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 .
.
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 (i.e., changing from AIA to pre-AIA ) 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, 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, 5-13, and 16-20 are rejected under 35 U.S.C. 103 as being unpatentable over Hu (US 20230419491 A1) in view of Cho et al. (US 20200003886 A1).
Regarding claim 1, method of claim 1 is performed by the system of claim 12. They recite similar limitations. Applicant is kindly advised to refer to rejection of claim 12 for the method claim 1.
Regarding claim 2, the modified Hu further teaches the method according to claim 1, further comprising performing, by the second processor, a fourth operation of outputting a second analysis result from the second batch by using the machine learning model (Fig. 1B and Pars. 25-27, combines weighting values of the first tile (batch) 135a and the second tile (batch) 135b) (first analysis result) and 135n (second analysis) (Fig. 1B and Pars. 25-27), wherein each node (layer) of an ANN is connected to each node of the preceding and/or subsequent layers of the ANN) (Par. 49)), wherein medical information associated with the medical image is generated based on the first analysis result and the second analysis result (Fig. 1B, classifications are then presented as evaluations (medical information) of the whole slide image)
Regarding claim 5, the modified Hu further teaches the method according to claim 1, wherein the at least one first patch and the at least one second patch are spatially associated with each other in the medical image (Par. 39, each tile (patch) includes some portion of pixels of the image that are included or overlapped (spatially associated) in at least one other tile (patch)).
Regarding claim 6, the modified Hu further teaches the method according to claim 5, wherein the at least one first patch and the at least one second patch are adjacent or overlapped in the medical image (Par. 39, each tile (patch) includes some portion of pixels of the image that are included or overlapped (spatially associated) in at least one other tile (patch)).
Regarding claim 7, the modified Hu further teaches the method according to claim 2, wherein: the at least one first patch and the at least one second patch include spatial information in the medical image (Par. 46, contextual information for the tile, such as the position of the tile (spatial information) within the digital pathology image); a processed image is generated based on the spatial information associated with the at least one first patch and the at least one second patch (Fig. 1B and Par. 26, classifications are then presented as evaluations (medical information) of the whole slide image (processed image)); and the medical information associated with the medical image is generated based on the first analysis result, the second analysis result, and the processed image (Fig. 1B and Par. 26).
Regarding claim 8, the modified Hu further teaches the method according to claim 2, wherein the generated medical information includes statistical information of the medical image (Par. 48, series of binary yes or no determinations with a confidence score (statistical information)).
Regarding claim 9, the modified Hu further teaches the method according to claim 1, wherein: the performing the first operation includes performing, by the first processor, image processing on the generated first batch (Fig. 1A and Pars. 41-42, generating the first tile (batch) 135a); the performing the second operation includes performing, by the first processor, image processing on the generated second batch (Fig. 1A and Pars. 41-42, generating the second tile (batch) 135b); and the image processing on the first batch and the image processing on the second batch include at least one of contrast adjustment, brightness adjustment, saturation adjustment, blur adjustment, noise injection, random crop, or sharpening (Par. 40).
Regarding claim 10, method of claim 10 is performed by the system of claim 12. They recite similar limitations. Applicant is kindly advised to refer to rejection of claim 12 for claim 10.
Regarding claim 11, non-transitory computer-readable recording medium of claim 11 is performed by the system of claim 12. They recite similar limitations. Applicant is kindly advised to refer to rejection of claim 12 for claim 11.
Regarding claim 12, Hu teaches an information processing system, comprising: a memory (Fig. 11); and a first processor and a second processor connected to the memory (Fig. 2, processing system 210 comprising plurality of processing modules) and configured to execute at least one computer-readable program included in the memory (Fig. 11), wherein the at least one computer-readable program includes instructions for: performing, by the first processor, a first operation of generating a first batch from at least one first patch extracted from a medical image (Fig. 1A and Pars. 41-42, embedding module 212 uses a neural network/embedding network (first processor) generating a first tile (batch) 135a as feature vector (first patch extracted) from a medical image 110) and providing the generated first batch to the second processor (Fig. 1B); performing, by the first processor, a second operation of generating a second batch from at least one second patch extracted from the medical image (Fig. 1A and Pars. 41-42, embedding module 212 uses a neural network (first processor) generating a second tile (batch) 135b as feature vector (second patch extracted) from a medical image 110) and providing the generated second batch to the second processor (Fig. 1B, image embedding module 214 or attention network 145a-145b); performing, by the second processor, a third operation of outputting a first analysis result from the first batch (Fig. 1B and Pars. 25-27, combines weighting values of the first tile (batch) 135a and the second tile (batch) 135b) by using a machine learning model (Pars. 49, 65, each node (layer) of an ANN is connected to each node (layer) of the preceding and/or subsequent layers of the ANN), wherein (Par. 39, each tile (patch) includes some portion of pixels of the image that are included or overlapped in at least one other tile (patch)).
Hu does not expressly disclose each of the first tile (batch) and the second tile (batch) corresponding to a time frame. However, this feature cannot be considered new or novel in the presence of Cho. Cho teaches the first model/convolutional neural network (CNN) 221 (first processor) including a convolutional layer corresponding to each time frame of each input data (batch) for extracting feature data and providing to each layer of the second model 222 (second processor), wherein feature data extraction for the previous time frame t-1 (first batch) and feature data extraction of time frame t are overlapped (Fig. 2 and 71, 92, 127). Further, Cho also teaches for each model (trained neural network) that each node (layer) is connected to each node (layer) of the preceding and/or subsequent layers (Fig. 2).
Therefore, it would have been obvious for one of ordinary skill in the art before the effective filing date of the claimed invention to incorporate the above teaching as taught by Cho into Hu to process the time series image.
Regarding claim 13, system of claim 13 is performed by the method of claim 2. They recite similar limitations. Applicant is kindly advised to refer to rejection of claim 2 for claim 13.
Regarding claim 16, system of claim 16 is performed by the method of claim 5. They recite similar limitations. Applicant is kindly advised to refer to rejection of claim 5 for claim 16.
Regarding claim 17, system of claim 17 is performed by the method of claim 6. They recite similar limitations. Applicant is kindly advised to refer to rejection of claim 6 for claim 17.
Regarding claim 18, system of claim 18 is performed by the method of claim 7. They recite similar limitations. Applicant is kindly advised to refer to rejection of claim 7 for claim 18.
Regarding claim 19, system of claim 19 is performed by the method of claim 8. They recite similar limitations. Applicant is kindly advised to refer to rejection of claim 8 for claim 19.
Regarding claim 20, system of claim 20 is performed by the method of claim 9. They recite similar limitations. Applicant is kindly advised to refer to rejection of claim 9 for claim 20.
Claims 3 and 14 are rejected under 35 U.S.C. 103 as being unpatentable over Hu (US 20230419491 A1) in view of Cho et al. (US 20200003886 A1) and in further view of Bu et al. (US 20210049758 A1).
Regarding claim 3, the modified Hu does not expressly disclose the method according to claim 1, wherein the medical image includes a 2D image or a 3D image obtained by scanning or capturing a human body.
However, this feature cannot be considered new or novel in the presence of Bu (Par. 70).
Therefore, it would have been obvious for one of ordinary skill in the art before the effective filing date of the claimed invention to incorporate the above teaching as taught by Bu into the modified Hu to scan target object.
Regarding claim 14, system of claim 14 is performed by the method of claim 3. They recite similar limitations. Applicant is kindly advised to refer to rejection of claim 3 for claim 14.
Claims 4 and 15 are rejected under 35 U.S.C. 103 as being unpatentable over Hu (US 20230419491 A1) in view of Cho et al. (US 20200003886 A1) and in further view of Godrich et al. (US 20230196583 A1).
Regarding claim 4, the modified Hu further teaches the method according to claim 1, wherein the medical image includes segmentation information for a specific object in the medical image (Pars. 22, 29, tile generating module 211 define a tile size depending on type of abnormality (specific object) being detected), and the performing the first operation includes extracting the at least one first patch that includes the specific object (Fig. 1A and Pars. 41-42, embedding module 212 uses a neural network/embedding network (first processor) generating a first tile (batch) 135a as feature vector, and generating the first batch including the extracted at least one first patch (Fig. 1A, 135a (first bach)).
Embedding is a form of feature extraction and Godrich is incorporated in here to provide such support evidence (See Godrich, Fig. 7 and Par. 84).
Therefore, it would have been obvious for one of ordinary skill in the art before the effective filing date of the claimed invention to incorporate the above teaching as taught by Godrich into the modified Hu to extract specific feature.
Regarding claim 15, system of claim 15 is performed by the method of claim 4. They recite similar limitations. Applicant is kindly advised to refer to rejection of claim 4 for claim 15.
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure:
Parvaneh et al. US 20220319654 A1 Each of the pre-trained models may include processor executable instructions.
Mun et al. US 20190088251 A1 [0117] Above is described an example of performing speech recognition by transmitting the personalization layer 611 of the device 610 to the speech signal recognition server 620, and training the personalization layer 611 with respect to speech features or aspects of a particular user by the speech signal recognition server 620. In an example, the device 610 may alternatively be provided the global model 625, and training based on the particular speech features or aspects of the particular user may be performed by the device 610 with respect to the personalization layer 611 and the received global model 625 by the device 610. Also, with respect to FIG. 6, the training data collector 612 and the speech signal recognition controller 613 are each representative of being one or more processors or hardware computing devices or are, in any combination, representative of respective configured operations of such one or more processors or hardware computing devices of the device 610. Similarly, the speech signal recognition controller 621, layer combiner 622, speech signal recognizer 623, and acoustic model trainer 624 are each representative of being one or more processors or hardware computing devices or are, in any combination, representative of respective configured operations of such one or more processors or hardware computing devices of the speech signal recognition server 620
Yerebakan et al. US 20220051805 A1 The image data may be radiology image data. The radiology image data may relate to two-dimensional image data providing two dimensions in space. Further, the radiology image data may relate to three-dimensional image data providing three dimensions in space. In general, the radiology image data depicts a body part of a patient in the sense that it contains two- or three-dimensional image data of the patient's body part.
Marki et al. (US 20180031662 A1)
Nie et al. US 20240079138 A1
Ocampo et al. US 20260120864 A1
Any inquiry concerning this communication or earlier communications from the examiner should be directed to CINDY HUYEN TRANDAI whose telephone number is (571)270-1914. The examiner can normally be reached 8am -4:30pm.
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, Wesley L. Kim can be reached at 571-272-7867. 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.
/Cindy Trandai/Primary Examiner, Art Unit 2648 9/3/2026
Prosecution Insights