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 Objections
Claim 5 is objected to because of the following informalities: The claim recites “. . . computer-implemented according to . . .” where Examiner interprets Applicant intended to recite “. . . computer-implemented method according to . . .” instead. Appropriate correction is required.
Claim Rejections - 35 USC § 112
The following is a quotation of 35 U.S.C. 112(b):
(b) CONCLUSION.—The specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the inventor or a joint inventor regards as the invention.
The following is a quotation of 35 U.S.C. 112 (pre-AIA ), second paragraph:
The specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the applicant regards as his invention.
Claim 10 is rejected under 35 U.S.C. 112(b) or 35 U.S.C. 112 (pre-AIA ), second paragraph, as being indefinite for failing to particularly point out and distinctly claim the subject matter which the inventor or a joint inventor (or for applications subject to pre-AIA 35 U.S.C. 112, the applicant), regards as the invention.
The term “in the vicinity” in claim 10 is a relative term which renders the claim indefinite. The term “in the vicinity” is not defined by the claim, the specification does not provide a standard for ascertaining the requisite degree, and one of ordinary skill in the art would not be reasonably apprised of the scope of the invention. Those of ordinary skill in the art would disagree as to which locations would be considered in the vicinity of the identified location.
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.
The factual inquiries for establishing a background for determining obviousness under 35 U.S.C. 103 are summarized as follows:
1. Determining the scope and contents of the prior art.
2. Ascertaining the differences between the prior art and the claims at issue.
3. Resolving the level of ordinary skill in the pertinent art.
4. Considering objective evidence present in the application indicating obviousness or nonobviousness.
This application currently names joint inventors. In considering patentability of the claims the examiner presumes that the subject matter of the various claims was commonly owned as of the effective filing date of the claimed invention(s) absent any evidence to the contrary. Applicant is advised of the obligation under 37 CFR 1.56 to point out the inventor and effective filing dates of each claim that was not commonly owned as of the effective filing date of the later invention in order for the examiner to consider the applicability of 35 U.S.C. 102(b)(2)(C) for any potential 35 U.S.C. 102(a)(2) prior art against the later invention.
Claim(s) 1, 11-13, and 16 is/are rejected under 35 U.S.C. 103 as being unpatentable over Hesse L.S. ET AL., "Primary Tumor Origin Classification of Lung Nodules in Spectral CT using Transfer Learning", ARXIV.ORG, CORNELL UNIVERSITY LIBRARY, 201 OLIN LIBRARY CORNELL UNIVERSITY ITHACA, NY 14853, 30 June 2020 (2020-06-30), XP081710678 (provided by Applicant with the IDS of 11/4/2024 and hereinafter referred to as “Hesse”) in view of Tushar, Fakrul Islam, et al. "Classification of multiple diseases on body CT scans using weakly supervised deep learning." Radiology: Artificial Intelligence 4.1 (2021): e210026 (hereinafter referred to as “Tushar”).
Regarding claim 1, Hesse discloses a computer-implemented method (see Hesse pg. 8, where the software was implemented on a “Titan XP GPU with 12 GB memory”) of processing spectral image data of a subject generated by a computed tomography scanner (see Hesse pg. 1, where “[i]n spectral CT (also called dual-energy CT) two CT scans are acquired simultaneously with different energy spectra”), the computer-implemented method comprising: processing the spectral image data to identify at least two regions of interest (see Hesse pg. 4, where the algorithm operates on a plurality of “patches”), each region of interest representing one or more organs (see Hesse pg. 6, where the patches correspond to a plurality of regions of the patient’s lungs); processing each region of interest using one or more machine-learning algorithms, each machine-learning algorithm being configured to process the region of interest to generate a predictive indicator that indicates a likelihood that the region of interest contains at least one representation of a predetermined pathology; and outputting at least one predictive indicator generated by the processing of each region of interest (see Hesse Figs. 1 and 2, and pgs. 4 and 5, where “suspicious places” are detected using a “probability” output from a CNN, and then those suspicious places are classified using a SVM to generate an output “malignancy score” and also output a prediction of the “location of the primary tumor”).
Hesse does not explicitly disclose each region of interest representing a different set of one or more organs; processing each region of interest using a respective one of a plurality of sets of one or more machine-learning algorithms.
However, Tushar discloses each region of interest representing a different set of one or more organs; processing each region of interest using a respective one of a plurality of sets of one or more machine-learning algorithms (see Tushar Fig. 2, and pgs. 3-5, where different neural networks are applied to a plurality of patches from a plurality of different organs); each machine-learning algorithm being configured to process the region of interest to generate a predictive indicator that indicates a likelihood that the region of interest contains at least one representation of a predetermined pathology; and outputting at least one predictive indicator generated by the processing of each region of interest (see Tushar Fig. 2 and 6, and pgs. 5 and 9, where a ”multi-label prediction” is output that indicates probabilities for a plurality of predetermined diseases).
It would have been obvious to one of ordinary skill in the art before the effective filing date to duplicate the machine learning techniques of Hesse in the manner taught and suggested by Tushar, resulting in a plurality of specialized expert machine learning models for each of a plurality of different organs, because it is predictable that doing so would reduce time and increase efficiency by detecting multiple potential diseases simultaneously using a single full body spectral CT scan.
Claims 13 and 16 are rejected under the same analysis as claim 1 above.
Regarding claim 11, Hesse discloses wherein each predetermined pathology comprises a disease, lesion, growth or abnormality in the respective region of interest (see Hesse Figs. 1 and 2, and pgs. 4 and 5, where “suspicious places” are detected, and then those suspicious places are classified with a “malignancy score” and “location of the primary tumor”).
Tushar also discloses wherein each predetermined pathology comprises a disease, lesion, growth or abnormality in the respective region of interest (see Tushar Fig. 2 and 6, and pgs. 5 and 9, where a “multi-label prediction” is output that indicates probabilities for a plurality of predetermined diseases).
Regarding claim 12, Hesse discloses wherein processing the spectral image data to identify at least two regions of interest comprises either: performing an image segmentation process on the spectral image data; or registering the spectral image data to an anatomical atlas that identifies expected regions of interest; and identifying regions of the registered spectral image data falling within the expected regions of interest of the anatomical atlas as the at least two regions of interest (see Hesse Fig. 3, and pgs. 5 and 6, where lung segmentation is performed using a lung mask).
Tushar also discloses wherein processing the image data to identify at least two regions of interest comprises either: performing an image segmentation process on the image data; or registering the spectral image data to an anatomical atlas that identifies expected regions of interest; and identifying regions of the registered image data falling within the expected regions of interest of the anatomical atlas as the at least two regions of interest (see Tushar Fig. 2, and pgs. 3 and 4, where image segmentation is performed using segmentation masks for all the organs).
It would have been obvious to one of ordinary skill in the art before the effective filing date to simply substitute the lung segmentation of Hesse with the multiple organ segmentation of Tushar, because it is predictable that doing so would reduce time and increase efficiency by detecting multiple potential diseases across multiple organs simultaneously using a single full body spectral CT scan.
Claim(s) 2 and 14 is/are rejected under 35 U.S.C. 103 as being unpatentable over Hesse in view of Tushar as applied to claim 1 above, and in further view of Villard et al., US 2022/0405927 A1 (hereinafter referred to as “Villard”).
Regarding claim 2, Hesse does not explicitly disclose wherein: each machine-learning algorithm is further configured to generate a confidence score for the predictive indicator generated by the machine-learning algorithm, the confidence score representing a confidence in the likelihood that the region of interest contains the at least one representation of the predetermined pathology; the computer-implemented method further comprises selecting at least one of the predictive indicators generated by the processing of each region of interest responsive to each confidence score generated by the sets of one or more machine-learning algorithms; and outputting at least one predictive indicator comprises outputting the selected at least one predictive indicator.
However, Villard discloses wherein: each machine-learning algorithm is further configured to generate a confidence score for the predictive indicator generated by the machine-learning algorithm, the confidence score representing a confidence in the likelihood that the region of interest contains the at least one representation of the predetermined pathology; the computer-implemented method further comprises selecting at least one of the predictive indicators generated by the processing of each region of interest responsive to each confidence score generated by the sets of one or more machine-learning algorithms; and outputting at least one predictive indicator comprises outputting the selected at least one predictive indicator (see Villard paras. 0029, 0030, 0052, 0053, 0038, and 0094, where confidence scores are calculated for multiple DD and IQA models and based on the confidence score either a disease prediction is output (high confidence) or an indication to retake images (low confidence)).
It would have been obvious to one of ordinary skill in the art before the effective filing date to use the techniques of Villard to calculate the confidence scores for the machine learning models of Hesse, as modified by Tushar, because it is predictable that doing so would improve disease diagnosis by retaking images in cases of low confidence and/or providing the physician with confidence levels for each disease prediction so that the physician is better informed before deciding how to proceed.
Claim 14 is rejected under the same analysis as claim 2 above.
Claim(s) 3 and 7-10 is/are rejected under 35 U.S.C. 103 as being unpatentable over Hesse in view of Tushar as applied to claim 1 above, and in further view of Flewelling et al., US 2006/0013454 A1 (hereinafter referred to as “Flewelling”).
Regarding claim 3, Hesse does not explicitly disclose wherein outputting the at least one predictive indicator comprises controlling a user interface to provide a visual representation responsive to the at least one predictive indicator.
However, Flewelling discloses wherein outputting the at least one predictive indicator comprises controlling a user interface to provide a visual representation responsive to the at least one predictive indicator (see Flewelling para. 1140, where color is added to the image to identify probable disease locations based on a minimum cut-off probability).
It would have been obvious to one of ordinary skill in the art before the effective filing date to use the color and location visual representations of Flewelling to indicate the probable diseases detected by Hesse, as modified by Tushar, because it is predictable that doing so would improve efficiency and accuracy of disease diagnosis by pointing the physician directly to the locations of concern in the images so that the physician can immediately confirm each disease diagnosis.
Regarding claim 7, Hesse does not explicitly disclose wherein at least one of the machine-learning algorithms comprises a location- identifying machine-learning algorithm, wherein: the location-identifying machine-learning algorithm is configured to process the region of interest to identify, within the region of interest, the location of any representation of the predetermined pathology predicted to be present with a likelihood greater than a predetermined likelihood; and the predictive indicator produced by the location-identifying machine-learning algorithm indicates any identified location within the region of interest.
However, Flewelling discloses wherein at least one of the machine-learning algorithms comprises a location- identifying machine-learning algorithm, wherein: the location-identifying machine-learning algorithm is configured to process the region of interest to identify, within the region of interest, the location of any representation of the predetermined pathology predicted to be present with a likelihood greater than a predetermined likelihood; and the predictive indicator produced by the location-identifying machine-learning algorithm indicates any identified location within the region of interest (see Flewelling para. 1140, where color is added to the image to identify probable disease locations based on a minimum cut-off probability).
It would have been obvious to one of ordinary skill in the art before the effective filing date to use the color and location visual representations of Flewelling to indicate the probable diseases detected by Hesse, as modified by Tushar, because it is predictable that doing so would improve efficiency and accuracy of disease diagnosis by pointing the physician directly to the locations of concern in the images so that the physician can immediately confirm each disease diagnosis.
Regarding claim 8, Hesse does not explicitly disclose wherein: outputting each predictive indicator comprises controlling a user interface to provide a visual representation responsive to the at least one predictive indicator; and for each location identified by any location-identifying machine-learning algorithm that provides a predictive indicator in the at least one predictive indicator, the visual representation comprises a visual representation responsive to the identified location.
However, Flewelling discloses wherein: outputting each predictive indicator comprises controlling a user interface to provide a visual representation responsive to the at least one predictive indicator; and for each location identified by any location-identifying machine-learning algorithm that provides a predictive indicator in the at least one predictive indicator, the visual representation comprises a visual representation responsive to the identified location (see Flewelling para. 1140, where color is added to the image to identify probable disease locations based on a minimum cut-off probability).
Regarding claim 9, Hesse discloses spectral image data (see Hesse pg. 1, where “[i]n spectral CT (also called dual-energy CT) two CT scans are acquired simultaneously with different energy spectra”).
Hesse does not explicitly disclose wherein controlling the user interface comprises, for each location identified by any location-identifying machine- learning algorithm that provides a predictive indicator in the at least one predictive indicator, overlaying a visual representation of the identified location over a visual representation of the image data.
However, Flewelling discloses wherein controlling the user interface comprises, for each location identified by any location-identifying machine- learning algorithm that provides a predictive indicator in the at least one predictive indicator, overlaying a visual representation of the identified location over a visual representation of the image data (see Flewelling para. 1140, where color is added to the image to identify probable disease locations based on a minimum cut-off probability).
Regarding claim 10, Hesse does not explicitly disclose wherein controlling the user interface comprises, for each location identified by any location- identifying machine-learning algorithm that provides a predictive indicator in the at least one predictive indicator, providing a visual representation of a portion of the image data in the vicinity of the identified location.
However, Flewelling discloses wherein controlling the user interface comprises, for each location identified by any location- identifying machine-learning algorithm that provides a predictive indicator in the at least one predictive indicator, providing a visual representation of a portion of the image data in the vicinity of the identified location (see Flewelling para. 1140, where color is added to the image to identify probable disease locations based on a minimum cut-off probability).
Claim(s) 4 is/are rejected under 35 U.S.C. 103 as being unpatentable over Hesse in view of Tushar and Flewelling as applied to claim 3 above, and in further view of Villard.
Regarding claim 4, Hesse does not explicitly disclose wherein: each machine-learning algorithm is further configured to generate a confidence score for the predictive indicator generated by the machine-learning algorithm, the confidence score representing a confidence in the likelihood that the region of interest contains the at least one representation of the predetermined pathology; and controlling the user interface is further responsive to each confidence score generated by the set of one or more machine-learning algorithms.
However, Villard discloses wherein: each machine-learning algorithm is further configured to generate a confidence score for the predictive indicator generated by the machine-learning algorithm, the confidence score representing a confidence in the likelihood that the region of interest contains the at least one representation of the predetermined pathology; and controlling the user interface is further responsive to each confidence score generated by the set of one or more machine-learning algorithms (see Villard paras. 0029, 0030, 0052, 0053, 0038, and 0094, where confidence scores are calculated for multiple DD and IQA models and based on the confidence score either a disease prediction is output (high confidence) or an indication to retake images (low confidence)).
It would have been obvious to one of ordinary skill in the art before the effective filing date to use the techniques of Villard to calculate the confidence scores for the machine learning models of Hesse, as modified by Tushar and Flewelling, because it is predictable that doing so would improve disease diagnosis by retaking images in cases of low confidence and/or providing the physician with confidence levels for each disease prediction so that the physician is better informed before deciding how to proceed.
Claim(s) 5 is/are rejected under 35 U.S.C. 103 as being unpatentable over Hesse in view of Tushar, Flewelling, and Villard as applied to claim 4 above, and in further view of Rodrigues-Diaz et al., US 2023/0260111 A1 (hereinafter referred to as “Rodrigues-Diaz”).
Regarding claim 5, Hesse does not explicitly disclose wherein controlling the user interface comprises providing greater visual emphasis to any of the at least one predictive indicator with higher confidence scores than any of the at least one predictive indicator with lower confidence scores.
However, Rodrigues-Diaz discloses wherein controlling the user interface comprises providing greater visual emphasis to any of the at least one predictive indicator with higher confidence scores than any of the at least one predictive indicator with lower confidence scores (see Rodrigues-Diaz para. 0083, where the image uses a green color to indicate a high confidence and a yellow color indicate a low confidence).
It would have been obvious to one of ordinary skill in the art before the effective filing date to use the visual emphasis technique of Rodrigues-Diaz to highlight the confidence scores of Villard within the images of Hesse, as modified by Tushar and Flewelling, because it is predictable that doing so would improve the efficiency and accuracy of disease diagnosis by emphasizing the locations and diseases whose predictions have the highest confidence. Furthermore, the color scheme takes advantage of human intuition and our everyday experience that green typically means act and yellow typically means caution.
Claim(s) 6 is/are rejected under 35 U.S.C. 103 as being unpatentable over Hesse in view of Tushar, Flewelling, and Villard as applied to claim 4 above, and in further view of Bernard et al., US 10,140,421 B1 (hereinafter referred to as “Bernard”).
Regarding claim 6, Hesse does not explicitly disclose wherein controlling the user interface comprises: ordering the at least one predictive indicator based on their confidence measures to determine an order of the at least one predictive indicator; and controlling the user interface to provide a visual representation of the determined order of the at least one predictive indicator.
However, Bernard discloses wherein controlling the user interface comprises: ordering the at least one predictive indicator based on their confidence measures to determine an order of the at least one predictive indicator; and controlling the user interface to provide a visual representation of the determined order of the at least one predictive indicator (see Bernard col. 41, lls. 9-45, where the scans are sorted and ordered based on confidence scores corresponding to detection of cardiomegaly of confidence score data).
It would have been obvious to one of ordinary skill in the art before the effective filing date to use the sorting and ordering technique Bernard based on the confidence scores of Villard to order the viewing of the image locations of Hesse, as modified by Tushar and Flewelling, because it is predictable that doing so would improve efficiency and accuracy by providing the physician with the most confident disease predictions first before viewing lower confident disease predictions that are unlikely to be correct.
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure.
Accomazzi et al., US 2020/0335198 A1, discloses selecting a neural network based on the anatomy in the image series (see Accomazzi paras. 0039, 0140, and 0400-0405).
Jin, Dan, et al. "Multiphase Dual-Energy Spectral CT-Based Deep Learning Method for the Noninvasive Prediction of Head and Neck Lymph Nodes Metastasis in Patients With Papillary Thyroid Cancer." Frontiers in Oncology 12 (2022): 869895, discloses processing spectral CT scans with a deep learning model to predict lymph node metastasis (see Jin pg. 1).
Choi, Won Seok, et al. "Spectral CT-based iodized oil quantification to predict tumor response following chemoembolization of hepatocellular carcinoma." Journal of Vascular and Interventional Radiology 32.1 (2021): 16-22, discloses processing spectral CT scans for monitoring tumors (see Choi Abstract).
Any inquiry concerning this communication or earlier communications from the examiner should be directed to ANDREW M MOYER whose telephone number is (571)272-9523. The examiner can normally be reached Monday-Friday 9-5 EST.
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.
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.
/ANDREW M MOYER/ Supervisory Patent Examiner, Art Unit 2675
Prosecution Insights