Notice of Pre-AIA or AIA Status
The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
Claim Status
Claims 1-20 are pending.
Priority
This application claims priority to U.S. Provisional Patent Application No. 63/601,479, filed November 21, 2023.
Information Disclosure Statement
The IDS filed 11/18/24, 06/10/25, 06/10/25 have been considered.
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.
Claims 1-2, 10-11, 15, and 19-20 are rejected under 35 U.S.C. 103 as being unpatentable over CHOI et al. (US 20210133572 A1 Hereinafter “CHOI”) in view of Karki et al. (“CT window trainable neural network for improving intracranial hemorrhage detection by combining multiple settings” Hereinafter “Karki”).
Regarding claim 1, CHOI teaches a system, comprising:
a processor that executes computer-executable components ([0071]: The processor 120 includes a first preprocessing module 121, a second preprocessing module 122, and a third preprocessing module 123. The processor 120 may control the first artificial neural network 110 and the reception interface 150”) stored in a non-transitory computer-readable memory ([0161]: “The image preprocessing, analysis support, and/or analysis method using machine learning-based artificial intelligence according to an embodiment of the present invention may be implemented in the form of program instructions executable via various computer means, and may be then recorded in a computer-readable storage medium”), wherein the computer-executable components comprise:
a receiving component that receives a set of “regions/volume of interest” images containing a plurality of organs (Fig. 1, [0073]: “The training input image 170 may be transferred to the first artificial neural network 110 via the reception interface 150”. The training image set comprises regions/volumes of interest ([0065]: “It was confirmed that the performance of the image processing (mainly image segmentation and region-of-interest detection in the present invention) of the artificial neural network was improved by training the artificial neural network and the preprocessing modules together”), of a plurality of organs which can be seen in Fig. 9, the processed result shows the different organs); and
an artificial intelligence deep learning neural network model component that automatically processes ([0078]: “Referring back to FIG. 1, the first artificial neural network 110 generates the preprocessing conditions used in the preprocessing processes performed by the plurality of preprocessing modules 121, 122 and 123. In this case, the first artificial neural network 110 may be a generative artificial neural network. The first artificial neural network 110 generates the preprocessing conditions based on the training input image 170”. The optimized organ images are generated based on these parameters; see Fig. 9 and accompanying description).
CHOI does not expressly disclose the renderer being an artificial intelligence deep learning neural network model component for processing and enhancing the image.
However, Karki teaches an artificial intelligence deep learning neural network model component for processing and enhancing the image (Page 4, section 2.4: “The scaling layer is a non-learnable layer which performs the windowing function. It transforms the original HU values based on the estimated window setting”. If a layer performs this operation, the layer must be part of a neural network, and neural networks are a kind of artificial intelligence deep learning neural network model).
At the time the invention was made, it would have been obvious to one of ordinary skill in the art to modify CHOI’s CT image enhancement method to include Karki’s use of a neural network to perform the windowing because such a modification is the result of applying a known technique to a known device ready for improvement to yield predictable results. More specifically, Karki’s use of a neural network to perform the windowing permits accurately applying ideal windowing parameters to CT images. This known benefit in Karki is applicable to CHOI’s CT image enhancement method as they both share characteristics and capabilities, namely, they are directed to enhancing CT images by determining ideal windowing parameters and applying them top the CT images. Therefore, it would have been recognized that modifying CHOI’s CT image enhancement method to include Karki’s use of a neural network to perform the windowing would have yielded predictable results because (i) the level of ordinary skill in the art demonstrated by the references applied shows the ability to incorporate Karki’s use of a neural network to perform the windowing in enhancing CT images by determining ideal windowing parameters and applying them top the CT images and (ii) the benefits of such a combination would have been recognized by those of ordinary skill in the art.
Regarding claim 2, the combination of CHOI and Karki teaches system of claim 1, in addition, CHOI further teaches wherein the artificial intelligence deep learning neural network model component predicts window level (WL) and window width (WW) maps per organ of a plurality of identified scanned organs ([0074]: “The first artificial neural network 110 may generate a first preprocessing condition, a second preprocessing condition, and a third preprocessing condition through inference on the training input image 170. The first preprocessing condition is intended for the first preprocessing module 121, the second preprocessing condition is intended for the second preprocessing module 122, and the third preprocessing condition is intended for the third preprocessing module 123. Each of the preprocessing conditions may be windowing information including the upper and lower limits of the brightness values of the training input image 170, and may be defined by a central level and a width instead of being bounded by the upper and lower limits”. These central levels and widths are the window level and width).
Regarding claim 10, the content of claim 10 is similar to the content of claim 1, therefore it is rejected for the same reasons of obviousness as claim 1.
Regarding claim 11, the content of claim 11 is similar to the content of claim 2, therefore it is rejected for the same reasons of obviousness as claim 2.
Regarding claim 15, the combination of CHOI and Karki teaches the method of claim 10, in addition, CHOI further teaches comprising using the (Fig. 9: Fig. 9 shows to organ-specific views generated based in part of contrast levels (which change based on windowing) and processing parameters (parameters for windowing seen above)).
Karki further teaches using an artificial intelligence deep learning neural network model component for processing and enhancing the image (Page 4, section 2.4: “The scaling layer is a non-learnable layer which performs the windowing function. It transforms the original HU values based on the estimated window setting”. If a layer performs this operation, the layer must be part of a neural network, and neural networks are a kind of artificial intelligence deep learning neural network model).
At the time the invention was made, it would have been obvious to one of ordinary skill in the art to modify CHOI’s CT image enhancement method to include Karki’s use of a neural network to perform the windowing because such a modification is the result of applying a known technique to a known device ready for improvement to yield predictable results. More specifically, Karki’s use of a neural network to perform the windowing permits accurately applying ideal windowing parameters to CT images. This known benefit in Karki is applicable to CHOI’s CT image enhancement method as they both share characteristics and capabilities, namely, they are directed to enhancing CT images by determining ideal windowing parameters and applying them top the CT images. Therefore, it would have been recognized that modifying CHOI’s CT image enhancement method to include Karki’s use of a neural network to perform the windowing would have yielded predictable results because (i) the level of ordinary skill in the art demonstrated by the references applied shows the ability to incorporate Karki’s use of a neural network to perform the windowing in enhancing CT images by determining ideal windowing parameters and applying them top the CT images and (ii) the benefits of such a combination would have been recognized by those of ordinary skill in the art.
Regarding claim 19, the content of claim 19 is similar to the content of claim 1, therefore it is rejected for the same reasons of obviousness as claim 1.
Regarding claim 20, the content of claim 20 is similar to the content of claim 2, therefore it is rejected for the same reasons of obviousness as claim 2.
Claims 4, 6-9, 13, and 16-18 are rejected under 35 U.S.C. 103 as being unpatentable over CHOI et al. (US 20210133572 A1 Hereinafter “CHOI”) in view of Karki et al. (“CT window trainable neural network for improving intracranial hemorrhage detection by combining multiple settings” Hereinafter “Karki”) in further view of NICKISCH et al. (US 20260011006 A1 Hereinafter “NICKISCH”).
Regarding claim 4, the combination of CHOI and Karki teaches system of claim 1, in addition, CHOI further teaches wherein the artificial intelligence deep learning neural network model is trained ([0065]: “The present invention was derived based on improvement in the performance of an artificial neural network and preprocessing modules achieved by training the artificial neural network and the preprocessing modules together”).
The combination of CHOI and Karki does not expressly disclose neural network model is trained in part by calculating a loss function between a desired organ image and a predicted organ image.
However, NICKISCH teaches neural network model is trained in part by calculating a loss function between a desired organ image and a predicted organ image ([0119]: “The process of training the neural network NN illustrated in FIG. 7 therefore includes adjusting its parameters. The parameters, or more particularly the weights and biases, control the operation of activation functions in the neural network. In supervised learning, the training process automatically adjusts the weights and the biases, such that when presented with the input data, the neural network accurately provides the corresponding expected output data. In order to do this, the value of the loss functions, or errors, are computed based on a difference between predicted output data and the expected output data”. This network performs contrast enhancement using techniques such as windowing and it is trained to better match the desired output which is the ground truth image “If the neural network is trained to identify which of multiple reconstruction techniques, or which of multiple windowing techniques to apply, in order to provide the contrast-adjusted image, the ground truth data may include an indication of a type of image reconstruction technique that should be used to generate the contrast-adjusted image, or an indication of the energy intervals that should be used to generate the contrast-adjusted image, or a type of windowing approach that should be used to generate the contrast-adjusted image”[0117]).
At the time the invention was made, it would have been obvious to one of ordinary skill in the art to modify the combination of CHOI and Karki’s CT image enhancement method to include NICKISCH’s training a neural network for contrast enhancement using a loss between predicted and ground truth images because such a modification is the result of applying a known technique to a known device ready for improvement to yield predictable results. More specifically, NICKISCH’s training a neural network for contrast enhancement using a loss between predicted and ground truth images permits training a model to accurately produce a contrast enhanced images by minimizing a loss between a ground truth image and predicted image. This known benefit in NICKISCH is applicable to the combination of CHOI and Karki’s CT image enhancement method as they both share characteristics and capabilities, namely, they are directed to enhancing contrast in CT images using neural networks. Therefore, it would have been recognized that modifying the combination of CHOI and Karki’s CT image enhancement method to include NICKISCH’s training a neural network for contrast enhancement using a loss between predicted and ground truth images would have yielded predictable results because (i) the level of ordinary skill in the art demonstrated by the references applied shows the ability to incorporate NICKISCH’s training a neural network for contrast enhancement using a loss between predicted and ground truth images in enhancing contrast in CT images using neural networks and (ii) the benefits of such a combination would have been recognized by those of ordinary skill in the art.
Regarding claim 6, the combination of CHOI, Karki, and NICKISCH teaches the system of claim 4, in addition, CHOI further teaches wherein the (Fig. 9: Fig. 9 shows to organ-specific views generated based in part of contrast levels (which change based on windowing) and processing parameters (parameters for windowing seen above)).
Karki further teaches using an artificial intelligence deep learning neural network model component for processing and enhancing the image (Page 4, section 2.4: “The scaling layer is a non-learnable layer which performs the windowing function. It transforms the original HU values based on the estimated window setting”. If a layer performs this operation, the layer must be part of a neural network, and neural networks are a kind of artificial intelligence deep learning neural network model).
At the time the invention was made, it would have been obvious to one of ordinary skill in the art to modify CHOI’s CT image enhancement method to include Karki’s use of a neural network to perform the windowing because such a modification is the result of applying a known technique to a known device ready for improvement to yield predictable results. More specifically, Karki’s use of a neural network to perform the windowing permits accurately applying ideal windowing parameters to CT images. This known benefit in Karki is applicable to CHOI’s CT image enhancement method as they both share characteristics and capabilities, namely, they are directed to enhancing CT images by determining ideal windowing parameters and applying them top the CT images. Therefore, it would have been recognized that modifying CHOI’s CT image enhancement method to include Karki’s use of a neural network to perform the windowing would have yielded predictable results because (i) the level of ordinary skill in the art demonstrated by the references applied shows the ability to incorporate Karki’s use of a neural network to perform the windowing in enhancing CT images by determining ideal windowing parameters and applying them top the CT images and (ii) the benefits of such a combination would have been recognized by those of ordinary skill in the art.
Regarding claim 7, the combination of CHOI and Karki teaches system of claim 1, in addition, CHOI further teaches wherein ([0078]: “Referring back to FIG. 1, the first artificial neural network 110 generates the preprocessing conditions used in the preprocessing processes performed by the plurality of preprocessing modules 121, 122 and 123. In this case, the first artificial neural network 110 may be a generative artificial neural network. The first artificial neural network 110 generates the preprocessing conditions based on the training input image 170”. The optimized organ images are generated based on these parameters).
Karki further teaches using an artificial intelligence deep learning neural network model component for processing and enhancing the image (Page 4, section 2.4: “The scaling layer is a non-learnable layer which performs the windowing function. It transforms the original HU values based on the estimated window setting”. If a layer performs this operation, the layer must be part of a neural network, and neural networks are a kind of artificial intelligence deep learning neural network model).
At the time the invention was made, it would have been obvious to one of ordinary skill in the art to modify CHOI’s CT image enhancement method to include Karki’s use of a neural network to perform the windowing because such a modification is the result of applying a known technique to a known device ready for improvement to yield predictable results. More specifically, Karki’s use of a neural network to perform the windowing permits accurately applying ideal windowing parameters to CT images. This known benefit in Karki is applicable to CHOI’s CT image enhancement method as they both share characteristics and capabilities, namely, they are directed to enhancing CT images by determining ideal windowing parameters and applying them top the CT images. Therefore, it would have been recognized that modifying CHOI’s CT image enhancement method to include Karki’s use of a neural network to perform the windowing would have yielded predictable results because (i) the level of ordinary skill in the art demonstrated by the references applied shows the ability to incorporate Karki’s use of a neural network to perform the windowing in enhancing CT images by determining ideal windowing parameters and applying them top the CT images and (ii) the benefits of such a combination would have been recognized by those of ordinary skill in the art.
The combination of CHOI and Karki does not expressly disclose the optimized organ image being generated based on comparison to ground truth images.
However, NICKISCH teaches neural network model is trained in part by calculating a loss function between a desired organ image (ground truth) and a predicted organ image ([0119]: “The process of training the neural network NN illustrated in FIG. 7 therefore includes adjusting its parameters. The parameters, or more particularly the weights and biases, control the operation of activation functions in the neural network. In supervised learning, the training process automatically adjusts the weights and the biases, such that when presented with the input data, the neural network accurately provides the corresponding expected output data. In order to do this, the value of the loss functions, or errors, are computed based on a difference between predicted output data and the expected output data”. This network performs contrast enhancement using techniques such as windowing and it is trained to better match the desired output which is the ground truth image “If the neural network is trained to identify which of multiple reconstruction techniques, or which of multiple windowing techniques to apply, in order to provide the contrast-adjusted image, the ground truth data may include an indication of a type of image reconstruction technique that should be used to generate the contrast-adjusted image, or an indication of the energy intervals that should be used to generate the contrast-adjusted image, or a type of windowing approach that should be used to generate the contrast-adjusted image”[0117]. By training the model in this way, the generation of optimized organ views is based on the comparison of the predicted image to the ground truth image).
At the time the invention was made, it would have been obvious to one of ordinary skill in the art to modify the combination of CHOI and Karki’s CT image enhancement method to include NICKISCH’s training a neural network for contrast enhancement using a loss between predicted and ground truth images because such a modification is the result of applying a known technique to a known device ready for improvement to yield predictable results. More specifically, NICKISCH’s training a neural network for contrast enhancement using a loss between predicted and ground truth images permits training a model to accurately produce a contrast enhanced images by minimizing a loss between a ground truth image and predicted image. This known benefit in NICKISCH is applicable to the combination of CHOI and Karki’s CT image enhancement method as they both share characteristics and capabilities, namely, they are directed to enhancing contrast in CT images using neural networks. Therefore, it would have been recognized that modifying the combination of CHOI and Karki’s CT image enhancement method to include NICKISCH’s training a neural network for contrast enhancement using a loss between predicted and ground truth images would have yielded predictable results because (i) the level of ordinary skill in the art demonstrated by the references applied shows the ability to incorporate NICKISCH’s training a neural network for contrast enhancement using a loss between predicted and ground truth images in enhancing contrast in CT images using neural networks and (ii) the benefits of such a combination would have been recognized by those of ordinary skill in the art.
Regarding claim 8, the combination of CHOI, Karki, and NICKISCH teaches system of claim 7, in addition, NICKISCH further teaches wherein ground truth images are generated utilizing optimized acquisition parameters ( ([0117]: ““If the neural network is trained to identify which of multiple reconstruction techniques, or which of multiple windowing techniques to apply, in order to provide the contrast-adjusted image, the ground truth data may include an indication of a type of image reconstruction technique that should be used to generate the contrast-adjusted image, or an indication of the energy intervals that should be used to generate the contrast-adjusted image, or a type of windowing approach that should be used to generate the contrast-adjusted image”. The “or” limitation means only one of the listed alternative needs met to make a case of obviousness).
The rationale for this combination is similar to the combination for claim 4 for NICKISCH due to similar method of combination (calculating loss between predicted and ground truth image, the parameters on how the ground truth image is acquired would be needed) and benefits (accurate contrast enhancement).
Regarding claim 9, the combination of CHOI, Karki, and NICKISCH teaches system of claim 6, in addition, CHOI further teaches wherein the ([0005]: “A CT number (also called a Hounsfield Unit) usually are mapped to greyscale to display for a radiologist. Mapping too wide range of CT numbers leads to smaller greyscale difference, which make difficult to distinguish the difference within interest organ”. Since CT images are mapped to grayscale values, changing window width and level of CT images performs remapping of the CT numbers, so the windowing process for CT images acts as the remapping algorithm of remapping the CT number).
Karki further teaches using an artificial intelligence deep learning neural network model component for performing the windowing (Page 4, section 2.4: “The scaling layer is a non-learnable layer which performs the windowing function. It transforms the original HU values based on the estimated window setting”. If a layer performs this operation, the layer must be part of a neural network, and neural networks are a kind of artificial intelligence deep learning neural network model).
At the time the invention was made, it would have been obvious to one of ordinary skill in the art to modify CHOI’s CT image enhancement method to include Karki’s use of a neural network to perform the windowing because such a modification is the result of applying a known technique to a known device ready for improvement to yield predictable results. More specifically, Karki’s use of a neural network to perform the windowing permits accurately applying ideal windowing parameters to CT images. This known benefit in Karki is applicable to CHOI’s CT image enhancement method as they both share characteristics and capabilities, namely, they are directed to enhancing CT images by determining ideal windowing parameters and applying them top the CT images. Therefore, it would have been recognized that modifying CHOI’s CT image enhancement method to include Karki’s use of a neural network to perform the windowing would have yielded predictable results because (i) the level of ordinary skill in the art demonstrated by the references applied shows the ability to incorporate Karki’s use of a neural network to perform the windowing in enhancing CT images by determining ideal windowing parameters and applying them top the CT images and (ii) the benefits of such a combination would have been recognized by those of ordinary skill in the art.
Regarding claim 13, the content of claim 13 is similar to the content of claim 4, therefore it is rejected for the same reasons of obviousness as claim 4.
Regarding claim 16, the content of claim 16 is similar to the content of claim 7, therefore it is rejected for the same reasons of obviousness as claim 7.
Regarding claim 17, the content of claim 17 is similar to the content of claim 8, therefore it is rejected for the same reasons of obviousness as claim 8.
Regarding claim 18, the content of claim 18 is similar to the content of claim 9, therefore it is rejected for the same reasons of obviousness as claim 9.
Claims 3 and 12 are rejected under 35 U.S.C. 103 as being unpatentable over CHOI et al. (US 20210133572 A1 Hereinafter “CHOI”) in view of Karki et al. (“CT window trainable neural network for improving intracranial hemorrhage detection by combining multiple settings” Hereinafter “Karki”) as evidenced by Geeks (“Regression in Machine Learning” Hereinafter “Geeks”).
Regarding claim 3, the combination of CHOI and Karki teaches system of claim 1, in addition, CHOI further teaches wherein the artificial intelligence deep learning neural network model predicts WL and WW maps using a regression technique (WL) and window width (WW) maps per organ of a plurality of identified scanned organs ([0074]: “The first artificial neural network 110 may generate a first preprocessing condition, a second preprocessing condition, and a third preprocessing condition through inference on the training input image 170. The first preprocessing condition is intended for the first preprocessing module 121, the second preprocessing condition is intended for the second preprocessing module 122, and the third preprocessing condition is intended for the third preprocessing module 123. Each of the preprocessing conditions may be windowing information including the upper and lower limits of the brightness values of the training input image 170, and may be defined by a central level and a width instead of being bounded by the upper and lower limits”. These central levels and widths are the window level and width. Predicting of parameters for window width and level are predictions of continuous numerical values (range of values of level and width) based on learning relationships between input variables (training image features of specific organs) and an output variable (Hounsfield units for those specific organs). This process itself of predicting window parameters is a form of regression, as the description of what regression is as evidence by Geeks “Regression is a supervised learning technique used to predict continuous numerical values by learning relationships between input variables (features) and an output variable (target)” (page 1)).
Regarding claim 12, the content of claim 12 is similar to the content of claim 3, therefore it is rejected for the same reasons of obviousness as claim 3.
Claims 5 and 14 are rejected under 35 U.S.C. 103 as being unpatentable over CHOI et al. (US 20210133572 A1 Hereinafter “CHOI”) in view of Karki et al. (“CT window trainable neural network for improving intracranial hemorrhage detection by combining multiple settings” Hereinafter “Karki”) in further view of Smith et al. (US 20200294288 A1 Hereinafter “Smith”).
Regarding claim 5, the combination of CHOI and Karki teaches system of claim 1, in addition, CHOI further teaches wherein the plurality of scanned organs respective images of the plurality of organs are obtained using at least one of([0072]: “training input image 170 is a medical image. Although a computed tomography (CT) image is mainly referred to below for ease of description, the training input image 170 is not limited to a CT image, but may be an image acquired by a modality including at least one of X-ray imaging, computed tomography (CT), magnetic resonance imaging (MRI), ultrasonic imaging, positron emission tomography (PET), and single photon emission computed tomography (SPECT)”).
CHOI does not expressly disclose the CT being single or dual energy.
However, Smith teaches use of single or dual energy CT image data ([0054]: “Embodiments of the present disclosure employ deep learning algorithms, such as those discussed above, to improve the diagnostic accuracy of routine single energy CT or dual energy CT in a variety of settings”. Due to the “or” nature of the claim only 1 of the listed alternatives need met for a case of obviousness)
At the time the invention was effectively filed, it would have been obvious to one of ordinary skill in the art to substitute the combination of CHOI and Karki’s CT image data with Smith’s single or dual energy CT data because such a modification is the result of simple substitution of one known element for another producing a predictable result. More specifically, Smith’s single or dual energy CT data teaches that single energy CT is the most common, but one could use either single or dual energy for obtaining medical image data ([0036]: “Computed tomography is a medical imaging technique that uses X-rays to image fine slices of a patient's body, thereby providing a window to the inside of a patient's body without invasive surgery. Radiologists use CT imaging to evaluate, diagnose, and/or treat any of the myriad internal maladies and dysfunctions. The majority of CT scanners Most CT is performed using single energy CT scanners”), and one of ordinary skill in the art would expect similar effects if substituted for the combination of CHOI and Karki’s CT image data with Smith’s single or dual energy CT data.
Regarding claim 14, the content of claim 14 is similar to the content of claim 5, therefore it is rejected for the same reasons of obviousness as claim 5.
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure:
Szczykutowicz (US 20200242815 A1) teaches processing medical images using window width and level
SCHULTZ et al. (US 20160343117 A1) teaches processing medical images using window width and level
Silverstein et al. (US 20090096807 A1) teaches processing medical images using window width and level
Any inquiry concerning this communication or earlier communications from the examiner should be directed to STEFANO A DARDANO whose telephone number is (703)756-4543. The examiner can normally be reached Monday - Friday 11:00 - 7:00.
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, Greg Morse can be reached at (571) 272-3838. 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.
/STEFANO ANTHONY DARDANO/ Examiner, Art Unit 2663
/SEAN M CONNER/ Primary Examiner, Art Unit 2663
Prosecution Insights