface 44 couples the imaging database 40 with the main server 20. It will be noted that the computing system of FIG. 1 incorporates parts that may be distinct from a computer, for example being part of an optical system and/or an electronics control system. However, these will all be considered part of the computing system for the purposes of this disclosure; Image data comprising a plurality of images of an examination area of a biological tissue (particularly a cervix of a subject) is received at a computing system. Each image is captured at different times during a period in which topical application of a pathology differentiating agent to the examination area of the tissue causes transient optical effects. In particular, the pathology differentiating agent may comprise acetic acid (typically dilute acetic acid, usually to 3-5%), such that the transient optical effects may comprise an aceto-whitening effect (although other differentiating agents and/or optical effects may be possible, for instance using molecular diagnostics). The examination area may be exposed to optical radiation, which may be broadband (across the majority or all of the optical spectrum) or narrowband (limited to a one or a range of specific wavelengths defining only one or a limited range of colours, possibly including ultraviolet and/or infrared), during the period of image capture. The images captured subsequent to application of the agent (for example, at predetermined and/or regular intervals) may therefore show progress of the transient optical effects. The received image data (which may have been subject to image processing, as discussed below) is provided as an input to a machine learning algorithm (operative on the computing system). The machine learning algorithm allocates one of a plurality of classifications to the tissue. The cervix may also be segmented, for example based on the application of one or more masks (for instance, defined by recognition of morphology or feature extraction) and/or on the basis of local classifications applied across the tissue. Thus, the tissue may be classified into discrete and defined sub-areas of the examination area of the tissue, in particular with a different classification allocated to each segment of the cervix. The classifications may be defined by a scale of values on a continuous range (such as from 0 to 1 or 0 to 100) or a set of discrete options, which might include a plurality of disease tags (for example: negative vs positive or for example: low risk; medium risk; high risk, specific disease states, for instance: CIN1, CIN2, CIN3, or the presence of one or more characteristics of morphology, such as presence of atypical vessels, sharp lesion borders or disease, such as persistent or dense aceto-whitening); tissue image is collected for examination)
processing, using the one or more processors, the histological image to generate a plurality of image patches; (Papagiannakis, [0011], discloses image data provided as an input to the machine learning algorithm may be derived from captured optical images, that is raw images, taken by an image collection module (which may form part of the computer system or it may be external). For example, the optical (raw) images may be scaled (for instance based on a focusing distance for the respective optical image). This may allow the plurality of images for one tissue to have the same scale as for another tissue. Each of the images may have the same pixel arrangement (that is, the same image size and shape). Alignment of the plurality of images may be achieved by applying one or more transformations to the optical images. Artefacts may be removed from the optical images by image analysis and/or processing. The images may be broken down or sub-divided into patches (for instance, a contiguous block of pixels, preferably two-dimensional), which may form the image data. The patches may be overlapping, for example patches created at a stride smaller than the patch size, which may increase resolution; obtained tissue images are broken down to patches depending on size and shape required)
providing, using the one or more processors, the plurality of image patches as input to a trained machine learning model, wherein the trained machine learning model is configured to (1) classify an image patch of the plurality of image patches as belonging to at least one of one or more feature categories and, based on a distribution of feature categories identified in the plurality of image patches, (Papagiannakis, [0012], [0017-0019], [0041], discloses An additional input to the machine learning algorithm may be based on processing of each of the plurality of images to extract tailored features, based on mathematical functions describing local colour, gradient and texture, although other characteristics may also be envisioned. This may be done separately on sub-portions of the image defined as patches of a block of pixels (for example, a square block of 8×8, 16×16, 32×32 pixels or other sizes, a rectangular block or another shape of block). Each image may be broken down to a number of patches with a stride between them, which can be at 4, 8, 16, 32 pixels (with other sizes also possible). As noted above, the patches may be provided as the image data provided as an input to the machine learning algorithm; machine learning algorithm is advantageously trained based on a respective plurality of images and a respective allocated classification (or classifications, if applicable) for each of a plurality of other biological tissues. The number of other biological tissues may be large, for instance at least 500, 1000, 2000 or 5000. The allocated classification may be such as that a specific region (or group of multiple regions) of each tissue is characterized by histopathology readings; machine learning algorithm may also be continually and/or dynamically trained (incremental learning) using methods such as transfer learning and selective re-training by providing a user-determined or database classification for the tissue (such as provided by a medical professional, for example from biopsy or excisional treatment with histology or subjective assessment) to the machine learning algorithm (and/or a version of the machine learning algorithm operative on a second computer system). Where the machine learning algorithm is provided in a distributed way, a first part may be provided local to the image collection module and a second part may be provided more remotely. Both parts may be capable of allocating classifications. The continual (dynamic) training may be applied only to the second part, particularly in batches and may incorporate data from multiple different first parts. The first part may be a fixed algorithm, which may be updated (at intervals, for instance after a plurality of classifications or a specific length of time); Beneficially, the machine learning algorithm allocates one of a plurality of classifications to each of one or more segments of the tissue. The segment or segments may be identified from the image data using the machine learning algorithm, for example to identify individual regions of interest or lesions. A portion of the images corresponding with the cervix may be identified in some embodiments, which may allow suitable segments to be determined. For example, the classifications may take the form of diagnostic tags. In another option, the classifications may be in the form of a ‘heat map’ of an image of the tissue (that is an output image, advantageously based on the plurality of images), in which the intensity and/or colour of each pixel is indicative of a classification for that pixel, preferably a probabilistic classification for the pixel. In another option the classification output may be in the form of a risk tag, whereby a tissue area is highlighted (for example by a bounding box) as no, low or high risk. Optionally, an overall classification for the tissue may also be allocated. This may be based on the classifications allocated to the segments or the result of a (separate) parallel machine learning model; machine learning algorithm is advantageously trained based on a respective plurality of images and a respective allocated classification (or classifications, if applicable) for each of a plurality of other biological tissues. The number of other biological tissues may be large, for instance at least 500, 1000, 2000 or 5000. The allocated classification may be such as that a specific region (or group of multiple regions) of each tissue is characterized by histopathology readings; feature categories based of the patches as color or texture are determined and further the distribution of feature categories tissue (heat map) is classified according to feature distribution into tissue being cervical abnormality or not)
(2) output a sex determination for the subject; (Papagiannakis, [0054-0055], discloses AI segmentation and classification results may be displayed as a probabilistic “heat-map” (a parametric pseudo-colour map), as an output of the AI. This is shown as an AI output 130 in FIG. 2. The heat-map output from the AI (which is different from the parametric pseudo-colour map produced by processing the images as described above and which may be used as an input to the AI) is then advantageously displayed in a graphic form as an overlay on the cervical images to the system operator during the examination (for instance via the local processor 15) to facilitate reading and clinical decisions. The resolution of the heat-map may be the same as the scaled images provided as input to the AI (such as 1024×768 or 2048×1536, for instance). This (or similar image processing) may allow overlaying of the AI heat-map output on an image of the cervix captured during examination (for instance, post-processing). Such a “heat-map” may be of significant clinical utility (such as for biopsy site identification or excisional treatment); AI segmentation and classification results may alternatively be displayed as a bounding box that indicates areas that achieve a classification score above a pre-defined threshold as an output of the AI. For example this may be as indication of no, low or high risk, or directly with a disease tag, for instance: Normal; CIN1; CIN2; CIN3; AIS; or Invasive cancer; result of the determination of tissue classification is output)
outputting, using the one or more processors, the determined sex of the subject. (Papagiannakis, [0029-0030], [0054-0055], discloses AI segmentation and classification results may be displayed as a probabilistic “heat-map” (a parametric pseudo-colour map), as an output of the AI. This is shown as an AI output 130 in FIG. 2. The heat-map output from the AI (which is different from the parametric pseudo-colour map produced by processing the images as described above and which may be used as an input to the AI) is then advantageously displayed in a graphic form as an overlay on the cervical images to the system operator during the examination (for instance via the local processor 15) to facilitate reading and clinical decisions. The resolution of the heat-map may be the same as the scaled images provided as input to the AI (such as 1024×768 or 2048×1536, for instance). This (or similar image processing) may allow overlaying of the AI heat-map output on an image of the cervix captured during examination (for instance, post-processing). Such a “heat-map” may be of significant clinical utility (such as for biopsy site identification or excisional treatment); AI segmentation and classification results may alternatively be displayed as a bounding box that indicates areas that achieve a classification score above a pre-defined threshold as an output of the AI. For example this may be as indication of no, low or high risk, or directly with a disease tag, for instance: Normal; CIN1; CIN2; CIN3; AIS; or Invasive cancer; Referring first to FIG. 1, there is shown a schematic diagram of a computing system in accordance with the disclosure. The computing system comprises: an image collection module 10; a local processor 15; a main server 20; an identity database 30; an imaging database 40. A local interface 12 couples the image collection module 10 with the local processor 15. A processing interface 22 couples the local processor 15 with the main server 20. A first identity interface 32 couples the identity database 30 with the local processor 15 and a second identity interface 34 couples the identity database 30 with the main server 20. A first image data interface 42 couples the imaging database 40 with the local processor 15 and a second image data interface 44 couples the imaging database 40 with the main server 20. It will be noted that the computing system of FIG. 1 incorporates parts that may be distinct from a computer, for example being part of an optical system and/or an electronics control system. However, these will all be considered part of the computing system for the purposes of this disclosure; image collection module 10 is a colposcopic imaging unit, for capturing and collection of optical images of an examination area, in particular a cervix uteri. Although the main embodiment of the present invention relates to a colposcopic system and there are significant and distinct advantages applicable to such a system, it will be understood that the implementation described herein may be used for other types of system for examination and/or imaging of biological tissue. The image collection module 10 is controlled by the local processor 15, which may include a user interface, for example comprising controls and/or display. The identity database 30 is used to store patient identity data. During an examination, the local processor may interface with the identity database 30 using the first identity interface 32, to retrieve identity data for the patient being examined. Images collected during the examination are stored in the imaging database 40 via the first image data interface 42. A patient identifier may be stored with the patent images to allow cross-referencing with the information stored in the identity database 30; result of the determination of tissue classification is output)
Papagiannakis does not explicitly disclose (2) output a sex determination for the subject;
Struble discloses (2) output a sex determination for the subject; (Struble, [0053], discloses methods for determining frequency of X and Y sequences in a maternal sample. These frequencies can be used, e.g., to determine fetal sex, and/or for identifying X chromosomal aneuploidies, Y chromosomal aneuploidies and/or sex chromosome mosaicisms. The samples are maternal samples comprising both maternal and fetal DNA such as maternal blood samples (i.e., whole blood, serum or plasma). The methods enrich and/or isolate and amplify one or, preferably, several to many selected nucleic acid regions in a maternal sample that correspond to the X and Y chromosomes and one or more autosomes that are used to determine the presence or absence and/or relative quantity or frequency of X and Y chromosomal sequences in view of the percent of fetal DNA present in the sample. As described in detail supra, the methods of the invention preferably employ one or more selective amplification, ligation or enrichment (e.g., using one or more nucleic acids that specifically hybridize to the selected nucleic acid regions) steps to enhance the content of the selected nucleic acid regions in the sample. The selective amplification, ligation and/or enrichment steps typically include mechanisms to engineer copies of the selected nucleic acid regions for further isolation, amplification and analysis. This selective approach is in direct contrast to the random amplification approach used by other techniques, e.g., massively parallel shotgun sequencing, as such techniques generally involve random amplification of all or a substantial portion of the genome; biological samples profiles are processed to determine frequencies of X or Y chromosome or nucleic acid DNA sequencing to determine fetal gender based on features of the biological samples collected)
Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention was made to combine the teachings of Papagiannakis in view of Struble having a method of classifying tissues images collected by processing the features and its distributions using machine learning algorithm , with the teachings of Struble having a system of classifying tissue based on number of X or Y chromosomes that determines the gender of fetal of which tissue is collected in order to determine gender of fetal in applications including early detection of risk factors.
Regarding Claim 2,
The combination of Papagiannakis and Struble further discloses wherein the histological image comprises a histopathology image. (Papagiannakis, [0018], discloses machine learning algorithm is advantageously trained based on a respective plurality of images and a respective allocated classification (or classifications, if applicable) for each of a plurality of other biological tissues. The number of other biological tissues may be large, for instance at least 500, 1000, 2000 or 5000. The allocated classification may be such as that a specific region (or group of multiple regions) of each tissue is characterized by histopathology readings; histopathology images are obtained). Additionally, the rational and motivation to combine the references Papagiannakis and Struble as applied in rejection of claim 1 apply to this claim.
Regarding Claim 5,
The combination of Papagiannakis and Struble further discloses wherein the one or more feature categories comprise at least one Barr body feature category. (Struble, [0028-0030], [0043], discloses term “chromosomal abnormality” refers to any genetic variant for all or part of a chromosome. The genetic variants may include but not be limited to any copy number variant such as duplications or deletions, translocations, inversions, and mutations; term “intersex mosaicism” or “sex chromosome mosaicism” or “sex chromosome mosaic” refers to the presence of two or more populations of cells with different sex chromosome genotypes in one individual. Intersex mosaicisms arise when some cells in an individual have, e.g., two X chromosomes (XX) and other cells in the individual have one X chromosome and one Y chromosome (XY); when some cells in an individual have one X chromosome (XO) and other cells in the individual have one X chromosome and one Y chromosome (XY); or when some cells in an individual have two X chromosomes and one Y chromosome (XXY) and other cells in the individual have one X chromosome and one Y chromosome (XY); term “selected nucleic acid region” as used herein refers to a nucleic acid region corresponding to an individual chromosome. Selected nucleic acid regions may be directly isolated and enriched from the sample for detection, e.g., based on hybridization and/or other sequence-based techniques, or they may be amplified using the sample as a template prior to detection of the sequence; gene mutations or alterations (barr body inactive cells nucleic acid chromosomes) sequence or patterns are determined as one of the features). Additionally, the rational and motivation to combine the references Papagiannakis and Struble as applied in rejection of claim 1 apply to this claim.
Regarding Claim 6,
The combination of Papagiannakis and Struble further discloses wherein the trained machine learning model is configured to predict whether an individual image patch is positive or negative for a presence of the at least one Barr body feature. (Struble, [0028-0030], [0043], [0141], discloses term “chromosomal abnormality” refers to any genetic variant for all or part of a chromosome. The genetic variants may include but not be limited to any copy number variant such as duplications or deletions, translocations, inversions, and mutations; term “intersex mosaicism” or “sex chromosome mosaicism” or “sex chromosome mosaic” refers to the presence of two or more populations of cells with different sex chromosome genotypes in one individual. Intersex mosaicisms arise when some cells in an individual have, e.g., two X chromosomes (XX) and other cells in the individual have one X chromosome and one Y chromosome (XY); when some cells in an individual have one X chromosome (XO) and other cells in the individual have one X chromosome and one Y chromosome (XY); or when some cells in an individual have two X chromosomes and one Y chromosome (XXY) and other cells in the individual have one X chromosome and one Y chromosome (XY); term “selected nucleic acid region” as used herein refers to a nucleic acid region corresponding to an individual chromosome. Selected nucleic acid regions may be directly isolated and enriched from the sample for detection, e.g., based on hybridization and/or other sequence-based techniques, or they may be amplified using the sample as a template prior to detection of the sequence; discloses percent fetal cell free DNA has been calculated, this data is combined with methods for detection and quantification of X and Y chromosome sequences to determine the likelihood that a fetus may be female, male, aneuploid for the X chromosome, aneuploid for the Y chromosome, an X chromosome mosaic, a Y chromosome mosaic. It can also be used in the determination of maternal aneuploidies, including mosaicism, or to identify whether the maternal sample being tested is contaminated; likelihood of feature of chromosomes are determined to predict presence or absence or positive or negative presence of X or Y chromosomes or their patterns). Additionally, the rational and motivation to combine the references Papagiannakis and Struble as applied in rejection of claim 1 apply to this claim.
Regarding Claim 7,
The combination of Papagiannakis and Struble further discloses wherein the trained machine learning model is configured to predict a sex for the subject based on a comparison of a number of Barr body positive image patches or a number of Barr body negative image patches to a predetermined threshold. (Struble, [0028-0030], [0043], [0141], discloses term “chromosomal abnormality” refers to any genetic variant for all or part of a chromosome. The genetic variants may include but not be limited to any copy number variant such as duplications or deletions, translocations, inversions, and mutations; term “intersex mosaicism” or “sex chromosome mosaicism” or “sex chromosome mosaic” refers to the presence of two or more populations of cells with different sex chromosome genotypes in one individual. Intersex mosaicisms arise when some cells in an individual have, e.g., two X chromosomes (XX) and other cells in the individual have one X chromosome and one Y chromosome (XY); when some cells in an individual have one X chromosome (XO) and other cells in the individual have one X chromosome and one Y chromosome (XY); or when some cells in an individual have two X chromosomes and one Y chromosome (XXY) and other cells in the individual have one X chromosome and one Y chromosome (XY); term “selected nucleic acid region” as used herein refers to a nucleic acid region corresponding to an individual chromosome. Selected nucleic acid regions may be directly isolated and enriched from the sample for detection, e.g., based on hybridization and/or other sequence-based techniques, or they may be amplified using the sample as a template prior to detection of the sequence; discloses percent fetal cell free DNA has been calculated, this data is combined with methods for detection and quantification of X and Y chromosome sequences to determine the likelihood that a fetus may be female, male, aneuploid for the X chromosome, aneuploid for the Y chromosome, an X chromosome mosaic, a Y chromosome mosaic. It can also be used in the determination of maternal aneuploidies, including mosaicism, or to identify whether the maternal sample being tested is contaminated; likelihood of feature of chromosomes are determined to predict presence or absence or positive or negative presence of X or Y chromosomes or their patterns). Additionally, the rational and motivation to combine the references Papagiannakis and Struble as applied in rejection of claim 1 apply to this claim.
Regarding Claim 8,
The combination of Papagiannakis and Struble further discloses wherein the trained machine learning model is configured to predict a sex for the sample based on a distribution of Barr body positive or Barr body negative image patches across the histopathology image. (Struble, [0028-0030], [0043], [0141], discloses term “chromosomal abnormality” refers to any genetic variant for all or part of a chromosome. The genetic variants may include but not be limited to any copy number variant such as duplications or deletions, translocations, inversions, and mutations; term “intersex mosaicism” or “sex chromosome mosaicism” or “sex chromosome mosaic” refers to the presence of two or more populations of cells with different sex chromosome genotypes in one individual. Intersex mosaicisms arise when some cells in an individual have, e.g., two X chromosomes (XX) and other cells in the individual have one X chromosome and one Y chromosome (XY); when some cells in an individual have one X chromosome (XO) and other cells in the individual have one X chromosome and one Y chromosome (XY); or when some cells in an individual have two X chromosomes and one Y chromosome (XXY) and other cells in the individual have one X chromosome and one Y chromosome (XY); term “selected nucleic acid region” as used herein refers to a nucleic acid region corresponding to an individual chromosome. Selected nucleic acid regions may be directly isolated and enriched from the sample for detection, e.g., based on hybridization and/or other sequence-based techniques, or they may be amplified using the sample as a template prior to detection of the sequence; discloses percent fetal cell free DNA has been calculated, this data is combined with methods for detection and quantification of X and Y chromosome sequences to determine the likelihood that a fetus may be female, male, aneuploid for the X chromosome, aneuploid for the Y chromosome, an X chromosome mosaic, a Y chromosome mosaic. It can also be used in the determination of maternal aneuploidies, including mosaicism, or to identify whether the maternal sample being tested is contaminated; likelihood of feature of chromosomes are determined to predict presence or absence or positive or negative presence of X or Y chromosomes or their patterns and fetal sex is determined based on the expressions of the genes in the images obtained). Additionally, the rational and motivation to combine the references Papagiannakis and Struble as applied in rejection of claim 1 apply to this claim.
Regarding Claim 9,
The combination of Papagiannakis and Struble further discloses wherein the one or more feature categories comprise at least one feature category associated with normal cells. (Papagiannakis, [0051], discloses AI (particularly the algorithm operative on the main server 20) is trained to classify the tissue from which the images were captured in some sense. Different datasets may be used to train different aspects of the AI. The type or types of data used for training may typically include any one or more of the type or types of data used for classification. In one implementation, the AI is configured to give a Cervical Intraepithelial Neoplasia (CIN) classification, based on training data comprising images and a relevant classification from a well characterized set of patient cases with known biopsy areas and histopathology outcomes. In particular, this is a set of cases with known sites that were biopsied and histology outcomes of the biopsies are advantageously known. Expert reviewer annotations of suspicious areas may also be available and these may be provided as further training data. In certain implementations, the set of cases have undergone excisional treatment and a detailed mapping of their histology is available, including multiple sections per treatment specimen, which can also be provided as training data. The AI may classify the cervix on a risk scale, where the different levels of this scale correspond to the patient's overall risk of having different grades of CIN (for example on a scale of 0 to 1, 0 to 100 or 1 to 100 on an integer or continuous scale). Different thresholds on this scale may be selected to fine-tune the final performance or provide a direct indication of no, low or high risk. In another embodiment the AI may directly provide the results in classifications, for example Normal, CIN1, CIN2, CIN3, AIS or Invasive cancer (one of a plurality of disease tags); normal cells are determined in image). Additionally, the rational and motivation to combine the references Papagiannakis and Struble as applied in rejection of claim 1 apply to this claim.
Regarding Claim 10,
The combination of Papagiannakis and Struble further discloses wherein the one or more feature categories comprise at least one feature category associated with cancer cells. (Papagiannakis, [0051], discloses AI (particularly the algorithm operative on the main server 20) is trained to classify the tissue from which the images were captured in some sense. Different datasets may be used to train different aspects of the AI. The type or types of data used for training may typically include any one or more of the type or types of data used for classification. In one implementation, the AI is configured to give a Cervical Intraepithelial Neoplasia (CIN) classification, based on training data comprising images and a relevant classification from a well characterized set of patient cases with known biopsy areas and histopathology outcomes. In particular, this is a set of cases with known sites that were biopsied and histology outcomes of the biopsies are advantageously known. Expert reviewer annotations of suspicious areas may also be available and these may be provided as further training data. In certain implementations, the set of cases have undergone excisional treatment and a detailed mapping of their histology is available, including multiple sections per treatment specimen, which can also be provided as training data. The AI may classify the cervix on a risk scale, where the different levels of this scale correspond to the patient's overall risk of having different grades of CIN (for example on a scale of 0 to 1, 0 to 100 or 1 to 100 on an integer or continuous scale). Different thresholds on this scale may be selected to fine-tune the final performance or provide a direct indication of no, low or high risk. In another embodiment the AI may directly provide the results in classifications, for example Normal, CIN1, CIN2, CIN3, AIS or Invasive cancer (one of a plurality of disease tags); abnormal (cancer) cells are determined in image). Additionally, the rational and motivation to combine the references Papagiannakis and Struble as applied in rejection of claim 1 apply to this claim.
Regarding Claim 11,
The combination of Papagiannakis and Struble further discloses wherein the trained machine learning model comprises a semantic segmentation computer vision model. (Papagiannaki, [0053], discloses a single classification for the tissue may be output from the AI, other options are also possible. In a specific implementation, the AI analyses and may segment the image of the cervix for each patient examined with the system. The image may be segmented in a predetermined way or based on the identification of lesions or disease risk and optionally may be done outside the machine learning algorithm. Each segment of the cervix is then classified on a risk scale for the estimated risk for different grades of CIN (as discussed above) or to provide a classification from one of a number of discrete disease states. Optionally, the AI may also classify each pixel and/or segment in accordance with a determined presence of a morphological characteristic. This may be an intermediate output of the AI, which may be used to determine further classifications but need not be provided as an output to a user; each pixel is classified into specific categories (semantic classification)). Additionally, the rational and motivation to combine the references Papagiannakis and Struble as applied in rejection of claim 1 apply to this claim.
Regarding Claim 15,
The combination of Papagiannakis and Struble further discloses wherein the trained machine learning model is trained using a supervised learning approach. (Papagiannakis, [0062], discloses the second machine learning algorithm may be trained by providing a user-determined or database classification for the tissue to the second machine learning algorithm (for instance from a biopsy with known histology, as discussed above). However, the first machine learning algorithm is optionally not trained by providing a user-determined or database classification for the tissue to the first machine learning algorithm. In this way, a (quick and/or lower complexity) machine learning algorithm may be provided without training (that is, a fixed algorithm), with a (slower and/or more sophisticated) machine learning algorithm provided with continuous dynamic training (incremental learning), for instance based on additional data being provided. In accordance with continuous dynamic training for example, the second machine learning algorithm may be provided with, for each of a plurality of examination areas of one or more biological tissues one or more of: a plurality of images of the examination area (as provided to a machine learning algorithm operative on a computer system); one or more biopsy locations (carried out for that examination area); the plurality of classifications to each of a plurality of segments of the tissue allocated by the first machine learning algorithm (operative on the computer system, that is, a local algorithm); and results of histopathology for the tissue. The machine learning algorithm without training may be local to the image capture and/or the machine learning algorithm with continuous dynamic training may be remote to the image capture. The process of continuous dynamic training is advantageously performed in batches. Beneficially, the process may incorporate data from multiple separate image capture devices (each with a respective, local machine learning algorithm). The first machine learning algorithm (the fixed algorithm) may be updated from time to time; the machine learning classifier may be trained on labelled data or unlabelled data continuously training with new data or keeping it fixed). Additionally, the rational and motivation to combine the references Papagiannakis and Struble as applied in rejection of claim 1 apply to this claim.
Regarding Claim 16,
The combination of Papagiannakis and Struble further discloses wherein the trained machine learning model is trained using at least one training data set comprising annotated image patch data that has been labeled by a pathologist as either including or not including a visible Barr body. (Papagiannakis, [0063], Experimental results will now be discussed. The experiments performed are explained with reference to FIG. 3, in which there is illustrated schematically a flowchart detailing a methodology for the experimental system. This flowchart indicates a working pipeline, subsequent to a step of selection of patient datasets that meet basic quality criteria (such as well-focused images, complete image sequence, no significant artefacts and known biopsy results). Firstly, annotation of the images for training 200 (marking of biopsy areas and appending disease labels to them) was carried out by reviewing the images and videos of the biopsy procedure for accurate placement of the labels on the tissue. This was followed by mask generation 210 comprising extraction of corresponding image masks. Extraction of patches 220 was then performed across 17 time-points. Feature extraction 230 comprises extracting features from each biopsy area and separately for all patches. A data imputation technique 240 was then performed to account for any missing values. Deep learning scheme step 250 comprises set-up and training of three different machine learning schemes for the calculation of probabilities for each patch. Finally, heat-map generation 260 results from the outputs of the deep learning schemes for the test cases. The test cases were prepared in a similar way to that described by the methodology of FIG. 3. The only difference was that the models did not know the disease status of the biopsy area (that is, in the annotation 200), but had to predict it instead; image data are labelled with different class categories (including feature or not including)). Additionally, the rational and motivation to combine the references Papagiannakis and Struble as applied in rejection of claim 1 apply to this claim.
Regarding Claim 17,
The combination of Papagiannakis and Struble further discloses wherein the trained machine learning model is trained using a weakly supervised approach. (Papagiannakis, [0062], discloses the second machine learning algorithm may be trained by providing a user-determined or database classification for the tissue to the second machine learning algorithm (for instance from a biopsy with known histology, as discussed above). However, the first machine learning algorithm is optionally not trained by providing a user-determined or database classification for the tissue to the first machine learning algorithm. In this way, a (quick and/or lower complexity) machine learning algorithm may be provided without training (that is, a fixed algorithm), with a (slower and/or more sophisticated) machine learning algorithm provided with continuous dynamic training (incremental learning), for instance based on additional data being provided. In accordance with continuous dynamic training for example, the second machine learning algorithm may be provided with, for each of a plurality of examination areas of one or more biological tissues one or more of: a plurality of images of the examination area (as provided to a machine learning algorithm operative on a computer system); one or more biopsy locations (carried out for that examination area); the plurality of classifications to each of a plurality of segments of the tissue allocated by the first machine learning algorithm (operative on the computer system, that is, a local algorithm); and results of histopathology for the tissue. The machine learning algorithm without training may be local to the image capture and/or the machine learning algorithm with continuous dynamic training may be remote to the image capture. The process of continuous dynamic training is advantageously performed in batches. Beneficially, the process may incorporate data from multiple separate image capture devices (each with a respective, local machine learning algorithm). The first machine learning algorithm (the fixed algorithm) may be updated from time to time; the machine learning classifier may be trained on labelled data or unlabelled data continuously training with new data or keeping it fixed). Additionally, the rational and motivation to combine the references Papagiannakis and Struble as applied in rejection of claim 1 apply to this claim.
Regarding Claim 18,
The combination of Papagiannakis and Struble further discloses wherein the trained machine learning model is trained using at least one training data set comprising both labeled and unlabeled image patch data. (Papagiannakis, [0063], Experimental results will now be discussed. The experiments performed are explained with reference to FIG. 3, in which there is illustrated schematically a flowchart detailing a methodology for the experimental system. This flowchart indicates a working pipeline, subsequent to a step of selection of patient datasets that meet basic quality criteria (such as well-focused images, complete image sequence, no significant artefacts and known biopsy results). Firstly, annotation of the images for training 200 (marking of biopsy areas and appending disease labels to them) was carried out by reviewing the images and videos of the biopsy procedure for accurate placement of the labels on the tissue. This was followed by mask generation 210 comprising extraction of corresponding image masks. Extraction of patches 220 was then performed across 17 time-points. Feature extraction 230 comprises extracting features from each biopsy area and separately for all patches. A data imputation technique 240 was then performed to account for any missing values. Deep learning scheme step 250 comprises set-up and training of three different machine learning schemes for the calculation of probabilities for each patch. Finally, heat-map generation 260 results from the outputs of the deep learning schemes for the test cases. The test cases were prepared in a similar way to that described by the methodology of FIG. 3. The only difference was that the models did not know the disease status of the biopsy area (that is, in the annotation 200), but had to predict it instead; image data are labelled with different class categories (including feature or not including)). Additionally, the rational and motivation to combine the references Papagiannakis and Struble as applied in rejection of claim 1 apply to this claim.
Regarding Claim 19,
The combination of Papagiannakis and Struble further discloses wherein the trained machine learning model is trained using a multiple instance learning approach.
Regarding Claim 22,
The combination of Papagiannakis and Struble further discloses wherein the determined sex of the subject is used to identify and/or correct a clerical error in clinical data associated with the histological image. (Papagiannakis, [0009], discloses automatic in vivo or in vitro classification of the biological tissue may be achieved. The use of multiple images taken over the progress of the transient optical effects may have a significant improvement on sensitivity and/or specificity of the classification over existing methods. Sensitivity and specificity may refer to the ability to identify cervical dysplasia and/or cervical neoplasia. Sensitivity thus refers to the ability to correctly identify tissues displaying cervical dysplasia and/or cervical neoplasia. Specificity thus refers to the ability to correctly identify tissues that do not display cervical dysplasia and/or cervical neoplasia. Depending upon the application context, the output may be focussed on maximising sensitivity or specificity, or operating at a threshold that may optimal for one or both. Although classification may be presented based on the subject/tissue as a whole, the invention permits regions of the tissue suspected to be precancerous or cancerous to be identified. This may be advantageous for directing biopsy or treatment, including surgical resection. These sites may be biopsied to confirm the identification. A successful implementation of such a system may render the taking of biopsy unusual in many or most cases. For example, a patient may be directly directed for discharge to routine screening or for treatment, based on the output of such a classification. Moreover, the output of the machine learning algorithm may be more clinically useful than for existing approaches, as will be discussed below. The technique may be implemented as a method, a computer program, programmable hardware, a computer system and/or in a system for tissue examination (such as a colposcopy system); specificity (more accurate that corrects any existing errors in classification) of classification of histological image is determined). (Struble, [0028-0030], [0043], [0141], discloses term “chromosomal abnormality” refers to any genetic variant for all or part of a chromosome. The genetic variants may include but not be limited to any copy number variant such as duplications or deletions, translocations, inversions, and mutations; term “intersex mosaicism” or “sex chromosome mosaicism” or “sex chromosome mosaic” refers to the presence of two or more populations of cells with different sex chromosome genotypes in one individual. Intersex mosaicisms arise when some cells in an individual have, e.g., two X chromosomes (XX) and other cells in the individual have one X chromosome and one Y chromosome (XY); when some cells in an individual have one X chromosome (XO) and other cells in the individual have one X chromosome and one Y chromosome (XY); or when some cells in an individual have two X chromosomes and one Y chromosome (XXY) and other cells in the individual have one X chromosome and one Y chromosome (XY); term “selected nucleic acid region” as used herein refers to a nucleic acid region corresponding to an individual chromosome. Selected nucleic acid regions may be directly isolated and enriched from the sample for detection, e.g., based on hybridization and/or other sequence-based techniques, or they may be amplified using the sample as a template prior to detection of the sequence; discloses percent fetal cell free DNA has been calculated, this data is combined with methods for detection and quantification of X and Y chromosome sequences to determine the likelihood that a fetus may be female, male, aneuploid for the X chromosome, aneuploid for the Y chromosome, an X chromosome mosaic, a Y chromosome mosaic. It can also be used in the determination of maternal aneuploidies, including mosaicism, or to identify whether the maternal sample being tested is contaminated; likelihood of feature of chromosomes are determined to predict presence or absence or positive or negative presence of X or Y chromosomes or their patterns and fetal sex is determined based on the expressions of the genes in the images obtained). Additionally, the rational and motivation to combine the references Papagiannakis and Struble as applied in rejection of claim 1 apply to this claim.
Regarding Claim 23,
The combination of Papagiannakis and Struble further discloses wherein the determined sex of the subject is used to identify and/or correct an error in metadata associated with the histological image. (Papagiannakis, [0009], discloses automatic in vivo or in vitro classification of the biological tissue may be achieved. The use of multiple images taken over the progress of the transient optical effects may have a significant improvement on sensitivity and/or specificity of the classification over existing methods. Sensitivity and specificity may refer to the ability to identify cervical dysplasia and/or cervical neoplasia. Sensitivity thus refers to the ability to correctly identify tissues displaying cervical dysplasia and/or cervical neoplasia. Specificity thus refers to the ability to correctly identify tissues that do not display cervical dysplasia and/or cervical neoplasia. Depending upon the application context, the output may be focussed on maximising sensitivity or specificity, or operating at a threshold that may optimal for one or both. Although classification may be presented based on the subject/tissue as a whole, the invention permits regions of the tissue suspected to be precancerous or cancerous to be identified. This may be advantageous for directing biopsy or treatment, including surgical resection. These sites may be biopsied to confirm the identification. A successful implementation of such a system may render the taking of biopsy unusual in many or most cases. For example, a patient may be directly directed for discharge to routine screening or for treatment, based on the output of such a classification. Moreover, the output of the machine learning algorithm may be more clinically useful than for existing approaches, as will be discussed below. The technique may be implemented as a method, a computer program, programmable hardware, a computer system and/or in a system for tissue examination (such as a colposcopy system); specificity (more accurate that corrects any existing errors in classification) of classification of histological image is determined). (Struble, [0028-0030], [0043], [0141], discloses term “chromosomal abnormality” refers to any genetic variant for all or part of a chromosome. The genetic variants may include but not be limited to any copy number variant such as duplications or deletions, translocations, inversions, and mutations; term “intersex mosaicism” or “sex chromosome mosaicism” or “sex chromosome mosaic” refers to the presence of two or more populations of cells with different sex chromosome genotypes in one individual. Intersex mosaicisms arise when some cells in an individual have, e.g., two X chromosomes (XX) and other cells in the individual have one X chromosome and one Y chromosome (XY); when some cells in an individual have one X chromosome (XO) and other cells in the individual have one X chromosome and one Y chromosome (XY); or when some cells in an individual have two X chromosomes and one Y chromosome (XXY) and other cells in the individual have one X chromosome and one Y chromosome (XY); term “selected nucleic acid region” as used herein refers to a nucleic acid region corresponding to an individual chromosome. Selected nucleic acid regions may be directly isolated and enriched from the sample for detection, e.g., based on hybridization and/or other sequence-based techniques, or they may be amplified using the sample as a template prior to detection of the sequence; discloses percent fetal cell free DNA has been calculated, this data is combined with methods for detection and quantification of X and Y chromosome sequences to determine the likelihood that a fetus may be female, male, aneuploid for the X chromosome, aneuploid for the Y chromosome, an X chromosome mosaic, a Y chromosome mosaic. It can also be used in the determination of maternal aneuploidies, including mosaicism, or to identify whether the maternal sample being tested is contaminated; likelihood of feature of chromosomes are determined to predict presence or absence or positive or negative presence of X or Y chromosomes or their patterns and fetal sex is determined based on the expressions of the genes in the images obtained). Additionally, the rational and motivation to combine the references Papagiannakis and Struble as applied in rejection of claim 1 apply to this claim.
Regarding Claim 24,
The combination of Papagiannakis and Struble further discloses wherein the determined sex of the subject is used to identify and/or correct a sample swap error in a pathology image-based determination of EGFR status. (Papagiannakis, [0009], discloses automatic in vivo or in vitro classification of the biological tissue may be achieved. The use of multiple images taken over the progress of the transient optical effects may have a significant improvement on sensitivity and/or specificity of the classification over existing methods. Sensitivity and specificity may refer to the ability to identify cervical dysplasia and/or cervical neoplasia. Sensitivity thus refers to the ability to correctly identify tissues displaying cervical dysplasia and/or cervical neoplasia. Specificity thus refers to the ability to correctly identify tissues that do not display cervical dysplasia and/or cervical neoplasia. Depending upon the application context, the output may be focussed on maximising sensitivity or specificity, or operating at a threshold that may optimal for one or both. Although classification may be presented based on the subject/tissue as a whole, the invention permits regions of the tissue suspected to be precancerous or cancerous to be identified. This may be advantageous for directing biopsy or treatment, including surgical resection. These sites may be biopsied to confirm the identification. A successful implementation of such a system may render the taking of biopsy unusual in many or most cases. For example, a patient may be directly directed for discharge to routine screening or for treatment, based on the output of such a classification. Moreover, the output of the machine learning algorithm may be more clinically useful than for existing approaches, as will be discussed below. The technique may be implemented as a method, a computer program, programmable hardware, a computer system and/or in a system for tissue examination (such as a colposcopy system); specificity (more accurate that corrects any existing errors in classification) of classification of histological image is determined). (Struble, [0028-0030], [0043], [0141], discloses term “chromosomal abnormality” refers to any genetic variant for all or part of a chromosome. The genetic variants may include but not be limited to any copy number variant such as duplications or deletions, translocations, inversions, and mutations; term “intersex mosaicism” or “sex chromosome mosaicism” or “sex chromosome mosaic” refers to the presence of two or more populations of cells with different sex chromosome genotypes in one individual. Intersex mosaicisms arise when some cells in an individual have, e.g., two X chromosomes (XX) and other cells in the individual have one X chromosome and one Y chromosome (XY); when some cells in an individual have one X chromosome (XO) and other cells in the individual have one X chromosome and one Y chromosome (XY); or when some cells in an individual have two X chromosomes and one Y chromosome (XXY) and other cells in the individual have one X chromosome and one Y chromosome (XY); term “selected nucleic acid region” as used herein refers to a nucleic acid region corresponding to an individual chromosome. Selected nucleic acid regions may be directly isolated and enriched from the sample for detection, e.g., based on hybridization and/or other sequence-based techniques, or they may be amplified using the sample as a template prior to detection of the sequence; discloses percent fetal cell free DNA has been calculated, this data is combined with methods for detection and quantification of X and Y chromosome sequences to determine the likelihood that a fetus may be female, male, aneuploid for the X chromosome, aneuploid for the Y chromosome, an X chromosome mosaic, a Y chromosome mosaic. It can also be used in the determination of maternal aneuploidies, including mosaicism, or to identify whether the maternal sample being tested is contaminated; likelihood of feature of chromosomes are determined to predict presence or absence or positive or negative presence of X or Y chromosomes or their patterns and fetal sex is determined based on the expressions of the genes in the images obtained). Additionally, the rational and motivation to combine the references Papagiannakis and Struble as applied in rejection of claim 1 apply to this claim.
Regarding Claim 25,
The combination of Papagiannakis and Struble further discloses wherein the determined sex of the subject is used to identify and/or correct errors in an image processing pipeline used to process histological images. (Papagiannakis, [0009], discloses automatic in vivo or in vitro classification of the biological tissue may be achieved. The use of multiple images taken over the progress of the transient optical effects may have a significant improvement on sensitivity and/or specificity of the classification over existing methods. Sensitivity and specificity may refer to the ability to identify cervical dysplasia and/or cervical neoplasia. Sensitivity thus refers to the ability to correctly identify tissues displaying cervical dysplasia and/or cervical neoplasia. Specificity thus refers to the ability to correctly identify tissues that do not display cervical dysplasia and/or cervical neoplasia. Depending upon the application context, the output may be focused on maximizing sensitivity or specificity, or operating at a threshold that may optimal for one or both. Although classification may be presented based on the subject/tissue as a whole, the invention permits regions of the tissue suspected to be precancerous or cancerous to be identified. This may be advantageous for directing biopsy or treatment, including surgical resection. These sites may be biopsied to confirm the identification. A successful implementation of such a system may render the taking of biopsy unusual in many or most cases. For example, a patient may be directly directed for discharge to routine screening or for treatment, based on the output of such a classification. Moreover, the output of the machine learning algorithm may be more clinically useful than for existing approaches, as will be discussed below. The technique may be implemented as a method, a computer program, programmable hardware, a computer system and/or in a system for tissue examination (such as a colposcopy system); specificity (more accurate that corrects any existing errors in classification) of classification of histological image is determined). (Struble, [0028-0030], [0043], [0141], discloses term “chromosomal abnormality” refers to any genetic variant for all or part of a chromosome. The genetic variants may include but not be limited to any copy number variant such as duplications or deletions, translocations, inversions, and mutations; term “intersex mosaicism” or “sex chromosome mosaicism” or “sex chromosome mosaic” refers to the presence of two or more populations of cells with different sex chromosome genotypes in one individual. Intersex mosaicisms arise when some cells in an individual have, e.g., two X chromosomes (XX) and other cells in the individual have one X chromosome and one Y chromosome (XY); when some cells in an individual have one X chromosome (XO) and other cells in the individual have one X chromosome and one Y chromosome (XY); or when some cells in an individual have two X chromosomes and one Y chromosome (XXY) and other cells in the individual have one X chromosome and one Y chromosome (XY); term “selected nucleic acid region” as used herein refers to a nucleic acid region corresponding to an individual chromosome. Selected nucleic acid regions may be directly isolated and enriched from the sample for detection, e.g., based on hybridization and/or other sequence-based techniques, or they may be amplified using the sample as a template prior to detection of the sequence; discloses percent fetal cell free DNA has been calculated, this data is combined with methods for detection and quantification of X and Y chromosome sequences to determine the likelihood that a fetus may be female, male, aneuploid for the X chromosome, aneuploid for the Y chromosome, an X chromosome mosaic, a Y chromosome mosaic. It can also be used in the determination of maternal aneuploidies, including mosaicism, or to identify whether the maternal sample being tested is contaminated; likelihood of feature of chromosomes are determined to predict presence or absence or positive or negative presence of X or Y chromosomes or their patterns and fetal sex is determined based on the expressions of the genes in the images obtained). Additionally, the rational and motivation to combine the references Papagiannakis and Struble as applied in rejection of claim 1 apply to this claim.
Regarding Claim 29,
The combination of Papagiannakis and Struble further discloses A system comprising: one or more processors; and a memory communicatively coupled to the one or more processors and configured to store instructions that, when executed by the one or more processors, cause the system to perform the method of claim 1. (Papagiannakis, [0029], Fig. 1, discloses schematic diagram of a computing system in accordance with the disclosure. The computing system comprises: an image collection module 10; a local processor 15; a main server 20; an identity database 30; an imaging database 40. A local interface 12 couples the image collection module 10 with the local processor 15. A processing interface 22 couples the local processor 15 with the main server 20. A first identity interface 32 couples the identity database 30 with the local processor 15 and a second identity interface 34 couples the identity database 30 with the main server 20. A first image data interface 42 couples the imaging database 40 with the local processor 15 and a second image data interface 44 couples the imaging database 40 with the main server 20. It will be noted that the computing system of FIG. 1 incorporates parts that may be distinct from a computer, for example being part of an optical system and/or an electronics control system. However, these will all be considered part of the computing system for the purposes of this disclosure).
Regarding Claim 30,
The combination of Papagiannakis and Struble further discloses A non-transitory computer-readable storage medium storing one or more programs, the one or more programs comprising instructions, which when executed by one or more processors of a system, cause the system to perform the method of claim 1. (Papagiannakis, [0029], Fig. 1, discloses schematic diagram of a computing system in accordance with the disclosure. The computing system comprises: an image collection module 10; a local processor 15; a main server 20; an identity database 30; an imaging database 40. A local interface 12 couples the image collection module 10 with the local processor 15. A processing interface 22 couples the local processor 15 with the main server 20. A first identity interface 32 couples the identity database 30 with the local processor 15 and a second identity interface 34 couples the identity database 30 with the main server 20. A first image data interface 42 couples the imaging database 40 with the local processor 15 and a second image data interface 44 couples the imaging database 40 with the main server 20. It will be noted that the computing system of FIG. 1 incorporates parts that may be distinct from a computer, for example being part of an optical system and/or an electronics control system. However, these will all be considered part of the computing system for the purposes of this disclosure).
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure:
US-20190292519-A1 (Mitalipov et al., Methods are provided of generating functional human oocytes following nuclear transfer of first polar body (PB1) genomes from metaphase II (MII) oocytes into enucleated donor MII cytoplasm (PBNT) and using mitochondrial replacement techniques to circumvent mother-to-child mtDNA disease transmission, Abstract; term “X chromosome inactivation” refers to the inactivation of one of each pair of X chromosomes to form the Barr body in female mammalian somatic cells. Thus tissues whose original zygote carried heterozygous X borne genes should have individual cells expressing one or other but not both of the X encoded gene products. The inactivation is thought to occur early in development and leads to mosaicism of expression of such genes in the body, [0112])
US-20150376612-A1 (Lee et al., This invention relates to methods and compositions for selectively reactivating or downregulating certain genes, e.g., genes regulated by zinc-finger protein CCCTC-binding factor (CTCF) on autosomes (e.g., imprinted genes, tumor suppressors, cancer) and the inactive X chromosome (Xi), e.g., genes associated with X-linked diseases, e.g., Rett Syndrome, Factor VIII or IX deficiency, Fragile X Syndrome, Duchenne muscular dystrophy, and PNH, in heterozygous females carrying a mutated allele, in addition to a functional wildtype or hypomorphic allele, Abstract; it was first noticed about half a century ago that X chromosome changes are often seen in female reproductive cancers. Some 70% of breast carcinomas lack a “Barr body’, the cytologic hallmark of the inactive X chromosome (Xi), and instead harbor two or more active Xs (Xa). Additional X's are also a risk factor for men, as XXY men (Klinefelter Syndrome) have a 20- to 50-fold increased risk of breast cancer in a BRCA1 background. The X is also known to harbor a number of oncogenes. Supernumerary Xa's correlate with a poor prognosis and stand as one of the most common cytogenetic abnormalities not only in reproductive cancers but also in leukemias, lymphomas, and germ cell tumors of both sexes, [0168]).
Any inquiry concerning this communication or earlier communications from the examiner should be directed to PINALBEN V PATEL whose telephone number is (571)270-5872. The examiner can normally be reached M-F: 10am - 8pm.
Examiner interviews are available via telephone, in-person, and video conferencing using a USPTO supplied web-based collaboration tool. To schedule an interview, applicant is encouraged to use the USPTO Automated Interview Request (AIR) at http://www.uspto.gov/interviewpractice.
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, Chineyere Wills-Burns can be reached at 571-272-9752. The fax phone number for the organization where this application or proceeding is assigned is 571-273-8300.
Information regarding the status of published or unpublished applications may be obtained from Patent Center. Unpublished application information in Patent Center is available to registered users. To file and manage patent submissions in Patent Center, visit: https://patentcenter.uspto.gov. Visit https://www.uspto.gov/patents/apply/patent-center for more information about Patent Center and https://www.uspto.gov/patents/docx for information about filing in DOCX format. For additional questions, contact the Electronic Business Center (EBC) at 866-217-9197 (toll-free). If you would like assistance from a USPTO Customer Service Representative, call 800-786-9199 (IN USA OR CANADA) or 571-272-1000.
/Pinalben Patel/Examiner, Art Unit 2673
Prosecution Insights