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 .
Election/Restrictions
Applicant’s election without traverse of Group III (claims 8-10) in the reply filed on 04/09/2026 is acknowledged.
Priority
Applicant’s claim for the benefit of a prior-filed application under 35 U.S.C. 119(e) or under 35 U.S.C. 120, 121, 365(c), or 386(c) is acknowledged. Applicant has not complied with one or more conditions for receiving the benefit of an earlier filing date under 35 U.S.C. 119(e) or under 35 U.S.C. 120, 121, 365(c), or 386(c) as follows:
The later-filed application must be an application for a patent for an invention which is also disclosed in the prior application (the parent or original nonprovisional application or provisional application). The disclosure of the invention in the parent application and in the later-filed application must be sufficient to comply with the requirements of 35 U.S.C. 112(a) or the first paragraph of pre-AIA 35 U.S.C. 112, except for the best mode requirement. See Transco Products, Inc. v. Performance Contracting, Inc., 38 F.3d 551, 32 USPQ2d 1077 (Fed. Cir. 1994).
The disclosure of the prior-filed application, Application No. 62/789,660, fails to provide adequate support or enablement in the manner provided by 35 U.S.C. 112(a) or pre-AIA 35 U.S.C. 112, first paragraph for one or more claims of this application. The provisional application does not support the following limitations of claim 8: “making an automatic adjustment to the segmentation; presenting the adjusted segmentation to the machine learning system; receiving, from the machine learning system, at least one second quality score for the adjusted segmentation; and outputting an indication when the at least one second quality score indicates an improved quality with respect to the at least one first quality score”.
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 8-10 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. The specification does not demonstrate that Applicant had possession of the full scope of the broadly claimed genus for each of, “quality” or “quality score”, “image”, “segmentation”, “machine learning system”, and “automatic adjustment to the segmentation”. Each of these is a broad genus comprising many species.
The very broad segmentation “quality” or “quality score” genus includes many different species including overlap-based metrics (such as Dice coefficient, Jaccard Index, intersection-over-union, true or false positive rates), distance-based metrics (such as Hausdorff distance), shape-based metrics (smoothness, porosity, continuity, connected component count), probabilistic metrics (such as confidence, posterior probability, uncertainty entropy), plausibility metrics (such as human-vs-machine likelihood), and categorical labels (such as accurate/inaccurate binary label). However, the specification only discloses a single species for segmentation quality being Dice coefficients. This single species is not a representative number of species spanning the full scope of the claimed genus.
The very broad “image” genus includes many different species including radiological (such as x-ray, MRI, ultrasound CT, PET, SPECT), microscopy, satellite, manufacturing inspection, remote sensing, wildlife and ecology, sports and entertainment, document, assistive driving, optical, radar, LIDAR, and spectrogram. However, the specification only discloses a single species of head MRI images, which is not a representative number of species spanning the full scope of the claimed genus.
The very broad “segmentation” genus includes many different species including manual, pixel-value thresholding, edge-detection, contouring, Active Contour models, region-based segmentation (such as region-growing or split and merge), watershed transformations, clustering-based segmentation, and graph-based segmentation. However, the specification only discloses manual or atlas-based head MRI segmentation for TTFields treatment planning, which are not a representative number of species spanning the full scope of the claimed genus.
The very broad “machine learning” genus includes many different species including support vector machines, regression, decision trees, random forests, K-nearest neighbors, K-means clustering, non-deep neural networks, and deep learning (such as deep vanilla neural networks, CNN, Transformers, GAN, Autoencoder, RNN GRU, LSTM, U-NET, ViT, and GNN, etc.). However, the specification only discloses a single species of a decision tree regressor, which is not a representative number of species spanning the full scope of the claimed genus.
The very broad genus of “automatic adjustment to the segmentation” includes many different species including CRF refinement (i.e. Conditional Random Field refinement), morphological operations, and GAN-based correction. However, the specification only discloses a single species of probability changes at inter-tissue borders, which is not a representative number of species spanning the full scope of the claimed genus.
Claims 8-10 are rejected under 35 U.S.C. 112(a) or 35 U.S.C. 112 (pre-AIA ), first paragraph, because the specification, while being enabling for a method of estimating the quality of a decision-tree-regressor based segmentation of head MR images for TTFields treatment planning using the DICE coefficient as the quality score and probability changes at the tissue borders as the automatic adjustment, does not reasonably provide enablement for the full scope of the claim including each genus mentioned above, including any quality measurement, any image modality, any segmentation type, any machine learning architecture, and any automatic adjustment methodology. The specification does not enable any person skilled in the art to which it pertains, or with which it is most nearly connected, to make and use the invention commensurate in scope with these claims. Wands factors for undue-experimentation consideration:
“Quality”
Breadth of the claims: The very broad segmentation “quality” or “quality score” genus includes many different species including overlap-based metrics (such as Dice coefficient, Jaccard Index, intersection-over-union, true or false positive rates), distance-based metrics (such as Hausdorff distance), shape-based metrics (smoothness, porosity, continuity, connected component count), probabilistic metrics (such as confidence, posterior probability, uncertainty entropy), plausibility metrics (such as human-vs-machine likelihood), and categorical labels (such as accurate/inaccurate binary label).
Nature of the invention: Machine learning segmentation quality estimation.
State of the prior art: At the effective filing date, there is not universal feature set that predicts the Level of ordinary skill: High. PHOSITA is a medical image analysis researcher with graduate training. However, designing a new quality predictor is a research-level undertaking, not a routine substitution.
Predictability in the art: Low predictability since different quality metrics rank segmentations differently.
Amount of direction or guidance: only provided for Dice coefficient score.
Presence of working examples: only provided for Dice coefficient score.
various types of quality listed above. High variability in design choices relative to the various types of quality.
Quantity of experimentation necessary: Substantial since PHOSITA seeking to predict any quality metric other than DICE would need to identify corresponding suitable predictive features, retrain the model, and revalidate for each of the different quality metrics.
Conclusion: The disclosure enables Dice-coefficient prediction, but the full scope of the broad genus of “Quality” is not enabled.
“Image”
Breadth of the claims: Very broad: See all of the various species listed above.
Nature of the invention: Same as above.
State of the prior art: At the effective filing date, cross image modality transfer of segmentations was a known challenge requiring substantial reengineering. Multi-modality quality predictors were not routine.
Level of ordinary skill: High, but PHOSITA’s domain expertise is usually modality specific. Switching between image modalities depends on machine learning expertise and knowledge of the corresponding imaging physics.
Predictability in the art: Low. Extracted image features will behave differently across image modalities. Images have different characteristics depending on the modality.
Amount of direction or guidance: Only for head MRI for TTFields treatment planning.
Presence of working examples: Only for head MRI for TTFields treatment planning.
Quantity of experimentation necessary: Substantial. Each different image modality, listed above, requires reengineering the feature set accounting for different image characteristics and retraining and revalidating the machine learning system.
Conclusion: The disclosure enables head MRI for TTFields treatment planning, but other image modalities within the full scope of the broad genus of “image” are not enabled.
“Segmentation”
Breadth of the claims: “Segmentation” is a very broad genus that includes the various species listed above.
Nature of the invention: Quality estimation for segmentation where features suitable for one segmentation type may not work well for other segmentation types.
State of the prior art: Quality estimation across various segmentation types was not standardized.
Level of ordinary skill: High, but image segmentation is tailored to the specific task one would not have the knowledge to implement all segmentation types for a given task
Predictability in the art: Different segmentation routines will behave differently for different tasks and trial-and-error would be necessary to determine which options would be successful.
Amount of direction or guidance: only for manual and atlas-based head segmentation.
Presence of working examples: only for manual and atlas-based head segmentation.
Quantity of experimentation necessary: Substantial for various different segmentation types. Different segmentation routines will behave differently for different tasks and trial-and-error would be necessary to determine which options would be successful.
Conclusion: The disclosure enables manual and atlas-based head segmentation. The full scope of the broad genus of “segmentation” is not enabled.
“Machine Learning”
Breadth of the claims: Very broad: See above for a list of the various species that fall within the broad scope of the genus “machine learning”.
Nature of the invention: See above.
State of the prior art: Applying all of the machine learning types to solve the problem of this invention would be unconventional.
Level of ordinary skill: High, but adapting the disclosed problem to the various machine learning species such as the various deep learning architectures requires careful design and effort.
Predictability in the art: It is unpredictable how the various machine learning species would fare at solving the disclosed problem since each as pros and cons.
Amount of direction or guidance: only for decision tree.
Presence of working examples: only for decision tree.
Quantity of experimentation necessary: Substantial for various different species that fall within “machine learning”. It is unpredictable how the various machine learning species would fare at solving the disclosed problem since each as pros and cons.
Conclusion: The disclosure enables decision tree based segmentation quality determination, but the full scope of the broad genus “machine learning” is not enabled.
“Automatic Adjustment”
Breadth of the claims: Very broad. See the list of the various species that fall within this genus (listed above).
Nature of the invention: Different segmentation adjustment methods will have different outcomes.
State of the prior art: At the effective filing date, the many adjustment methods were not standardized as universally applicable and interchangeable.
Level of ordinary skill: High, but implementing and adapting various different segmentation adjustment methods is non-trivial.
Predictability in the art: Different segmentation adjustment methods would require empirical testing for implementation on a machine learning routine.
Amount of direction or guidance: only for probability changes at inter-tissue boundaries.
Presence of working examples: only for probability changes at inter-tissue boundaries.
Quantity of experimentation necessary: Substantial. Each different segmentation adjustment method has its own implementation, parameters, and convergence properties.
Conclusion: The disclosure enables probability changes at inter-tissue boundaries, but does not enable the full scope of the broad genus of automatic segmentation adjustment.
Double Patenting
The nonstatutory double patenting rejection is based on a judicially created doctrine grounded in public policy (a policy reflected in the statute) so as to prevent the unjustified or improper timewise extension of the “right to exclude” granted by a patent and to prevent possible harassment by multiple assignees. A nonstatutory double patenting rejection is appropriate where the conflicting claims are not identical, but at least one examined application claim is not patentably distinct from the reference claim(s) because the examined application claim is either anticipated by, or would have been obvious over, the reference claim(s). See, e.g., In re Berg, 140 F.3d 1428, 46 USPQ2d 1226 (Fed. Cir. 1998); In re Goodman, 11 F.3d 1046, 29 USPQ2d 2010 (Fed. Cir. 1993); In re Longi, 759 F.2d 887, 225 USPQ 645 (Fed. Cir. 1985); In re Van Ornum, 686 F.2d 937, 214 USPQ 761 (CCPA 1982); In re Vogel, 422 F.2d 438, 164 USPQ 619 (CCPA 1970); In re Thorington, 418 F.2d 528, 163 USPQ 644 (CCPA 1969).
A timely filed terminal disclaimer in compliance with 37 CFR 1.321(c) or 1.321(d) may be used to overcome an actual or provisional rejection based on nonstatutory double patenting provided the reference application or patent either is shown to be commonly owned with the examined application, or claims an invention made as a result of activities undertaken within the scope of a joint research agreement. See MPEP § 717.02 for applications subject to examination under the first inventor to file provisions of the AIA as explained in MPEP § 2159. See MPEP § 2146 et seq. for applications not subject to examination under the first inventor to file provisions of the AIA . A terminal disclaimer must be signed in compliance with 37 CFR 1.321(b).
The filing of a terminal disclaimer by itself is not a complete reply to a nonstatutory double patenting (NSDP) rejection. A complete reply requires that the terminal disclaimer be accompanied by a reply requesting reconsideration of the prior Office action. Even where the NSDP rejection is provisional the reply must be complete. See MPEP § 804, subsection I.B.1. For a reply to a non-final Office action, see 37 CFR 1.111(a). For a reply to final Office action, see 37 CFR 1.113(c). A request for reconsideration while not provided for in 37 CFR 1.113(c) may be filed after final for consideration. See MPEP §§ 706.07(e) and 714.13.
The USPTO Internet website contains terminal disclaimer forms which may be used. Please visit www.uspto.gov/patent/patents-forms. The actual filing date of the application in which the form is filed determines what form (e.g., PTO/SB/25, PTO/SB/26, PTO/AIA /25, or PTO/AIA /26) should be used. A web-based eTerminal Disclaimer may be filled out completely online using web-screens. An eTerminal Disclaimer that meets all requirements is auto-processed and approved immediately upon submission. For more information about eTerminal Disclaimers, refer to www.uspto.gov/patents/apply/applying-online/eterminal-disclaimer.
Claim 8 is rejected on the ground of nonstatutory double patenting as being unpatentable over claim 6 of U.S. Patent No. 11276171. Although the claims at issue are not identical, they are not patentably distinct from each other because the limitations of the Application claim are all present in the listed Patent claim.
Claim Rejections - 35 USC § 102
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 the appropriate paragraphs of 35 U.S.C. 102 that form the basis for the rejections under this section made in this Office action:
A person shall be entitled to a patent unless –
(a)(2) the claimed invention was described in a patent issued under section 151, or in an application for patent published or deemed published under section 122(b), in which the patent or application, as the case may be, names another inventor and was effectively filed before the effective filing date of the claimed invention.
Claim(s) 8 and 9 are rejected under 35 U.S.C. 102(a)(2) as being anticipated by US 20210220670 A1 (Li).
As per claim 8, Li teaches a method of estimating a quality of a segmentation of an image, the method comprising (Li:
Para 38: “The systems and methods described in the present disclosure can also automatically and rapidly identify good-quality contours generated by auto-segmentation using image texture analysis, and can automatically correct contours that have been identified as being inaccurate.”;
Para 40: “the systems and methods described in the present disclosure provide for contour correction, contour verification, or both. As mentioned above, and described in detail below, contours can be automatically identified as being good quality, and those contours identified as being inaccurate can be automatically corrected.”;
Para 88: “The systems and methods described in the present disclosure provide a fast and automated patient-specific contour validation approach using quantitative image features (e.g., image texture features) of medical image.”;
Para 94: “It is also contemplated that the differences of these image textures between the structure and the surrounding tissues can be used to identify if the contour is valid or not.”;
Fig. 7: 704: “Contour Accuracy Acceptable?”;
Fig. 5: 524: “Contour Re-evaluation”;
any way of grading segmentation quality):
presenting a new image and a segmentation of the new image to a machine learning system (Li:
Fig. 7 (shown below): “Testing Data” (which is comprised of image and contour data) input into 704 (which is machine learning contour evaluation);
“[0098] Contour data are accessed with a computer system, as indicated at step 706. The contour data generally include images of the subject and corresponding accurate contours and inaccurate contours. For instance, the contour data may include auto-generated organ contours as well as the manual generated ground truth contours. As an example, accurate contours can be identified based on multiple criterion, such as having Dice Similarity Coefficient 0.85, mean distance to agreement 1.5 mm, and 95 percent of distance to agreement 5 mm.”
Para 99: “accessing the contour data may include acquired images with a medical imaging system and communicating or otherwise transferring the images to the computer system. Additionally or alternatively, accessing the contour data may include generating contours from medical images and communicating or otherwise transferring those contours to the computer system.”;
“[0100] The contour data are then separated into training data and testing data, as indicated at step 708. The training data will be used to train an AI-based model for evaluating the testing data. As an example, the contours in the contour data can be randomly separated such that a first percentage of the contour data are assigned as training data and a second percentage of the contour data are assigned as testing data. For instance, two-thirds of the contour data may be used as training data and one-third of the contour data may be used as testing data.”;
Para 95 (shown below): “An artificial intelligence (“AI”)-based model can be constructed and trained to access the texture attributes for the core regions, inner shell regions, and outer shell regions. As an example, the AI-based model may be a machine learning (“ML”)-based model that is constructed and trained based on supervised learning. As another example, the AI-based model can be a decision tree model. For instance, the ML-based model can implement a recursive random forest classification algorithm to rank image features and to select top-ranked features. Supervised classification models can also be implemented for contour validation.”;
PNG
media_image1.png
1557
994
media_image1.png
Greyscale
“[0097] Referring now to FIG. 7, a flowchart is illustrated as setting forth the steps of an example method for automated contour validation. The method includes a model training stage 702, which may be an off-line model training stage, and a contour evaluation stage 704.”
PNG
media_image2.png
806
1361
media_image2.png
Greyscale
provide an image and its contours into a machine learning (ML) system),
wherein the machine learning system has been trained to estimate a quality of a segmentation of an image based on a plurality of reference images and at least one quality score that has been assigned to each of the reference images (Li:
Fig. 7 (shown above): mainly 702, 708-718;
paras 98-100 (provided above);
para 98 (provided above): “The contour data generally include images of the subject and corresponding accurate contours and inaccurate contours. For instance, the contour data may include auto-generated organ contours as well as the manual generated ground truth contours. As an example, accurate contours can be identified based on multiple criterion, such as having Dice Similarity Coefficient 0.85, mean distance to agreement 1.5 mm, and 95 percent of distance to agreement 5 mm.”;
Para 110: “An AI-based model is then constructed and used to evaluate the image feature attributes for the core, inner shell, and outer shell regions, as indicated at step 716. As noted above, in some instances the AI-based model may be a ML-based model that implements supervised learning. As one non-limiting example, the AI-based model is a decision tree model, which can be trained to access the texture attributes for core regions, inner shell regions, and outer shell regions. An example of a decision tree model is shown in FIG. 8. Decision tree models provide an effective method for data mining. In general, a decision tree model partitions data into subsets by posing a series of questions about the attributes associated with the data. Each node on the tree contains one question and has one “yes” and one “no” child node. In one example, shown in FIG. 8, a three-level decision tree is established to evaluate the texture feature attributes for core, inner shell, and outer shell regions. Contours that meet all of the attribute questions (i.e., “yes” nodes) from the three levels are identified as accurate; otherwise, the contour will be determined as inaccurate and reported for further verification.”;
the ML system was trained on two or more reference images that each had a quality score assigned. The quality score of Li is binary choice of accurate or inaccurate);
receiving, from the machine learning system, at least one first quality score for the segmentation of the new image (Li:
Fig. 7 (shown above): “Contour Accuracy Acceptable? Y N”;
Fig. 8 (shown below): “inaccurate contour” or “accurate contour”;
Para 110: “Contours that meet all of the attribute questions (i.e., “yes” nodes) from the three levels are identified as accurate; otherwise, the contour will be determined as inaccurate and reported for further verification.”;
PNG
media_image3.png
712
892
media_image3.png
Greyscale
the ML system returns one or more initial quality score for the new image segmentation);
outputting the at least one first quality score for the segmentation of the new image (Li:
Fig. 7 (shown above): 704: mainly: decision “Contour Accuracy Acceptable?” determination is provided to either “Contour Auto-Refinement”, “Accurate Contour”-“Treatment Plan Generation”, or “Contour Manual Editing”;
“[0126] In the end, the model features are ready to be used for automatic contour quality assurance. At the contour evaluation stage 704, the identified clinically acceptable good contours can be used immediately for further re-planning and treatment, while reported inaccurate contours can be sent to physicians for manual modification, or corrected using the contour correction methods described in the present disclosure.”;
send the initial quality scope somewhere);
making an automatic adjustment to the segmentation (Li:
Fig. 7 (shown above): mainly 704: “Contour Auto-Refinment”;
Fig. 5 (shown below): mainly 516-522: “Automatic Contour Correction”;
Para 67: “The systems and methods described in the present disclosure provide for automatically correcting inaccurate contours generated from auto-segmentation using image texture information.”;
Para 68: “The inaccurate contours are re-processed using a texture-based automatic contour correction method. The procedure generally includes two steps: calculating voxel-based texture feature map (e.g., a GLCM-cluster shade texture feature map using 3×3 block size inside a region-of-interest (“ROI”) created from the initial contour), and incorporating the texture feature map into an active contour algorithm as an external force to find correct structure boundaries.”;
“[0073] The workflow of an example texture-based automatic contour correction method is shown in FIG. 5.”;
“[0075] The automatic contour correction method 516 starts with the GLCM texture feature map calculation, which may be calculated for an ROI created from initial contour at step 518.”;
“[0077] The feature map is then integrated into an active contour algorithm as external force to find correct structure boundaries as indicated at step 520.”;
“[0087] The image-texture based automatic contour correction approach described above can be used to improve the overall contour accuracy and the efficiency of contour modification.”;
“[0085] After active contouring, morphological operations are performed to refine the contours at step 522.”;
Para 86: “A representative example is shown in FIGS. 6A-6C to illustrate the texture based contour correction method described above…
The evolving contours are illustrated with cyan lines, while ground truth contour, initial contour, and corrected contour are highlighted with different colors (i.e., green, blue, and red) in FIG. 6C. The contour successfully converged to the ground truth after correction.”;
“[0087] The image-texture based automatic contour correction approach described above can be used to improve the overall contour accuracy and the efficiency of contour modification.”;
“[0126] In the end, the model features are ready to be used for automatic contour quality assurance. At the contour evaluation stage 704, the identified clinically acceptable good contours can be used immediately for further re-planning and treatment, while reported inaccurate contours can be sent to physicians for manual modification, or corrected using the contour correction methods described in the present disclosure.”;
PNG
media_image4.png
1300
985
media_image4.png
Greyscale
PNG
media_image5.png
1353
895
media_image5.png
Greyscale
automatically adjust the segmentation in any way);
presenting the adjusted segmentation to the machine learning system (Li:
Fig. 7 (shown above): mainly 704: after the “Contour Auto-Refinement”, the flow returns to “Contour Accuracy Acceptable?”;
Fig. 8 (shown above); Para 110 (shown above);
provide the automatically adjusted contour to the ML system);
receiving, from the machine learning system, at least one second quality score for the adjusted segmentation (Li:
Fig. 7 (shown above): mainly 704: second “Contour Accuracy Acceptable? Y N”;
Fig. 8 (shown above); Para 110 (shown above);
Para 86 (referenced above): “The contour successfully converged to the ground truth after correction”;
the ML system returns a new score for the updated contour); and
outputting an indication when the at least one second quality score indicates an improved quality with respect to the at least one first quality score (Li:
Fig. 7 (shown above): mainly 704: when “Y” for “Contour Accuracy Acceptable”, the flow goes to “Accurate Contour”. The “N” branch leads to Contour Manual Editing;
Figs. 6A-6C (shown above): paras 85-86 (both referenced above);
Also, see citations and arguments provided for the preceding clauses.
indicate when the new score is better than the initial score. The “Accurate Contour”, after “Contour Auto-Refinement” due to unacceptable first contour is an indication that the auto-refined contour is improved relative to the initial contour.).
As per claim 9, Li teaches the method of claim 8, further comprising training the machine learning system to estimate the quality of a segmentation of an image based on the plurality of reference images and at least one quality score that has been assigned to each of the reference images, wherein the training occurs prior to the presenting of the new image (Li: See arguments and citations offered in rejecting claim 8 above;
Fig. 7 separates model training (702) using separated training data and from a contour evaluation stage (704);
Paras 97, 100).
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.
Claim(s) 10 is rejected under 35 U.S.C. 103 as being unpatentable over Li as applied to claim 8 above, and further in view of HUANG et al., "Fully Automated Whole-Head Segmentation with Improved Smoothness and Continuity, with Theory Reviewed," PlosOne, Vol. 10, No. 5, e0125477, May 2015 (Huang).
As per claim 10, Li teaches the method of claim 8, wherein the machine learning system has been trained to estimate the quality of segmentation of the new image based on image quality and tissue's shape properties (Li: See arguments and citations offered in rejecting claim 8 above).
Li does not teach for extra-cranial tissues
Huang teaches segmentation of the new image based on image quality and tissue's shape properties for extra-cranial tissues (Huang:
Page 5, para 2:
PNG
media_image6.png
939
869
media_image6.png
Greyscale
Beginning in last para on page 5:
PNG
media_image7.png
695
1102
media_image7.png
Greyscale
Page 9, para 1:
PNG
media_image8.png
672
1151
media_image8.png
Greyscale
Page 10, last para
PNG
media_image9.png
531
1723
media_image9.png
Greyscale
PNG
media_image10.png
886
1207
media_image10.png
Greyscale
PNG
media_image11.png
1307
1110
media_image11.png
Greyscale
page 6, para 2:
PNG
media_image12.png
612
1153
media_image12.png
Greyscale
The smoothness of the segmentation, artifacts in the air, scalp shape correction, and bias-field correction are all image quality parameters).
Thus, it would have been obvious for one of ordinary skill in the art, prior to filing, to implement the teachings of Huang into Li since both Li and Huang suggest a practical solution and field of endeavor of a automated segmentation of medical images of the head obtaining accurate, robust segmentations in general and Huang additionally provides teachings that can be incorporated into Li in that segmentation of the new image based on image quality and tissue's shape properties for extra-cranial tissues since “It is becoming increasingly clear that individual anatomy including complete cerebrospinal fluid (CSF), skull and scalp are key to obtain meaningful source localization in EEG and targeting of TES for individual subjects” (Huang: Introduction). The teachings of Huang can be incorporated into Li in that segmentation of the new image based on image quality and tissue's shape properties for extra-cranial tissues. Furthermore, one of ordinary skill in the art could have combined the elements as claimed by known methods and, in combination, each component functions the same as it does separately. One of ordinary skill in the art would have recognized that the results of the combination would be predictable.
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure. US 20220383504 A1 (Nordstrom) also teaches all limitations of claims 8 and 9.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to Atiba Fitzpatrick whose telephone number is (571) 270-5255. The examiner can normally be reached on M-F 10:00am-6pm.
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, Andrew Bee can be reached on (571) 270-5183. The fax phone number for Atiba Fitzpatrick is (571) 270-6255.
Information regarding the status of an application may be obtained from the Patent Application Information Retrieval (PAIR) system. Status information for published applications may be obtained from either Private PAIR or Public PAIR. Status information for unpublished applications is available through Private PAIR only. For more information about the PAIR system, see http://pair-direct.uspto.gov. Should you have questions on access to the Private PAIR system, contact the Electronic Business Center (EBC) at 866-217-9197 (toll-free). If you would like assistance from a USPTO Customer Service Representative or access to the automated information system, call 800-786-9199 (IN USA OR CANADA) or 571-272-1000.
Atiba Fitzpatrick
/ATIBA O FITZPATRICK/
Primary Examiner, Art Unit 2677
Prosecution Insights