DETAILED ACTION
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 .
Notice to Applicant
Receipt of Applicant’s Amendment filed December 22, 2025 is acknowledged.
Response to Amendment
Claims 1 and 10 have been amended. Claims 2-9 and 11-15 have not been modified. Claim 16 has been added. Claims 1-16 are pending and are provided to be examined upon their merits.
Response to Arguments
Applicant’s arguments filed December 22, 2025 have been fully considered but they are not persuasive. A response is provided below.
Applicant argues 35 U.S.C. §101 Rejections, pg. 6 of Remarks:
Examiner acknowledges Applicant amendment and withdraws the §101 rejection.
Applicant argues 35 U.S.C. §102 & 103 Rejections, pg. 6 of Remarks:
Applicant argues that the confidence score of Robinson is not associated with the probability of success of a treatment, as recited in claim 1.
Examiner respectfully disagrees. Robinson teaches wherein confidence score is related to the success of a treatment ([0111], “the machine learning environment can be used to predict success of an ablation procedure using past information such as patient success related to their confidence score.”). As the confidence score is representative of success rate ([0008], “an expected success rate with non-invasive therapy, or an expected success rate with alternative treatment modalities.”), Examiner maintains that the confidence score teaches the claim limitation.
Applicant further argues that Robinson teaches the opposite of determining the confidence score based on machine learning.
[0124] of Robinson recites: “Machine learning tools and predictive analytics can be integrated within method 200 to create a clinical decision support infrastructure such as the decision support module.”
[0117] of Robinson: “Further provided herein is a decision support module to provide an informed ablation plan to the physician and provide metrics for support about success and risks of various treatment option and opportunities to improve patient outcomes. After a target has been identified, the identified target(s) may be presented to a physician in the form of a decision support module. In addition, the decision support module may include one or more of a description of the scar pattern, a description of the scar burden size and location, a suggested volume to achieve ablation (e.g. full-thickness ablation, partial-thickness ablation, etc.), a confidence score of the combined mappings, a listing of at-risk structures, general recommendations, expected success with SBRT, expected success with alternative treatment modalities (e.g., catheter RF, antiarrhythmic drug (e.g. amiodarone), etc.), or combinations thereof.”
As recited, the confidence score is provided by the decision support module, which is created through machine learning tools and predictive analytics. Thus, the confidence score is an output of machine learning.
The input that Applicant refers to may be the data input for training the machine learning model for the subsequent processing of the confidence score and success metrics ([0111], “The neural network can continue to receive input (e.g., training data) over a period of time until the neural network is trained. The input into the neural network can also include data related to the success of a treatment, such as a success outcome, survival, side effects, etc.”).
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)(1) the claimed invention was patented, described in a printed publication, or in public use, on sale, or otherwise available to the public before the effective filing date of the claimed invention.
Claims 1-2, 5-7, and 12-13 are rejected under 35 U.S.C. 102(a)(1) as being unpatentable over Robinson (US 20210137384).
Regarding claim 1, Robinson teaches an electronic device comprising at least one processor and a computer memory storing program code instructions which, when they are executed by said at least one processor, configure said at least one processor to implement a method for selecting at least one optimal treatment from among several available treatments for the percutaneous ablation of a lesion within an anatomy of interest of a patient ([0141], “The storage device 430 can include software services, servers, services, etc., that when the code that defines such software is executed by the processor 410, it causes the system to perform a function. In some examples, a hardware service that performs a particular function can include the software component stored in a computer-readable medium in connection with the necessary hardware components, such as processor 410, connection 405, output device 435, etc., to carry out the function.” [0069], “improved identification of a cardiac arrhythmia target volume and treatment plan for noninvasive and invasive arrhythmia ablation and improved decision support for the selection of therapy for arrhythmia ablation”), said method comprising:
obtaining of a medical image previously acquired on the patient and on which the lesion can be seen within the anatomy of interest of the patient ([0005], “The method may include receiving one or more mappings selected from an electrical mapping, an anatomic mapping, a functional mapping, and combinations thereof; identifying an abnormality in the one or more mappings;… the electrical mapping may be an electrocardiograph image, the anatomic mapping may be at least one of a computer tomography image or a magnetic resonance image, and the functional mapping may be at least one of a photo emission computed tomographic image, a positron emission tomography image or an echocardiogram image.” [0115], “In an example, over time, the method can predict consistent utilization of invasive catheter ablation for small endocardial lesions in the left ventricular apex in older males.”),
calculating using a machine learning algorithm, from the medical image, for each of the different available ablation treatments, a confidence index whose value is representative of a probability of success of said treatment for the ablation of the lesion in the anatomy of interest of the patient ([0124], “Machine learning tools and predictive analytics can be integrated within method 200 to create a clinical decision support infrastructure such as the decision support module.”” [0117], “After a target has been identified, the identified target(s) may be presented to a physician in the form of a decision support module. In addition, the decision support module may include one or more of a description of the scar pattern,… a confidence score of the combined mappings, a listing of at-risk structures, general recommendations, expected success with SBRT, expected success with alternative treatment modalities (e.g., catheter RF, antiarrhythmic drug (e.g. amiodarone), etc.), or combinations thereof.” [0111], “the machine learning environment can be used to predict success of an ablation procedure using past information such as patient success related to their confidence score.” [0008], “an expected success rate with non-invasive therapy, or an expected success rate with alternative treatment modalities.”). As recited, the confidence score is provided by the decision support module, which is created through machine learning tools and predictive analytics. Thus, the confidence score is an output of machine learning. Furthermore, the confidence score is related to the success of the treatment, as explained by [0111], and said predicted success may refer to a rate, as shown in [0008], Examiner interprets the confidence score of Robinson to encompass the confidence index of the instant application.
the machine learning algorithm having been trained beforehand from a set of training elements ([0093], “The neural network can continue to receive input (e.g., training data) over a period of time until the neural network is trained. The input into the neural network can also include data related to the success of a treatment, such as a success outcome, survival, side effects, etc.”), each training element comprising:
a medical image on which a lesion can be seen within the anatomy of interest of another patient ([0142], “ training data can be one or more images and a label (e.g., VT or no VT) can be associated with each image. In some examples, the labeled data can be actual target probabilities (e.g., each segment as the weighted average over all received images for each patent).”),
an identification of a treatment chosen selected from among the different ablation treatments available for treating said other patient ([0093], “As additional patients are treated, the previous treatments can inform future treatments and can be used to adjust the target of the current patient (for example, by training the neural network). In another example, the target may be adjusted based on a weighting of input mappings or the confidence score of the combined mappings.” [0111], “The recommendations may be provided from prior patients... In an example, over time, the method can predict consistent utilization of invasive catheter ablation for small endocardial lesions in the left ventricular apex in older males. Subsequent outcomes data reveals high rates of toxicity in this population and low rates of toxicity with similar outcomes in patients treated with noninvasive ablation. While the decision support module may initially suggest further catheter ablation based on the multiplicity of prior treatments, an updated decision support module would suggest noninvasive ablation.”), and
a confidence index of said treatment on said other patient ([0111], “In an example, the machine learning environment can be used to predict success of an ablation procedure using past information such as patient success related to their confidence score.”), and
selecting at least one optimal treatment as a function of the confidence indices calculated for the different available treatments ([0132], “In an example, over time, the method can predict consistent utilization of invasive catheter ablation for small endocardial lesions in the left ventricular apex in older males. Subsequent outcomes data reveals high rates of toxicity in this population and low rates of toxicity with similar outcomes in patients treated with noninvasive ablation. While the decision support module may initially suggest further catheter ablation based on the multiplicity of prior treatments, an updated decision support module would suggest noninvasive ablation.” [0142], “The computing device(s) 502 can continue to receive data from the one or more data sources 506 until the neural network 508 (e.g., convolution neural networks, deep convolution neural networks, artificial neural networks, learning algorithms, etc.) of the computing device(s) 502 are trained (e.g., have had sufficient unbiased data to respond to new incoming data requests and provided an autonomous or near autonomous recommended course of actions, and/or actually provide input to perform the course of action).”). Examiner interprets performing a course of action/recommendation encompasses selecting said recommendation.
wherein the method comprises transforming the medical image of the patient, the medical image of each training element having undergone a similar transformation before being used to train the machine learning algorithm ([0075], “At block 110, the method 100 can define or identify one or more abnormalities in the one or more input mappings. The abnormalities can be identified, for example, in an MRI by the scar location, an abnormality in PET/SPECT may be regions that are not viable, an abnormality in an electrical mapping may be where a VT originates, etc. In various examples, the abnormality can be defined by selecting the abnormality on each mapping, manually or automatically segmenting the mapping and manually selecting the abnormality, automatically contouring the mapping and manually selecting the abnormality, automatically contouring the mapping and automatically selecting the abnormality, or combinations thereof. In some examples, the identification of the abnormalities can be used in training a neural network, for example, by using supervised or reinforcement learning.” [0132], “the method can predict consistent utilization of invasive catheter ablation for small endocardial lesions in the left ventricular apex in older males.”),
wherein transforming the medical image comprises:
segmenting the lesion on the medical image, and/or
segmenting the anatomy of interest on the medical image ([0076], “Defining or identifying one or more abnormalities in the one or more input mappings can include segmenting one or more of the input mappings. In some examples, the one or more of the input mappings can be segmented using a segmentation model. The at least one processor can determine one or more abnormality in one or more cardiac arrhythmia target segments.”), and/or
segmenting blood vessels on the medical image, and/or
when the anatomy of interest segmenting hepatic ducts on the medical image.
Regarding claim 2, Robinson teaches the electronic device of claim 1. Robinson further teaches wherein each available treatment is an ablation technology selected from among radiofrequencies, microwaves, laser, cryotherapy, electroporation or brachytherapy ([0069], “noninvasive ablation methods can include, but are not limited to stereotactic body radiotherapy, stereotactic ablative radiotherapy, stereotactic radiosurgery, fractionated radiotherapy, hypofractionated radiotherapy, high-frequency/focused ultrasound, or lasers.”).
Regarding claim 5, Robinson teaches the electronic device of claim 1. Robinson further teaches wherein, for each training element, the confidence index is defined as a function:
of the observation or non-observation of a recurrence for said other patient with the chosen treatment, and/or
of a duration of the period elapsed between the end of the treatment and a current date without a recurrence having been observed for said other patient, and/or
of a duration of the period elapsed between the end of the treatment and a recurrence for said other patient, and/or
of an estimation, by one or more medical experts, of a probability of success of the chosen treatment on said other patient ([0069], “After treatment of the patient, method 200 may further determine the success of the treatment. The database may then be updated with the success information, including any side effects or problems encountered. The success information may adjust treatment recommendations or selection of segments for ablation for future patients.”).
Regarding claim 6, Robinson teaches the electronic device of claim 1. Robinson further teaches wherein the method comprises associating a set of parameters with the medical image of the patient, said set of parameters comprising one or more parameters relating to the lesion, to the anatomy of interest and/or to characteristics of the patient ([0005], “identifying an abnormality in the one or more mappings; combining the one or more mappings; and defining the one or more cardiac arrhythmia targets based on an overlap of the identified abnormality in the combined one or more mappings.” [0065], “expert-defined targets can be used to train a model, such as a neural network, with inputs being information from the multimodality images or the images themselves.” [0074], “ the clinical mapping can be, but are not limited to: demographics (e.g., age, gender, NYHA, CKD, lungs, PVD, Charlson vs. Seattle HF model, etc.), surgical history (e.g., cardiac surgery, etc.), knowledge about clinical VT (e.g., MMVT or multiple VTs) and/or prior electroanatomical mapping/ablations.”),
each training element comprising a set of similar parameters relating to the lesion, to the anatomy of interest and/or to characteristics of the other patient for which the medical image corresponding to said reference element was obtained ([0074], “ the clinical mapping can be, but are not limited to: demographics (e.g., age, gender, NYHA, CKD, lungs, PVD, Charlson vs. Seattle HF model, etc.), surgical history (e.g., cardiac surgery, etc.), knowledge about clinical VT (e.g., MMVT or multiple VTs) and/or prior electroanatomical mapping/ablations.”).
Regarding claim 7, Robinson teaches the electronic device of claims 1 and 6. Robinson further teaches wherein the set of parameters comprises one or more parameters selected from:
the age, the sex, the weight and/or the height of the patient for whom the medical image was obtained ([0074], “ the clinical mapping can be, but are not limited to: demographics (e.g., age, gender, NYHA, CKD, lungs, PVD, Charlson vs. Seattle HF model, etc.), surgical history (e.g., cardiac surgery, etc.), knowledge about clinical VT (e.g., MMVT or multiple VTs) and/or prior electroanatomical mapping/ablations.”),
a comorbidity presented by the patient for whom the medical image was obtained,
a value representative of the size and/or of a volume of the lesion,
a distance between the lesion and a capsule of the anatomy of interest,
a distance between the lesion and a blood vessel close to the lesion, and/or
when the anatomy of interest is the liver, a distance between the lesion and a hepatic duct close to the lesion.
Regarding claim 12, Robinson teaches the electronic device of claim 1. Robinson further teaches wherein the medical images were acquired by tomodensitometry, by magnetic resonance imaging or by ultrasound ([0064], “The imaging modality may be noninvasive. Noninvasive imaging modalities may include, but are not limited to CT, MRI, PET, SPECT, ECGI, and 12-lead EKG.”). Examiner notes tomodensitometry is a synonym for computed tomography (CT) scanning.
Regarding claim 13, Robinson teaches the electronic device of claim 1. Robinson further teaches wherein the medical images are three-dimensional ([0078], “The input mapping may be a 3D input mapping.” [0084], “an input mapping with a segmentation model and a contour of at least one 3D input mapping can be combined. For example, a segmentation model from a 12-lead ECG can be co-registered with the identified contours from a 3D geometry (e.g., ECGI, MRI, CT, PET).”).
Claim Rejections - 35 USC § 103
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
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.
Claims 3 and 4 are rejected under 35 U.S.C. 103 as being unpatentable over Robinson (US 20210137384) in view of Cornelis (Cornelis; F.H., Microwave ablation of renal tumors: A narrative review of technical considerations and clinical results, April 2017, Diagnostic and Interventional Imaging, Volume 98, Issue 4, Pages 287-297).
Regarding claim 3, Robinson teaches the electronic device of claims 1 and 2. Robinson further teaches wherein each available treatment uses a strategy of energy deposition by one or more applicators ([0057], “Based on symptoms, severity, and cause of the arrhythmia, treatment options include, but are not limited to, antiarrhythmic drugs, placement of a pacemaker/defibrillator, surgical ablation, catheter-based ablation (endocardial, epicardial) using radiofrequency energy to create thermal injury, non-invasive ablation with SBRT, and/or a combination thereof.”). Examiner interprets a catheter used in catheter-based ablation to encompass an applicator.
Robinson does not explicitly teach wherein said strategy is selected from an energy deposition of centrifugal type or an energy deposition of centripetal convergent type.
However, the combination of Robinson in view of Cornelis does teach wherein said strategy is selected from an energy deposition of centrifugal type or an energy deposition of centripetal convergent type (Cornelius, pg. 291, “Several antenna designs are available to reduce return loss and focused energy radiation [37]. Then, using a centrifugal method, energy dissipates towards the periphery from a single applicator inserted into the center of the targeted tumor [9]. In some cases, an overlapping ablation strategy may be required in order to treat the whole lesion with sufficient margins. This may be achieved by multiple reinsertion of a single device or by using several applicators. Multiple antennas can be operated simultaneously in close proximity without switching [34]. As antennas may be positioned and phased to exploit overlap of the electromagnetic field, ablation strategy involving centripetal convergence of energy from the periphery towards the center of the tumor may be performed [9], [12].” Robinson, [0142], “The computing device(s) 502 can continue to receive data from the one or more data sources 506 until the neural network 508 (e.g., convolution neural networks, deep convolution neural networks, artificial neural networks, learning algorithms, etc.) of the computing device(s) 502 are trained (e.g., have had sufficient unbiased data to respond to new incoming data requests and provided an autonomous or near autonomous recommended course of actions, and/or actually provide input to perform the course of action).”).
Robinson in view of Cornelis are considered analogous to the claimed invention because they are in the field of laser ablation. Therefore, it would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to have modified the possible ablation treatment types of Robinson to specifically include treatment types that use energy deposition of centrifugal and/or centripetal convergence types as in Cornelis for the advantage of “identify[ing] the specific technical considerations to adequately perform MWA” (microwave ablations) (Cornelis; pg. 288).
Regarding claim 4, Robinson in view of Cornelis teaches the electronic device of claims 1-3. Robinson does not teach wherein each available treatment is characterized by a number of applicators, and/or by the positions of said applicators with respect to the lesion.
However, Cornelis does teach wherein each available treatment is characterized by a number of applicators (Pg. 292, “As a routine, the MW generator is initially powered at 50–65 W for a single antenna for a prescribed time of 5–10 min [50], [53], [54], which may be secondarily adjusted. Initial settings depend on tumor location, tumor size and the expected margins. When phased constructively, heating increases proportional to the square of the number of antennas, allowing more efficient heating and generation of higher temperatures when compared to a single antenna [12]. Therefore, for a MWA procedure with several antennas, the power of each single antenna has to be decreased in order to obtain an overall power concordant with the tumor volume.”), and/or
by the positions of said applicators with respect to the lesion (Pg. 291, “As antennas may be positioned and phased to exploit overlap of the electromagnetic field, ablation strategy involving centripetal convergence of energy from the periphery towards the center of the tumor may be performed [9], [12].”).
Robinson in view of Cornelis are considered analogous to the claimed invention because they are in the field of laser ablation. Therefore, it would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to have modified Robinson with Cornelis for the advantage of “identify[ing] the specific technical considerations to adequately perform MWA” (microwave ablations) (Cornelis; pg. 288).
Claims 10 and 15-16 are rejected under 35 U.S.C. 103 as being unpatentable over Robinson (US 20210137384) in view of Chen (CN 112132808 A).
Regarding claim 10, Robinson teaches the electronic device of claim 1. Robinson does not teach wherein transforming the medical image comprises reframing of the image around the lesion according to a frame whose dimensions are predetermined, said frame being common to all the medical images of the training elements, the position of the lesion with respect to the frame being variable from one medical image to the other.
However, Chen does teach wherein transforming the medical image comprises reframing of the image around the lesion according to a frame whose dimensions are predetermined (Pg. 33, “from each normal breast X-ray image of the breast area randomly extracting 300 size is 256 * 256 pixel of the image block, namely normal image block… according to the medical expert manual marking the lesion position, from the abnormal breast X-ray image extracting comprises a lesion of the square image block, namely the lesion image block, as the positive sample of the test stage. Note that each image block only contains a lesion; the size of the image block is determined by the larger dimension of the lesion boundary frame… In the picture, the inner and outer frames respectively represent the original lump lesion frame and the adjusted square frame for extracting the image block. It should be noted that, using the square frame after adjusting extracting lesion image block instead of directly using the lesion frame, the purpose is when adjusting the lesion image block is a square image block of specified size suitable for CNN model, it can ensure the original shape and shape of the lesion such as not changed.” ),
said frame being common to all the medical images of the training elements (Pg. 33, “the purpose is when adjusting the lesion image block is a square image block of specified size suitable for CNN model, it can ensure the original shape and shape of the lesion such as not changed.” Pg. 27, “using the normalized part of the normal area image block as the training set; inputting to the dual-depth convolutional neural network model for training; taking the trained dual-depth convolutional neural network model as a normal model;”),
the position of the lesion with respect to the frame being variable from one medical image to the other (Pg. 33, “when extracting the image block, the lesion is located at the centre of the image block, at the same time, the background of the image block is minimum; However, the lesion near the breast or image boundary may not be located in the center of the extracted image block so as to minimize the background contained.”).
Robinson in view of Chen are considered analogous to the claimed invention because they are in the field of medical image processing for lesions. Therefore, it would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to have modified the image transformation of Robinson with the reframing of lesions as taught by Chen for the advantage of ensuring that images are of a “specified size suitable for CNN model” (Chen; pg. 33).
Regarding claim 15, Robinson teaches the electronic device of claim 1. Robinson does not teach wherein the machine learning algorithm is further trained beforehand with training elements whose medical images do not show any lesion in the anatomy of interest.
However, Chen does teach wherein the machine learning algorithm is further trained beforehand with training elements whose medical images do not show any lesion in the anatomy of interest (Pg. 30, “FIG. 6 shows a schematic diagram of 20 square image blocks extracted from an X-ray image of a normal mammary gland according to an embodiment of the present invention;” Pg. 33, “As one embodiment of the present invention, from each normal breast X-ray image of the breast area randomly extracting 300 size is 256 * 256 pixel of the image block, namely normal image block, as the negative sample of the training and test stage. the normal area image block only comprises a breast area, and by setting a redundant threshold value, the redundancy between each two normal area image blocks does not exceed the redundancy threshold value; that is, preferably only selecting completely located in the breast area of the image block, and limiting the overlapping degree of the image block to reduce the redundancy of the sample set. extracting 20 image blocks from 300 image blocks, as shown in FIG. 6.”). It would be obvious to one of ordinary skill in the art that normal tissue used as a negative sample would indicate images without lesions.
Robinson in view of Chen are considered analogous to the claimed invention because they are in the field of medical image processing for lesions. Therefore, it would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to have modified Robinson with Chen for the advantage of including a “negative sample [in] the training” (Chen; pg. 33).
Regarding claim 16, Robinson teaches an electronic device comprising at least one processor and a computer memory storing program code instructions which, when they are executed by said at least one processor, configure said at least one processor to implement a method for selecting at least one optimal treatment from among several available treatments for the percutaneous ablation of a lesion within an anatomy of interest of a patient ([0141], “The storage device 430 can include software services, servers, services, etc., that when the code that defines such software is executed by the processor 410, it causes the system to perform a function. In some examples, a hardware service that performs a particular function can include the software component stored in a computer-readable medium in connection with the necessary hardware components, such as processor 410, connection 405, output device 435, etc., to carry out the function.” [0069], “improved identification of a cardiac arrhythmia target volume and treatment plan for noninvasive and invasive arrhythmia ablation and improved decision support for the selection of therapy for arrhythmia ablation”), said method comprising:
obtaining a medical image previously acquired on the patient and on which the lesion can be seen within the anatomy of interest of the patient ([0005], “The method may include receiving one or more mappings selected from an electrical mapping, an anatomic mapping, a functional mapping, and combinations thereof; identifying an abnormality in the one or more mappings;… the electrical mapping may be an electrocardiograph image, the anatomic mapping may be at least one of a computer tomography image or a magnetic resonance image, and the functional mapping may be at least one of a photo emission computed tomographic image, a positron emission tomography image or an echocardiogram image.” [0115], “In an example, over time, the method can predict consistent utilization of invasive catheter ablation for small endocardial lesions in the left ventricular apex in older males.”),
calculating using a machine learning algorithm, from the medical image, for each of the different available ablation treatments, a confidence index whose value is representative of a probability of success of said treatment for the ablation of the lesion in the anatomy of interest of the patient ([0124], “Machine learning tools and predictive analytics can be integrated within method 200 to create a clinical decision support infrastructure such as the decision support module.”” [0117], “After a target has been identified, the identified target(s) may be presented to a physician in the form of a decision support module. In addition, the decision support module may include one or more of a description of the scar pattern,… a confidence score of the combined mappings, a listing of at-risk structures, general recommendations, expected success with SBRT, expected success with alternative treatment modalities (e.g., catheter RF, antiarrhythmic drug (e.g. amiodarone), etc.), or combinations thereof.” [0111], “the machine learning environment can be used to predict success of an ablation procedure using past information such as patient success related to their confidence score.” [0008], “an expected success rate with non-invasive therapy, or an expected success rate with alternative treatment modalities.”). As recited, the confidence score is provided by the decision support module, which is created through machine learning tools and predictive analytics. Thus, the confidence score is an output of machine learning. Furthermore, the confidence score is related to the success of the treatment, as explained by [0111], and said predicted success may refer to a rate, as shown in [0008], Examiner interprets the confidence score of Robinson to encompass the confidence index of the instant application.
the machine learning algorithm having been trained beforehand from a set of training elements ([0093], “The neural network can continue to receive input (e.g., training data) over a period of time until the neural network is trained. The input into the neural network can also include data related to the success of a treatment, such as a success outcome, survival, side effects, etc.”),
each training element comprising:
a medical image on which a lesion can be seen within the anatomy of interest of another patient ([0142], “ training data can be one or more images and a label (e.g., VT or no VT) can be associated with each image. In some examples, the labeled data can be actual target probabilities (e.g., each segment as the weighted average over all received images for each patent).”),
an identification of a treatment selected from among the different ablation treatments available for treating said other patient ([0093], “As additional patients are treated, the previous treatments can inform future treatments and can be used to adjust the target of the current patient (for example, by training the neural network). In another example, the target may be adjusted based on a weighting of input mappings or the confidence score of the combined mappings.” [0111], “The recommendations may be provided from prior patients... In an example, over time, the method can predict consistent utilization of invasive catheter ablation for small endocardial lesions in the left ventricular apex in older males. Subsequent outcomes data reveals high rates of toxicity in this population and low rates of toxicity with similar outcomes in patients treated with noninvasive ablation. While the decision support module may initially suggest further catheter ablation based on the multiplicity of prior treatments, an updated decision support module would suggest noninvasive ablation.”), and
a confidence index of said treatment on said other patient ([0111], “In an example, the machine learning environment can be used to predict success of an ablation procedure using past information such as patient success related to their confidence score.”), and
selecting at least one optimal treatment as a function of the confidence indices calculated for the different available treatments ([0132], “In an example, over time, the method can predict consistent utilization of invasive catheter ablation for small endocardial lesions in the left ventricular apex in older males. Subsequent outcomes data reveals high rates of toxicity in this population and low rates of toxicity with similar outcomes in patients treated with noninvasive ablation. While the decision support module may initially suggest further catheter ablation based on the multiplicity of prior treatments, an updated decision support module would suggest noninvasive ablation.” [0142], “The computing device(s) 502 can continue to receive data from the one or more data sources 506 until the neural network 508 (e.g., convolution neural networks, deep convolution neural networks, artificial neural networks, learning algorithms, etc.) of the computing device(s) 502 are trained (e.g., have had sufficient unbiased data to respond to new incoming data requests and provided an autonomous or near autonomous recommended course of actions, and/or actually provide input to perform the course of action).”);
wherein the method comprises transforming the medical image of the patient, the medical image of each training element having undergone a transformation before being used to train the machine learning algorithm ([0075], “At block 110, the method 100 can define or identify one or more abnormalities in the one or more input mappings. The abnormalities can be identified, for example, in an MRI by the scar location, an abnormality in PET/SPECT may be regions that are not viable, an abnormality in an electrical mapping may be where a VT originates, etc. In various examples, the abnormality can be defined by selecting the abnormality on each mapping, manually or automatically segmenting the mapping and manually selecting the abnormality, automatically contouring the mapping and manually selecting the abnormality, automatically contouring the mapping and automatically selecting the abnormality, or combinations thereof. In some examples, the identification of the abnormalities can be used in training a neural network, for example, by using supervised or reinforcement learning.” [0132], “the method can predict consistent utilization of invasive catheter ablation for small endocardial lesions in the left ventricular apex in older males.”).
Robinson does not teach wherein transforming the medical image comprises reframing of the image around the lesion according to a frame whose dimensions are predetermined, said frame being common to all the medical images of the training elements, the position of the lesion with respect to the frame being variable from one medical image to the other.
However, Chen does teach wherein transforming the medical image comprises reframing of the image around the lesion according to a frame whose dimensions are predetermined (Pg. 33, “from each normal breast X-ray image of the breast area randomly extracting 300 size is 256 * 256 pixel of the image block, namely normal image block… according to the medical expert manual marking the lesion position, from the abnormal breast X-ray image extracting comprises a lesion of the square image block, namely the lesion image block, as the positive sample of the test stage. Note that each image block only contains a lesion; the size of the image block is determined by the larger dimension of the lesion boundary frame… In the picture, the inner and outer frames respectively represent the original lump lesion frame and the adjusted square frame for extracting the image block. It should be noted that, using the square frame after adjusting extracting lesion image block instead of directly using the lesion frame, the purpose is when adjusting the lesion image block is a square image block of specified size suitable for CNN model, it can ensure the original shape and shape of the lesion such as not changed.” ),
said frame being common to all the medical images of the training elements (Pg. 33, “the purpose is when adjusting the lesion image block is a square image block of specified size suitable for CNN model, it can ensure the original shape and shape of the lesion such as not changed.” Pg. 27, “using the normalized part of the normal area image block as the training set; inputting to the dual-depth convolutional neural network model for training; taking the trained dual-depth convolutional neural network model as a normal model;”),
the position of the lesion with respect to the frame being variable from one medical image to the other (Pg. 33, “when extracting the image block, the lesion is located at the centre of the image block, at the same time, the background of the image block is minimum; However, the lesion near the breast or image boundary may not be located in the center of the extracted image block so as to minimize the background contained.”).
Robinson in view of Chen are considered analogous to the claimed invention because they are in the field of medical image processing for lesions. Therefore, it would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to have modified the image transformation of Robinson with the reframing of lesions as taught by Chen for the advantage of ensuring that images are of a “specified size suitable for CNN model” (Chen; pg. 33).
Claim 11 is rejected under 35 U.S.C. 103 as being unpatentable over Robinson (US 20210137384) in view of Krishnan (US 20050049497).
Regarding claim 11, Robinson teaches the electronic device of claim 1. Robinson further teaches wherein the method comprises: for each training element corresponding to the selected optimal treatment, calculating a similarity value representative of the similarity between the medical image of the patient to be treated and the medical image of the training element, and selecting a reference training element from among the set of the training elements as a function of the calculated similarity values.
However, Krishnan does teach wherein the method comprises:
for each training element corresponding to the selected optimal treatment, calculating a similarity value representative of the similarity between the medical image of the patient to be treated and the medical image of the training element ([0072], “Another exemplary approach is as follows. Assume that there are m cases in a training set. A new case will be compared to these m cases using the n features described above. Based on this comparison, p cases are selected as being most "similar" to the current case, wherein similarity can be defined in one of various ways. For instance, one approach is to consider the Euclidean distance in the n-dimensional feature space. Other well-known distance measures can also be employed. It is to be appreciated that the above process can also be used to select exemplar cases from a library of cases for display as well.” [0023], “The input to the CAD system (10) comprises various sources of patient information including image data (1) in one or more imaging modalities (e.g., ultrasound image data, X-ray mammography image data, MRI etc.) and non-image data” [0038], “The automated detection module (29) implements methods for processing ultrasound image data (3) of breast tissue to detect and segment potential lesions in the imaged breast tissue.”), and
selecting a reference training element from among the set of the training elements as a function of the calculated similarity values ([0072], “Based on this comparison, p cases are selected as being most "similar" to the current case, wherein similarity can be defined in one of various ways. For instance, one approach is to consider the Euclidean distance in the n-dimensional feature space.” [0010], “CAD systems and methods for breast imaging implement machine-learning techniques which use training data that is obtained (learned) from a database of previously diagnosed (labeled) patient cases in one or more relevant clinical domains and/or expert interpretations of such data to enable the CAD systems to "learn" to properly and accurately analyze patient data and make proper diagnostic and/or therapeutic assessments and decisions for assisting physician workflow.”).
Robinson in view of Krishnan are considered analogous to the claimed invention because they are in the field of medical image processing for lesions. Therefore, it would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to have modified Robinson with Krishnan for the advantage of “studying similar cases to see how other patients responded to different treatment options… to assess the efficacy of these options for the current patient” (Krishnan; [0029]).
Claim 14 is rejected under 35 U.S.C. 103 as being unpatentable over Robinson (US 20210137384) in view of Wibowo (US 20210038314).
Regarding claim 14, Robinson teaches the electronic device of claim 1. Robinson further teaches wherein the anatomy of interest is the liver, a kidney or a lung ([0064], “FIG. 9B shows serial thoracic CT scans after treatment in Patient 1. The treatment area is shown in blue. At 3 months, there were adjacent local inflammatory changes in the lung parenchyma, effects that had nearly resolved at 12 months.”).
Robinson does not teach wherein the lesion is a tumor or a cyst.
However, Wibowo does teach wherein the lesion is a tumor or a cyst ([0007], “inputs to the dose prediction model include the geometry of the target tumor, physical properties of the tissue, dimensions of the ablation device's probe/applicator, position of the probe/applicator within the lesion, and extracted tissue properties” [0029], “The settings allow the method to detect more tumors and pre-compute radiomics features for each detected lesion.” [0026], “Additionally, although the invention has particular use for planning ablation for lung tumors and nodules, the invention is not so limited. Examples of other tumor types for which ablation zone models and dose may be computed include, without limitation, lung, breast, liver, kidney, colon, prostrate, and ovarian.”).
Robinson in view of Wibowo are considered analogous to the claimed invention because they are in the field of medical image processing for lesions. Therefore, it would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to have modified Robinson with Wibowo for the advantage of promoting “ablation planning for treatment of cancerous tissue” (Wibowo; [0002]).
Conclusion
THIS ACTION IS MADE FINAL. Applicant is reminded of the extension of time policy as set forth in 37 CFR 1.136(a).
A shortened statutory period for reply to this final action is set to expire THREE MONTHS from the mailing date of this action. In the event a first reply is filed within TWO MONTHS of the mailing date of this final action and the advisory action is not mailed until after the end of the THREE-MONTH shortened statutory period, then the shortened statutory period will expire on the date the advisory action is mailed, and any nonprovisional extension fee (37 CFR 1.17(a)) pursuant to 37 CFR 1.136(a) will be calculated from the mailing date of the advisory action. In no event, however, will the statutory period for reply expire later than SIX MONTHS from the mailing date of this final action.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to whose telephone number is 571-272-3931. The examiner can normally be reached M-Th: 10:30-8:00 ET.
Examiner interviews are available via telephone, in-person, and video conferencing using a USPTO supplied web-based collaboration tool. To schedule an interview, applicant is encouraged to use the USPTO Automated Interview Request (AIR) at http://www.uspto.gov/interviewpractice.
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, Shahid Merchant can be reached on (571)270-1360. 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.
/D.C./Examiner, Art Unit 3684
/Shahid Merchant/Supervisory Patent Examiner, Art Unit 3684