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 .
Interview Statement
Applicant’s summary of the interview on October 15, 2025, has been reviewed and is complete and accurate.
Response to Amendments
Applicant's amendments and remarks, filed 10/20/2025, are acknowledged. Applicant's arguments have been fully considered. The following rejections and/or objections are either reiterated or newly applied. They constitute the complete set presently being applied to the instant application. Rejections and/or objections not reiterated from the previous office actions are hereby withdrawn.
Status of Claims
Claims 1-2, 4-6, 8-19 are currently under examination. Claims 3, 7 and 20 are cancelled.
Information Disclosure Statement
The information disclosure statement (IDS) document(s) submitted on 10/21/2025, is/are in compliance with the provisions of 37 CFR 1.97. Accordingly, the IDS document(s) has/have been fully considered by the examiner.
Priority
This application is a continuation of US application 18/508,098, filed 11/13/2023, which is a CIP of 18/179,921, filed 03/07/2023.
Applicant' s claim for the benefit of priority under 35 U.S.C. 119(e) for application 18/179,921 to provisional application 63/381,210, filed 10/27/2022, 63/368,293, filed 07/13/2022, 63/365,381, filed 05/26/2022, 63/364,078, filed 05/03/2022, 63/364,084, filed 05/03/2022, 63/362,856, filed 04/12/2022, 63/362,108, filed 03/29/2022, 63/269,136, filed 03/10/2022,
Applicant' s claim for the benefit of priority under 35 U.S.C. 119(e) for application 18/508,098 to provisional application 63/477,985, filed 12/30/2022, 63/478,076, filed 12/30/2022, 63/477,961, filed 12/30/2022, 63/478,084, filed 12/30/2022, 63/477,656, filed 12/29/2022, 63/477,638, filed 12/29/2022, 63/477,640, filed 12/29/2022, 63/476,251, filed 12/20/2022, 63/476,245, filed 12/20/2022, 63/476,255, filed 12/20/2022, 63/386,376, filed 12/07/2022, 63/386,297, filed 12/06/2022, 63/385,472, filed 11/30/2022, 63/385,179, filed 11/28/2022, 63/383,904, filed 11/15/2022, 63/383,632, filed 11/14/2022.
The examiner notes that only the provisional application 65/477,656 is disclosing the use of NLP for system and method as claimed in the instant claimed invention and therefore only this application is considered for priority of the claims providing a priority date on 12/29/2022.
Response to Arguments
Applicant’s arguments filed 10/20/2025 have been fully considered.
Applicant amended the independent claims with subject matter not previously prosecuted and therefore are modifying the scope of the claims necessitating new grounds of rejection.
Applicant argues regarding the claim rejections under 35 U.S.C. 101 that the amendments are placing the independent claims in patent eligible.
Applicant first argues that the claims do not recite a judicial exception as a mental process since the claims recite the generation of primary and secondary image-derived variables that cannot be determined by mental activity from the image being accessed and analyzing the identified relevant scientific article via natural language processing to generate a summary of the data related to the medical condition of the subject cannot also be performed by mental activity arguing the amount of the publications being presented and analyzed limiting in time the analysis of the data.
In response, the examiner is considering the determination of the secondary image-derived variables as mentally accessible such as the blood vessel length, segment length or lesion length. The examiner would consider that accessing mentally from the medical images the four primary variables as possible within some uncertainty limit to the determination depending on the types of medical images being accessed for the level of pre-processing being provided as interpretation of results from radiologist is based on medical images as regarding the size of the different regions of the plaques as essential and visible due to their intensity contrast.
Applicant further argues that under the Step 2A prong Two, the claim is integrating the judicial exception into a practical application wherein the claim is directed to analyze the subject medical images then extract specific variables characterizing the medical status of the subject from which the invention is directed to generate syntaxes descriptive of these variables from the medical images to be used for performing searches within medical and research database for scientific and medical publications which are analyzed using natural language processing or AI for generating a summary of the published data as related to the medical status of the subject therefore the claim being directed to a specific technical approach for performing searches for information within relevant scientific and medical publication using the medical image analysis results related to the subject to generate an update of the medical knowledge related to the medical status of the subject as assessed from his medical images.
In response, the examiner is considering that the integration of the judicial exception into a practical application is found within the amended limitation with the generation of the summary of the knowledge as directly related to the medical images of the subject within a specific medical context. Therefore the examiner is considering the amended claims as integrating the judicial exception into a practical application.
Applicant further argues that under the Step 2B, the amended limitations as described are not routine, well-understood and/or conventional activity and therefore present an inventive step.
In response, the examiner is considering that the steps are not inventive since extracting features from medical images and using them to provide keywords for searching and performing automatic summarization via AI of articles/publications as based on the keywords related to the specific medical condition related to the subject are known in the art.
In conclusion, since the examiner is considering the claims are integrating the judicial exemption into practical application, the claim is patent eligible. The claim rejection under 35 U.S.C. 101 is therefore withdrawn.
Regarding the claim rejections under 35 U.S.C. 103, Applicant amended the independent claims with subject matter not previously prosecuted necessitating new grounds of rejection. Applicant is claiming that the references of record do not teach all the amended limitations.
In response, the examiner is considering the arguments as moot since they are directed to subject matter necessitating new grounds of rejection.
Claim Rejections - 35 USC § 112
The following is a quotation of 35 U.S.C. 112(d):
(d) REFERENCE IN DEPENDENT FORMS.—Subject to subsection (e), a claim in dependent form shall contain a reference to a claim previously set forth and then specify a further limitation of the subject matter claimed. A claim in dependent form shall be construed to incorporate by reference all the limitations of the claim to which it refers.
The following is a quotation of pre-AIA 35 U.S.C. 112, fourth paragraph:
Subject to the following paragraph [i.e., the fifth paragraph of pre-AIA 35 U.S.C. 112], a claim in dependent form shall contain a reference to a claim previously set forth and then specify a further limitation of the subject matter claimed. A claim in dependent form shall be construed to incorporate by reference all the limitations of the claim to which it refers.
Claims 4-6 and 8 are rejected under 35 U.S.C. 112(d) or pre-AIA 35 U.S.C. 112, 4th paragraph, as being of improper dependent form for failing to further limit the subject matter of the claim upon which it depends, or for failing to include all the limitations of the claim upon which it depends. Claims 4-6 and 8 appears to be directly or indirectly dependent from claim 3 which has been cancelled. Clarification is required via amendments.
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, 9, 11-14, are rejected under 35 U.S.C. 103 as being unpatentable over Min (USPN 20200237329 A1; Pub.Date 07/30/2020; Fil.Date 01/23/2020) in view of Min et al. (USPN 20210319558 A1; Pub.Date 10/14/2021; Fil.Date 01/17/2021) thereafter Min’2021 in view of Yang et al. (2021 arXiv:2106-06471v1 10 pages; Pub.Date 05/21/2021) in view of Mridha et al. (2021 IEEEAccess 9:156043-156070; Pub.Date 11/22/2021).
Regarding independent claim 1, Min teaches systems and method for characterizing plaques and characterizing the efficacy of the treatment based on the classified risk presented by the plaques based on matching the radiodensity of the plaques within a 3D representation of the coronary arteries (Title, abstract and [0018], [0076]) and providing patients reports including patient medical images and comparison of data/information with other patients’ data representation ([0069]) via computer/processor executing the methods ([0010]) therefore teaching a computer-implemented method of automatically searching and curating data related to a medical condition of a subject based at least in part on one or more variables derived from image-based analysis of the subject, the method comprising:
accessing, by a computer system, a medical image of a subject, the medical image comprising a representation of a portion of one or more coronary arteries ([0068]-[0069] plurality of images of a patient's coronary arteries and patient information/characteristics may be provided and may include 2D or 3D image data);;
analyzing, by the computer system, the medical image to identify one or more coronary arteries, the one or more coronary arteries comprising one or more regions of plaque ([0077] identify coronary arteries via segmenting and smoothing image data then providing a 2D or 3D representation of the coronary arteries [0103]);
analyzing, by the computer system [… using at least one machine learning model…], the identified one or more coronary arteries and the one or more regions of plaque to generate a plurality of primary image-derived variables, the plurality of primary image-derived variables comprising: severity of stenosis, total plaque volume, low-density non-calcified plaque volume, non-calcified plaque volume, and calcified plaque volume ([0010], [0014], [0019], [0043], [0101] volumetric analysis of the plaque for characterization with Figs. 6-7 and [0104]-[0110] with characterization of the different types of plaques according to their radiodensity/density with the determination of each plaque boundary or contour according to the Figs. 6-7, their density according their radiodensity directly related to their physical density [0110], their volume according to Fig. 15, including plaque volume [0101], [0109] with the different types of plaques dependent on their attenuations corresponding to the low density non-calcified, non-calcified and calcified plaque [0104]-[0110], Figs. 6-7)
analyzing, by the computer system, [… using the at least one machine learning model…], the identified one or more coronary arteries and the one or more regions of plaque to generate a plurality of secondary image-derived variables, the plurality of image-derived variables comprising one or more of percent atheroma volume, necrotic core volume, vessel length, segment length, lesion length, number of chronic total occlusion (CTO), number of stenoses, remodeling index, minimum lumen diameter, maximum lumen diameter, or perivascular fat attenuation, (0010], [0014], [0019], [0043], [0101] volumetric analysis of the blood vessels and plaque including baseline anatomy for vessel and segments characterization including length ([0101], [0154], [0158], [0162]) or perivascular fat characterization via CT imaging ([0010]) reading on attenuation).
[…generating, by the computer system, one or more syntaxes descriptive of the medical image based at least in part on the plurality primary image-derived variables and the plurality of secondary image-derived variables;
generating, by the computer system, one or more search queries for searching a plurality of external medical literature databases for data related to a medical condition of the subject based at least in part on the generated one or more syntaxes, wherein the plurality of external medical literature databases comprise scientific articles relating to coronary artery disease, the scientific articles comprising one or more medical images with associated syntaxes derived therefrom;
causing, by the computer system, execution of the one or more search queries against the plurality of external medical literature databases to identify one or more relevant scientific articles, wherein the relevant scientific articles are identified based at least in part on syntaxes derived from medical images included in the scientific articles
wherein the syntaxes are based at least in part primary image-derived variables and secondary image-derived variables;
analyzing, by the computer system using a natural language processing (NLP) algorithm, the identified relevant scientific articles to generate a summary of the data related to the medical condition of the subject; and …];
and causing, by the computer system, generation of a display of the […summary of the …] data related to the medical condition of the subject, ([0069] using the processor system to generate patient reports including graphical depictions of the patient coronary arteries including plaque containing arteries ([0075] generating 3D depiction of coronary arteries and plaque contained within the coronary arteries, [0076] as mapping pixels of the 2D image to a 3D mapping image of the coronary arteries with the radiodensity to generate a graphical depiction of the plaques within the coronary arteries, [0105] and Fig. 5A step 515 to generate a 3D representation of coronary arteries with plaques leading to Figs. 6 and 7 as representation of the image data as graphical representation of where the plaques are localized) wherein the computer system comprises a computer processor and an electronic storage medium ([0082]-[0083] processor 404 and computer readable storage media or main memory 406 for storing program instructions to be executed by the processor 404).
Min does not specifically teach using at least one machine learning model for generating the primary and secondary image-derived variables, generating, by the computer system, one or more syntaxes descriptive of the medical image based at least in part on the plurality primary image-derived variables and the plurality of secondary image-derived variables; generating, by the computer system, one or more search queries for searching a plurality of external medical literature databases for data related to a medical condition of the subject based at least in part on the generated one or more syntaxes, wherein the plurality of external medical literature databases comprise scientific articles relating to coronary artery disease, the scientific articles comprising one or more medical images with associated syntaxes derived therefrom; causing, by the computer system, execution of the one or more search queries against the plurality of external medical literature databases to identify one or more relevant scientific articles, wherein the relevant scientific articles are identified based at least in part on syntaxes derived from medical images included in the scientific articles, wherein the syntaxes are based at least in part primary image-derived variables and secondary image-derived variables; analyzing, by the computer system using a natural language processing (NLP) algorithm, the identified relevant scientific articles to generate a summary of the data related to the medical condition of the subject; and generating a summary of the data as in claim 1.
However, Min’2021 teaches within the same field of endeavor of analyzing CCTA medical images for characterizing plaque (Title and abstract) the use of machine learning for extracting anatomical/blood vessels and plaque variables from the medical images ([0008], [0167], [0174], [0405]-[0444] for plaque and vessel characterization, and also image reconstruction with necessary determination of the geometric characteristics of the vessels and plaques [0449]) therefore reading on using at least one machine learning model for generating the primary and secondary image-derived variables.
Therefore it would have been obvious for a person of ordinary skill in the art before the effective filling date of the invention to have modified the method of Min such that the method further comprises: using at least one machine learning model for generating the primary and secondary image-derived variables, since one of ordinary skill in the art would recognize that extracting image features related to the patient medical images such as CCTA in order to provide blood vessels and plaque characterization with associated variables was known in the art as taught by Min’2021. One of ordinary skill in the art would have expected that this modification could have been made with predictable results since Min and Min’2021 teach analyzing image data for characterizing the medical images for blood vessels and plaque. The motivation would have been to automate the characterization of the plaque and coronary arteries, as suggested by Min’2021 (abstract).
Min and Min’2021 do not specifically teach generating, by the computer system, one or more syntaxes descriptive of the medical image based at least in part on the plurality primary image-derived variables and the plurality of secondary image-derived variables; generating, by the computer system, one or more search queries for searching a plurality of external medical literature databases for data related to a medical condition of the subject based at least in part on the generated one or more syntaxes, wherein the plurality of external medical literature databases comprise scientific articles relating to coronary artery disease, the scientific articles comprising one or more medical images with associated syntaxes derived therefrom; causing, by the computer system, execution of the one or more search queries against the plurality of external medical literature databases to identify one or more relevant scientific articles, wherein the relevant scientific articles are identified based at least in part on syntaxes derived from medical images included in the scientific articles, wherein the syntaxes are based at least in part primary image-derived variables and secondary image-derived variables; analyzing, by the computer system using a natural language processing (NLP) algorithm, the identified relevant scientific articles to generate a summary of the data related to the medical condition of the subject;
However, Yang teaches within the same field of endeavor of providing medical report generation from medical images (Title and abstract and Fig.1) to retrieve the most relevant medical reports for the given medical images of the patient (abstract) wherein the software MedWriter initially predicts disease labeled as word syntax from the medical X-ray images (p.2 col.1 1st ¶ “predicts disease labels based on the visual features (Figs. 1-2)). Additionally, in order to generate an accurate medical report, Yang teaches the automatic retrieval of prior medical reports (p.4-6 ¶ 3.2 LLR module pretraining and 3.3 Retrieval-based report generation) for matching medical image features and sentences by semantics matching in order to generate the patient medical report therefore teaching using the variables already taught by Min with knowing that the medical reports and scientific/medical publications already used for training and validating the machine learning model as taught by Min’2021 as being at least included within the set of publications/reports to be searched within the inquiries therefore teaching generating, by the computer system, one or more search queries for searching a plurality of external medical literature databases for data related to a medical condition of the subject based at least in part on the generated one or more syntaxes, wherein the plurality of external medical literature databases comprise scientific articles relating to coronary artery disease, the scientific articles comprising one or more medical images with associated syntaxes derived therefrom; causing, by the computer system, execution of the one or more search queries against the plurality of external medical literature databases to identify one or more relevant scientific articles, wherein the relevant scientific articles are identified based at least in part on syntaxes derived from medical images included in the scientific articles, wherein the syntaxes are based at least in part primary image-derived variables and secondary image-derived variables; analyzing, by the computer system using a natural language processing (NLP) algorithm, the identified relevant scientific articles as claimed.
Therefore it would have been obvious for a person of ordinary skill in the art before the effective filling date of the invention to have modified the method of Min such that the method further comprises: generating, by the computer system, one or more search queries for searching a plurality of external medical literature databases for data related to a medical condition of the subject based at least in part on the generated one or more syntaxes, wherein the plurality of external medical literature databases comprise scientific articles relating to coronary artery disease, the scientific articles comprising one or more medical images with associated syntaxes derived therefrom; causing, by the computer system, execution of the one or more search queries against the plurality of external medical literature databases to identify one or more relevant scientific articles, wherein the relevant scientific articles are identified based at least in part on syntaxes derived from medical images included in the scientific articles, wherein the syntaxes are based at least in part primary image-derived variables and secondary image-derived variables; analyzing, by the computer system using a natural language processing (NLP) algorithm, the identified relevant scientific articles, since one of ordinary skill in the art would recognize that extracting image features related to the patient medical conditions in order to provide disease labels attached to the medical image was known in the art as taught by Yang considering the extraction of coronary artery disease variables was also known in the art as taught by Min and since the use of medical images from previous medical publications and reports with reference image analysis was also known in the art as taught by Min’2021 for machine learning development and using the results and the corresponding variables for searching prior medical reports/data for retrieving medical information was also known in the art as taught by Yang. One of ordinary skill in the art would have expected that this modification could have been made with predictable results since Min and Yang teach analyzing image data for characterizing the medical status of the patient and generating a patient medical report including the description of the disease and additional related comparative information. The motivation would have been to automate and improve the generation of the medical image report while minimizing the time burden of the practitioner necessary to generate that medical report, as suggested by Yang (abstract and p.2 col.1 1st-4th ¶).
Min, Min’2021 and Yang do not specifically teach analyzing, by the computer system using a natural language processing (NLP) algorithm, the identified relevant scientific articles to generate a summary of the data related to the medical condition of the subject as in claim 1.
However, Mridha reviews within the same field of endeavor data analysis including machine learning and deep learning model (Title, abstract, Table 7) as directed to scientific and medical publications analysis (p.156046 col.1 3rd ¶) and teaches the automatic summarization of documents such as medical and scientific documents (p.156046 ¶ IV. Structure of ATS, Figs.1-3, and p.156054-156055 ¶ IX. 2) Deep Learning Algorithm) therefore combined with the teachings of Min, Min’2021 and Yang directed to medical image characterization, Mridha teachings read on analyzing, by the computer system using a natural language processing (NLP) algorithm, the identified relevant scientific articles to generate a summary of the data related to the medical condition of the subject as claimed.
Therefore it would have been obvious for a person of ordinary skill in the art before the effective filling date of the invention to have modified the method of Min as modified by Min’2021 and Yang such that the method further comprises: analyzing, by the computer system using a natural language processing (NLP) algorithm, the identified relevant scientific articles to generate a summary of the data related to the medical condition of the subject, since one of ordinary skill in the art would recognize that analyzing via NLP algorithm selected scientific and medical documents to provide a summary of the relevant data from the documents was known in the art as taught by Mridha. One of ordinary skill in the art would have expected that this modification could have been made with predictable results since Yang and Mridha teach the review and analysis of reports and/or publications for diagnosis. The motivation would have been to automatically and efficiently provide the common knowledge related to a specific medical condition as corresponding to the subject, as suggested by Mridha (abstract).
Regarding the dependent claims 2, 9, 11-14, all the elements of these claims are instantly disclosed or fully envisioned by the combination of Min, Min’2021, Yang and Mridha.
Regarding claim 2, as discussed above, Min teaches the use of other patients’ medical images for comparison and classification of the medical conditions shown by the patient medical image ([0069]) therefore reading on the data related to the medical condition of the subject comprises one or more images of other subjects with similar medical conditions to the subject as claimed.
Regarding claim 9, as discussed above, Min teaches the use of machine learning trained with other patients medical images for identifying information related to the medical condition of the patient using comparative analysis ([0069]) therefore teaching the plurality of image-derived variables is generated using one or more of an artificial intelligence (AI) or machine learning (ML) algorithm trained on a dataset comprising a plurality of medical images with known image-derived variables from a plurality of other subjects as claimed.
Regarding dependent claims 11-13, Min as discussed above teaches the medical image is obtained using a Computed Tomography (CT) image ([0013] CT scans of arteries) as in claim 11, therefore also as in claim 12 the medical image is obtained using coronary CT angiography (CCTA) ([0053]) and for claim 13 for the medical image is obtained using an imaging technique comprising one or more of CT, x-ray, ultrasound, echocardiography, MR imaging, optical coherence tomography (OCT), nuclear medicine imaging, positron-emission tomography (PET), single photon emission computed tomography (SPECT), or near-field infrared spectroscopy (NIRS).
Regarding claim 14, as discussed above, Min teaches the CT medical image ([0013]) wherein Min teaches “the one or more plaque parameters comprises classification of the one or more regions of plaque” with the consideration of separating the plaques in different classes such as low density non-calcified plaque, non-calcified plaque, and calcified plaque according to the corresponding radiodensity within the respective Hounsfield Unit ranges ([0008], [0054]-[0055], [0117]-[0124] and Fig.7 with perivascular fat region as low density (HU closer to -100), low density plaque presenting high risk and high density calcified plaque as classified presenting HU in range 130-4000 presenting low risk ) therefore teaching the different types of plaques classified according to their radiodensity as correlated to their physical density as wherein low-density non-calcified plaque comprises a region of plaque comprising a radiodensity value between about -189 and about 30 Hounsfield units, wherein non-calcified plaque comprises a region of plaque comprising a radiodensity value between about 31 and about 350 Hounsfield units, and wherein calcified plaque comprises a region of plaque comprising a radiodensity value between about 351 and 2500 Hounsfield units as claimed.
Claims 4-5, 10 are rejected under 35 U.S.C. 103 as being unpatentable over Min (USPN 20200237329 A1; Pub.Date 07/30/2020; Fil.Date 01/23/2020) in view of Min et al. (USPN 20210319558 A1; Pub.Date 10/14/2021; Fil.Date 01/17/2021) thereafter Min’2021 in view of Yang et al. (2021 arXiv:2106-06471v1 10 pages; Pub.Date 05/21/2021) in view of Mridha et al. (2021 IEEEAccess 9:156043-156070; Pub.Date 11/22/2021) as applied to claim 1 and further in view of Yang et al. (2023 IEEE Trans. on Multimedia 25:167-178; First Pub.Date 11-08-2021) hereafter Yang’2021 in view of Costa et al. (2021 MedRxiv 6 pages; Pub.Date 2021).
Min, Min’2021, Yang and Mridha teach a method as set forth above.
Min, Min’2021, Yang and Mridha do not specifically teach generating, by the computer system, a listing of the one or more scientific articles by one or more of relevance or date.
However Yang’2021 teaches within the same field of endeavor of generating medical report for medical images (Title and abstract) the extraction of image features leading to the image embedding of MeSH as related medical tags (abstract, Fig. 2 and p.168 col.1 2nd ¶ part 1-3), additionally, Costa teaches within the same field of endeavor of generating medical reports (Title, abstract and p.5 col.2 ¶ 7, for the generation of medical record using the MeSH classifier for integrating health related information from health-related text such as from the specific dataset as from MEDLINE as database known for providing access to medical/biomedical/scientific articles using MESH keywords/semantics (p.2-3 ¶ 3. Construction of the dataset) within medical reports as Costa teaches the integration of the information from scientific articles from MEDLINE within medical reports to improve the accuracy of the reports wherein one of ordinary skill in the art would recognize as common practice for any scientific related publication to provide a reference list of any scientific articles from which relevant information is being presented within the medical report therefore making obvious generating, by the computer system, a listing of the one or more scientific articles by one or more of relevance or date as claimed in claim 4.
Therefore it would have been obvious for a person of ordinary skill in the art before the effective filling date of the invention to have modified the method of Min as modified by Min’2021, Yang and Mridha such that the method further comprises: generating, by the computer system, a listing of the one or more scientific articles by one or more of relevance or date, since one of ordinary skill in the art would recognize that extracting image features related to the patient medical conditions including the corresponding MESH for characterizing the patient medical image providing disease labels attached to the medical image was known in the art as taught by Yang’2021 and using them for searching prior medical reports/scientific data for retrieving medical information from MEDLINE was also known in the art as taught by Yang. One of ordinary skill in the art would have expected that this modification could have been made with predictable results since Min and Yang’2021 teach analyzing image data for characterizing the medical status of the patient and generating a patient medical report including the description of the disease and additional related comparative information. The motivation would have been to automate and improve the generation of the medical image report while minimizing the time burden of the practitioner necessary to generate that medical report, as suggested by Yang (abstract and p.2 col.1 1st-4th ¶).
Regarding the dependent claim 5, all the elements of these claims are instantly disclosed or fully envisioned by the combination of Min, Min’2021, Yang, Mridha, Yang’2021 and Costa.
Regarding claim 5, as discussed above, Yang’2021 and Costa teach the development of MeSH classifier for characterizing the medical conditions of concern based on the considered medical images for searching MedLine. Since Yang’2021 teaches the use of MeSH keywords/syntaxes for characterizing the patient medical images and Costa is teaching the use also of MeSH for searching the corresponding scientific articles, therefore Yang’2021 and Costa teach the relevance of the one or more scientific articles is determined by: deriving, by the computer system, one or more syntaxes from the one or more scientific articles; determining, by the computer system, an overlap between the one or more syntaxes derived from the one or more scientific articles and the generated one or more syntaxes descriptive of the medical image; and determining, by the computer system, relevance of the one or more scientific articles based at least in part on the determined overlap as claimed.
Regarding claim 10 as dependent from claim 1, Min does not teach wherein the one or more syntaxes descriptive of the medical image are generated based at least in part on a database comprising a plurality of predetermined syntaxes generated from a plurality of medical images with known image-derived variables from a plurality of other subjects as in claim 10.
However, as discussed above, Yang’2021 and Costa teach the development of MeSH classifier for characterizing the medical conditions of concern based on the considered medical images for searching MedLine, with specifically Yang’2021 teaching the embedding of MeSH keywords or syntaxes with the medical images for referring to the proper diagnosis. Therefore since Min is teaching the identification of other patients’ medical images for performing the analysis of the patient medical image, these other patients’ medical images are also taught to have MeSH keywords or syntaxes attached or embedded for recognition and comparative analysis with the patient medical image, therefore teaching wherein the one or more syntaxes descriptive of the medical image are generated based at least in part on a database comprising a plurality of predetermined syntaxes generated from a plurality of medical images with known image-derived variables from a plurality of other subjects as claimed.
Therefore it would have been obvious for a person of ordinary skill in the art before the effective filling date of the invention to have modified the method of Min as modified by Min, Min’2021, Yang, Mridha, such that the method further comprises: wherein the one or more syntaxes descriptive of the medical image are generated based at least in part on a database comprising a plurality of predetermined syntaxes generated from a plurality of medical images with known image-derived variables from a plurality of other subjects, since one of ordinary skill in the art would recognize that embedding a MeSH keywords/syntaxe with any medical image for diagnosis was known in the art as taught by Yang’2021 and since extracting image features related to the different patient medical conditions including the corresponding MESH for characterizing these medical image providing disease labels attached to the medical image was known in the art as taught by Yang’2021 and using other patient medical images for machine learning training and identification for diagnosis was also known in the art as taught by Min, searching for prior medical reports for retrieving medical information from MEDLINE or other database such as image medical records/texts was also known in the art as taught by Costa and Yang. One of ordinary skill in the art would have expected that this modification could have been made with predictable results since Min and Yang’2021 teach analyzing image data for characterizing the medical status of the patient and generating a patient medical report including the description of the disease and additional related comparative information. The motivation would have been to automate and improve the generation of the medical image report to generate more accurate medical report, as suggested by Yang (abstract and p.2 col.1 1st-4th ¶).
Claims 6, 8 are rejected under 35 U.S.C. 103 as being unpatentable over Min (USPN 20200237329 A1; Pub.Date 07/30/2020; Fil.Date 01/23/2020) in view of Min et al. (USPN 20210319558 A1; Pub.Date 10/14/2021; Fil.Date 01/17/2021) thereafter Min’2021 in view of Yang et al. (2021 arXiv:2106-06471v1 10 pages; Pub.Date 05/21/2021) in view of Mridha et al. (2021 IEEEAccess 9:156043-156070; Pub.Date 11/22/2021) in view of Yang et al. (2023 IEEE Trans. on Multimedia 25:167-178; First Pub.Date 11-08-2021) hereafter Yang’2021 in view of Costa et al. (2021 MedRxiv 6 pages; Pub.Date 2021) as applied to claim 5 with evidential reference Fontelo et al. (2005 BMC Medical Informatics and Decision Making 5:article 5 6 pages; Pub.Date 2005).
Min, Min’2021, Yang, Mridha, Yang’2021 and Costa teach a method as set forth above.
Regarding claim 6, as discussed above, Yang teaches the use of machine learning/neural network for a word/sentence decoder for natural language (p.5 col. 2 1st ¶) with Yang’2021 teaching also the medical image report generation method using automatic generation of diagnosis description with language natural sentences based on the given medical images (abstract and p.167 col.2 1st ¶). Additionally, Costa teaches the development of MeSH classifier for characterizing the medical conditions of concern based on the considered medical images for searching MedLine and considered scientific/medical articles as being relevant to the medical conditions. MedLine is additionally known in the art to be search using natural language processing according to Fontelo (Title, abstract) for search engine askMedLine based on MeSH keywords and MeSH descriptors (p.1-3 ¶ Background and Implementation) using text-only format for natural language, therefore providing for the scientific/medical article natural language description of the article relevance. Therefore Min, Yang, Yang’2021 and Costa considering the evidential reference Fontelo teach wherein the one or more syntaxes from the one or more scientific articles is derived using natural language processing (NLP) in order to assess the relevance of the article to the patient medical image via the MeSH keywords and sentences as descriptors for the scientific/medical articles.
Therefore it would have been obvious for a person of ordinary skill in the art before the effective filling date of the invention to have modified the method of Min as modified by Min’2021, Yang, Mridha, Yang’2021 and Costa, such that the method further comprises: wherein the one or more syntaxes from the one or more scientific articles is derived using natural language processing (NLP), since one of ordinary skill in the art would recognize that searching and assessing the relevance of scientific/medical articles from Medline was known in the art to use plain text-format access via natural language processing as taught by Yang, Yang’2021 and Costa considering the evidential reference Fontelo. One of ordinary skill in the art would have expected that this modification could have been made with predictable results since Min and Yang’2021 teach analyzing image data for characterizing the medical status of the patient and generating a patient medical report including the description of the disease and additional related comparative information. The motivation would have been to automate and improve the generation of the medical image report to generate more accurate medical report, as suggested by Yang (abstract and p.2 col.1 1st-4th ¶).
Regarding claim 8, Min teaches generating a patient report comprising a recommended treatment for a patient based on the characterization of the coronary plaque ([0011]). Since, as discussed above, Min, Min’2021, Yang, Mridha, Yang’2021 and Costa with evidential reference Fontelo teach the use of NLP for processing the search and providing relevant description of medical/scientific articles to the patient medical image diagnosis for the appropriate treatment as taught by Min, it would have been therefore obvious to generate the medical image report using NLP for recommending a treatment according to the recommendations already taught by Min, therefore teaching extracting, by the computer system, from the one or more scientific articles one or more recommended treatments for the medical condition of the subject using NLP as claimed.
Claims 15-18 are rejected under 35 U.S.C. 103 as being unpatentable over Min (USPN 20200237329 A1; Pub.Date 07/30/2020; Fil.Date 01/23/2020) in view of Min et al. (USPN 20210319558 A1; Pub.Date 10/14/2021; Fil.Date 01/17/2021) thereafter Min’2021 in view of Yang et al. (2021 arXiv:2106-06471v1 10 pages; Pub.Date 05/21/2021) in view of Yang et al. (2023 IEEE Trans. on Multimedia 25:167-178; First Pub.Date 11-08-2021) hereafter Yang’2021 in view of Costa et al. (2021 MedRxiv 6 pages; Pub.Date 2021) in view of Mridha et al. (2021 IEEEAccess 9:156043-156070; Pub.Date 11/22/2021).
Regarding independent claim 15, Min teaches systems and method for characterizing plaques and characterizing the efficacy of the treatment based on the classified risk presented by the plaques based on matching the radiodensity of the plaques within a 3D representation of the coronary arteries (Title, abstract and [0018], [0076]) and providing patients reports including patient medical images and comparison of data/information with other patients’ data representation ([0069]) via computer/processor executing the methods ([0010]) therefore teaching a computer-implemented method of automatically searching and curating data related to a medical condition of a subject based at least in part on one or more variables derived from image-based analysis of the subject, the method comprising:
accessing, by a computer system, a medical image of a subject, the medical image comprising a representation of a portion of one or more coronary arteries ([0068]-[0069] plurality of images of a patient's coronary arteries and patient information/characteristics may be provided and may include 2D or 3D image data);;
analyzing, by the computer system, the medical image to identify one or more coronary arteries, the one or more coronary arteries comprising one or more regions of plaque ([0077] identify coronary arteries via segmenting and smoothing image data then providing a 2D or 3D representation of the coronary arteries [0103]);;
analyzing, by the computer system [… using at least one machine learning model…], the identified one or more coronary arteries and the one or more regions of plaque to generate a plurality of primary image-derived variables, the plurality of primary image-derived variables comprising: severity of stenosis, total plaque volume, low-density non-calcified plaque volume, non-calcified plaque volume, and calcified plaque volume ([0010], [0014], [0019], [0043], [0101] volumetric analysis of the plaque for characterization with Figs. 6-7 and [0104]-[0110] with characterization of the different types of plaques according to their radiodensity/density with the determination of each plaque boundary or contour according to the Figs. 6-7, their density according their radiodensity directly related to their physical density [0110], their volume according to Fig. 15, including plaque volume [0101], [0109] with the different types of plaques dependent on their attenuations corresponding to the low density non-calcified, non-calcified and calcified plaque [0104]-[0110], Figs. 6-7)
analyzing, by the computer system, [… using the at least one machine learning model…], the identified one or more coronary arteries and the one or more regions of plaque to generate a plurality of secondary image-derived variables, (0010], [0014], [0019], [0043], [0101] volumetric analysis of the blood vessels and plaque including baseline anatomy for vessel and segments characterization including length ([0101], [0154], [0158], [0162]) or perivascular fat characterization via CT imaging ([0010]) reading on attenuation)
[…generating, by the computer system, one or more syntaxes descriptive of the medical image based at least in part on the plurality of image-derived variables;
automatically searching, by the computer system, a database of medical literature for data related to a medical condition of the subject based at least in part on the generated one or more syntaxes,
wherein the database comprises one or more scientific articles and one or more syntaxes generated from the one or more scientific articles, wherein the searching of the database of medical literature related to the medical condition of the subject is based at least in part on determining an overlap in the one or more syntaxes generated from the medical image and the one or more syntaxes generated from the one or more scientific articles; causing, by the computer system, execution of the one or more search queries against the plurality of external medical literature databases to identify one or more relevant scientific articles wherein the relevant scientific articles are identified based at least in part on syntaxes derived from medical images included in the scientific articles, wherein the syntaxes are based at least in part primary image-derived variables and secondary image-derived variables;
analyzing, by the computer system using a natural language processing (NLP) algorithm, the identified relevant scientific articles to generate a summary of the data related to the medical condition of the subject;…];
and causing, by the computer system, generation of a display of the data related to the medical condition of the subject, ([0069] using the processor system to generate patient reports including graphical depictions of the patient coronary arteries including plaque containing arteries ([0075] generating 3D depiction of coronary arteries and plaque contained within the coronary arteries, [0076] as mapping pixels of the 2D image to a 3D mapping image of the coronary arteries with the radiodensity to generate a graphical depiction of the plaques within the coronary arteries, [0105] and Fig. 5A step 515 to generate a 3D representation of coronary arteries with plaques leading to Figs. 6 and 7 as representation of the image data as graphical representation of where the plaques are localized) wherein the computer system comprises a computer processor and an electronic storage medium ([0082]-[0083] processor 404 and computer readable storage media or main memory 406 for storing program instructions to be executed by the processor 404).
Min does not specifically teach using at least one machine learning model for generating the primary and secondary image-derived variables, generating, by the computer system, one or more syntaxes descriptive of the medical image based at least in part on the plurality of primary image-derived variables and the plurality of secondary image derived variables; generating, by the computer system, one or more search queries for searching a plurality of external medical literature databases for data related to a medical condition of the subject based at least in part on the generated one or more syntaxes, wherein the database comprises one or more scientific articles and one or more syntaxes generated from the one or more scientific articles, wherein the searching of the database of medical literature related to the medical condition of the subject is based at least in part on determining an overlap in the one or more syntaxes generated from the medical image and the one or more syntaxes generated from the one or more scientific articles; causing, by the computer system, execution of the one or more search queries against the plurality of external medical literature databases to identify one or more relevant scientific articles wherein the relevant scientific articles are identified based at least in part on syntaxes derived from medical images included in the scientific articles, wherein the syntaxes are based at least in part primary image-derived variables and secondary image-derived variables; analyzing, by the computer system using a natural la