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 .
Status of claims: claims 1-30 are pending below.
Continued Examination Under 37 CFR 1.114
A request for continued examination under 37 CFR 1.114, including the fee set forth in 37 CFR 1.17(e), was filed in this application after final rejection. Since this application is eligible for continued examination under 37 CFR 1.114, and the fee set forth in 37 CFR 1.17(e) has been timely paid, the finality of the previous Office action has been withdrawn pursuant to 37 CFR 1.114. Applicant's submission filed on April 21, 2026 has been entered.
Information Disclosure Statement
The information disclosure statement (IDS) submitted on April 22th, 2026 was filed and considered. The submission is in compliance with the provisions of 37 CFR 1.97. Accordingly, the information disclosure statement is being considered by the examiner.
Response to Arguments
Applicant's arguments filed April 21st, 2026 have been fully considered but they are not persuasive.
Applicant remarks – (pages 9-10) Applicant argued the lack of teaching regarding lack of teaching for new claim element “value representation” and further argued that region or bounding box representational by pixels and vectors differed from the claimed value. Please see the Remarks for further detail. As argument applies to all independent and their dependent claims.
Examiner response – Examiner respectfully disagree. A review of KAUFMAN et al (US 2020/0226748) detail values that are inputted by user and extracted by neural network in paragraph 0049, 0050, 0051, 0056-0057: values such as intensity, location, volume, textural, spatial information. Such that the combine teaching of KAUFMAN et al (US 2020/0226748) in view of Yip et al (US 2021/0166381) addresses the claim instant invention. Please see Office Action below for more information.
Claim Rejections - 35 USC § 112
The following is a quotation of the first paragraph of 35 U.S.C. 112(a):
(a) IN GENERAL.—The specification shall contain a written description of the invention, and of the manner and process of making and using it, in such full, clear, concise, and exact terms as to enable any person skilled in the art to which it pertains, or with which it is most nearly connected, to make and use the same, and shall set forth the best mode contemplated by the inventor or joint inventor of carrying out the invention.
The following is a quotation of the first paragraph of pre-AIA 35 U.S.C. 112:
The specification shall contain a written description of the invention, and of the manner and process of making and using it, in such full, clear, concise, and exact terms as to enable any person skilled in the art to which it pertains, or with which it is most nearly connected, to make and use the same, and shall set forth the best mode contemplated by the inventor of carrying out his invention.
Claims 1-30 are rejected under 35 U.S.C. 112(a) or 35 U.S.C. 112 (pre-AIA ), first paragraph, as failing to comply with the written description requirement. The claim(s) contains subject matter which was not described in the specification in such a way as to reasonably convey to one skilled in the relevant art that the inventor or a joint inventor, or for applications subject to pre-AIA 35 U.S.C. 112, the inventor(s), at the time the application was filed, had possession of the claimed invention.
Independent claims 1, 7, 13, 19 and 25, and all their dependent claims, recited new claim language “value representation”, where and review of the specification of the instant invention does not support this new claim element. Specification of the instant invention disclosed “two-dimensional representation” in 0048, “semi- and self-supervised representation learning with probabilistic weak supervision” in 0052, “learned representation” in 0057, “input data…representation” in 0352, but not disclose or define that claim elements, thus it is view as new matter. Please direct to support specification or amend to overcome rejection. For compact prosecution, Examiner will examine the claim element “value representation” in reasonable interpretation.
Claim Rejections - 35 USC § 103
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, 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 1-30 are rejected under 35 U.S.C. 103 as being unpatentable over KAUFMAN et al (US 2020/0226748) in view of Yip et al (US 2021/0166381).
Claim 1:
KAUFMAN et al (US 2020/0226748) teach the following subject matter:
One or more processors (figure 21 and 0137 teaches one or more microprocessors with memory), comprising: circuitry to identify boundaries of one or more objects within an image based, at least
in part, on one or more indications of one or more locations of the one or more objects in the image (0010-0011 teaches neural network generating outline (boundary) for foreground and background to visualized lesion with location and shape; figure 19 and 0132 teaches outline manually generated by expert (user generated) of object (pancreas, liver, spleen)) and
at least in part, on one or more values representative of a proportion of the image that comprises the one or more objects (0049 detail classifier 420 (neural network) use for values for shape, intensity, location, outline; 0050 detail use of classifier for values such as 3D volume, 0051 detail use of convolutional neural network (CNN) for textural and spatial information; 0056-0057 detail CNN for quantitative feature, location, intensity, shape and texture information values; figure 1 and 0046-0047 teaches scan to identify various regions of tissue type and organs or lesion, where one ordinary skill in the arts understand the “various regions” is part or proportion of one or more objects of interest; paragraph 0110 teaches utilizing probabilistic atlases and statistical shape model of registered volumetric image in order to recognize variation of size, shape and location of organs(objects); where one ordinary skill in the arts where probabilistic atlases and statistical shape, are portions of object of interest to be identify due to overlapping/layer of organs during scanning.).
KAUFMAN et al teaches all the subject matter above with the use of neural network, but not the following:
determine one or more weights of the one or more neural network; adjusts the one or more weights of the one or more neural networks based; and causes the one or more neural networks comprising the adjusted weights to generate output identifying the boundaries of the one or more objects.
Yip et al (US 2021/0166381) teaches the following subject matter regarding boundary of region of interest to a whole slide image (abstract):
determine one or more weights of the one or more neural network; adjusts the one or more weights of the one or more neural networks based; and causes the one or more neural networks comprising the adjusted weights to generate output identifying the boundaries of the one or more objects (figure 3 and 0336 teaches training of the deep learning where the weights in the layers (determine weight with in the neural network) are further adjusted to accurately label the boundaries for classification, with help of human analyst (around the region of interest) to the image (whole image)).
KAUFMAN et al and Yip et al are both in the field of image analysis, especially the use of neural network for with human input for address region of interest process in conjunction with the whole image such the combine outcome is predictable.
Therefore it would have been obvious to one having ordinary skill before the effective filing date to modify KAUFMAN et al by Yip et al such that the training set of matrices provide an accurate labeling for the tissue classifier as disclosed by Yip et al in 0036. Where 0254 detail the metric between proportion of tissue by user-selected threshold to the digital image, 0256 technician dissecting to isolate tumor to non-tumor tissue in slide (whole image slide), and 0435 detail percentage of tumor content boundary dissected by pathologist to the whole slide content through the GUI (graphic user interface). All this is used for training of the deep learning network (neural network) that would determine the layer of weight and adjustment of those weights.
Claim 2:
KAUFMAN et al further teaches:
The one or more processors of claim 1, wherein the circuitry is further to adjust the one or more weights based on a user-generated outline that is a polygon with points located proximate the boundaries of the one or more objects (0051 detail user input and CNN for the segmentation of the lesion/object for textural and spatial information; 0010-0011 teaches neural network generating outline (boundary) for foreground and background to visualized lesion with location and shape, and further teach outlines due to characteristic of the foreground with first and second classifiers; 0047-0049 teaches outline refined (proximate) by lesions; figure 1 part 110 show polygon around object; above teaches user generated).
Claim 3:
KAUFMAN et al further teaches:
The one or more processors of claim 1, wherein the one or more values representative of the proportion of the image that comprises the one or more objects are provided by a user (0050-0051 detail user combine with CNN for textural and spatial information of the lesion; 0074 teaches consideration of average size of lesions; 0102 detail user adjust parameters to edit the feature such as volume (value representative); 0077 teaches CNN classifier for size of lesion; 0106 teaches size; 0109-0110 teaches segmentation of structures in size, position, shape and location; 0134).
Claim 4:
KAUFMAN et al further teaches:
The one or more processors of claim 3, wherein the one or more values comprise that comprises (0049 detail values such as shape, intensity, location, outline; 0050 detail value such as volume; 0051 detail value such as spatial information; 0056-0057 detail value such as quantitative feature, location, intensity, shape and texture information values; 0070 teaches general shape of accurate diagnosis define with percentage of lesion pixel).
Claim 5:
KAUFMAN et al further teaches:
The one or more processors of claim 1, whereinare trained using semi- Page 2 of 12 supervised and self-supervised representation learning, in a first stage, with probabilistic weak supervision in a second stage (0126 teaches two stage convnet-based, where 0009 detail first stage classifiers (neural network) with Random Forest classifier (supervised) and second stage classifier to be convolutional neural network classifier; 0013 detail multi-label segmentation using convolutional neural network, where multi-label is supervised learning).
Claim 6:
KAUFMAN et al further teaches:
The one or more processors of claim 1, wherein the one or more objects [[is]] comprise a tumor and the image is a histopathologic image (0119 teaches tumor; 0044 teaches lesions for the histopathological).
Claim 7:
KAUFMAN et al (US 2020/0226748) teach the following subject matter:
A system (figure 1 and 0046 teaches system) comprising: one or more processors (figure 21 and 0137 teaches one or more microprocessors with memory) to identify [[the]] boundaries of the object one or more objects within [[the]] an image based, at least in part, on one or more indications of one or more locations of the one or more objects in the image (0010-0011 teaches neural network generating outline (boundary) for foreground and background to visualized lesion with location and shape; figure 19 and 0132 teaches outline manually generated by expert (user generated) of object (pancreas, liver, spleen)) and information about a size of the object provided by a user one or more values representative of a proportion of the image that comprises the one or more objects (0049 detail classifier 420 (neural network) use for values for shape, intensity, location, outline; 0050 detail use of classifier for values such as 3D volume, 0051 detail use of convolutional neural network (CNN) for textural and spatial information; 0056-0057 detail CNN for quantitative feature, location, intensity, shape and texture information values; figure 1 and 0046-0047 teaches scan to identify various regions of tissue type and organs or lesion, where one ordinary skill in the arts understand the “various regions” is part or proportion of one or more objects of interest; paragraph 0110 teaches utilizing probabilistic atlases and statistical shape model of registered volumetric image in order to recognize variation of size, shape and location of organs(objects); where one ordinary skill in the arts where probabilistic atlases and statistical shape, are portions of object of interest to be identify due to overlapping/layer of organs during scanning).
KAUFMAN et al teaches all the subject matter above with the use of neural network, but not the following:
determine one or more weights of the one or more neural network; adjusts the one or more weights of the one or more neural networks based; and causes the one or more neural networks comprising the adjusted weights to generate output identifying the boundaries of the one or more objects.
Yip et al (US 2021/0166381) teaches the following subject matter regarding boundary of region of interest to a whole slide image (abstract):
determine one or more weights of the one or more neural network; adjusts the one or more weights of the one or more neural networks based; and causes the one or more neural networks comprising the adjusted weights to generate output identifying the boundaries of the one or more objects (figure 3 and 0336 teaches training of the deep learning where the weights of the layers (determine weight with in the neural network) are further adjusted to accurately label the boundaries for classification, with help of human analyst (around the region of interest) to the image (whole image)).
KAUFMAN et al and Yip et al are both in the field of image analysis, especially the use of neural network for with human input for address region of interest process in conjunction with the whole image such the combine outcome is predictable.
Therefore it would have been obvious to one having ordinary skill before the effective filing date to modify KAUFMAN et al by Yip et al such that the training set of matrices provide an accurate labeling for the tissue classifier as disclosed by Yip et al in 0036. Where 0254 detail the metric between proportion of tissue by user-selected threshold to the digital image, 0256 technician dissecting to isolate tumor to non-tumor tissue in slide (whole image slide), and 0435 detail percentage of tumor content boundary dissected by pathologist to the whole slide content through the GUI (graphic user interface). All this is used for training of the deep learning network (neural network) that would determine the layer of weight and adjustment of those weights.
Claim 8:
KAUFMAN et al further teaches:
The system of claim 7, wherein (0051 detail user input and CNN for the segmentation of the lesion/object for textural and spatial information; 0010-0011 teaches neural network generating outline (boundary) for foreground and background to visualized lesion with location and shape, and further teach outlines due to characteristic of the foreground with first and second classifiers; 0047-0049 teaches outline refined (proximate) by lesions; figure 1 part 110 show polygon around object; above teaches user generated).
Claim 9:
KAUFMAN et al further teaches:
The system of claim 7, wherein the one or more processors are further to (0047 detail user interface tool for region of interest with outline and boundary).
Claim 10:
KAUFMAN et al further teaches:
The system of claim 9, wherein the user-generated outline is a polygon with points located proximate the boundaries of the one or more objects. (0070 teaches general shape of accurate diagnosis define with percentage of lesion pixel).
Claim 11:
KAUFMAN et al further teaches:
The system of claim 7, whereithe one or more neural networks are trained using semi-supervised and self-supervised representation learning, in a first stage, with probabilistic weak supervision in a second stage. (0126 teaches two stage convnet-based, where 0009 detail first stage classifiers (neural network) with Random Forest classifier (supervised) and second stage classifier to be convolutional neural network classifier; 0013 detail multi-label segmentation using convolutional neural network, where multi-label is supervised learning).
Claim 12:
KAUFMAN et al further teaches:
The system of claim 7, wherein the one or more objects [[is]] comprise a tumor and the image is a histopathologic image. (0119 teaches tumor; 0044 teaches lesions for the histopathological).
Claim 13:
KAUFMAN et al (US 2020/0226748) teach the following subject matter:
A method (abstract teach method) comprising: identifying boundaries of (0010-0011 teaches neural network generating outline (boundary) for foreground and background to visualized lesion with location and shape; figure 19 and 0132 teaches outline manually generated by expert (user generated) of object (pancreas, liver, spleen)) and (0049 detail classifier 420 (neural network) use for values for shape, intensity, location, outline; 0050 detail use of classifier for values such as 3D volume, 0051 detail use of convolutional neural network (CNN) for textural and spatial information; 0056-0057 detail CNN for quantitative feature, location, intensity, shape and texture information values; figure 1 and 0046-0047 teaches scan to identify various regions of tissue type and organs or lesion, where one ordinary skill in the arts understand the “various regions” is part or proportion of one or more objects of interest; paragraph 0110 teaches utilizing probabilistic atlases and statistical shape model of registered volumetric image in order to recognize variation of size, shape and location of organs(objects); where one ordinary skill in the arts where probabilistic atlases and statistical shape, are portions of object of interest to be identify due to overlapping/layer of organs during scanning).
KAUFMAN et al teaches all the subject matter above with the use of neural network, but not the following:
determine one or more weights of the one or more neural network; adjusts the one or more weights of the one or more neural networks based; and causes the one or more neural networks comprising the adjusted weights to generate output identifying the boundaries of the one or more objects.
Yip et al (US 2021/0166381) teaches the following subject matter regarding boundary of region of interest to a whole slide image (abstract):
determine one or more weights of the one or more neural network; adjusts the one or more weights of the one or more neural networks based; and causes the one or more neural networks comprising the adjusted weights to generate output identifying the boundaries of the one or more objects (figure 3 and 0336 teaches training of the deep learning where the weights of the layers (determine weight with in the neural network) are further adjusted to accurately label the boundaries for classification, with help of human analyst (around the region of interest) to the image (whole image)).
KAUFMAN et al and Yip et al are both in the field of image analysis, especially the use of neural network for with human input for address region of interest process in conjunction with the whole image such the combine outcome is predictable.
Therefore it would have been obvious to one having ordinary skill before the effective filing date to modify KAUFMAN et al by Yip et al such that the training set of matrices provide an accurate labeling for the tissue classifier as disclosed by Yip et al in 0036. Where 0254 detail the metric between proportion of tissue by user-selected threshold to the digital image, 0256 technician dissecting to isolate tumor to non-tumor tissue in slide (whole image slide), and 0435 detail percentage of tumor content boundary dissected by pathologist to the whole slide content through the GUI (graphic user interface). All this is used for training of the deep learning network (neural network) that would determine the layer of weight and adjustment of those weights.
Claim 14:
KAUFMAN et al further teaches:
The method of claim 13, wherein the one or more indications of the one or more locations is a user-generated outline with points located proximate the boundaries of the one or more objects.. (0010-0011 teaches neural network generating outline (boundary) for foreground and background to visualized lesion with location and shape, and further teach outlines due to characteristic of the foreground with first and second classifiers; 0047-0049 teaches outline refined (proximate) by lesions; figure 1 part 110 show polygon around object; above teaches user generated).
Claim 15:
KAUFMAN et al further teaches:
The method of claim 13, wherein the one or more values representative of the proportion of the image that comprises the one or more objects are provided by a user (000047 detail user interface tool for region of interest with outline and boundary).
Claim 16:
KAUFMAN et al further teaches:
The method of claim 15, wherein the one or more values comprise an estimated percentage of the image that comprises the one or more objects (0049 detail values such as shape, intensity, location, outline; 0050 detail value such as volume; 0051 detail value such as spatial information; 0056-0057 detail value such as quantitative feature, location, intensity, shape and texture information values; 0070 teaches general shape of accurate diagnosis define with percentage of lesion pixel).
Claim 17:
KAUFMAN et al further teaches:
The method of claim 13, whereinthe one or more neural networks are trained using semi-supervised and self- supervised representation learning, in a first stage, with probabilistic weak supervision in a second stage.. (0126 teaches two stage convnet-based, where 0009 detail first stage classifiers (neural network) with Random Forest classifier (supervised) and second stage classifier to be convolutional neural network classifier; 0013 detail multi-label segmentation using convolutional neural network, where multi-label is supervised learning).
Claim 18:
KAUFMAN et al further teaches:
The method of claim 13, wherein the one or more objects [[is]] comprise a tumor and the image is a histopathologic image (0119 teaches tumor; 0044 teaches lesions for the histopathological).
Claim 19:
KAUFMAN et al (US 2020/0226748) teach the following subject matter:
A machine-readable medium (abstract teaches computer-accessible medium; figure 21 and 0138 teaches hard disk, CD-ROM, RAM, ROM) having stored thereon a set of instructions, which if performed by one or more processors, cause the one or more processors to at least: identify [[the]] boundaries of one or more objects within [[the]] an image based, at least in part, on one or more indications of one or more locations of the one or more objects in the image (0010-0011 teaches neural network generating outline (boundary) for foreground and background to visualized lesion with location and shape; figure 19 and 0132 teaches outline manually generated by expert (user generated) of object (pancreas, liver, spleen)) and one or more value representative of a proportion of the image that comprises the one or more objects (0049 detail classifier 420 (neural network) use for values for shape, intensity, location, outline; 0050 detail use of classifier for values such as 3D volume, 0051 detail use of convolutional neural network (CNN) for textural and spatial information; 0056-0057 detail CNN for quantitative feature, location, intensity, shape and texture information values; figure 1 and 0046-0047 teaches scan to identify various regions of tissue type and organs or lesion, where one ordinary skill in the arts understand the “various regions” is part or proportion of one or more objects of interest; paragraph 0110 teaches utilizing probabilistic atlases and statistical shape model of registered volumetric image in order to recognize variation of size, shape and location of organs(objects); where one ordinary skill in the arts where probabilistic atlases and statistical shape, are portions of object of interest to be identify due to overlapping/layer of organs during scanning).
KAUFMAN et al teaches all the subject matter above with the use of neural network, but not the following:
determine one or more weights of the one or more neural network; adjusts the one or more weights of the one or more neural networks based; and causes the one or more neural networks comprising the adjusted weights to generate output identifying the boundaries of the one or more objects.
Yip et al (US 2021/0166381) teaches the following subject matter regarding boundary of region of interest to a whole slide image (abstract):
determine one or more weights of the one or more neural network; adjusts the one or more weights of the one or more neural networks based; and causes the one or more neural networks comprising the adjusted weights to generate output identifying the boundaries of the one or more objects (figure 3 and 0336 teaches training of the deep learning where the weights of the layers (determine weight with in the neural network) are further adjusted to accurately label the boundaries for classification, with help of human analyst (around the region of interest) to the image (whole image)).
KAUFMAN et al and Yip et al are both in the field of image analysis, especially the use of neural network for with human input for address region of interest process in conjunction with the whole image such the combine outcome is predictable.
Therefore it would have been obvious to one having ordinary skill before the effective filing date to modify KAUFMAN et al by Yip et al such that the training set of matrices provide an accurate labeling for the tissue classifier as disclosed by Yip et al in 0036. Where 0254 detail the metric between proportion of tissue by user-selected threshold to the digital image, 0256 technician dissecting to isolate tumor to non-tumor tissue in slide (whole image slide), and 0435 detail percentage of tumor content boundary dissected by pathologist to the whole slide content through the GUI (graphic user interface). All this is used for training of the deep learning network (neural network) that would determine the layer of weight and adjustment of those weights.
Claim 20:
KAUFMAN et al further teaches:
The machine-readable medium of claim 19, whereinthe one or more values comprise an estimated percentage of the image that comprises more objects (0051 detail user input and CNN for the segmentation of the lesion/object for textural and spatial information; 0010-0011 teaches neural network generating outline (boundary) for foreground and background to visualized lesion with location and shape, and further teach outlines due to characteristic of the foreground with first and second classifiers; 0047-0049 teaches outline refined (proximate) by lesions; figure 1 part 110 show polygon around object; above teaches user generated).
Claim 21:
KAUFMAN et al further teaches:
The machine-readable medium of claim 19, wherein the one or more processors are further to generate the output identifyingthe one or more objects within [[an]] the image based, at least in part, on a user-generated outline of only a portion of the one or more objects. (074 teaches consideration of average size of lesions; 0077 teaches CNN classifier for size of lesion; 0106 teaches size; 0109-0110 teaches segmentation of structures in size, position, shape and location; 0134).
Claim 22:
KAUFMAN et al teaches:
The machine-readable medium of The machine-readable medium of wherein the user-generated outline is a polygon with points located proximate the boundaries of the one or more objects. (0049 detail values such as shape, intensity, location, outline; 0050 detail value such as volume; 0051 detail value such as spatial information; 0056-0057 detail value such as quantitative feature, location, intensity, shape and texture information values; 0070 teaches general shape of accurate diagnosis define with percentage of lesion pixel).
Claim 23:
KAUFMAN et al teaches:
The machine-readable medium of claim 19, whereinthe one or more neural networks are trained using semi- supervised and self-supervised representation learning, in a first stage, with probabilistic weak supervision in a second stage.. (0126 teaches two stage convnet-based, where 0009 detail first stage classifiers (neural network) with Random Forest classifier (supervised) and second stage classifier to be convolutional neural network classifier; 0013 detail multi-label segmentation using convolutional neural network, where multi-label is supervised learning).
Claim 24:
KAUFMAN et al teaches:
The machine-readable medium of claim 19, wherein the one or more objects [[is]] comprise a tumor and the image is a histopathologic image. (0119 teaches tumor; 0044 teaches lesions for the histopathological).
Claim 25:
KAUFMAN et al (US 2020/0226748) teach the following subject matter:
An image annotation system (0104 teaches ground truth annotation; 0106 teaches expert annotation; figure 1 and 0046 teaches system), comprising: one or more processors (figure 21 and 0137 teaches one or more microprocessors with memory) to identify, using one or more neural networks, boundaries of one or more objects within an image based, at least in part, on one or more indications of one or more locations of the one or more objects in the image (0011 teaches neural network generating outline (boundary) for foreground and background to visualized lesion with location and shape; figure 19 and 0132 teaches outline manually generated by expert (user generated) of object (pancreas, liver, spleen)) and one or more value representative of a proportion of the image that comprises the one or more objects (0049 detail classifier 420 (neural network) use for values for shape, intensity, location, outline; 0050 detail use of classifier for values such as 3D volume, 0051 detail use of convolutional neural network (CNN) for textural and spatial information; 0056-0057 detail CNN for quantitative feature, location, intensity, shape and texture information values; figure 1 and 0046-0047 teaches scan to identify various regions of tissue type and organs or lesion, where one ordinary skill in the arts understand the “various regions” is part or proportion of one or more objects of interest; paragraph 0110 teaches utilizing probabilistic atlases and statistical shape model of registered volumetric image in order to recognize variation of size, shape and location of organs(objects); where one ordinary skill in the arts where probabilistic atlases and statistical shape, are portions of object of interest to be identify due to overlapping/layer of organs during scanning); and memory for storing network parameters for the one or more neural networks (figure 21 and 0137 teaches one or more microprocessors with memory).
KAUFMAN et al teaches all the subject matter above with the use of neural network, but not the following:
determine one or more weights of the one or more neural network; adjusts the one or more weights of the one or more neural networks based; and causes the one or more neural networks comprising the adjusted weights to generate output identifying the boundaries of the one or more objects.
Yip et al (US 2021/0166381) teaches the following subject matter regarding boundary of region of interest to a whole slide image (abstract):
determine one or more weights of the one or more neural network; adjusts the one or more weights of the one or more neural networks based; and causes the one or more neural networks comprising the adjusted weights to generate output identifying the boundaries of the one or more objects (figure 3 and 0336 teaches training of the deep learning where the weights of the layers (determine weight with in the neural network) are further adjusted to accurately label the boundaries for classification, with help of human analyst (around the region of interest) to the image (whole image)).
KAUFMAN et al and Yip et al are both in the field of image analysis, especially the use of neural network for with human input for address region of interest process in conjunction with the whole image such the combine outcome is predictable.
Therefore it would have been obvious to one having ordinary skill before the effective filing date to modify KAUFMAN et al by Yip et al such that the training set of matrices provide an accurate labeling for the tissue classifier as disclosed by Yip et al in 0036. Where 0254 detail the metric between proportion of tissue by user-selected threshold to the digital image, 0256 technician dissecting to isolate tumor to non-tumor tissue in slide (whole image slide), and 0435 detail percentage of tumor content boundary dissected by pathologist to the whole slide content through the GUI (graphic user interface). All this is used for training of the deep learning network (neural network) that would determine the layer of weight and adjustment of those weights.
Claim 26:
KAUFMAN et al teaches:
The image annotation system of claim 25, wherein the one or more indications of the one or more locations comprises a polygon with points located proximate the boundaries of the one or more objects (0051 detail user input and CNN for the segmentation of the lesion/object for textural and spatial information; 0010-0011 teaches neural network generating outline (boundary) for foreground and background to visualized lesion with location and shape, and further teach outlines due to characteristic of the foreground with first and second classifiers; 0047-0049 teaches outline refined (proximate) by lesions; figure 1 part 110 show polygon around object; above teaches user generated).
Claim 27:
KAUFMAN et al teaches:
The image annotation system of claim 25, whereinthe one or more values representative of the proportion of the image that comprises the one or more objects are provided by a user. (0050-0051 detail user combine with CNN for textural and spatial information of the lesion; 0074 teaches consideration of average size of lesions; 0102 detail user adjust parameters to edit the feature such as volume (value representative); 0077 teaches CNN classifier for size of lesion; 0106 teaches size; 0109-0110 teaches segmentation of structures in size, position, shape and location; 0134).
Claim 28:
KAUFMAN et al teaches:
The image annotation system of claim 27, whereinthe one or more values comprise an estimated percentage of the image that comprises (0049 detail values such as shape, intensity, location, outline; 0050 detail value such as volume; 0051 detail value such as spatial information; 0056-0057 detail value such as quantitative feature, location, intensity, shape and texture information values; 0070 teaches general shape of accurate diagnosis define with percentage of lesion pixel).
Claim 29:
KAUFMAN et al teaches:
The image annotation system of claim 25, whereinthe one or more neural networks are trained using semi- supervised and self-supervised representation learning, in a first stage, with probabilistic weak supervision in a second stage (0126 teaches two stage convnet-based, where 0009 detail first stage classifiers (neural network) with Random Forest classifier (supervised) and second stage classifier to be convolutional neural network classifier; 0013 detail multi-label segmentation using convolutional neural network, where multi-label is supervised learning).
Claim 30:
KAUFMAN et al teaches:
The image annotation system of claim 25, wherein the one or more objects [[is]] comprise a tumor and the image is a histopathologic image (0119 teaches tumor; 0044 teaches lesions for the histopathological).
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure.
Yaroslavsky et al (US 2013/0324846) teaches DEVICES AND METHODS FOR OPTICAL PATHOLOGY – 0061 teaches manually outlined and 0064 teaches tumor with histopathology with tumor (outlined).
Any inquiry concerning this communication or earlier communications from the examiner should be directed to TSUNG-YIN TSAI whose telephone number is (571)270-1671. The examiner can normally be reached 7am-4pm.
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, Bhavesh Mehta can be reached at (571) 272-7453. 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.
/TSUNG YIN TSAI/Primary Examiner, Art Unit 2656
Prosecution Insights