Software imaging for composition placement

Abstract

Provided herein are methods of classifying one or more objects in an image and training a machine learning algorithm for performance thereof. Also provided herein are methods of displacing a gel composition, methods of reducing a dose of radiation applied to one or more organs within a compartment of a human subject, and methods of evaluating placement of a composition. Also provided herein are methods of determining an object placement in a compartment of a human subject.

Description

SOFTWARE IMAGING FOR COMPOSITION PLACEMENTCROSS-REFERENCE

[0001] This application claims the benefit of U.S. Provisional Application No. 63 / 512,860, filed July 10, 2023, which is hereby incorporated by reference in its entirety.INCORPORATION BY REFERENCE

[0002] All publications, patents, and patent applications mentioned in this specification are herein incorporated by reference to the same extent as if each individual publication, patent, or patent application was specifically and individually indicated to be incorporated by reference.BACKGROUND

[0003] Generally, radiotherapy is considered as a palliative treatment option for severe or uncommon cases of melanoma. However there has recently been an increased demand for new systems and methods of melanoma management. Brachytherapy techniques also involve balloon or strut multi catheter brachytherapy, with the applicator being placed in the surgical cavity by the breast surgeon at the time of or shortly after the wide local excision.SUMMARY

[0004] In one aspect, disclosed herein is a method of classifying one or more objects in an image, comprising: receiving an image of a compartment of a human subject, wherein the image contains one or more objects of the compartment; detecting, using a machine learning model, the one or more objects of the compartment; determining, using the machine learning model, a location and a value of a metric of at least a first object of the one or more objects; identifying, using the machine learning model, the first object of the one or more objects as a class of a plurality of classes based on the value and the location, wherein the plurality of classes comprises an organ, a tumor, and a spacer; and displaying the location and the class of the first object on a user interface.

[0005] In some embodiments, the metric is: a distance between the first object and at least a second object of the one or more objects; a radiosensitivity of the first object; a contrast to noise ratio of the first object; a shape of the first object; a volume of the first object; or any combination thereof. In some embodiments, the metric is the radiosensitivity of the first object. In some embodiments, the class of the first object is a spacer. In some embodiments, the method further comprises receiving user input from the computing device. In some embodiments, the identifying is further based on the user input. In some embodiments, the user input comprises: the location of the first object in the image; a portion of the human subject associated with theimage; a condition or disorder of the human subject; or any combination thereof. In some embodiments, the identifying is further based on a comparison of the value of the metric associated with first object to a value of the metric associated with a second object of the one or more objects. In some embodiments, the method further comprises: receiving feedback from the computing device; and updating the machine learning model based on the feedback. In some embodiments, updating the machine learning model comprises adjusting one or more parameters of the machine learning model. In some embodiments, the feedback comprises: an updated class of the first object; a class of a second object of the one or more objects; a location of the second object of the one or more objects; an updated class of the first object; an updated location of the first object; or any combination thereof.

[0006] Another aspect provided herein is a method of training a machine learning model comprising one or more parameters, comprising: receiving an image of a compartment of a human, wherein the image contains one or more objects of the compartment; determining, using the machine learning model, a location and a value of a metric of each object of the one or more objects; identifying, using the machine learning model, each object of the one or more objects as a class of a plurality of classes; receiving an updated class of the plurality of class for at least one object of the one or more objects; and updating the one or more parameters of the machine learning model based on the updated class, thereby training the machine learning model. In some embodiments, the method further comprises detecting, using the machine learning model, the one or more objects. In some embodiments, the metric is: a distance between each object of the one or more objects; a radiosensitivity of each object of the one or more objects; a contrast to noise ratio of each object of the one or more objects; a shape of each object of the one or more objects; a volume of each object of the one or more objects; or any combination thereof. In some embodiments, the plurality of classes comprises an organ, a tumor, or a spacer. In some embodiments, the updated type of the at least one object of the one or more objects indicates a spacer. In some embodiments, the identifying is further based on a comparison of the value of the metric associated with first object to a value of the metric associated with a second object of the one or more objects.

[0007] Another aspect provided herein is a method of displacing a gel composition, comprising: detecting one or more organs and a tumor on an image of a compartment of the human subject; identifying, using a machine learning model, a plurality of spaces of the compartment for administering the gel composition based on image information of the one or more organs and the tumor; selecting, using the machine learning model a first space of the plurality of spaces based on a relationship between the first space and the one or more organs and the tumor; and displacing the gel composition in the first space.

[0008] In some embodiments, the relationship is based on: a distance between the first space and the one or more organs; a distance between the first space and the tumor; a shape of the gel composition associated with the first space; a volume of the gel compositions associated with the first space or any combination thereof. In some embodiments, selecting on an image, using the machine learning model, the first space of the plurality of spaces is further based on a relationship between a second space and the one or more organs and the tumor. In some embodiments, the method further comprises determining, using the machine learning model: the distance between the first space and the one or more organs; the distance between the first space and the tumor; the shape of the gel composition associated with the first space; the volume of the gel compositions associated with the first space; the amount of contrast (signal) on the image; a distance between the second space and the one or more organs; a distance between the second space and the tumor; a shape of the gel composition associated with the second space; a volume of the gel compositions associated with the second space or any combination thereof. In some embodiments, the distance between the first space and the one or more organs is associated with a reduced dose of radiation applied to the one or more organs relative to another distance between another space of the plurality of spaces and the one or more organs. In some embodiments, the distance between the first space and the tumor is associated with a therapeutically effective dose of radiation applied to the tumor. In some embodiments, the other space is the second space. In some embodiments, at least two spaces of the plurality of spaces overlap. In some embodiments, at least two spaces of the plurality of spaces do not overlap. In some embodiments, the at least two spaces comprise the first space. In some embodiments, the method further comprises displaying the first space. In some embodiments, the first space comprises displaying the image of the first space within the compartment on a user interface. In some embodiments, the user interface is on a computing device. In some embodiments, the machine model determines one or more of the following based on a signal to noise ratio of one or more portions of the image: the distance between the first space and the one or more organs; the distance between the first space and the tumor; the shape of the gel composition associated with the first space; the volume of the gel compositions associated with the first space; the amount of contrast (signal) on the image; the distance between the second space and the one or more organs; the distance between the second space and the tumor; the shape of the gel composition associated with the second space; or the volume of the gel compositions associated with the second space. In some embodiments, identifying, using a machine learning model, a plurality of spaces of the compartment comprises generating the plurality of spaces in the compartment.

[0009] Another aspect provided herein is a method of reducing a dose of radiation applied to one or more organs within a compartment of a human subject, comprising: detecting the one or more organs within the compartment and a tumor in the compartment; generating a plurality of beam markers based on the one or more organs within the compartment and the tumor in the compartment, wherein a beam marker of the plurality of beam markers indicates a point of administration of a dose of radiation to the human subject; and selecting a first beam marker of the plurality of beam markers based on a relationship of the first beam marker to the one or more organs displaced in the compartment and the tumor in the compartment, wherein the first beam marker indicates a reduced dose of radiation applied to the one or more organs relative to a reference point of administration, and wherein the first beam marker indicates a therapeutically effective dose of radiation applied to the tumor relative to the reference point of administration. In some embodiments, the method further comprises detecting a spacer displaced in the compartment, wherein generating the plurality of beam markers is further based on the spacer and selecting the first beam marker of the plurality of beam markers is further based on the spacer. In some embodiments, generating the plurality of beam markers based on the one or more organs displaced in said compartment and the spacer displaced in said compartment comprises generating, using a machine learning model, the plurality of beam markers based on the one or more organs displaced in said compartment and the spacer displaced in said compartment. In some embodiments, selecting the first beam of the plurality of beams based on the relationship of the first beam to at least one of the one or more organs and the spacer comprises wherein selecting, using a machine learning model, the first beam of the plurality of beams based on the relationship of the first beam to at least one of the one or more organs and the spacer. In some embodiments, prior to detecting the one or more organs displaced in the compartment, the spacer displaced in the compartment, the tumor displaced in the compartment, or any combination thereof, displacing the spacer adjacent to the one or more organs and the tumor In some embodiments, determining the shape is further based on one or more criteria associated with the plurality of beam markers. In some embodiments, the tumor is in need of radiation therapy. In some embodiments, the method further comprises ranking the beam markers of the plurality of beam markers based on the one or more criteria and a plurality of relationships between the plurality of beam markers and the one or more organs, wherein the selecting is based on the ranking and the plurality of relationships comprises the relationship of the first beam. In some embodiments, the one or more criteria are associated with the one or more organs. In some embodiments, the one or more organs are visible in the image, and wherein the one or more criteria are based on locations of the one or more organs in the image. In some embodiments, at least one of the relationship is associated with a distance from an organof the one or more organs. In some embodiments, the relationship comprises: a distance from one or more organs; overlap of a beam marker of the plurality of beam markers with one or more organs; or both. In some embodiments, each beam marker of the plurality of beam markers overlap. In some embodiments, each beam marker of the plurality of beam markers has a respective boundary. In some embodiments, each beam marker of the plurality of beam markers share a common location. In some embodiments, the method further comprises displaying an indication of the first beam marker. In some embodiments, the selecting the first beam marker of the plurality of beam markers is based on the first beam marker having a highest ranking. In some embodiments, the first beam marker has a first topography and a first boundary, wherein the first topography and the first boundary result in less radiation applied to organs of the subject than respective topographies and respective boundaries of each other beam marker of the plurality of beam markers. In some embodiments, the method further comprises applying the dose of radiation based on the first beam marker. In some embodiments, the dose of radiation is associated with a dose radiation gradient, wherein an organ disposed within the dose radiation gradient receives a tolerable amount of radiation. In some embodiments, the method further comprises displaying the plurality of beam markers on a user interface. In some embodiments, the reduced dose of radiation applied to the one or more organs is no more than 90% of the therapeutically effective dose of radiation applied to the tumor. In some embodiments, the reduced dose of radiation applied to the one or more organs is no more than 50% of the therapeutically effective dose of radiation applied to the tumor. In some embodiments, the reduced dose of radiation applied to the one or more organs is no more than 30% of the therapeutically effective dose of radiation applied to the tumor. In some embodiments, the point of administration is indicated by the first beam marker.

[0010] Another aspect provided herein is a method of evaluating placement of a composition, comprising: detecting one or more organs and a tumor within a compartment of a human subject; determining, using a machine learning model, a first space of the compartment based on the one or more organs and the tumor; detecting the composition displaced in a second space of the compartment; comparing the first space to the second space; determining a score based on the comparing; and providing the score to a user.

[0011] In some embodiments, the method further comprises displacing the composition in the second section. In some embodiments, determining the first space based on the one or more organs and the tumor is based on a relationship between the first space and the one or more organs and the tumor. In some embodiments, the method further comprises, determining, using the machine learning model: a distance between the first space and the one or more organs; a distance between the first space and the tumor; a shape associated with the first space; a volumeassociated with the first space; a distance between the second space and the one or more organs; a distance between the second space and the tumor; a shape associated with the second space; a volume associated with the second space; or any combination thereof. In some embodiments, the relationship between the first space and the one or more organs and the tumor is based on one of more of the distance between the first space and the one or more organs; the distance between the first space and the tumor; the shape associated with the first space; or the volume associated with the first space. In some embodiments, the method further comprises determining a relationship between the second space and the one or more organs and the tumor. In some embodiments, determining the relationship between the second space and the one or more organs and the tumor is based on: the distance between the second space and the one or more organs; the distance between the second space and the tumor; the shape associated with the second space; the volume associated with the second space; or any combination thereof. In some embodiments, the first space and the second space at least partially overlap. In some embodiments, the first space and the second space do not overlap. In some embodiments, the first space is selected from a plurality of spaces based on: the first space being associated with a reduced dose of radiation applied to the one or more organs relative to another dose of radiation applied to the one or more organs associated with another space of the plurality of spaces; the first space being associated with a therapeutically effective dose of radiation applied to the tumor; or both. In some embodiments, the method further comprises displaying one or more of the first space or the second space on a user interface. In some embodiments, the one or more metrics are used to determine: the distance between the first space and the one or more organs; the distance between the first space and the tumor; the shape of the gel composition associated with the first space; the volume of the gel compositions associated with the first space; the distance between the second space and the one or more organs; the distance between the second space and the tumor; the shape of the gel composition associated with the second space; the volume of the gel compositions associated with the second space; or any combination thereof. In some embodiments, the one or more metrics comprise: a method of training a machine learning model, comprising: detecting one or more organs and a tumor in a compartment in a human subject; generating a plurality of spaces of the compartment based on the one or more organs and the tumor in the compartment; determining one or more relationships between each space of the plurality of spaces and the one or more organs and the tumor based on one or more of: a distance between each space of the plurality of spaces and the one or more organs; a distance between each space of the plurality of spaces and the tumor; a shape of each space of the plurality of spaces; or a volume of each space of the plurality of spaces; receiving a first space of the plurality of spaces associated with a first relationship of the one or more relationships; andadjusting the one or more relationships based on receiving the first space, thereby training the machine learning model. In some embodiments, the method further comprises ranking the spaces of the plurality of spaces based on the one or more relationships. In some embodiments, each space of the plurality of spaces is associated with a dose of radiation applied to the one or more organs. In some embodiments, the first space is associated with a reduced dose of radiation applied to the one or more organs relative to doses of radiation applied to the one or more organs associated with other spaces of the plurality of spaces. In some embodiments, the first space is associated with a therapeutically effective dose of radiation applied to the tumor. In some embodiments, training the machine learning model comprises training the machine learning model to select a space associated with a reduced dose of radiation applied to the one or more organs and a therapeutically effective dose of radiation applied to the tumor. In some embodiments, at least two spaces of the plurality of spaces overlap. In some embodiments, at least two spaces of the plurality of spaces do not overlap. In some embodiments, the at least two spaces comprises the first space. In some embodiments, the composition comprises: a viscoelastic medium; and a first visual additive, wherein said first visual additive comprises a metal, and wherein said metal has a particle diameter of greater than or equal to 80 micrometers (pm). In some embodiments, said metal is a precious metal. In some embodiments, said metal or said precious metal is chosen from the group consisting of gold (Au), iodine (I), gadolinium (Gd), iron (Fe), barium (Ba), calcium (Ca), magnesium (Mg), and combinations thereof. In some embodiments, said metal or precious metal is an isotope of gold (Au), iodine (I), gadolinium (Gd), iron (Fe), barium (Ba), calcium (Ca), magnesium (Mg), or combinations thereof. In some embodiments, said metal or precious metal is a powder. In some embodiments, said first visual additive has a radiographic density of between about 1.0 g / cm3 and 2.0 g / cm3. In some embodiments, half of said first visual additive is configured to disperse within a tissue within nine months. In some embodiments, said particle diameter of said metal or precious metal is between about 80 pm and about 120 pm. In some embodiments, said particle diameter of said metal or precious metal is about 100 pm. In some embodiments, said first visual additive further comprises one or more microbubbles. In some embodiments, the method further comprises a second visual additive. In some embodiments, said second visual additive is different than said first visual additive. In some embodiments, said second visual additive comprises one or more microbubbles. In some embodiments, said first visual additive has a concentration within said viscoelastic medium of greater than 5 milligrams per milliliter (mg / ml). In some embodiments, said first visual additive has a concentration within said viscoelastic medium of less than 90 mg / ml. In some embodiments, said first visual additive has a concentration within said viscoelastic medium between 15 mg / ml and 30 mg / ml. In some embodiments, said metal isbetween about 0.5 wt% and about 9.0 wt% of said composition. In some embodiments, said metal or precious metal is between about 0.015 wt% and about 1.5 wt% of said composition. In some embodiments, said viscoelastic medium comprises a volume of about 1 milliliter (ml) to about 50 ml. In some embodiments, said composition is configured to be biodegradable. In some embodiments, said composition is configured to be present on an imaging modality for at least 9 months. In some embodiments, said composition is configured to not substantially migrate prior to or during imaging. In some embodiments, said first visual additive is configured to not substantially migrate prior to or during imaging. In some embodiments, said composition is configured to be disposed within a subject. In some embodiments, said subject is in need of radiography. In some embodiments, said composition is configured to be disposed through injection. In some embodiments, said composition is configured to be disposed subcutaneously or subepidermally. In some embodiments, said composition is configured to be disposed within a compartment. In some embodiments, said compartment comprises one or more of, a fat tissue, a muscle tissue, and organ tissue, or a combination thereof. In some embodiments, said composition is configured to be imaged on one or more modalities. In some embodiments, said one or more modalities comprise X-Ray, MRI, CT, CBCT, ultrasound, PET, SPECT or a combination thereof. In some embodiments, said imaging comprises real-time imaging. In some embodiments, said composition is configured to be imaged within 30 min, within 90 min, within 4 hours, within 8 hours, or within 4 days of disposition. In some embodiments, said viscoelastic medium comprises hyaluronic acid, polyethylene glycol, or dextranomers. In some embodiments, said viscoelastic medium comprises hyaluronic acid, polyethylene glycol, or dextranomers at a concentration between about 5 mg / ml to about 100 mg / ml. In some embodiments, said viscoelastic medium comprises gel particles at a size range of about 0.08 mm to about 5 mm. In some embodiments, said viscoelastic medium comprises non-animal stabilized hyaluronic acid (“NASHA”). In some embodiments, said viscoelastic medium expands within said compartment to less than 10% of an original disposition volume. In some embodiments, said viscoelastic medium is injected one time every six months. In some embodiments, said viscoelastic medium is completely resorbed within 20 months. In some embodiments, said viscoelastic medium is completely resorbed within 16 months. In some embodiments, said viscoelastic medium is completely resorbed within 12 months.

[0012] Another aspect provided herein is a method of determining an object placement in a compartment of a human subject, comprising: receiving an image of the compartment, wherein the image contains one or more objects of the compartment; determining, using a machine learning model, a location and a value of a metric of the one or more objects of the compartment; determining, using the machine learning model, a placement of a new object to beplaced in the compartment based on the location and the value of the metric of the one or more objects; providing an indication of the placement of the new object in the compartment. In some embodiments, the placement comprises a location of the new object. In some embodiments, the placement comprises a shape of the new object. In some embodiments, the metric comprises proximity of the one or more objects to each other, shape of the one or more objects, size measurements of the one or more objects at one or more locations, contours of the one or more objects, volume of the one or more objects, dosimetric impact on the one or more objects. In some embodiments, the shape comprises a measure of symmetry. In some embodiments, the method further comprises calculating the measure of symmetry by: identifying a set of locations and a direction to measure on the one or more objects; assessing a measurement at the set of locations in the direction; calculating a measure of symmetry based at least in part on the measurement at the set of locations. In some embodiments, the set of locations comprises an indication of the prostate mid-gland, an indication approximately 1 cm superior to the indication of the prostate mid-glad, an indication approximately 1 cm inferior to the indication of the prostate mid-gland. In some embodiments, the direction is a transverse slice. In some embodiments, the method further comprises injecting a composition at the placement of the new object in the compartment. In some embodiments, the composition comprises a gel spacer. In some embodiments, the determining the placement of the new object is based at least in part on user input. In some embodiments, the user input comprises at least one of an image, a text string indicating a portion of the human subject, an indication of desired new position of the one or more objects. In some embodiments, the machine learning model takes as input at least one of a dose-volume histogram set, an extracted contour set, or a set of metadata. In some embodiments, the set of metadata is based, at least in part, on at least one of DICOM RT struct files or RTdose files. In some embodiments, the extracted contour set comprises rectal volume. In some embodiments, the volume of the one or more objects comprises a volume of the rectum. In some embodiments, the volume of the rectum is determined through a measurement derived from the image. In some embodiments, the measurement derived from the image comprises an anatomical region from the closer of rectosigmoid flexure or bottom of sacroiliac joint to the inferior extent of the ischial tuberosities. In some embodiments, the contours of one or more objects comprises a set of spacer contours. In some embodiments, the set of locations further comprises a location at a variable distance from the other locations in the set of locations, the distance based at least in part on the size of the prostate. In some embodiments, a mean superiorinferior length of the new object is calculated from the set of locations. In some embodiments, the method further comprises providing instruction for printing the indication. In some embodiments, the printing comprises 3d printing.

[0013] Another aspect of the present disclosure provides a non-transitory computer readable medium comprising machine executable code that, upon execution by one or more computer processors, implements any of the methods above or elsewhere herein.

[0014] Another aspect of the present disclosure provides a system comprising one or more computer processors and computer memory coupled thereto. The computer memory comprises machine executable code that, upon execution by the one or more computer processors, implements any of the methods above or elsewhere herein.BRIEF DESCRIPTION OF THE DRAWINGS

[0015] The novel features of the disclosure are set forth with particularity in the appended claims. A better understanding of the features and advantages of the present disclosure will be obtained by reference to the following detailed description that sets forth illustrative embodiments, in which the principles of the disclosure are utilized, and the accompanying drawings of which:

[0016] FIG. 1 shows a diagram of an exemplary method of classifying one or more objects in an image, in accordance with some embodiments.

[0017] FIG. 2 shows a diagram of an exemplary method of training a machine learning model, in accordance with some embodiments.

[0018] FIG. 3 shows a diagram of an exemplary method of displacing a gel composition, in accordance with some embodiments.

[0019] FIG. 4 shows a diagram of an exemplary method of reducing a dose of radiation applied to one or more organs within a compartment of a human subject, in accordance with some embodiments.

[0020] FIG. 5 shows a diagram of an exemplary method of evaluating placement of a composition, in accordance with some embodiments.

[0021] FIG. 6A shows an illustration of a rectum and a prostate before injection of a spacer, in accordance with some embodiments.

[0022] FIG. 6B shows an illustration of a rectum and a radiated prostate after injection of a spacer, in accordance with some embodiments.

[0023] FIG. 7 shows a cross-sectional illustration of male anatomy after injection of a spacer, in accordance with some embodiments.

[0024] FIG. 8A shows an image of a first exemplary computerized tomography (CT) scan of a rectum and a prostate with a spacer injected therebetween, in accordance with some embodiments.

[0025] FIG. 8B shows an image of a second exemplary computerized tomography (CT) scan of a rectum and a prostate with a spacer injected therebetween, in accordance with some embodiments.

[0026] FIG. 9A shows an image of a third exemplary computerized tomography (CT) scan of a rectum and a prostate with a spacer injected therebetween, in accordance with some embodiments.

[0027] FIG. 9B shows a window and level optimized image of the third exemplary computerized tomography (CT) scan, in accordance with some embodiments.

[0028] FIG. 10A shows an image of a fourth exemplary computerized tomography (CT) scan of a rectum and a prostate with a spacer injected therebetween, in accordance with some embodiments;

[0029] FIG. 10B shows a window and level optimized image of the fourth exemplary computerized tomography (CT) scan, in accordance with some embodiments.

[0030] FIG. 11 shows an image of a fourth exemplary computerized tomography (CT) scan of a rectum and a prostate with dose gradients, in accordance with some embodiments.

[0031] FIG. 12 shows an image of a fifth exemplary computerized tomography (CT) scan of a rectum and a prostate with dose gradients, in accordance with some embodiments.

[0032] FIG. 13 shows a non-limiting example of a computing device; in this case, a device with one or more processors, memory, storage, and a network interface, in accordance with some embodiments.

[0033] FIG. 14 shows a non-limiting example of a web / mobile application provision system; in this case, a system providing browser-based and / or native mobile user interfaces, in accordance with some embodiments.

[0034] FIG. 15 shows a non-limiting example of a cloud-based web / mobile application provision system; in this case, a system comprising an elastically load balanced, auto-scaling web server and application server resources as well synchronously replicated databases, in accordance with some embodiments.

[0035] FIG. 16 is a schematic diagram illustrating a portion of an exemplary assessment model based on a Random Forest classifier, in accordance with some embodiments.

[0036] FIG. 17 is an exemplary operational flow of a prediction module as described herein, in accordance with some embodiments, in accordance with some embodiments.

[0037] FIG. 18 shows a non-limiting example of a schematic of the interconnectivity of computing devices, devices and servers.

[0038] FIG. 19 shows an image of a sixth exemplary computerized tomography (CT) scan of a rectum and a prostate with a spacer injected therebetween, in accordance with some embodiments, lines are indicative of isodoses.

[0039] FIG. 20 shows an exemplary method for determining a desired placement of a new object within an image.

[0040] FIG. 21 shows an exemplary method for predicting a desired gel composition placement.DETAILED DESCRIPTION

[0041] Provided herein are methods for decreasing the toxicity of advanced ablative cancer therapies on neighboring organs. The methods herein provide spacing between single or multiple tumor cites and immediate healthy organs while maintaining or increasing patient quality of life. Such toxicity isolation can be performed by inserting a spacer around the one or more tumor cites, which can be performed concurrently with fiducial marker placement.

[0042] In conventional methods, after inserting the spacer, the spacer may still be difficult to identify in images of the subject that include the spacer due to poor contrast that may be generated between the spacer and surrounding tissue.

[0043] The systems and methods described herein provide an increased ability to view the spacer and surrounding tissue as well as an ability to efficiently determine what is in the image at large (e.g., such as tumors and organs, in addition to the spacer.) For example, an image with poor contrast between the spacer and other objects in the image may be received, but the image may then be refined (e.g., by improving the contrast between objects or increasing the visibility of borders of one or more objects) in order to improve the contrast between the objects and improve overall visibility. Even further, a machine learning model can be used to identify which objects are present in the image and can classify each of them before passing the refined image, along with the classifications, for the subject or a person associated with the subject to review. By improving the details of the image, along with classifying the objects, analyzing the image and classifying its contents becomes a more efficient and less resource intensive-process over conventional methods, as the time to analyze the image could be reduced from hours to second or minutes.

[0044] Therefore, with the systems and methods described herein include an improved method of analysis of an image to determine where a spacer is located within a subject along with the organs and / or tumors of the subject by using machine learning to refine images and / or classify the spacer, organs, and / or tumors.

[0045] Additionally to above, qualities of the spacer such as shape symmetry and temporal stability of the space may be improved to improve the benefit of the spacer such as decreased exposure to radiation of the neighboring tissues / organs, improved quality of life and decreased toxicity. Improved materials may also be used to produce the spacer may also be used to improve quality of life through less negative interaction with tissues and organs such as chafing.

[0046] Advantageously, these improvements may be implemented using the systems and methods as described herein. In some embodiments, the aspects of spacers may be based on a scan of the patient, allowing for qualities such as shape and symmetry of the spacer to be determined more precisely and quicker than previously possible, thus allowing for greater improvements in the quality of life of patients and saving those patients from potential future discomfort.Subcutaneous Spacer Materials

[0047] The subcutaneous spacer materials herein are configured to form a cavity adjacent to prevent radiation or toxicity damage to organs proximal to or in contact with a treatment organ. The subcutaneous spacer materials herein may comprise a viscoelastic media comprising hyaluronic acid particles. The particle size and concentration of the hyaluronic acid within the spacer material can be tuned to exhibit a hardness, density, or both to enable consistent and uniform injection and cavity formation.

[0048] In some embodiments, the implant comprises particles of one or more viscoelastic media dispersed in a physiological salt buffer, a suitable physiological salt solvent, or both. In some embodiments, the implant further comprises other additives, such as local anesthetics, antiinflammatory drugs, antibiotics and supportive medications (e.g. bone growth factors or cells). In some embodiments, there may also be included a viscoelastic medium which may be formed of same material as the particles or a different material than the particles. In some embodiments, the viscoelastic medium is not present as particles.

[0049] Viscoelastic media according to embodiments can herein include gels, dispersions, solutions, suspensions, slurries and mixtures thereof. In some embodiments, the medium is present as a dispersion of gel or gel-like particles. The viscoelastic medium provided herein can be more resistant to biodegradation in vivo than natural hyaluronic acid. The prolonged presence of the stable viscoelastic substance is advantageous for the patient, since the time between treatments is increased. The viscoelastic media herein can be biocompatible, sterile, and present as particles.

[0050] Advantageously, the viscoelastic media herein are stable within, but can be impermanent under, physiological conditions. In some embodiments, about 70% to about 90%, of theviscoelastic medium remains for at least two weeks in vivo. In some embodiments, at least 70%, of the viscoelastic medium remains for at about two weeks and two years in vivo. In some embodiments, at least 90%, of the viscoelastic medium remains for at about two weeks and two years in vivo. The viscoelastic medium can degrade automatically after five years or more in vivo.

[0051] Viscoelastic media include, without being limited thereto, polysaccharides and derivatives thereof. Suitable viscoelastic media include stabilized starch and derivatives thereof. Suitable viscoelastic media can also be selected from stabilized glycosaminoglycans and derivatives thereof, such as stabilized hyaluronic acid, stabilized chondroitin sulfate, stabilized heparin, and derivatives thereof. Suitable viscoelastic media also include stabilized dextran and derivatives thereof, such as dextranomer. In some embodiments, the dextranomer (Dx) has a molecular weight of about 40 kDa to about 70 kDa. In some embodiments, the Dx has a molecular weight of at most about 70 kDa. In some embodiments, the Dx has a molecular weight of at least about 40 kDa.. In one example the viscoelastic media comprises hyaluronic acid and Dx. In another example the viscoelastic media comprises non-animal stabilized hyaluronic acid (NASHA) / Dx gel with de. In some embodiments, the non / Dx gel has a sufficiently low viscosity such that it can be injected through a syringe by finger pressure alone. In some embodiments, the NASHA / Dx gel has a sufficiently high viscosity to avoid leakage from the injection site. In some embodiments, the NASHA / Dx gel has a long degradation time which enables and stabilizes the natural formation of connective tissue at the site of the implant. In some embodiments, the viscoelastic medium further comprises carbon-coated zirconium beads, calcium hydroxylapatite, or both. In some embodiments, the viscoelastic medium is cross-linked hyaluronic acid, or a derivatives thereof. An example of a viscoelastic medium is NASHA. One type of suitable cross-linked hyaluronic acid is obtainable by cross-linking of hyaluronic acid. The viscoelastic medium may also be a combination of two or more of the suitable viscoelastic media listed herein or otherwise known to the art. The viscoelastic medium may be of non-animal origin.

[0052] In some embodiments, the viscoelastic medium comprises a hydrogel. In some embodiments, the hydrogel is formed from natural, synthetic, or biosynthetic polymers. In some embodiments, the natural polymer comprises glycosminoglycans, polysaccharides, proteins, or any combination thereof. In some embodiments, the glycosaminoglycan is dermatan sulfate, hyaluronic acid, chondroitin sulfate, chitin, heparin, keratan sulfate, keratosulfate, or any combination thereof. In some embodiments, the hydrogel comprises an acidic carboxy polymer, an acrylic acid-based polymer, a polyacrylamide, a starch graft copolymer, an acrylate polymer, or any combination thereof. In some embodiments, the hydrogel comprises allylpentaerythritol,polyacrylic acid, ester cross-linked polyglucan, or any combination thereof. In some embodiments, the viscoelastic media is hydrophilic.

[0053] The size of the gel particles can depend upon the ionic strength of the buffer, the solution, carrier, or any combination thereof that is included in and / or surrounding the gel particles. As such, given particle sizes can assume physiological conditions, particularly isotonic conditions. In some embodiments, the gel particles contain and are dispersed in a physiological salt solution. In some embodiments, the gel particles are temporarily brought to different sizes by subjecting the gel particles to a solution of another tonicity. Particle sizes within the given ranges under physiological conditions when implanted subepidermally in the body or when subjected to a physiological, or isotonic, salt solution (i.e. a solution with the same tonicity as the relevant biological fluids, such as an iso-osmotic with serum).

[0054] In some embodiments, the particles have a specific tuned size. The size of the particles can be achieved by producing a gel made of a viscoelastic medium at a desired concentration and subjecting the gel to a physical disruption. The physical disruption can comprise: mincing, mashing filtering, or any combination thereof. The resulting gel particles can be dispersed in a physiological salt solution, resulting in a gel dispersion or slurry with particles of desired size. Particle size may be determined in any suitable way, such as by laser diffraction, microscopy, or filtration, etc. In some embodiments, the specific shape of the gel particles is not critical. The size of a spherical particle can equal its diameter. The size may be measured as an average size, a median size, a maximum size, or a minimum size.

[0055] In some embodiments, the particles have a size in the range of from 1 to 2.5 mm, such as from 1.5 to 2 mm, in the presence of a physiological salt solution. In some embodiments, the particles have a size in the range of from 2.5 to 5 mm, such as from 3 to 4 mm, in the presence of a physiological salt solution. At least 50% (v / v) of the particles can have a size of at least about 1 mm. At least 50% (v / v) of the particles can have a size of about 1-5 mm in the presence of a physiological salt solution. In some embodiments, more than 70% (v / v) of the particles are within the given size limits under physiological conditions. In some embodiments, more than 90% (v / v) of the particles are within the given size limits under physiological conditions. Administration of the implant employing the method according to an embodiment herein prevents or diminishes migration and / or displacement of the implant, which comprises or consists of the 1-5 mm large particles under physiological conditions. Large particles can exhibit less in vitro migration and can be more easily removed. In some embodiments, the viscoelastic medium is not present as particles of a size smaller than 0.1 mm. In some embodiments, the Dx is composed of microspheres. In some embodiments, the microspheres have a diameter of about 80 pm to about 250 pm. In some embodiments, the microspheres have a diameter of at leastabout 80 pm. In some embodiments, the microspheres have a diameter of at most about 250 pm. In some embodiments, the Dx is composed of microspheres. In some embodiments, the microspheres have a diameter of about 80 pm to about 250 pm. In some embodiments, the microspheres have a diameter of at least about 80 pm. In some embodiments, the microspheres have a diameter of at most about 250 pm.

[0056] In some embodiments, the particles have a specific tuned density, hardness or both. The gel particle density can be regulated by adjusting the concentration of the viscoelastic medium, the amount and type of cross-linking agent, or both. Harder particles can be achieved by increased concentration of the viscoelastic medium in the gel. Harder particles can be less viscoelastic and exhibit a longer half-life in vivo than softer particles. The particles herein should retain enough viscoelastic properties that they can be safely injected. In some embodiments, the implant comprises both soft gel particles and harder gel particles. The soft and hard gel particles may be made of the same or different viscoelastic media. The resulting mixture of gel particles combines desirable properties of softness / hardness for use in radiative protection and long durability in vivo.

[0057] Hyaluronic acid is a natural component of biologic extracellular and connective matrix for the maintenance of proper structure and function of tissues by creating volume, lubricating tissues and enhancing mobility. The physiological function of hyaluronic acid stems from the large size and hydrodynamic volume of its hydrophilic molecular network. This glycosaminoglycan-based polymer forms a mesh with the capacity to hold large amounts of water. The unique aspect of compositions described herein is the way the hyaluronic acid molecules are joined and entangled to create an especially stable 3-dimensional molecular network, which defines it as a gel.

[0058] The gel may be injected into the patient’s perirectal space under transrectal ultrasound (TRUS) guidance. The gel appears dark (hypoechoic) and is easily visible on TRUS. In some embodiments, the gel remains malleable and may be sculpted with the needle to achieve an even and symmetric implant around the posterior prostate. In some embodiments, the goal of injecting the gel is to create approximately 1 cm or more of separation from base to apex, and in some cases, separation around the seminal vesicles may be desired. Precise measurements of the space may be taken during the procedure on the TRUS image, which, in addition to the total volume injected, may be of use to the treatment planning team as they contour.

[0059] Since the gel is composed of biodegradable material, it may be completely absorbed by the patient’s body over time. It has been formulated specifically to maintain stable space for the entire course of prostate radiotherapy treatment.

[0060] FIGS. 6A and 6B show illustrations of a rectum 620 and a prostate 610 before and after injection of a spacer 630. As shown, in this example, the spacer 630 protects the rectum 620 from a zone of radiation 640 directed to the prostate 610. Further, as shown in FIG. 7, the spacer 630 protects the rectum 620, the bladder 650, and the penis 660 from radiation applied to the prostate 610.Methods of Forming a Spacer Material

[0061] The subcutaneous spacer materials herein are configured to form a cavity adjacent to prevent radiation or toxicity damage to organs proximal to or in contact with a treatment organ. The subcutaneous spacer materials herein may comprise a viscoelastic media comprising hyaluronic acid particles. The particle size and concentration of the hyaluronic acid within the spacer material can be tuned to exhibit a hardness, density, or both to enable consistent and uniform injection and cavity formation.

[0062] Provided herein are methods of forming a spacer material comprising forming an aqueous solution comprising: a water soluble cross-linkable polysaccharide; initiating a crosslinking of the polysaccharide in the presence of a polyfunctional cross-linking agent; sterically hindering the cross-linking reaction from terminating before gelation occurs to generate activated polysaccharide; and reintroducing the sterically unhindered conditions for the activated polysaccharide to continue the cross-linking thereof up to a viscoelastic gel. In some embodiments, the initial cross-linking reaction in the presence of a polyfunctional cross-linking agent can be performed at varying pH values, primarily depending on whether ether or ester reactions should be promoted.

[0063] The cross-linking agent can be any previously known cross-linking agent useful in connection with polysaccharides that are biocompatibile. However, the cross-linking agent comprises: aldehydes, epoxides, polyaziridyl compounds, glycidyl ethers, di vinyl sulfones, or any combination thereof. Glycidyl ethers represent an group, of which 1,4-butanediol di glycidyl ether can be advantageous. In some embodiments, the spacer material comprises a hydrogel comprising a glycosaminoglycan that is extracted from a natural source that is purified and derivatized. In some embodiments, the glycosaminoglycan is synthetically produced or synthesized by modified microorganisms such as bacteria. In some embodiments, the glycosaminoglycan is modified synthetically from a naturally soluble state to a partially soluble or water swellable or hydrogel state.

[0064] A suitable way of obtaining a desired particle size involves producing a gel made of cross-linked hyaluronic acid at a desired concentration and subjecting the gel to physical disruption, such as mincing, mashing or allowing the gel to pass through a filter with suitableparticle size. The resulting gel particles are dispersed in a physiological salt solution, resulting in a gel dispersion or slurry with particles of desired size. The size of the particles can be achieved by producing a gel made of a viscoelastic medium at a desired concentration, and subjecting the gel to a physical disruption. The physical disruption can comprise: mincing, mashing filtering, or any combination thereof. The resulting gel particles can be dispersed in a physiological salt solution, resulting in a gel dispersion or slurry with particles of desired size.

[0065] In some embodiments, the particles have a specific tuned density, hardness or both. The gel particle density can be regulated by adjusting the concentration of the viscoelastic medium, the amount and type of cross-linking agent, or both. Harder particles can be achieved by increased concentration of the viscoelastic medium in the gel. By varying the hyaluronic acid concentrations to, for example, 20, 25, 40, 50 and 100 mg / ml gel particles of varying hardness can be obtained. Harder particles can be less viscoelastic and exhibit a longer half-life in vivo than softer particles. The particles herein should retain enough viscoelastic properties that they can be safely injected.

[0066] In some embodiments, the implant comprises both soft gel particles and harder gel particles. The soft and hard gel particles may be made of the same or different viscoelastic media. The resulting mixture of gel particles combines desirable properties of softness / hardness for use in radiative protection and long durability in vivo. In one embodiment the soft gel particles comprise 15-22 mg / ml of the cross-linked hyaluronic acid, and the hard gel particles comprise 22-30 mg / ml of the cross-linked hyaluronic acid.

[0067] When the injectable medium is a hyaluronic acid medium, the hyaluronic acid concentration can be at least about 5 mg / ml. In some embodiments, the hyaluronic acid concentration is about 5 mg / ml to about 100 mg / ml. In some embodiments, the hyaluronic acid concentration is about 10 to about 50 mg / ml. In some embodiments, the hyaluronic acid concentration is about 20 mg / ml. The cross-linked hyaluronic acid can be present as particles or beads of any form.

[0068] In some embodiments, the method further comprises adding an image enhancement agent described herein to the spacer material. In some embodiments, the method further comprises mixing in an image enhancement agent described herein to the spacer material.Methods of Injecting a Spacer Material

[0069] The subcutaneous spacer materials herein are configured to form a cavity adjacent to prevent radiation or toxicity damage to organs proximal to or in contact with a treatment organ. The subcutaneous spacer materials herein may comprise a viscoelastic media comprising hyaluronic acid particles. The particle size and concentration of the hyaluronic acid within thespacer material can be tuned to exhibit a hardness, density, or both to enable consistent and uniform injection and cavity formation. Further, a specific needle size can be used to deliver the spacer material to its intended in vivo location based on the particle size, hardness, density, and concentration of the hyaluronic acid.

[0070] Provided herein is a method of injecting a spacing material. The spacing material can comprise a viscoelastic medium for therapeutic radiative protection in a mammal, including man. The spacing material can be suitable for subepidermal administration at a site in said mammal where therapeutic soft tissue protection is required from radiation or other toxic sources. In particular, the particles are suitable for administration to tissues covered by publicly exposed skin, such as facial tissue, as the particles do not cause bruises or other discolorations. The particles herein are suitable for administration into deep subcutaneous or to submuscular / supraperiostal tissue, optionally in more than one layer. Deep subcutaneous or submuscular / supraperiostal administration can further prevent or diminished migration of the particles away from the desired site.

[0071] The spacing material can be administered by injection under the epidermis in any suitable way. By way of example, a dermal incision can be made with a scalpel or a sharp injection needle to facilitate transdermal insertion of a larger cannula for administration of the implant at the desired site.

[0072] The implant, consisting of particles of a viscoelastic medium and optionally other suitable ingredients, may be administered as a single aliquot or as layers of multiple aliquots. Optionally, the viscoelastic medium may be replaced, refilled or replenished by a subsequent injection of the same or another viscoelastic medium. The injected volume is determined by the size of the desired cavity.

[0073] In some embodiments, a volume of the spacer material that is injected is about 1 ml to about 500 ml. In some embodiments, a volume of the spacer material that is injected is about 1 ml to about 5 ml, about 1 ml to about 10 ml, about 1 ml to about 25 ml, about 1 ml to about 50 ml, about 1 ml to about 100 ml, about 1 ml to about 150 ml, about 1 ml to about 200 ml, about 1 ml to about 250 ml, about 1 ml to about 300 ml, about 1 ml to about 400 ml, about 1 ml to about 500 ml, about 5 ml to about 10 ml, about 5 ml to about 25 ml, about 5 ml to about 50 ml, about 5 ml to about 100 ml, about 5 ml to about 150 ml, about 5 ml to about 200 ml, about 5 ml to about 250 ml, about 5 ml to about 300 ml, about 5 ml to about 400 ml, about 5 ml to about 500 ml, about 10 ml to about 25 ml, about 10 ml to about 50 ml, about 10 ml to about 100 ml, about 10 ml to about 150 ml, about 10 ml to about 200 ml, about 10 ml to about 250 ml, about 10 ml to about 300 ml, about 10 ml to about 400 ml, about 10 ml to about 500 ml, about 25 ml to about 50 ml, about 25 ml to about 100 ml, about 25 ml to about 150 ml, about 25 ml to about 200 ml,about 25 ml to about 250 ml, about 25 ml to about 300 ml, about 25 ml to about 400 ml, about 25 ml to about 500 ml, about 50 ml to about 100 ml, about 50 ml to about 150 ml, about 50 ml to about 200 ml, about 50 ml to about 250 ml, about 50 ml to about 300 ml, about 50 ml to about 400 ml, about 50 ml to about 500 ml, about 100 ml to about 150 ml, about 100 ml to about 200 ml, about 100 ml to about 250 ml, about 100 ml to about 300 ml, about 100 ml to about 400 ml, about 100 ml to about 500 ml, about 150 ml to about 200 ml, about 150 ml to about 250 ml, about 150 ml to about 300 ml, about 150 ml to about 400 ml, about 150 ml to about 500 ml, about 200 ml to about 250 ml, about 200 ml to about 300 ml, about 200 ml to about 400 ml, about 200 ml to about 500 ml, about 250 ml to about 300 ml, about 250 ml to about 400 ml, about 250 ml to about 500 ml, about 300 ml to about 400 ml, about 300 ml to about 500 ml, or about 400 ml to about 500 ml. In some embodiments, a volume of the spacer material that is injected is about 1 ml, about 5 ml, about 10 ml, about 25 ml, about 50 ml, about 100 ml, about 150 ml, about 200 ml, about 250 ml, about 300 ml, about 400 ml, or about 500 ml. In some embodiments, a volume of the spacer material that is injected is at least about 1 ml, about 5 ml, about 10 ml, about 25 ml, about 50 ml, about 100 ml, about 150 ml, about 200 ml, about 250 ml, about 300 ml, or about 400 ml. In some embodiments, a volume of the spacer material that is injected is at most about 5 ml, about 10 ml, about 25 ml, about 50 ml, about 100 ml, about 150 ml, about 200 ml, about 250 ml, about 300 ml, about 400 ml, or about 500 ml.

[0074] Administration may be performed in any suitable way, such as via injection from standard cannula and needles of appropriate sizes. The administration is performed where the radiative protection is desired, such as the chin, cheeks or elsewhere in the face or body.

[0075] The spacing material herein is injectable through standard needles used in medicine, such as 20 gauge or larger needles. Alternatively, the spacing material comprising hyaluronic acid can be injected using any of the following sized needles:

[0076] In some embodiments, an interior surface of the needle comprises a protrusion, a mesh, a constriction, or any combination thereof. In some embodiments, injecting the spacing material past the protrusion, the mesh, the constriction, or any combination thereof produces a gas bubble in the spacing material. In some embodiments, the gas bubble is a microbubble. In some embodiments, the mesh has a mesh spacing of about 20 pm to about 300 pm. In some embodiments, a size of the mesh spacing determines a size of the microbubbles produced thereby.

[0077] In some embodiments of the systems and methods described herein, aspects of the spacer may be determined because the spacer is injected into a subject. For example, as described herein, an image (e.g., a scan) of an area of the subject may be captured, where the area is includes where the spacer is intended to be input. The scan may show one or more objects within the area (e.g., organs and / or tumors within the subject). Based on the one or more objects, aspects of the spacer, such as the shape and symmetry, may be determined using one or more techniques, with the goal of improving the quality of life of the subject. For example, the shape and symmetry of the spacer may be determined in order to prevent chafing of the spacer against one or more organs after the spacer has been injected. In some embodiments, the one or more techniques may include machine learning techniques.Methods of Classifying One or More Objects in an Image

[0078] In one aspect, disclosed herein is a method of classifying one or more objects in an image. In some embodiments, the one or more objects may include a spacer, as described above. In some embodiments, the one or more objects may include at least one organ and / or at least one tumor within a subject, where the image may be captured before a spacer is injected into the subject. In some embodiments, the one or more objects may include at least one organ, at least one spacer, and / or at least one tumor within a subject, where the image may be captured after the spacer is injected into the subject. In some embodiments, per FIG. 1, the method comprises receiving an image of a compartment of a human subject, wherein the image contains one or more objects of the compartment 101. In some embodiments, the method further comprises detecting, using a machine learning model, the one or more objects of the compartment 102. In some embodiments, the method further comprises determining, using the machine learning model, a location and a value of a metric of at least a first object of the one or more objects 103. In some embodiments, the method further comprises identifying, using the machine learning model, the first object of the one or more objects as a class of a plurality of classes based on the value and the location, wherein the plurality of classes comprises an organ, a tumor, and a spacer104. In some embodiments, the method further comprises displaying the location and the class of the first object on a user interface 105.

[0079] In some embodiments, the metric is: a distance between the first object and at least a second object of the one or more objects; a radiosensitivity of the first object; a contrast to noise ratio of the first object; a shape of the first object; a volume of the first object; or any combination thereof. In some embodiments, the metric is the radiosensitivity of the first object. In some embodiments, the class of the first object is a spacer. In some embodiments, the method further comprises receiving user input from the computing device. In some embodiments, the identifying is further based on the user input.

[0080] In some embodiments, the user input comprises: the location of the first object in the image; a portion of the human subject associated with the image; a condition or disorder of the human subject; or any combination thereof. In some embodiments, the identifying is further based on a comparison of the value of the metric associated with first object to a value of the metric associated with a second object of the one or more objects. In some embodiments, the method further comprises receiving feedback from the computing device. In some embodiments, the method further comprises updating the machine learning model based on the feedback. In some embodiments, updating the machine learning model comprises adjusting one or more parameters of the machine learning model. In some embodiments, the feedback comprises: an updated class of the first object; a class of a second object of the one or more objects; a location of the second object of the one or more objects; an updated class of the first object; an updated location of the first object; or any combination thereof.Method of Training a Machine Learning Model

[0081] Another aspect provided herein is a method of training a machine learning model. In some embodiments, the machine learning model comprises one or more parameters. In some embodiments, per FIG. 2, the method comprises: receiving an image of a compartment of a human, wherein the image contains one or more objects of the compartment 201. In some embodiments, the method further comprises determining, using the machine learning model, a location and a value of a metric of each object of the one or more objects 202. In some embodiments, the method further comprises identifying, using the machine learning model, each object of the one or more objects as a class of a plurality of classes 203. In some embodiments, the method further comprises receiving an updated class of the plurality of class for at least one object of the one or more objects 204. In some embodiments, the method further comprises updating the one or more parameters of the machine learning model based on the updated class, thereby training the machine learning model 205.

[0082] In some embodiments, the machine learning model comprises one or more parameters.

[0083] FIG. 20 depicts an example method for determining a desired object placement. The method begins at step 2001 with receiving an image of a compartment of a human, wherein the image contains one or more objects of the compartment. In this depicted embodiment, the method further comprises determining, using the machine learning model, a location, and a value of a metric of each object of the one or more objects at step 2002. In this depicted embodiment, the method further comprises determining, using the machine learning model, a placement of a new object to be placed in the compartment based on the location and the value of the metric of the one or more objects at 2003. In this depicted embodiments, the method further comprises providing an indication of the placement of the new object in the compartment at step 2004. In some embodiments, the image of the compartment may be received as user input. In some embodiments, the image of the compartment may be received from a device that captured the image of the compartment. In some embodiments, the placement comprises a location and / or a shape of the new object. In some embodiments, the metric comprises a proximity of the one or more objects to each other, shape of the one or more objects, size measurements of the one or more objects at one or more locations, contours of the one or more objects, volume of the one or more objects, or dosimetric impact on the one or more objects. In some embodiments, the shape includes a measure of symmetry. In some embodiments, the measure of symmetry is calculated by identifying a set of locations and a direction to measure for the one or more objects, assessing a measurement at the set of locations using the direction of the one or more objects, and calculating the measure of symmetry based at least in part on the measurement at the set of locations. In some embodiments, the set of locations include a location of the prostate mid-gland, a location approximately 1 cm superior to the indication of the prostate mid-glad, and a location approximately 1 cm inferior to the indication of the prostate mid-gland. In some embodiments, the set of locations include a location of the prostate midgland, a location up to 5 cm superior to the indication of the prostate mid-glad, and a location up to 5 cm inferior to the indication of the prostate mid-gland. In some embodiments, the direction is a transverse slice of the new object. In some embodiments, the method further comprises injecting a composition to the subject at the location of the new object. In some embodiments, the composition comprises a gel spacer, as described further herein. In some embodiments, determining the placement of the new object is based on user input. In some embodiments, the user input may comprise at least one of an image (e.g., the image of the compartment), a text string indicating a portion of the human subject, or an indication of desired new position of the one or more objects. In some embodiments, the machine learning model receives at least one of a dose-volume histogram set, an extracted contour set, or a set of metadata as input. In someembodiments, the set of metadata is based, at least in part, on at least one of DICOM RT struct files or RTdose files. In some embodiments, the extracted contour set comprises rectal volume. In some embodiments, the volume of the rectum is determined through a measurement derived from the image. In some embodiments, the measurement derived from the image comprises an anatomical region from the closer of rectosigmoid flexure or bottom of sacroiliac joint to the inferior extent of the ischial tuberosities. In some embodiments, the contours of one or more objects comprises a set of spacer contours. In some embodiments, the set of locations further comprises a location at a variable distance from the other locations in the set of locations, the distance based at least in part on the size of the prostate. In some embodiments, a mean superiorinferior length of the new object is calculated from the set of locations.

[0084] In some embodiments, the method further comprises detecting, using the machine learning model, the one or more objects. In some embodiments, the metric is: a distance between each object of the one or more objects; a radiosensitivity of each object of the one or more objects; a contrast to noise ratio of each object of the one or more objects; a shape of each object of the one or more objects; a volume of each object of the one or more objects; or any combination thereof. In some embodiments, the plurality of classes comprises an organ, a tumor, or a spacer. In some embodiments, the updated type of the at least one object of the one or more objects indicates a spacer. In some embodiments, the identifying is further based on a comparison of the value of the metric associated with first object to a value of the metric associated with a second object of the one or more objects.

[0085] A machine learning model such as those disclosed here may be comprised of parameters, such as weights and biases, one or more processing steps, one or more outputs and one or more inputs. During training the machine learning model may calculate a loss useful for calculating the error between the real output of the model and the expected output of the model. Some set of the model parameters may be updated based at least in part on the loss calculation. The model may perform multiple rounds, or epochs, of training wherein an input or set of inputs is given and processed by the model which then produces an output or set of outputs which may then the basis for updating the weights. The new weights may be used in the next epoch. This is the basic schema for training a machine learning model and some embodiments may comprise more steps. Training may occur in different environments such as supervised, unsupervised, semisupervised, self-supervised or some combination thereof.

[0086] A machine learning model may be trained as a classifier or regression model. In both cases a model is trained as described above. A classifier is trained to give a value corresponding to a given class or set of classes. When one class is used the model is said to be a binary classifier. As an example a binary classifier may give a single value between 1 and 0 where 1indicates the presence of the desired class in the input and 0 indicates the absence of the desired class in the input. A value between 0 and 1 may indicate a probability of the desired class. In some cases the output value of a model may be compared against a threshold value, when the output is below the threshold the classifier outputs an indication that the desired class is not present, when the output is above the classifier outputs an indication that the desired class is present.

[0087] A classifier may also perform multiclass classification where more than one class is indicated in a binary fashion. In this case the model will output a single class as present and all other classes with be absent.

[0088] A classifier may be a multiclass multilabel, where more than one class may be output as present at one time. This may be useful in settings where classes may co-exist in the input. For example, an image segmentation model or object detection model may indicate the presence of multiple objects in a picture such as a rectum, a prostate, and a spacer. In this example the model may take as input an image without a spacer, the model would be expected to identify the rectum and the prostate while making no indication of a spacer being present.

[0089] A regression model may be used in a predictive fashion, whereas a classifier is used to place input or portions of input into classes that are predefined. Regression models may take an input and output a continuous value as a prediction or forecast score. As an example, a regression model may take an image and predict a desired set of values describing a shape of a new object to be placed in the image. In this example the output, or a portion of the output, of a regression may be used as an input to another model.

[0090] Once training is completed, a model may be used to infer on a set of inputs. The model output may be the desired output for the use of the model or there may be some portion of the model that is used for a desired output different than the output that was used during training time. At inference time the model’s weights may be static, whereas during training they may be dynamic as discussed previously.

[0091] A model may be trained again after first being trained. Such cases may include transfer learning applications, fine-tuning of the model, integration of the model into a larger model, or some combination thereof. During fine-tuning a trained model may be trained on a different set of data, a subset of the original data or some combination thereof to cause the model to improve its performance on a given task or subtask to that on which it was previously trained. In fine tuning a set of model parameters may be untrainable often so that the previously learned information represented by those parameters is maintained through the fine-tuning process.During transfer learning a model may be trained to improve performance on a task similar to the task the model was previously trained on, for example; a model may be trained to detect a firstset of objects (e.g., common objects in a set of pictures comprising images of the outdoors). The model in this example may then be trained to detect another set of objects such as organs in body scan images that may not have been present in the first set of objects. Although body scan images may not have been present in its initial data set the model had previously learned information about pictures that was transferrable to the body scan images. This example may be useful to a person or organization who hopes to build an organ detection model but who has very little body scan data, the information present in the far more abundant outdoor image data allow the model to learn a large amount about images without overtraining on the small body scan dataset.

[0092] A model may be integrated into another model. In those cases, a model may be trained and appended to another model that may or may not be trained. The new model, which includes the previously trained original model, may be trained such that the original model’s parameters are static (the original model is in inference mode), while the rest of the new model may be trainable, or partially trainable. Such an example may seem like a transfer learning scenario, and it may be used that way, but it may also be used to connect two models via the trained model using it as a stable intermodal processor. The method may also be used to convert an input into a format that is easier for the new model to process. As such different tasks of a model’s behavior may be trained independent of each other and then appended and trained in an end-to end fashion allowing the new model to benefit from the improved training regime of the independent training and to require potentially less data or time to train that might have before.Neural networks

[0093] Neural networks are a class of machine learning models which use artificial neurons an individual processing units. These artificial neurons may comprise an input, a set of weights, a set of biases, a summation step and an activation function. When an artificial neuron receives an input, which may be one or more values, the input may have a weight, or a set of weights, and a bias or set of biases applied to it. The transformed input may have a function applied that combines them into one value, such as summation, in the case that multiple inputs are given to an artificial neuron, and / or processed by an activation function which outputs a value.Multiple artificial neurons may be used to create a layer of neurons that takes in the same input and outputs a number of values equal to the number of neurons in that layer. A neural network may be composed of multiple layers. Layers may take as input data, or output from other layers or some other values such as a random value. Layers may be smaller, larger or the same size as the input they take. Layers may be of various types such as, but not limited to, the following layer types; dense, convolutional, pooling, recurrent, preprocessing, normalization,regularization, attention, reshaping, merging, or activation. When a layer’s output is received as input by another layer the two layers are connected, layers may be connected to any layer that follows.Model types

[0094] Layer connectivity may define the model’s architecture. Choice of a model’s architecture may be directed by the task being carried out by the layer or set of layers. Layer architectures may then be described by their function.Object detection

[0095] Object detection involves localizing and / or classifying objects within an input such as an image. It may identify specific objects and may output bounding boxes for identified objects. Object detection may comprise components for generating potential object proposals, feature extraction networks for analyzing proposals, and object classification networks for assigning class labels. Examples of object detection networks may include, but are not limited to, Faster R-CNN, YOLO (You Only Look Once), and SSD (Single Shot MultiBox Detector).Segmentation

[0096] Segmentation networks may partition an input, such as an image, into meaningful regions to identify and differentiate objects or regions of interest. Segmentation networks may be used for tasks such as understanding object boundaries, and extracting fine-grained information. Segmentation techniques may include semantic segmentation, which may assign class labels to each pixel, and instance segmentation, which may identify individual instances of objects. Panoptic segmentation may combine semantic and instance segmentation where it may label all pixels while distinguishing different instances.Image generation and multimodal models

[0097] Images may be generated using neural network architectures that incorporate multiple types of models. Using networks such as diffusion model images may be generated from multiple inputs. Inputs may include data such as text or other images. These models may be conditioned on an inputs such text to alter an existing image. For example, an image of a photograph may be input along with the text conditioning of “in the style of Van Gogh” where the output would be the original image transformed to be in the style of Van Gogh. In this case the model may be said to be multimodal as it takes two modes of data, image and text, to generate its output.Example 1 : Methods of Predicting a desired Gel Composition Arrangement

[0098] Another aspect provided herein is a method of predicting a desired gel composition arrangement (e.g., the shape and symmetry of the gel composition, and also referred to herein as a “placement”). To place a gel composition in a way that minimizes exposure of a tissue to radiation therapy when that tissue is adjacent to the target of the radiation therapy may be assisted by machine learning. A machine learning model may be used to determine spatial qualities (also referred to herein as metrics) of one or more objects in an image (e.g., of a compartment of a human). The spatial qualities may be, but are not limited to, size, position, shape, symmetry, or any combination thereof. The model may perform image segmentation, object detection, classification or any combination thereof, on an image to detect objects of interest. The one or more objects may be anatomical. The one or more objects may be organs, tissues, cavities, tumors, or any combination thereof. The model may produce an indication from an image comprising a portion of the body with target tissue. The output may be one or more segmentation masks, they may include segment classifications, each mask may be labeled with a segment class. Segment classes may be determined by the user. Segment classes may be pretrained, i.e. they may be classes the model was trained on during some portion of its training method.

[0099] In some embodiments, a new machine learning model may take as input the image. In some embodiments, a new machine learning model may take as input the segmentation masks. In some embodiments, a new machine learning model may take as input a new machine learning model may take as input an indication of the target tissue. In some embodiments, a new machine learning model may take as input an indication of distance within the image. In some embodiments, a new machine learning model may take as input an indication of desired maximum radiation level. In some embodiments, a new machine learning model may take as input any combination of the image, the segmentation masks. In some embodiments, the target tissue, an indication of distance within the image, an indication of desired maximum radiation level. The new machine learning model may be a multimodal model, a generative model, a diffusion model, a large language model, a convolutional neural network, an encoder, an autoencoder, a variational autoencoder, a deconvolutional network or any combination thereof.

[0100] The new machine learning model may take user input comprising at least one of an indication of target tissue, an indication of desired maximum radiation level. The indication of target tissue may be a shaped selection (such as a circle or a square), or an outline of the tissue for example. An indication of desired maximum radiation level may be any measurement of radiation such as but not limited to radiation absorbed dose or greys scale. The new machine learning model may use any of its inputs as conditioning. Conditioning may be thought of as,but not limited to, an information framework used to constrain or ‘condition’ the model. Conditioning may be thought of as a way to control the output by the user or through programmatic means (such as through heuristics or the output of a model). At least one segmentation mask may then be adjusted by the new machine learning model to position the adjacent tissue where it will be exposed to a radiation level no higher than the desired maximum radiation level.

[0101] The model may then generate a new version of the image from the adjusted segmentation mask. The new image may comprise an indication of a new object. The new object may be a segmentation mask. The segmentation mask may indicate a gel composition. The gel composition may be a spacer. In some embodiments the new object may be indicated in an output image wherein the original tissues are shifted and a gel composition is generated to fit between at least two of the original tissues.

[0102] In some embodiments the model may output a file used to render a 3d image. In some embodiments the model may output a file used to create a 3d model. The output may then be processed into a file suitable for 3d printing. In some embodiments the 3d printed object may be constructed of a material suitable for use in a medical setting or implantation into an organism. In some cases the organism is a mammal. In some cases the organism is human. In some embodiments the 3d printed object may be used to cast a mold. The mold may then be used to produce an object constructed of a material suitable for implantation in an organism which may be mammal, which may be human.

[0103] The indication of the new object may be used as the basis to manufacture a physical object with the same or similar size and shape. The manufacture process may include 3d printing.Example 2: Methods of Displacing a Gel Composition

[0104] Another aspect provided herein is a method of displacing a gel composition. In some embodiments, per FIG. 3, the method comprises: detecting one or more organs and a tumor on an image of a compartment of the human subject 301. In some embodiments, the method further comprises identifying, using a machine learning model, a plurality of spaces of the compartment for administering the gel composition based on image information of the one or more organs and the tumor 302. In some embodiments, the method further comprises selecting, using the machine learning model a first space of the plurality of spaces based on a relationship between the first space and the one or more organs and the tumor 303. In some embodiments, the method further comprises providing an indicator to dispose the gel composition in the first space 304.

[0105] In some embodiments, the relationship is based on: a distance between the first space and the one or more organs; a distance between the first space and the tumor; a shape of the gel composition associated with the first space; a volume of the gel compositions associated with the first space or any combination thereof. In some embodiments, selecting on an image, using the machine learning model, the first space of the plurality of spaces is further based on a relationship between a second space and the one or more organs and the tumor. In some embodiments, the method further comprises determining, using the machine learning model: the distance between the first space and the one or more organs; the distance between the first space and the tumor; the shape of the gel composition associated with the first space; the volume of the gel compositions associated with the first space; the amount of contrast (signal) on the image; a distance between the second space and the one or more organs; a distance between the second space and the tumor; a shape of the gel composition associated with the second space; a volume of the gel compositions associated with the second space, or any combination thereof. In some embodiments, the distance between the first space and the one or more organs is associated with a reduced dose of radiation applied to the one or more organs relative to another distance between another space of the plurality of spaces and the one or more organs. In some embodiments, the distance between the first space and the tumor is associated with a therapeutically effective dose of radiation applied to the tumor.

[0106] In some embodiments, the other space is the second space. In some embodiments, at least two spaces of the plurality of spaces overlap. In some embodiments, at least two spaces of the plurality of spaces do not overlap. In some embodiments, the at least two spaces comprise the first space. In some embodiments, the method further comprises displaying the first space. In some embodiments, the first space comprises displaying the image of the first space within the compartment on a user interface. In some embodiments, the user interface is on a computing device.

[0107] In some embodiments, the machine learning model determines one or more of the following based on a signal to noise ratio of one or more portions of the image: the distance between the first space and the one or more organs; the distance between the first space and the tumor; the shape of the gel composition associated with the first space; the volume of the gel compositions associated with the first space; the amount of contrast (signal) on the image; the distance between the second space and the one or more organs; the distance between the second space and the tumor; the shape of the gel composition associated with the second space; or the volume of the gel compositions associated with the second space. In some embodiments, identifying, using a machine learning model, a plurality of spaces of the compartment comprises generating the plurality of spaces in the compartment.Preventing or Minimizing Tissue Damage by the Gel Composition

[0108] A gel composition may be placed such that it comes in contact with one or more tissues (as shown in a non-limiting example of FIG. 19). As the contacting tissues move, the gel composition may rub or apply pressure to those tissues which, over time, may cause the tissue to become damaged. Examples of damage may include, but are not limited to; inflammation, pertusion, chafing, causing a hole to form in the tissue, causing the tissue to thin, causing scarring of the tissue, etc. To prevent damage of the tissue, a gel spacer may be made to move in a similar way as the surrounding tissues. This may be done by causing the material of the gel spacer, either through a manufacturing means or through the materials used to manufacture the gel spacer, to have similar physical properties as the surrounding tissue. Such properties may include but are not limited to, density, smoothness, measures of flexibility, coefficient of thermal expansion, elasticity, or viscosity.

[0109] The physical properties of materials, such as a gel composition and tissue, may relate to how subjects respond to physical interactions and changes in the environment. The tissues of a subject’s body are not static, and regular movement such as changing body position, walking, exercise, and breathing may cause movements in some tissues. Internal changes in the body may cause movement of tissues, as an example, the passage of food through the digestive track causes the tissues to change shape and apply pressure to surrounding tissues. As tissues change shape or move they may interact with those tissues surrounding them. Physical properties such as elasticity and overall flexibility (torsion) may govern how pliable an organ or tissue is. Properties such as smoothness, viscosity, and density may govern how easily a tissue or organ is manipulated by forces outside of itself such as pressure from outside the body or from a neighboring tissue or organ. Tissues may also respond to changes in temperature locally or globally. Many materials expand to some degree or another due to rising temperature and contract to some degree to lowering temperature, which is a property known as thermal expansion, and may be described by a coefficient. For example, if two materials interact and have different coefficients of thermal expansion they will pull apart or stretch at different rates. This may, over time, cause damage such as shearing or chafing. Similarly, if any of the physical properties differ when two material interact they may be prone to damaging each other. In tissues and organs this can lead to pain, disfunction, infection or death when unattended. A gel composition inserted between two organs may cause damage to those organs or tissues it contacts leading to a decrease in quality of life, illness or death of the patient who has received the gel composition.

[0110] By altering or carefully selecting the materials used in the gel composition (e.g., a spacer) and the placement of the comosition, the toxicity of damaging or negative interactionswith neighboring tissues or organs may be reduced thereby decreasing the toxicity of long term implantation of the gel composition. For example, one or more aspects of the spacer, such as the shape and / or symmetry, may be determined by the systems and methods described herein in order to cause less damage to tissue and organs in a subject. Less damage to surrounding tissues or organs leads to less risk of dangerous complications such as tissue ruptures, necrosis or clotting over time providing huge benefits to the patient and their families or caregivers by alleviating pain or discomfort as well as risk. Additionally, there may be a savings in time, in determining the aspects of the spacer using the systems and methods as described herein, and patients may further not need to expend extra costs of seeing a doctor due to the discomfort caused by other gel compositions or implants.Example 3: Method of Reducing a Dose of Radiation Applied to One or More Organs[OHl] Another aspect provided herein is a method of reducing a dose of radiation applied to one or more organs within a compartment of a human subject. In some embodiments, per FIG.4, the method comprises: detecting the one or more organs within the compartment and a tumor in the compartment 401. In some embodiments, the method further comprises generating a plurality of beam markers based on the one or more organs within the compartment and the tumor in the compartment, wherein a beam marker of the plurality of beam markers indicates a point of administration of a dose of radiation to the human subject 402. In some embodiments, the method further comprises selecting a first beam marker of the plurality of beam markers based on a relationship of the first beam marker to the one or more organs displaced in the compartment and the tumor in the compartment, wherein the first beam marker indicates a reduced dose of radiation applied to the one or more organs relative to a reference point of administration, and wherein the first beam marker indicates a therapeutically effective dose of radiation applied to the tumor relative to the reference point of administration 403.

[0112] In some embodiments, the method further comprises detecting a spacer displaced in the compartment. In some embodiments, generating the plurality of beam markers is further based on the spacer. In some embodiments, selecting the first beam marker of the plurality of beam markers is further based on the spacer. In some embodiments, generating the plurality of beam markers based on the one or more organs displaced in said compartment. In some embodiments, the spacer displaced in said compartment comprises generating, using a machine learning model, the plurality of beam markers based on the one or more organs displaced in said compartment. In some embodiments, the spacer displaced in said compartment.

[0113] In some embodiments, selecting the first beam of the plurality of beams based on the relationship of the first beam to at least one of the one or more organs and the spacer comprisesselecting, using a machine learning model, the first beam of the plurality of beams based on the relationship of the first beam to at least one of the one or more organs and the spacer. In some embodiments, prior to detecting the one or more organs displaced in the compartment, the spacer displaced in the compartment, the tumor displaced in the compartment, or any combination thereof, displacing the spacer adjacent to the one or more organs and the tumor.

[0114] In some embodiments, the tumor is in need of radiation therapy. In some embodiments, the method further comprises ranking the beam markers of the plurality of beam markers based on the one or more criteria and a plurality of relationships between the plurality of beam markers and the one or more organs, wherein the selecting is based on the ranking and the plurality of relationships comprises the relationship of the first beam. In some embodiments, the one or more criteria are associated with the one or more organs. In some embodiments, the one or more organs are visible in the image, and wherein the one or more criteria are based on locations of the one or more organs in the image.

[0115] In some embodiments, at least one of the relationship is associated with a distance from an organ of the one or more organs. In some embodiments, the relationship comprises: a distance from one or more organs; overlap of a beam marker of the plurality of beam markers with one or more organs; or both. In some embodiments, each beam marker of the plurality of beam markers overlap. In some embodiments, each beam marker of the plurality of beam markers has a respective boundary. In some embodiments, each beam marker of the plurality of beam markers share a common location. In some embodiments, the method further comprises displaying an indication of the first beam marker. In some embodiments, the selecting the first beam marker of the plurality of beam markers is based on the first beam marker having a highest ranking. In some embodiments, the first beam marker has a first topography and a first boundary, wherein the first topography and the first boundary result in less radiation applied to organs of the subject than respective topographies and respective boundaries of each other beam marker of the plurality of beam markers.

[0116] In some embodiments, the method further comprises applying the dose of radiation based on the first beam marker. In some embodiments, the dose of radiation is associated with a dose radiation gradient, wherein an organ disposed within the dose radiation gradient receives a tolerable amount of radiation. In some embodiments, the method further comprises displaying the plurality of beam markers on a user interface.

[0117] In some embodiments, the reduced dose of radiation applied to the one or more organs is no more than 90%, 85%, 80%, 70%, 75%, 70%, or less of the therapeutically effective dose of radiation applied to the tumor, including increments therein. In some embodiments, the reduced dose of radiation applied to the one or more organs is no more than 50%, 55%, 60%, 65%, 70%,75%, 80%, 85%, or less of the therapeutically effective dose of radiation applied to the tumor, including increments therein. In some embodiments, the reduced dose of radiation applied to the one or more organs is no more than 30%, 35%, 40%, 45%, 50%, 55%, or more of the therapeutically effective dose of radiation applied to the tumor, including increments therein. In some embodiments, the point of administration is indicated by the first beam marker.

[0118] After the gel is inserted, the additional distance provided by the gel at the prostaterectum interface makes planning with usual constraints easier.

[0119] If an MR image with the gel inserted is present in addition to the simulation CT, the gel may be fused to fiducial markers first. The prostate, bladder, and rectal interfaces may be aligned. If the fusion is well-matched through this volume, an MR can be used to guide the contouring of the gel. The prostate and seminal vesicles may also be contoured on the MR, with or without refinement on the CT.

[0120] For CT-only planning, if there is no MR for the patient, the prostate may be contoured first, and then the CTV and PTV target volumes may be added. A slightly less dense (dark) layer, usually a flattened oval mass of “space,” between the anterior rectal wall and the posterior prostate wall may be shown as a buffer zone of 1 cm or more. In some embodiments, the exact prostate border may be difficult to infer. In those embodiments, the prostrate border may it be extended slightly into the gel region. The length of the rectum may also be contoured in relationship to the PTV, with the contours close to anatomic borders of the rectal wall. In some embodiments, the exact dose of the gel may not be important but may still be defined as a structure. An ideal injection will lead to an approximately symmetric implant centered on the midline, with a flattened oval cross section 1-2 cm wide at the middle (minor axis). The injection’s shape should taper off smoothly with rounded edges to each side on the axial view, per FIG. 6B, extending on all the slices from prostate base to apex. While the overall volume of space created should be contoured, distinct boluses, per FIG. 8B, should not be contoured. The volume of the prostate and volume of the gel can be obtained from the TRUS images.

[0121] The best results may be obtained for VMAT with two arcs, one with CCW rotation and one CW rotation. In some embodiments, unequal collimator angles (e.g., 350° and 40°) may yield more flexibility with dose control than beams with collimator angles closer to each other (e.g., 15° and 345°). In some embodiments, collimator angles of 45° and 315° create the greatest amount of low-dose leaf leakage. For static IMRTs, multiple beams may be used to collimate along the slope of the rectum while avoiding collimator angles of 0-5° to avoid overlap of interleaf leakage doses. In some embodiments, 7-9 beams may be used. Further, 30% / 50% / 90% isodoses flow parallel to posterior prostate contour and bisecting the rectum with at least the 50% isodose line, and if possible, the 30% isodose line are recommended.

[0122] In some embodiments, recommended planning structure priority is: Prostate V100 > 98%; Minimize Rectum D90; and Bladder V60 < 5%. In some embodiments, a reasonable objective for well-optimized prostate plans is that the dose gradient from the target should decrease roughly 10% per mm of distance.

[0123] Window and level optimization is important and can help bring out subtle density and texture differences in CT scans. FIG. 8A shows an image of a first exemplary computerized tomography (CT) scan of a rectum and a prostate with a spacer injected therebetween. FIG. 8B shows an image of a second exemplary computerized tomography (CT) scan of a rectum and a prostate with a spacer injected therebetween. FIG. 9A shows an image of a third exemplary computerized tomography (CT) scan of a rectum and a prostate with a spacer injected therebetween. FIG. 9B shows a window and level optimized image of the third exemplary computerized tomography (CT) scan, wherein adjusting window & level settings enhances contrast and texture. FIG. 10A shows an image of a fourth exemplary computerized tomography (CT) scan of a rectum and a prostate with a spacer injected therebetween. FIG. 10B shows a window and level optimized image of the fourth exemplary computerized tomography (CT) scan, wherein adjusting window & level settings enhances contrast and texture.

[0124] Presets for Liver / Cerebellum, per FIGS. 9A-10B, may enhance the gel borders and the rectal wall. The gel may be slightly darker gray than surrounding tissues in the axial view. Some imprecision in defining the border of the gel is tolerable, as the quality of the plan may be driven by the target coverage and rectal sparing.

[0125] Optimizing the structure of the composition enables improved dose gradients that takes advantage of the space between the prostate and rectum to maximize rectal sparing. FIGS. 11 and 12 show images of fourth and fifth exemplary computerized tomography (CT) scans of a rectum and a prostate with dose gradients. The 30% isodose 1110, the 50% isodose 1120, and the 100% isodose 1130 in FIG. 11, show a suboptimal radiation plan and composition structure, whereas the 30% isodose 1210, the 50% isodose 1220, and the 100% isodose 1230 in FIG. 12, show a more optimal plan.Example 4: Method of Evaluating a Placement of a Composition

[0126] Another aspect provided herein is a method of evaluating a placement of a composition in a compartment. In this example, a composition may have been placed in the compartment, and the physical placement of the composition in the compartment may be compared to a desired placement (e.g., as determined by methods described herein). In some embodiments, per FIG. 5, the method comprises: detecting one or more organs and a tumor within a compartment of a human subject 501. In some embodiments, the method comprises determining, using amachine learning model, a first space of the compartment based on the one or more organs and the tumor 502 (e.g., a desired placement as described above with respect to example 1). In some embodiments, the method comprises detecting the composition displaced in a second space of the compartment 503. In some embodiments, the method comprises comparing the first space to the second space 504. In some embodiments, the method comprises determining a score based on the comparing 505; and providing the score to a user 506.

[0127] In some embodiments, determining the first space based on the one or more organs and the tumor is based on a relationship between the first space and the one or more organs and the tumor. In some embodiments, the method further comprises, determining, using the machine learning model: a distance between the first space and the one or more organs; a distance between the first space and the tumor; a shape associated with the first space; a volume associated with the first space; a distance between the second space and the one or more organs; a distance between the second space and the tumor; a shape associated with the second space; a volume associated with the second space; or any combination thereof.

[0128] In some embodiments, the relationship between the first space and the one or more organs and the tumor is based on one of more of the distance between the first space and the one or more organs; the distance between the first space and the tumor; the shape associated with the first space; or the volume associated with the first space. In some embodiments, the method further comprises determining a relationship between the second space and the one or more organs and the tumor. In some embodiments, determining the relationship between the second space and the one or more organs and the tumor is based on: the distance between the second space and the one or more organs; the distance between the second space and the tumor; the shape associated with the second space; the volume associated with the second space; or any combination thereof. In some embodiments, the first space and the second space at least partially overlap. In some embodiments, the first space and the second space do not overlap. In some embodiments, the first space is selected from a plurality of spaces based on: the first space being associated with a reduced dose of radiation applied to the one or more organs relative to another dose of radiation applied to the one or more organs associated with another space of the plurality of spaces; the first space being associated with a therapeutically effective dose of radiation applied to the tumor; or both. In some embodiments, the method further comprises displaying one or more of the first space or the second space on a user interface. In some embodiments, the one or more metrics are used to determine: the distance between the first space and the one or more organs; the distance between the first space and the tumor; the shape of the gel composition associated with the first space; the volume of the gel compositions associated with the first space; the distance between the second space and the one or more organs; the distancebetween the second space and the tumor; the shape of the gel composition associated with the second space; the volume of the gel compositions associated with the second space; or any combination thereof.

[0129] In some embodiments, the method further comprises ranking the spaces of the plurality of spaces based on the one or more relationships. In some embodiments, each space of the plurality of spaces is associated with a dose of radiation applied to the one or more organs. In some embodiments, the first space is associated with a reduced dose of radiation applied to the one or more organs relative to doses of radiation applied to the one or more organs associated with other spaces of the plurality of spaces. In some embodiments, the first space is associated with a therapeutically effective dose of radiation applied to the tumor. In some embodiments, training the machine learning model comprises training the machine learning model to select a space associated with a reduced dose of radiation applied to the one or more organs and a therapeutically effective dose of radiation applied to the tumor. In some embodiments, at least two spaces of the plurality of spaces overlap. In some embodiments, at least two spaces of the plurality of spaces do not overlap. In some embodiments, the at least two spaces comprises the first space.

[0130] In some embodiments, the composition comprises: a viscoelastic medium; and a first visual additive, wherein said first visual additive comprises a metal, and wherein said metal has a particle diameter of greater than or equal to 80 micrometers (pm) 90 pm, 100 pm, 125 pm, 150 pm, 200 pm, 250 pm, 300 pm, or more, including increments therein. In some embodiments, said metal is a precious metal. In some embodiments, said metal or said precious metal is chosen from the group consisting of gold (Au), iodine (I), gadolinium (Gd), iron (Fe), barium (Ba), calcium (Ca), magnesium (Mg), and combinations thereof. In some embodiments, said metal or precious metal is an isotope of gold (Au), iodine (I), gadolinium (Gd), iron (Fe), barium (Ba), calcium (Ca), magnesium (Mg), or combinations thereof. In some embodiments, said metal or precious metal is a powder.

[0131] In some embodiments, said first visual additive has a radiographic density of about 1 g / cm3to about 2 g / cm3. In some embodiments, said first visual additive has a radiographic density of about 1 g / cm3to about 1.1 g / cm3, about 1 g / cm3to about 1.2 g / cm3, about 1 g / cm3to about 1.3 g / cm3, about 1 g / cm3to about 1.4 g / cm3, about 1 g / cm3to about 1.5 g / cm3, about 1 g / cm3to about 1.6 g / cm3, about 1 g / cm3to about 1.7 g / cm3, about 1 g / cm3to about 1.8 g / cm3, about 1 g / cm3to about 1.9 g / cm3, about 1 g / cm3to about 2 g / cm3, about 1.1 g / cm3to about 1.2 g / cm3, about 1.1 g / cm3to about 1.3 g / cm3, about 1.1 g / cm3to about 1.4 g / cm3, about 1.1 g / cm3to about 1.5 g / cm3, about 1.1 g / cm3to about 1.6 g / cm3, about 1.1 g / cm3to about 1.7 g / cm3, about 1.1 g / cm3to about 1.8 g / cm3, about 1.1 g / cm3to about 1.9 g / cm3, about 1.1 g / cm3to about 2g / cm3, about 1.2 g / cm3to about 1.3 g / cm3, about 1.2 g / cm3to about 1.4 g / cm3, about 1.2 g / cm3to about 1.5 g / cm3, about 1.2 g / cm3to about 1.6 g / cm3, about 1.2 g / cm3to about 1.7 g / cm3, about1.2 g / cm3to about 1.8 g / cm3, about 1.2 g / cm3to about 1.9 g / cm3, about 1.2 g / cm3to about 2 g / cm3, about 1.3 g / cm3to about 1.4 g / cm3, about 1.3 g / cm3to about 1.5 g / cm3, about 1.3 g / cm3to about 1.6 g / cm3, about 1.3 g / cm3to about 1.7 g / cm3, about 1.3 g / cm3to about 1.8 g / cm3, about1.3 g / cm3to about 1.9 g / cm3, about 1.3 g / cm3to about 2 g / cm3, about 1.4 g / cm3to about 1.5 g / cm3, about 1.4 g / cm3to about 1.6 g / cm3, about 1.4 g / cm3to about 1.7 g / cm3, about 1.4 g / cm3to about 1.8 g / cm3, about 1.4 g / cm3to about 1.9 g / cm3, about 1.4 g / cm3to about 2 g / cm3, about 1.5 g / cm3to about 1.6 g / cm3, about 1.5 g / cm3to about 1.7 g / cm3, about 1.5 g / cm3to about 1.8 g / cm3, about 1.5 g / cm3to about 1.9 g / cm3, about 1.5 g / cm3to about 2 g / cm3, about 1.6 g / cm3to about 1.7 g / cm3, about 1.6 g / cm3to about 1.8 g / cm3, about 1.6 g / cm3to about 1.9 g / cm3, about 1.6 g / cm3to about 2 g / cm3, about 1.7 g / cm3to about 1.8 g / cm3, about 1.7 g / cm3to about 1.9 g / cm3, about 1.7 g / cm3to about 2 g / cm3, about 1.8 g / cm3to about 1.9 g / cm3, about 1.8 g / cm3to about 2 g / cm3, or about 1.9 g / cm3to about 2 g / cm3, including increments therein. In some embodiments, said first visual additive has a radiographic density of about 1 g / cm3, about 1.1 g / cm3, about 1.2 g / cm3, about 1.3 g / cm3, about 1.4 g / cm3, about 1.5 g / cm3, about 1.6 g / cm3, about 1.7 g / cm3, about 1.8 g / cm3, about 1.9 g / cm3, or about 2 g / cm3. In some embodiments, said first visual additive has a radiographic density of at least about 1 g / cm3, about 1.1 g / cm3, about 1.2 g / cm3, about 1.3 g / cm3, about 1.4 g / cm3, about 1.5 g / cm3, about 1.6 g / cm3, about 1.7 g / cm3, about 1.8 g / cm3, or about 1.9 g / cm3. In some embodiments, said first visual additive has a radiographic density of at most about 1.1 g / cm3, about 1.2 g / cm3, about 1.3 g / cm3, about 1.4 g / cm3, about 1.5 g / cm3, about 1.6 g / cm3, about 1.7 g / cm3, about 1.8 g / cm3, about 1.9 g / cm3, or about 2 g / cm3.

[0132] In some embodiments, said particle diameter of said metal or precious metal is about 80 pm to about 120 pm. In some embodiments, said particle diameter of said metal or precious metal is about 80 pm to about 85 pm, about 80 pm to about 90 pm, about 80 pm to about 95 pm, about 80 pm to about 100 pm, about 80 pm to about 105 pm, about 80 pm to about 110 pm, about 80 pm to about 115 pm, about 80 pm to about 120 pm, about 85 pm to about 90 pm, about 85 pm to about 95 pm, about 85 pm to about 100 pm, about 85 pm to about 105 pm, about 85 pm to about 110 pm, about 85 pm to about 115 pm, about 85 pm to about 120 pm, about 90 pm to about 95 pm, about 90 pm to about 100 pm, about 90 pm to about 105 pm, about 90 pm to about 110 pm, about 90 pm to about 115 pm, about 90 pm to about 120 pm, about 95 pm to about 100 pm, about 95 pm to about 105 pm, about 95 pm to about 110 pm, about 95 pm to about 115 pm, about 95 pm to about 120 pm, about 100 pm to about 105 pm, about 100 pm to about 110 pm, about 100 pm to about 115 pm, about 100 pm to about 120 pm, about 105 pm to about 110 pm, about 105 pm to about 115 pm, about 105 pm to about 120 pm,about 110 pm to about 115 pm, about 110 pm to about 120 pm, or about 115 pm to about 120 pm, including increments therein. In some embodiments, said particle diameter of said metal or precious metal is about 80 pm, about 85 pm, about 90 pm, about 95 pm, about 100 pm, about 105 pm, about 110 pm, about 115 pm, or about 120 pm. In some embodiments, said particle diameter of said metal or precious metal is at least about 80 pm, about 85 pm, about 90 pm, about 95 pm, about 100 pm, about 105 pm, about 110 pm, or about 115 pm. In some embodiments, said particle diameter of said metal or precious metal is at most about 85 pm, about 90 pm, about 95 pm, about 100 pm, about 105 pm, about 110 pm, about 115 pm, or about 120 pm. In some embodiments, half of said first visual additive is configured to disperse within a tissue within nine months.

[0133] In some embodiments, said first visual additive further comprises one or more microbubbles. In some embodiments, the method further comprises a second visual additive. In some embodiments, said second visual additive is different than said first visual additive. In some embodiments, said second visual additive comprises one or more microbubbles.

[0134] In some embodiments, said first visual additive has a concentration within said viscoelastic medium of greater than 5 milligrams per milliliter (mg / ml) 6 mg / ml, 7 mg / ml, 8 mg / ml, 9 mg / ml, 10 mg / ml, or more, including increments therein. In some embodiments, said first visual additive has a concentration within said viscoelastic medium of less than 90 mg / ml, 80 mg / ml, 70 mg / ml, 60 mg / ml, 50 mg / ml, or fewer, including increments therein.

[0135] In some embodiments, said first visual additive has a concentration within said viscoelastic medium of about 15 mg / ml to about 30 mg / ml. In some embodiments, said first visual additive has a concentration within said viscoelastic medium of about 15 mg / ml to about 17 mg / ml, about 15 mg / ml to about 19 mg / ml, about 15 mg / ml to about 21 mg / ml, about 15 mg / ml to about 23 mg / ml, about 15 mg / ml to about 25 mg / ml, about 15 mg / ml to about 27 mg / ml, about 15 mg / ml to about 30 mg / ml, about 17 mg / ml to about 19 mg / ml, about 17 mg / ml to about 21 mg / ml, about 17 mg / ml to about 23 mg / ml, about 17 mg / ml to about 25 mg / ml, about 17 mg / ml to about 27 mg / ml, about 17 mg / ml to about 30 mg / ml, about 19 mg / ml to about 21 mg / ml, about 19 mg / ml to about 23 mg / ml, about 19 mg / ml to about 25 mg / ml, about 19 mg / ml to about 27 mg / ml, about 19 mg / ml to about 30 mg / ml, about 21 mg / ml to about 23 mg / ml, about 21 mg / ml to about 25 mg / ml, about 21 mg / ml to about 27 mg / ml, about 21 mg / ml to about 30 mg / ml, about 23 mg / ml to about 25 mg / ml, about 23 mg / ml to about 27 mg / ml, about 23 mg / ml to about 30 mg / ml, about 25 mg / ml to about 27 mg / ml, about 25 mg / ml to about 30 mg / ml, or about 27 mg / ml to about 30 mg / ml, including increments therein. In some embodiments, said first visual additive has a concentration within said viscoelastic medium of about 15 mg / ml, about 17 mg / ml, about 19 mg / ml, about 21 mg / ml, about 23 mg / ml, about 25mg / ml, about 27 mg / ml, or about 30 mg / ml. In some embodiments, said first visual additive has a concentration within said viscoelastic medium of at least about 15 mg / ml, about 17 mg / ml, about 19 mg / ml, about 21 mg / ml, about 23 mg / ml, about 25 mg / ml, or about 27 mg / ml. In some embodiments, said first visual additive has a concentration within said viscoelastic medium of at most about 17 mg / ml, about 19 mg / ml, about 21 mg / ml, about 23 mg / ml, about 25 mg / ml, about 27 mg / ml, or about 30 mg / ml.

[0136] In some embodiments, said composition has a concentration of the metal of about 0.5 wt% to about 9 wt%. In some embodiments, , said composition has a concentration of the metal of about 0.5 wt% to about 1 wt%, about 0.5 wt% to about 2 wt%, about 0.5 wt% to about 3 wt%, about 0.5 wt% to about 4 wt%, about 0.5 wt% to about 5 wt%, about 0.5 wt% to about 6 wt%, about 0.5 wt% to about 7 wt%, about 0.5 wt% to about 8 wt%, about 0.5 wt% to about 9 wt%, about 1 wt% to about 2 wt%, about 1 wt% to about 3 wt%, about 1 wt% to about 4 wt%, about 1 wt% to about 5 wt%, about 1 wt% to about 6 wt%, about 1 wt% to about 7 wt%, about 1 wt% to about 8 wt%, about 1 wt% to about 9 wt%, about 2 wt% to about 3 wt%, about 2 wt% to about 4 wt%, about 2 wt% to about 5 wt%, about 2 wt% to about 6 wt%, about 2 wt% to about 7 wt%, about 2 wt% to about 8 wt%, about 2 wt% to about 9 wt%, about 3 wt% to about 4 wt%, about 3 wt% to about 5 wt%, about 3 wt% to about 6 wt%, about 3 wt% to about 7 wt%, about 3 wt% to about 8 wt%, about 3 wt% to about 9 wt%, about 4 wt% to about 5 wt%, about 4 wt% to about 6 wt%, about 4 wt% to about 7 wt%, about 4 wt% to about 8 wt%, about 4 wt% to about 9 wt%, about 5 wt% to about 6 wt%, about 5 wt% to about 7 wt%, about 5 wt% to about 8 wt%, about 5 wt% to about 9 wt%, about 6 wt% to about 7 wt%, about 6 wt% to about 8 wt%, about 6 wt% to about 9 wt%, about 7 wt% to about 8 wt%, about 7 wt% to about 9 wt%, or about 8 wt% to about 9 wt%, including increments therein. In some embodiments, said composition has a concentration of the metal of about 0.5 wt%, about 1 wt%, about 2 wt%, about 3 wt%, about 4 wt%, about 5 wt%, about 6 wt%, about 7 wt%, about 8 wt%, or about 9 wt%. In some embodiments, said composition has a concentration of the metal of at least about 0.5 wt%, about 1 wt%, about 2 wt%, about 3 wt%, about 4 wt%, about 5 wt%, about 6 wt%, about 7 wt%, or about 8 wt%. In some embodiments, , said composition has a concentration of the metal of at most about 1 wt%, about 2 wt%, about 3 wt%, about 4 wt%, about 5 wt%, about 6 wt%, about 7 wt%, about 8 wt%, or about 9 wt%.

[0137] In some embodiments, , said composition has a concentration of the precious metal of about 0.015 wt% to about 0.15 wt%. In some embodiments, , said composition has a concentration of the precious metal of about 0.015 wt% to about 0.025 wt%, about 0.015 wt% to about 0.05 wt%, about 0.015 wt% to about 0.075 wt%, about 0.015 wt% to about 0.1 wt%, about 0.015 wt% to about 0.125 wt%, about 0.015 wt% to about 0.15 wt%, about 0.025 wt% to about0.05 wt%, about 0.025 wt% to about 0.075 wt%, about 0.025 wt% to about 0.1 wt%, about 0.025 wt% to about 0.125 wt%, about 0.025 wt% to about 0.15 wt%, about 0.05 wt% to about 0.075 wt%, about 0.05 wt% to about 0.1 wt%, about 0.05 wt% to about 0.125 wt%, about 0.05 wt% to about 0.15 wt%, about 0.075 wt% to about 0.1 wt%, about 0.075 wt% to about 0.125 wt%, about 0.075 wt% to about 0.15 wt%, about 0.1 wt% to about 0.125 wt%, about 0.1 wt% to about 0.15 wt%, or about 0.125 wt% to about 0.15 wt%, including increments therein. In some embodiments, , said composition has a concentration of the precious metal of about 0.015 wt%, about 0.025 wt%, about 0.05 wt%, about 0.075 wt%, about 0.1 wt%, about 0.125 wt%, or about 0.15 wt%. In some embodiments, , said composition has a concentration of the precious metal of at least about 0.015 wt%, about 0.025 wt%, about 0.05 wt%, about 0.075 wt%, about 0.1 wt%, or about 0.125 wt%. In some embodiments, , said composition has a concentration of the precious metal of at most about 0.025 wt%, about 0.05 wt%, about 0.075 wt%, about 0.1 wt%, about 0.125 wt%, or about 0.15 wt%.

[0138] In some embodiments, said viscoelastic medium comprises a volume of about 1 ml to about 50 ml. In some embodiments, said viscoelastic medium comprises a volume of about 1 ml to about 5 ml, about 1 ml to about 10 ml, about 1 ml to about 15 ml, about 1 ml to about 20 ml, about 1 ml to about 25 ml, about 1 ml to about 30 ml, about 1 ml to about 35 ml, about 1 ml to about 40 ml, about 1 ml to about 45 ml, about 1 ml to about 50 ml, about 5 ml to about 10 ml, about 5 ml to about 15 ml, about 5 ml to about 20 ml, about 5 ml to about 25 ml, about 5 ml to about 30 ml, about 5 ml to about 35 ml, about 5 ml to about 40 ml, about 5 ml to about 45 ml, about 5 ml to about 50 ml, about 10 ml to about 15 ml, about 10 ml to about 20 ml, about 10 ml to about 25 ml, about 10 ml to about 30 ml, about 10 ml to about 35 ml, about 10 ml to about 40 ml, about 10 ml to about 45 ml, about 10 ml to about 50 ml, about 15 ml to about 20 ml, about 15 ml to about 25 ml, about 15 ml to about 30 ml, about 15 ml to about 35 ml, about 15 ml to about 40 ml, about 15 ml to about 45 ml, about 15 ml to about 50 ml, about 20 ml to about 25 ml, about 20 ml to about 30 ml, about 20 ml to about 35 ml, about 20 ml to about 40 ml, about 20 ml to about 45 ml, about 20 ml to about 50 ml, about 25 ml to about 30 ml, about 25 ml to about 35 ml, about 25 ml to about 40 ml, about 25 ml to about 45 ml, about 25 ml to about 50 ml, about 30 ml to about 35 ml, about 30 ml to about 40 ml, about 30 ml to about 45 ml, about 30 ml to about 50 ml, about 35 ml to about 40 ml, about 35 ml to about 45 ml, about 35 ml to about 50 ml, about 40 ml to about 45 ml, about 40 ml to about 50 ml, or about 45 ml to about 50 ml, including increments therein. In some embodiments, said viscoelastic medium comprises a volume of about 1 ml, about 5 ml, about 10 ml, about 15 ml, about 20 ml, about 25 ml, about 30 ml, about 35 ml, about 40 ml, about 45 ml, or about 50 ml. In some embodiments, said viscoelastic medium comprises a volume of at least about 1 ml, about 5 ml, about 10 ml, about15 ml, about 20 ml, about 25 ml, about 30 ml, about 35 ml, about 40 ml, or about 45 ml. In some embodiments, said viscoelastic medium comprises a volume of at most about 5 ml, about 10 ml, about 15 ml, about 20 ml, about 25 ml, about 30 ml, about 35 ml, about 40 ml, about 45 ml, or about 50 ml.

[0139] In some embodiments, said composition is configured to be biodegradable. In some embodiments, said composition is configured to be present on an imaging modality for at least 9 months. In some embodiments, said composition is configured to not substantially migrate prior to or during imaging. In some embodiments, said first visual additive is configured to not substantially migrate prior to or during imaging.

[0140] In some embodiments, said composition is configured to be disposed within a subject. In some embodiments, said subject is in need of radiography. In some embodiments, said composition is configured to be disposed through injection. In some embodiments, said composition is configured to be disposed subcutaneously or subepidermally. In some embodiments, said composition is configured to be disposed within a compartment. In some embodiments, said compartment comprises one or more of, a fat tissue, a muscle tissue, and organ tissue, or a combination thereof. In some embodiments, said composition is configured to be imaged on one or more modalities. In some embodiments, said one or more modalities comprise X-Ray, MRI, CT, CBCT, ultrasound, PET, SPECT or a combination thereof. In some embodiments, said imaging comprises real-time imaging. In some embodiments, said composition is configured to be imaged within 30 min, within 90 min, within 4 hours, within 8 hours, or within 4 days of disposition.

[0141] The gel location, shape, and volume will remain consistent in the body from shortly after the implant to the duration of the radiotherapy course and some months beyond. In some embodiments, patients will be instructed to have a comfortably full bladder and empty rectum for the CT simulation image set, MRI simulation image (if used) and for each treatment session, or as per the standard bowel preparation in your institutional practice.

[0142] In some embodiments, a post-implant MRI image is not required but can enhance the ability to distinguish prostate and gel borders for ease in contouring. In some embodiments, a T2 weighted sequence is optimal for gel visualization and anatomic localization, which may cause the gel to appear bright white. In some embodiments, the parameters of the MRI include an Axial T2 proton weighted acquisition to include iliac crests through the perineum with a slice thickness of 3-5 mm.

[0143] In some embodiments, said viscoelastic medium comprises hyaluronic acid, polyethylene glycol, or dextranomers. In some embodiments, said viscoelastic medium comprises hyaluronic acid, polyethylene glycol, or dextranomers at a concentration of about 5 mg / ml to about 100mg / ml. In some embodiments, said viscoelastic medium comprises hyaluronic acid, polyethylene glycol, or dextranomers at a concentration of about 5 mg / ml to about 10 mg / ml, about 5 mg / ml to about 20 mg / ml, about 5 mg / ml to about 30 mg / ml, about 5 mg / ml to about 40 mg / ml, about 5 mg / ml to about 50 mg / ml, about 5 mg / ml to about 60 mg / ml, about 5 mg / ml to about 70 mg / ml, about 5 mg / ml to about 80 mg / ml, about 5 mg / ml to about 90 mg / ml, about 5 mg / ml to about 100 mg / ml, about 10 mg / ml to about 20 mg / ml, about 10 mg / ml to about 30 mg / ml, about 10 mg / ml to about 40 mg / ml, about 10 mg / ml to about 50 mg / ml, about 10 mg / ml to about 60 mg / ml, about 10 mg / ml to about 70 mg / ml, about 10 mg / ml to about 80 mg / ml, about 10 mg / ml to about 90 mg / ml, about 10 mg / ml to about 100 mg / ml, about 20 mg / ml to about 30 mg / ml, about 20 mg / ml to about 40 mg / ml, about 20 mg / ml to about 50 mg / ml, about 20 mg / ml to about 60 mg / ml, about 20 mg / ml to about 70 mg / ml, about 20 mg / ml to about 80 mg / ml, about 20 mg / ml to about 90 mg / ml, about 20 mg / ml to about 100 mg / ml, about 30 mg / ml to about 40 mg / ml, about 30 mg / ml to about 50 mg / ml, about 30 mg / ml to about 60 mg / ml, about 30 mg / ml to about 70 mg / ml, about 30 mg / ml to about 80 mg / ml, about 30 mg / ml to about 90 mg / ml, about 30 mg / ml to about 100 mg / ml, about 40 mg / ml to about 50 mg / ml, about 40 mg / ml to about 60 mg / ml, about 40 mg / ml to about 70 mg / ml, about 40 mg / ml to about 80 mg / ml, about 40 mg / ml to about 90 mg / ml, about 40 mg / ml to about 100 mg / ml, about 50 mg / ml to about 60 mg / ml, about 50 mg / ml to about 70 mg / ml, about 50 mg / ml to about 80 mg / ml, about 50 mg / ml to about 90 mg / ml, about 50 mg / ml to about 100 mg / ml, about 60 mg / ml to about 70 mg / ml, about 60 mg / ml to about 80 mg / ml, about 60 mg / ml to about 90 mg / ml, about 60 mg / ml to about 100 mg / ml, about 70 mg / ml to about 80 mg / ml, about 70 mg / ml to about 90 mg / ml, about 70 mg / ml to about 100 mg / ml, about 80 mg / ml to about 90 mg / ml, about 80 mg / ml to about 100 mg / ml, or about 90 mg / ml to about 100 mg / ml, including increments therein. In some embodiments, said viscoelastic medium comprises hyaluronic acid, polyethylene glycol, or dextranomers at a concentration of about 5 mg / ml, about 10 mg / ml, about 20 mg / ml, about 30 mg / ml, about 40 mg / ml, about 50 mg / ml, about 60 mg / ml, about 70 mg / ml, about 80 mg / ml, about 90 mg / ml, or about 100 mg / ml. In some embodiments, said viscoelastic medium comprises hyaluronic acid, polyethylene glycol, or dextranomers at a concentration of at least about 5 mg / ml, about 10 mg / ml, about 20 mg / ml, about 30 mg / ml, about 40 mg / ml, about 50 mg / ml, about 60 mg / ml, about 70 mg / ml, about 80 mg / ml, or about 90 mg / ml. In some embodiments, said viscoelastic medium comprises hyaluronic acid, polyethylene glycol, or dextranomers at a concentration of at most about 10 mg / ml, about 20 mg / ml, about 30 mg / ml, about 40 mg / ml, about 50 mg / ml, about 60 mg / ml, about 70 mg / ml, about 80 mg / ml, about 90 mg / ml, or about 100 mg / ml.

[0144] In some embodiments, said viscoelastic medium comprises gel particles at a size range of about 0.1 mm to about 5 mm. In some embodiments, said viscoelastic medium comprises gel particles at a size range of about 0.8 mm to about 0.1 mm, about 0.8 mm to about 0.25 mm, about 0.8 mm to about 0.5 mm, about 0.8 mm to about 0.75 mm, about 0.8 mm to about 1 mm, about 0.8 mm to about 2 mm, about 0.8 mm to about 3 mm, about 0.8 mm to about 4 mm, about 0.8 mm to about 5 mm, about 0.1 mm to about 0.25 mm, about 0.1 mm to about 0.5 mm, about 0.1 mm to about 0.75 mm, about 0.1 mm to about 1 mm, about 0.1 mm to about 2 mm, about 0.1 mm to about 3 mm, about 0.1 mm to about 4 mm, about 0.1 mm to about 5 mm, about 0.25 mm to about 0.5 mm, about 0.25 mm to about 0.75 mm, about 0.25 mm to about 1 mm, about 0.25 mm to about 2 mm, about 0.25 mm to about 3 mm, about 0.25 mm to about 4 mm, about 0.25 mm to about 5 mm, about 0.5 mm to about 0.75 mm, about 0.5 mm to about 1 mm, about 0.5 mm to about 2 mm, about 0.5 mm to about 3 mm, about 0.5 mm to about 4 mm, about 0.5 mm to about 5 mm, about 0.75 mm to about 1 mm, about 0.75 mm to about 2 mm, about 0.75 mm to about 3 mm, about 0.75 mm to about 4 mm, about 0.75 mm to about 5 mm, about 1 mm to about 2 mm, about 1 mm to about 3 mm, about 1 mm to about 4 mm, about 1 mm to about 5 mm, about 2 mm to about 3 mm, about 2 mm to about 4 mm, about 2 mm to about 5 mm, about 3 mm to about 4 mm, about 3 mm to about 5 mm, or about 4 mm to about 5 mm, including increments therein. In some embodiments, said viscoelastic medium comprises gel particles at a size range of about 0.8 mm, about 0.1 mm, about 0.25 mm, about 0.5 mm, about 0.75 mm, about 1 mm, about 2 mm, about 3 mm, about 4 mm, or about 5 mm. In some embodiments, said viscoelastic medium comprises gel particles at a size range of at least about 0.8 mm, about 0.1 mm, about 0.25 mm, about 0.5 mm, about 0.75 mm, about 1 mm, about 2 mm, about 3 mm, or about 4 mm. In some embodiments, said viscoelastic medium comprises gel particles at a size range of at most about 0.1 mm, about 0.25 mm, about 0.5 mm, about 0.75 mm, about 1 mm, about 2 mm, about 3 mm, about 4 mm, or about 5 mm.

[0145] In some embodiments, said viscoelastic medium comprises non-animal stabilized hyaluronic acid (“NASHA”). In some embodiments, said viscoelastic medium expands within said compartment to less than 10% of an original disposition volume. In some embodiments, said viscoelastic medium is injected one time every six months. In some embodiments, said viscoelastic medium is completely resorbed within 20 months. In some embodiments, said viscoelastic medium is completely resorbed within 16 months. In some embodiments, said viscoelastic medium is completely resorbed within 12 months.Example 5: Methods of Predicting a Desired Gel Composition Placement

[0146] An non-limiting example of a system configured to predict a gel composition placement is shown in FIG. 21. The system may be used to predict where a desired gel composition, such as a spacer, should be placed in a predicted cavity of an anatomical space. For the purposes of this example, it may be assumed that the anatomical cavity is a predicted space between the prostate and the colon of a subject.

[0147] For the purposes of this example, the analyzing component (2114) may be configured to predict the gel composition placement may be placed based on shape and symmetry of the gel composition as well as anatomical features present in the image of a subject (2122) such as those shown in FIG.s 8-12.

[0148] The device (2120) could be any device capable of storing and transmitting an image but for the purposes of this example it is considered a medical imaging device with communicative connectivity to the computing device (2110). For the purposes of this example, it is assumed the device (2120) is an MRI device connected to a local network where it may communicate with computing device (2110).

[0149] Upon receiving the image (2122), the computing device (2110) processes the image through the imaging component (2112) where the image will be preprocessed to prepare it for the analyzing component (2114). The imaging component (2112), for the purposes of this example, is assumed to be a software configured to resize the image to fit the input size of the analyzing component (2114) and apply normalizations to the image to remove noise and artifacts from the image (2122). The imaging component (2112) may either store the modified image to the computing device (2110) for later retrieval by a user, pass the modified image to the analyzing component (2114) or both. In the event the modified image is sent to the analyzing component (2114) it will be processed there and an output will be given based upon the function of the analyzing component (2114). For the purposes of this example, it is assumed the analyzing component is a machine learning model comprising multiple sub models connected and trained simultaneously (I.E. in an end-to end fashion) to process the image into semantic features (herein defined as features that may be associated with lexical items such as “tissue borders”, “cavity”, or anatomical features such as “pancreas”, “prostate”, “intestine”, “colon”, “transverse flexure”, for example) then producing, through generative means, such as through a diffusion model, a new image based upon some parameters such as target tissue, an indication of desired maximum radiation level, and a level of acceptable dosimetric impact on one or more objects such at tissues or organs which are assumed to be provided by the user as input to the analyzing component (2114) through the computing device (2110) for the purposes of this example. More details about such a system is described in the present disclosure. The generated image is a new version of the input image received by the analyzing component (2114) from theimaging component (2112) where the target tissue is moved to create a cavity where the spacer may be arranged such that the predicted level of radiation at a neighboring tissue to the target tissue is at or below the desired maximum radiation level. The generated image may then be processed by another sub model of the analyzing component (2114) to generate a new image with the spacer in the cavity of the previous generated image. The analyzing component may then analyze the image and output a series of measurements for the spacer that may be transmitted to a printing device, for the purposes of this example is a 3d printer, to produce a physical version of the spacer. The final generated image (2134) will, for the purposes of this example, comprise the new arrangement of tissues and the spacer. Example schematics of different parts of the analyzing component functionality are shown in FIG.s 1 - 5 & 20.

[0150] The final image may be saved to the computing device (2110) and / or be transmitted to the computing device (2130), which for the purposes of this example is a personal computer communicatively connected to the computing device through a local network. The computing device (2130) will then display the final image (2134) on a UI component which, for the purposes of this example, is assumed to be a program configured to at least display the image on a screen for a user.

[0151] In other embodiments, of this example, computing device (2110) may comprise a UI component. In those embodiments, the UI component of the computing device (2110) may perform the same functions as the UI component (2132) of computing device (2130). In those embodiments, the computing device (2110) may not provide the image (2134) to another computing device, and may instead display image (2134) through the UI component of the computing device (2110).Exemplary Embodiments

[0152] The embodiments listed below are exemplary embodiments of the systems and methods described herein, and do not limit the description above:1. A method of determining an object placement in a compartment of a human subject, comprising:(a) receiving an image of the compartment, wherein the image contains one or more objects of the compartment;(b) determining, using a machine learning model, a location and a value of a metric of the one or more objects of the compartment;(c) determining, using the machine learning model, a placement of a new object to be placed in the compartment based on the location and the value of the metric of the one or more objects;(d) providing an indication of the placement of the new object in the compartment. The method of embodiment 1, wherein the placement comprises a location of the new object. The method of embodiment 1 or 2, wherein the placement comprises a shape of the new object. The method of any one of embodiments 1 to 3, wherein the metric comprises a proximity of the one or more objects to each other, shape of the one or more objects, size measurements of the one or more objects at one or more locations, contours of the one or more objects, volume of the one or more objects, dosimetric impact on the one or more objects. The method of embodiment 3 or 4, wherein the shape comprises a measure of symmetry. The method of embodiment 5, further comprising calculating the measure of symmetry by:(a) identifying a set of locations and a direction of the one or more objects;(b) assessing a measurement at the set of locations using the direction of the one or more objects;(c) calculating the measure of symmetry based at least in part on the measurement at the set of locations. The method of embodiment 6, wherein the set of locations comprises an indication of the prostate mid-gland, an indication approximately 1 cm superior to the indication of the prostate mid-glad, and an indication approximately 1 cm inferior to the indication of the prostate mid-gland. The method of embodiment 6 or 7, wherein the direction is a transverse slice of the new object. The method of any one of embodiments 1 to 8, further comprising injecting a composition at the placement of the new object in the compartment. The method of embodiment 9, wherein the composition comprises a gel spacer. The method of any one of embodiments 1 to 10, wherein the determining the placement of the new object is based at least in part on user input. The method of embodiment 11, wherein the user input comprises at least one of an image, a text string indicating a portion of the human subject, an indication of desired new position of the one or more objects. The method of any one of embodiments 1 to 12, wherein the machine learning model takes as input at least one of a dose-volume histogram set, an extracted contour set, or a set of metadata.The method of embodiment 13, wherein the set of metadata is based, at least in part, on at least one of DICOM RTstruct files or RTdose files. The method of embodiment 13 or 14, wherein the extracted contour set comprises rectal volume. The method of any one of embodiments 4 to 15, wherein the volume of the one or more objects comprises a volume of the rectum. The method of embodiment 16, wherein the volume of the rectum is determined through a measurement derived from the image. The method of embodiment 17, wherein the measurement derived from the image comprises an anatomical region from the closer of rectosigmoid flexure or bottom of sacroiliac joint to the inferior extent of the ischial tuberosities. The method of any one of embodiments 4 to 18, wherein the contours of one or more objects comprises a set of spacer contours. The method of any one of embodiments 6 to 19, wherein the set of locations further comprises a location at a variable distance from the other locations in the set of locations, the distance based at least in part on the size of the prostate. The method of any one of embodiments 6 to 20, wherein a mean superior-inferior length of the new object is calculated from the set of locations. The method of any one of embodiments 1 to 21, wherein a composition is injected at the placement of the new object in the compartment. The method of any one of embodiments 1 to 22, further comprising providing instruction for printing the indication. The method of embodiment 23, wherein the printing comprises 3d printing. A method of classifying one or more objects in an image, comprising:(a) receiving an image of a compartment of a human subject, wherein the image contains one or more obj ects of the compartment;(b) detecting, using a machine learning model, the one or more objects of the compartment;(c) determining, using the machine learning model, a location and a value of a metric of at least a first object of the one or more objects;(d) identifying, using the machine learning model, the first object of the one or more objects as a class of a plurality of classes based on the value and the location, wherein the plurality of classes comprises an organ, a tumor, and a spacer; and(e) displaying the location and the class of the first object on a user interface. The method of embodiment 25, wherein the metric is:(a) a distance between the first object and at least a second object of the one or more objects;(b) a radiosensitivity of the first object;(c) a contrast to noise ratio of the first object;(d) a shape of the first object;(e) a volume of the first object; or(f) any combination thereof. The method of embodiment 25 or 26, wherein the metric is the radiosensitivity of the first object. The method of any one of embodiments 1 to 27, wherein the class of the first object is a spacer. The method of any one embodiments 1 to 28, further comprising receiving user input from the computing device. The method of embodiment 29, wherein the identifying is further based on the user input. The method of embodiment 29 or 30, wherein the user input comprises:(a) the location of the first object in the image;(b) a portion of the human subject associated with the image;(c) a condition or disorder of the human subject; or(d) any combination thereof. The method of any one of embodiments 25 to 31, wherein the identifying is further based on a comparison of the value of the metric associated with first object to a value of the metric associated with a second object of the one or more objects. The method of any one of embodiments 25 to 32, further comprising:(a) receiving feedback from the computing device; and(b) updating the machine learning model based on the feedback. The method of embodiment 33, wherein updating the machine learning model comprises adjusting one or more parameters of the machine learning model. The method of embodiment 33 or 34, wherein the feedback comprises:(a) an updated class of the first object;(b) a class of a second object of the one or more objects;(c) a location of the second object of the one or more objects;(d) an updated class of the first object;(e) an updated location of the first object; or(f) any combination thereof.A method of training a machine learning model comprising one or more parameters, comprising:(a) receiving an image of a compartment of a human, wherein the image contains one or more objects of the compartment;(b) determining, using the machine learning model, a location and a value of a metric of each object of the one or more objects;(c) identifying, using the machine learning model, each object of the one or more objects as a class of a plurality of classes;(d) receiving an updated class of the plurality of class for at least one object of the one or more objects; and(e) updating the one or more parameters of the machine learning model based on the updated class, thereby training the machine learning model. The method of embodiment 36, further comprising detecting, using the machine learning model, the one or more objects. The method of embodiment 36 or 37, wherein the metric is:(a) a distance between each object of the one or more objects;(b) a radiosensitivity of each object of the one or more objects;(c) a contrast to noise ratio of each object of the one or more objects;(d) a shape of each object of the one or more objects;(e) a volume of each object of the one or more objects; or(f) any combination thereof. The method of any one of embodiments 36 to 38, wherein the plurality of classes comprises an organ, a tumor, or a spacer. The method of any one of embodiments 36 to 39, wherein the updated type of the at least one object of the one or more objects indicates a spacer. The method of any one of embodiments 36 to 40, wherein the identifying is further based on a comparison of the value of the metric associated with first object to a value of the metric associated with a second object of the one or more objects. A method of displacing a gel composition, comprising:(a) detecting one or more organs and a tumor on an image of a compartment of the human subject;(b) identifying, using a machine learning model, a plurality of spaces of the compartment for administering the gel composition based on image information of the one or more organs and the tumor;-SO-(c) selecting, using the machine learning model a first space of the plurality of spaces based on a relationship between the first space and the one or more organs and the tumor; and(d) displacing the gel composition in the first space. The method of embodiment 42, wherein the relationship is based on:(a) a distance between the first space and the one or more organs;(b) a distance between the first space and the tumor;(c) a shape of the gel composition associated with the first space;(d) a volume of the gel compositions associated with the first space or(e) any combination thereof. The method of embodiment 42 or 43, wherein selecting on an image, using the machine learning model, the first space of the plurality of spaces is further based on a relationship between a second space and the one or more organs and the tumor. The method of embodiment 43 or 44, further comprising, determining, using the machine learning model:(a) the distance between the first space and the one or more organs;(b) the distance between the first space and the tumor;(c) the shape of the gel composition associated with the first space;(d) the volume of the gel compositions associated with the first space;(e) the amount of contrast (signal) on the image;(f) a distance between the second space and the one or more organs;(g) a distance between the second space and the tumor;(h) a shape of the gel composition associated with the second space;(i) a volume of the gel compositions associated with the second space; or(j) any combination thereof. The method of any one of embodiments 42 to 45, wherein the distance between the first space and the one or more organs is associated with a reduced dose of radiation applied to the one or more organs relative to another distance between another space of the plurality of spaces and the one or more organs. The method of any one of embodiments 42 to 46, wherein the distance between the first space and the tumor is associated with a therapeutically effective dose of radiation applied to the tumor. The method of embodiment 46 or 47, wherein the other space is the second space. The method of any one of embodiments 42 to 48, wherein at least two spaces of the plurality of spaces overlap.The method of any one of embodiments 42 to 49, wherein at least two spaces of the plurality of spaces do not overlap. The method of embodiment 49 or 50, wherein the at least two spaces comprise the first space. The method of any one of embodiments 1 to 51, further comprising displaying the first space. The method of embodiment 52, wherein displaying the first space comprises displaying the image of the first space within the compartment on a user interface. The method of embodiment 53, wherein the user interface is on a computing device. The method of any one of claims 45 to 54, wherein the machine model determines one or more of the following based on a signal to noise ratio of one or more portions of the image:(a) the distance between the first space and the one or more organs;(b) the distance between the first space and the tumor;(c) the shape of the gel composition associated with the first space;(d) the volume of the gel compositions associated with the first space;(e) the amount of contrast (signal) on the image;(f) the distance between the second space and the one or more organs;(g) the distance between the second space and the tumor;(h) the shape of the gel composition associated with the second space; or(i) the volume of the gel compositions associated with the second space. The method of any one of embodiments 42 to 55, wherein identifying, using a machine learning model, a plurality of spaces of the compartment comprises generating the plurality of spaces in the compartment. A method of reducing a dose of radiation applied to one or more organs within a compartment of a human subject, comprising:(a) detecting the one or more organs within the compartment and a tumor in the compartment;(b) generating a plurality of beam markers based on the one or more organs within the compartment and the tumor in the compartment, wherein a beam marker of the plurality of beam markers indicates a point of administration of a dose of radiation to the human subject; and(c) selecting a first beam marker of the plurality of beam markers based on a relationship of the first beam marker to the one or more organs displaced in the compartment and the tumor in the compartment, wherein the first beam marker indicates a reduced dose of radiation applied to the one or more organs relative toa reference point of administration, and wherein the first beam marker indicates a therapeutically effective dose of radiation applied to the tumor relative to the reference point of administration. The method of embodiment 57, further comprising detecting a spacer displaced in the compartment, wherein generating the plurality of beam markers is further based on the spacer and selecting the first beam marker of the plurality of beam markers is further based on the spacer. The method of embodiment 57 or 58, wherein generating the plurality of beam markers based on the one or more organs displaced in said compartment and the spacer displaced in said compartment comprises generating, using a machine learning model, the plurality of beam markers based on the one or more organs displaced in said compartment and the spacer displaced in said compartment. The method of any one of embodiments 57 to 59, wherein selecting the first beam of the plurality of beams based on the relationship of the first beam to at least one of the one or more organs and the spacer comprises wherein selecting, using a machine learning model, the first beam of the plurality of beams based on the relationship of the first beam to at least one of the one or more organs and the spacer. The method of any one of embodiments 57 to 60, prior to detecting the one or more organs displaced in the compartment, the spacer displaced in the compartment, the tumor displaced in the compartment, or any combination thereof, displacing the spacer adjacent to the one or more organs and the tumor. The method of any one of embodiments 57 to 61, wherein determining the shape is further based on one or more criteria associated with the plurality of beam markers. The method of any one of embodiments 57 to 62, wherein the tumor is in need of radiation therapy. The method of any one of embodiments 57 to 63, further comprising ranking the beam markers of the plurality of beam markers based on the one or more criteria and a plurality of relationships between the plurality of beam markers and the one or more organs, wherein the selecting is based on the ranking and the plurality of relationships comprises the relationship of the first beam. The method of embodiment 64, wherein the one or more criteria are associated with the one or more organs. The method of embodiment 65, wherein the one or more organs are visible in the image, and wherein the one or more criteria are based on locations of the one or more organs in the image.The method of embodiment 66, wherein at least one of the relationship is associated with a distance from an organ of the one or more organs. The method of any one of embodiments 65 to 67, wherein the relationship comprises:(a) a distance from one or more organs;(b) overlap of a beam marker of the plurality of beam markers with one or more organs; or(c) both. The method of any one of embodiments 57 to 68, wherein each beam marker of the plurality of beam markers overlap. The method of any one of embodiments 57 to 69, wherein each beam marker of the plurality of beam markers has a respective boundary. The method of any one of embodiments 57 to 70, wherein each beam marker of the plurality of beam markers share a common location. The method of any one of embodiments 57 to 71, further comprising displaying an indication of the first beam marker. The method of embodiment 64, wherein the selecting the first beam marker of the plurality of beam markers is based on the first beam marker having a highest ranking. The method of embodiment 64, wherein the first beam marker has a first topography and a first boundary, wherein the first topography and the first boundary result in less radiation applied to organs of the subject than respective topographies and respective boundaries of each other beam marker of the plurality of beam markers. The method of any one of embodiments 57 to 74, further comprising applying the dose of radiation based on the first beam marker. The method of embodiment 75, wherein the dose of radiation is associated with a dose radiation gradient, wherein an organ disposed within the dose radiation gradient receives a tolerable amount of radiation. The method of any one of embodiments 57 to 76, further comprising displaying the plurality of beam markers on a user interface. The method of any one of embodiments 57 to 77, wherein the reduced dose of radiation applied to the one or more organs is no more than 90% of the therapeutically effective dose of radiation applied to the tumor. The method of any one of embodiments 57 to 78, wherein the reduced dose of radiation applied to the one or more organs is no more than 50% of the therapeutically effective dose of radiation applied to the tumor.The method of any one of embodiments 57 to 79, wherein the reduced dose of radiation applied to the one or more organs is no more than 30% of the therapeutically effective dose of radiation applied to the tumor. The method of any one of embodiments 57 to 80, wherein the point of administration is indicated by the first beam marker. A method of evaluating placement of a composition, comprising:(a) detecting one or more organs and a tumor within a compartment of a human subject;(b) determining, using a machine learning model, a first space of the compartment based on the one or more organs and the tumor;(c) detecting the composition displaced in a second space of the compartment;(d) comparing the first space to the second space;(e) determining a score based on the comparing; and(f) providing the score to a user. The method of embodiment 82, further comprising displacing the composition in the second section. The method of embodiment 82 or 83, wherein determining the first space based on the one or more organs and the tumor is based on a relationship between the first space and the one or more organs and the tumor. The method of any one of embodiments 82 to 84, further comprising, determining, using the machine learning model:(a) a distance between the first space and the one or more organs;(b) a distance between the first space and the tumor;(c) a shape associated with the first space;(d) a volume associated with the first space;(e) a distance between the second space and the one or more organs;(f) a distance between the second space and the tumor;(g) a shape associated with the second space;(h) a volume associated with the second space; or(i) any combination thereof. The method of embodiment 85, wherein the relationship between the first space and the one or more organs and the tumor is based on one of more of:(a) the distance between the first space and the one or more organs;(b) the distance between the first space and the tumor;(c) the shape associated with the first space; or(d) the volume associated with the first space. The method of embodiment 85 or 86, further comprising determining a relationship between the second space and the one or more organs and the tumor. The method of embodiment 87, wherein determining the relationship between the second space and the one or more organs and the tumor is based on:(a) the distance between the second space and the one or more organs;(b) the distance between the second space and the tumor;(c) the shape associated with the second space;(d) the volume associated with the second space; or(e) any combination thereof. The method of any one of embodiments 82 to 88, wherein the first space and the second space at least partially overlap. The method of any one of embodiments 82 to 88, wherein the first space and the second space do not overlap. The method of any one of embodiments 82 to 90, wherein the first space is selected from a plurality of spaces based on:(a) the first space being associated with a reduced dose of radiation applied to the one or more organs relative to another dose of radiation applied to the one or more organs associated with another space of the plurality of spaces;(b) the first space being associated with a therapeutically effective dose of radiation applied to the tumor; or(c) both. The method of any one of embodiments 82 to 91, further comprising displaying one or more of the first space or the second space on a user interface. The method of any one of embodiments 82 to 92, wherein the one or more metrics are used to determine:(a) the distance between the first space and the one or more organs;(b) the distance between the first space and the tumor;(c) the shape of the gel composition associated with the first space;(d) the volume of the gel compositions associated with the first space;(e) the distance between the second space and the one or more organs;(f) the distance between the second space and the tumor;(g) the shape of the gel composition associated with the second space;(h) the volume of the gel compositions associated with the second space; or(i) any combination thereof.The method of embodiment 93, wherein the one or more metrics comprise:(a) a method of training a machine learning model, comprising:(i) detecting one or more organs and a tumor in a compartment in a human subject;(ii) generating a plurality of spaces of the compartment based on the one or more organs and the tumor in the compartment;(iii) determining one or more relationships between each space of the plurality of spaces and the one or more organs and the tumor based on one or more of:(1) a distance between each space of the plurality of spaces and the one or more organs;(2) a distance between each space of the plurality of spaces and the tumor;(3) a shape of each space of the plurality of spaces; or(4) a volume of each space of the plurality of spaces;(iv) receiving a first space of the plurality of spaces associated with a first relationship of the one or more relationships; and(v) adjusting the one or more relationships based on receiving the first space, thereby training the machine learning model. The method of embodiment 94, further comprising ranking the spaces of the plurality of spaces based on the one or more relationships. The method of embodiment 94 or 95, wherein each space of the plurality of spaces is associated with a dose of radiation applied to the one or more organs. The method of embodiment 96, wherein the first space is associated with a reduced dose of radiation applied to the one or more organs relative to doses of radiation applied to the one or more organs associated with other spaces of the plurality of spaces. The method of any one of embodiments 94 to 97, wherein the first space is associated with a therapeutically effective dose of radiation applied to the tumor. The method of any one of embodiments 94 to 98, wherein training the machine learning model comprises training the machine learning model to select a space associated with a reduced dose of radiation applied to the one or more organs and a therapeutically effective dose of radiation applied to the tumor. The method of any one of embodiments 94 to 99, wherein at least two spaces of the plurality of spaces overlap.The method of any one of embodiments 94 to 100, wherein at least two spaces of the plurality of spaces do not overlap. The method of embodiment 100, or 101, wherein the at least two spaces comprises the first space. The method of any one of the preceding embodiments, wherein, the composition comprises:(a) a viscoelastic medium; and(b) a first visual additive, wherein said first visual additive comprises a metal, and wherein said metal has a particle diameter of greater than or equal to 80 micrometers (pm). The method of any one of the preceding embodiments, wherein said metal is a precious metal. The method of any one of the preceding embodiments, wherein said metal or said precious metal is chosen from the group consisting of gold (Au), iodine (I), gadolinium (Gd), iron (Fe), barium (Ba), calcium (Ca), magnesium (Mg), and combinations thereof. The method of any one of the preceding embodiments, wherein said metal or precious metal is an isotope of gold (Au), iodine (I), gadolinium (Gd), iron (Fe), barium (Ba), calcium (Ca), magnesium (Mg), or combinations thereof. The method of any one of the preceding embodiments, wherein said metal or precious metal is a powder. The method of any one of the preceding embodiments, wherein said first visual additive has a radiographic density of between about 1.0 g / cm3 and 2.0 g / cm3. The method of any one of the preceding embodiments, wherein half of said first visual additive is configured to disperse within a tissue within nine months. The method of any one of the preceding embodiments, wherein said particle diameter of said metal or precious metal is between about 80 pm and about 120 pm. The method of any one of the preceding embodiments, wherein said particle diameter of said metal or precious metal is about 100 pm. The method of any one of the preceding embodiments, wherein said first visual additive further comprises one or more microbubbles. The method of any one of the preceding embodiments, further comprising a second visual additive. The method of any one of the preceding embodiments, wherein said second visual additive is different than said first visual additive.The method of any one of the preceding embodiments, wherein said second visual additive comprises one or more microbubbles. The method of any one of the preceding embodiments, wherein said first visual additive has a concentration within said viscoelastic medium of greater than 5 milligrams per milliliter (mg / ml). The method of any one of the preceding embodiments, wherein said first visual additive has a concentration within said viscoelastic medium of less than 90 mg / ml. The method of any one of the preceding embodiments, wherein said first visual additive has a concentration within said viscoelastic medium between 15 mg / ml and 30 mg / ml. The method of any one of the preceding embodiments, wherein said metal is between about 0.5 wt% and about 9.0 wt% of said composition. The method of any one of the preceding embodiments, wherein said metal or precious metal is between about 0.015 wt% and about 1.5 wt% of said composition. The method of any one of the preceding embodiments, wherein said viscoelastic medium comprises a volume of about 1 milliliter (ml) to about 50 ml. The method of any one of the preceding embodiments, wherein said composition is configured to be biodegradable. The method of any one of the preceding embodiments, wherein said composition is configured to be present on an imaging modality for at least 9 months. The method of any one of the preceding embodiments, wherein said composition is configured to not substantially migrate prior to or during imaging. The method of any one of the preceding embodiments, wherein said first visual additive is configured to not substantially migrate prior to or during imaging. The method of any one of the preceding embodiments, wherein said composition is configured to be disposed within a subject. The method of embodiment 126, wherein said subject is in need of radiography. The method of any one of the preceding embodiments, wherein said composition is configured to be disposed through injection. The method of any one of the preceding embodiments, wherein said composition is configured to be disposed subcutaneously or subepidermally. The method of any one of the preceding embodiments, wherein said composition is configured to be disposed within a compartment. The method of embodiment 130, wherein said compartment comprises one or more of, a fat tissue, a muscle tissue, and organ tissue, or a combination thereof.The method of any one of the preceding embodiments, wherein said composition is configured to be imaged on one or more modalities. The method of embodiment 132, wherein said one or more modalities comprise X-Ray, MRI, CT, CBCT, ultrasound, PET, SPECT or a combination thereof. The method of any one of the preceding embodiments, wherein said imaging comprises real-time imaging. The method of any one of the preceding embodiments, wherein said composition is configured to be imaged within 30 min, within 90 min, within 4 hours, within 8 hours, or within 4 days of disposition. The method of any one of the preceding embodiments, wherein said viscoelastic medium comprises hyaluronic acid, polyethylene glycol, or dextranomers. The method of any one of the preceding embodiments, wherein said viscoelastic medium comprises hyaluronic acid, polyethylene glycol, or dextranomers at a concentration between about 5 mg / ml to about 100 mg / ml. The method of any one of the preceding embodiments, wherein said viscoelastic medium comprises gel particles at a size range of about 0.08 mm to about 5 mm. The method of any one of the preceding embodiments, wherein said viscoelastic medium comprises non-animal stabilized hyaluronic acid (“NASHA”). The method of any one of the preceding embodiments, wherein said viscoelastic medium expands within said compartment to less than 10% of an original disposition volume. The method of any one of the preceding embodiments, wherein said viscoelastic medium is injected one time every six months. The method of any one of the preceding embodiments, wherein said viscoelastic medium is completely resorbed within 20 months. The method of any one of the preceding embodiments, wherein said viscoelastic medium is completely resorbed within 16 months. The method of any one of the preceding embodiments, wherein said viscoelastic medium is completely resorbed within 12 months. A non-transitory computer-readable medium comprising executable instructions that, when executed by one or more processors, cause the one or more processors to perform the method according to any one of embodiments 1 to 144. A computer system comprising: a memory comprising executable instructions; andat least one processor configured to execute the instructions, wherein when the at least one processor executes the instructions, the at least one processor causes the system to perform method according to any one of embodiments 1 to 145.Terms and Definitions

[0153] Unless otherwise defined, all technical terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this disclosure belongs.

[0154] As used herein, the singular forms “a,” “an,” and “the” include plural references unless the context clearly dictates otherwise. Any reference to “or” herein is intended to encompass “and / or” unless otherwise stated.

[0155] As used herein, the term “about” in some cases refers to an amount that is approximately the stated amount.

[0156] As used herein, the term “about” refers to an amount that is near the stated amount by 10%, 5%, or 1%, including increments therein.

[0157] As used herein, the term “about” in reference to a percentage refers to an amount that is greater or less the stated percentage by 10%, 5%, or 1%, including increments therein.

[0158] As used herein, the phrases “at least one,” “one or more,” and “and / or” are open-ended expressions that are both conjunctive and disjunctive in operation. For example, each of the expressions “at least one of A, B and C”, “at least one of A, B, or C”, “one or more of A, B, and C”, “one or more of A, B, or C” and “A, B, and / or C” means A alone, B alone, C alone, A and B together, A and C together, B and C together, or A, B and C together.Computing system

[0159] Referring to FIG. 13, a block diagram is shown depicting an exemplary machine that includes a computer system 1300 (e.g., a processing or computing system) within which a set of instructions can execute for causing a device to perform or execute any one or more of the aspects and / or methodologies for static code scheduling of the present disclosure. The components in FIG. 13 are examples only and do not limit the scope of use or functionality of any hardware, software, embedded logic component, or a combination of two or more such components implementing particular embodiments.

[0160] Computer system 1300 may include one or more processors 1301, a memory 1303, and a storage 1308 that communicate with each other, and with other components, via a bus 1340. The bus 1340 may also link a display 1332, one or more input devices 1333 (which may, for example, include a keypad, a keyboard, a mouse, a stylus, etc.), one or more output devices 1334, one or more storage devices 1335, and various tangible storage media 1336. All of theseelements may interface directly or via one or more interfaces or adaptors to the bus 1340. For instance, the various tangible storage media 1336 can interface with the bus 1340 via storage medium interface 1326. Computer system 1300 may have any suitable physical form, including but not limited to one or more integrated circuits (ICs), printed circuit boards (PCBs), mobile handheld devices (such as mobile telephones or PDAs), laptop or notebook computers, distributed computer systems, computing grids, or servers.

[0161] Computer system 1300 includes one or more processor(s) 1301 (e.g., central processing units (CPUs) or general purpose graphics processing units (GPGPUs)) that carry out functions. Processor(s) 1301 optionally contains a cache memory unit 1302 for temporary local storage of instructions, data, or computer addresses. Processor(s) 1301 are configured to assist in execution of computer readable instructions. Computer system 1300 may provide functionality for the components depicted in FIG. 13 as a result of the processor(s) 1301 executing non-transitory, processor-executable instructions embodied in one or more tangible computer-readable storage media, such as memory 1303, storage 1308, storage devices 1335, and / or storage medium 1336. The computer-readable media may store software that implements particular embodiments, and processor(s) 1301 may execute the software. Memory 1303 may read the software from one or more other computer-readable media (such as mass storage device(s) 1335, 1336) or from one or more other sources through a suitable interface, such as network interface 1320. The software may cause processor(s) 1301 to carry out one or more processes or one or more steps of one or more processes described or illustrated herein. Carrying out such processes or steps may include defining data structures stored in memory 1303 and modifying the data structures as directed by the software.

[0162] The memory 1303 may include various components (e.g., machine readable media) including, but not limited to, a random access memory component (e.g., RAM 1304) (e.g., static RAM (SRAM), dynamic RAM (DRAM), ferroelectric random access memory (FRAM), phasechange random access memory (PRAM), etc.), a read-only memory component (e.g., ROM 1305), and any combinations thereof. ROM 1305 may act to communicate data and instructions unidirectionally to processor(s) 1301, and RAM 1304 may act to communicate data and instructions bidirectionally with processor(s) 1301. ROM 1305 and RAM 1304 may include any suitable tangible computer-readable media described below. In one example, a basic input / output system 1306 (BIOS), including basic routines that help to transfer information between elements within computer system 1300, such as during start-up, may be stored in the memory 1303.

[0163] Fixed storage 1308 is connected bidirectionally to processor(s) 1301, optionally through storage control unit 1307. Fixed storage 1308 provides additional data storage capacity and mayalso include any suitable tangible computer-readable media described herein. Storage 1308 may be used to store operating system 1309, executable(s) 1310, data 1311, applications 1312 (application programs), and the like. Storage 1308 can also include an optical disk drive, a solid- state memory device (e.g., flash-based systems), or a combination of any of the above. Information in storage 1308 may, in appropriate cases, be incorporated as virtual memory in memory 1303.

[0164] In one example, storage device(s) 1335 may be removably interfaced with computer system 1300 (e.g., via an external port connector (not shown)) via a storage device interface 1325. Particularly, storage device(s) 1335 and an associated machine-readable medium may provide non-volatile and / or volatile storage of machine-readable instructions, data structures, program modules, and / or other data for the computer system 1300. In one example, software may reside, completely or partially, within a machine-readable medium on storage device(s) 1335. In another example, software may reside, completely or partially, within processor(s) 1301

[0165] Bus 1340 connects a wide variety of subsystems. Herein, reference to a bus may encompass one or more digital signal lines serving a common function, where appropriate. Bus 1340 may be any of several types of bus structures including, but not limited to, a memory bus, a memory controller, a peripheral bus, a local bus, and any combinations thereof, using any of a variety of bus architectures. As an example and not by way of limitation, such architectures include an Industry Standard Architecture (ISA) bus, an Enhanced ISA (EISA) bus, a Micro Channel Architecture (MCA) bus, a Video Electronics Standards Association local bus (VLB), a Peripheral Component Interconnect (PCI) bus, a PCI-Express (PCI-X) bus, an Accelerated Graphics Port (AGP) bus, HyperTransport (HTX) bus, serial advanced technology attachment (SATA) bus, and any combinations thereof.

[0166] Computer system 1300 may also include an input device 1333. In one example, a user of computer system 1300 may enter commands and / or other information into computer system 1300 via input device(s) 1333. Examples of an input device(s) 1333 include, but are not limited to, an alpha-numeric input device (e.g., a keyboard), a pointing device (e.g., a mouse or touchpad), a touchpad, a touch screen, a multi-touch screen, a joystick, a stylus, a gamepad, an audio input device (e.g., a microphone, a voice response system, etc.), an optical scanner, a video or still image capture device (e.g., a camera), and any combinations thereof. In some embodiments, the input device is a Kinect, Leap Motion, or the like. Input device(s) 1333 may be interfaced to bus 1340 via any of a variety of input interfaces 1323 (e.g., input interface 1323) including, but not limited to, serial, parallel, game port, USB, FIREWIRE, THUNDERBOLT, or any combination of the above.

[0167] In particular embodiments, when computer system 1300 is connected to network 1330, computer system 1300 may communicate with other devices, specifically mobile devices and enterprise systems, distributed computing systems, cloud storage systems, cloud computing systems, and the like, connected to network 1330. Communications to and from computer system 1300 may be sent through network interface 1320. For example, network interface 1320 may receive incoming communications (such as requests or responses from other devices) in the form of one or more packets (such as Internet Protocol (IP) packets) from network 1330, and computer system 1300 may store the incoming communications in memory 1303 for processing. Computer system 1300 may similarly store outgoing communications (such as requests or responses to other devices) in the form of one or more packets in memory 1303 and communicated to network 1330 from network interface 1320. Processor(s) 1301 may access these communication packets stored in memory 1303 for processing.

[0168] Examples of the network interface 1320 include, but are not limited to, a network interface card, a modem, and any combination thereof. Examples of a network 1330 or network segment 1330 include, but are not limited to, a distributed computing system, a cloud computing system, a wide area network (WAN) (e.g., the Internet, an enterprise network), a local area network (LAN) (e.g., a network associated with an office, a building, a campus or other relatively small geographic space), a telephone network, a direct connection between two computing devices, a peer-to-peer network, and any combinations thereof. A network, such as network 1330, may employ a wired and / or a wireless mode of communication. In general, any network topology may be used.

[0169] Information and data can be displayed through a display 1332. Examples of a display 1332 include, but are not limited to, a cathode ray tube (CRT), a liquid crystal display (LCD), a thin film transistor liquid crystal display (TFT-LCD), an organic liquid crystal display (OLED) such as a passive-matrix OLED (PMOLED) or active-matrix OLED (AMOLED) display, a plasma display, and any combinations thereof. The display 1332 can interface to the processor(s) 1301, memory 1303, and fixed storage 1308, as well as other devices, such as input device(s) 1333, via the bus 1340. The display 1332 is linked to the bus 1340 via a video interface 1322, and transport of data between the display 1332 and the bus 1340 can be controlled via the graphics control 1321. In some embodiments, the display is a video projector. In some embodiments, the display is a head-mounted display (HMD) such as a VR headset. In further embodiments, suitable VR headsets include, by way of non-limiting examples, HTC Vive, Oculus Rift, Samsung Gear VR, Microsoft HoloLens, Razer OSVR, FOVE VR, Zeiss VR One, Avegant Glyph, Freefly VR headset, and the like. In still further embodiments, the display is a combination of devices such as those disclosed herein.

[0170] In addition to a display 1332, computer system 1300 may include one or more other peripheral output devices 1334 including, but not limited to, an audio speaker, a printer, a storage device, and any combinations thereof. Such peripheral output devices may be connected to the bus 1340 via an output interface 1324. Examples of an output interface 1324 include, but are not limited to, a serial port, a parallel connection, a USB port, a FIREWIRE port, a THUNDERBOLT port, and any combinations thereof.

[0171] In addition or as an alternative, computer system 1300 may provide functionality as a result of logic hardwired or otherwise embodied in a circuit, which may operate in place of or together with software to execute one or more processes or one or more steps of one or more processes described or illustrated herein. Reference to software in this disclosure may encompass logic, and reference to logic may encompass software. Moreover, reference to a computer-readable medium may encompass a circuit (such as an IC) storing software for execution, a circuit embodying logic for execution, or both, where appropriate. The present disclosure encompasses any suitable combination of hardware, software, or both.

[0172] Those of skill in the art will appreciate that the various illustrative logical blocks, modules, circuits, and algorithm steps described in connection with the embodiments disclosed herein may be implemented as electronic hardware, computer software, or combinations of both. To clearly illustrate this interchangeability of hardware and software, various illustrative components, blocks, modules, circuits, and steps have been described above generally in terms of their functionality.

[0173] The various illustrative logical blocks, modules, and circuits described in connection with the embodiments disclosed herein may be implemented or performed with a general purpose processor, a digital signal processor (DSP), an application specific integrated circuit (ASIC), a field programmable gate array (FPGA) or other programmable logic device, discrete gate or transistor logic, discrete hardware components, or any combination thereof designed to perform the functions described herein. A general purpose processor may be a microprocessor, but in the alternative, the processor may be any conventional processor, controller, microcontroller, or state machine. A processor may also be implemented as a combination of computing devices, e.g., a combination of a DSP and a microprocessor, a plurality of microprocessors, one or more microprocessors in conjunction with a DSP core, or any other such configuration.

[0174] The steps of a method or algorithm described in connection with the embodiments disclosed herein may be embodied directly in hardware, in a software module executed by one or more processor(s), or in a combination of the two. A software module may reside in RAM memory, flash memory, ROM memory, EPROM memory, EEPROM memory, registers, harddisk, a removable disk, a CD-ROM, or any other form of storage medium known in the art. An exemplary storage medium is coupled to the processor such the processor can read information from, and write information to, the storage medium. In the alternative, the storage medium may be integral to the processor. The processor and the storage medium may reside in an ASIC. The ASIC may reside in a user terminal. In the alternative, the processor and the storage medium may reside as discrete components in a user terminal.

[0175] In accordance with the description herein, suitable computing devices include, by way of non-limiting examples, server computers, desktop computers, laptop computers, notebook computers, sub-notebook computers, netbook computers, netpad computers, set-top computers, media streaming devices, handheld computers, Internet appliances, mobile smartphones, tablet computers, personal digital assistants, video game consoles, and vehicles. Those of skill in the art will also recognize that select televisions, video players, and digital music players with optional computer network connectivity are suitable for use in the system described herein. Suitable tablet computers, in various embodiments, include those with booklet, slate, and convertible configurations, known to those of skill in the art.

[0176] In some embodiments, the computing device includes an operating system configured to perform executable instructions. The operating system is, for example, software, including programs and data, which manages the device’s hardware and provides services for execution of applications. Those of skill in the art will recognize that suitable server operating systems include, by way of non-limiting examples, FreeBSD, OpenBSD, NetBSD®, Linux, Apple® Mac OS X Server®, Oracle® Solaris®, Windows Server®, and Novell® NetWare®. Those of skill in the art will recognize that suitable personal computer operating systems include, by way of non-limiting examples, Microsoft® Windows®, Apple® Mac OS X®, UNIX®, and UNIX- like operating systems such as GNU / Linux®. In some embodiments, the operating system is provided by cloud computing. Those of skill in the art will also recognize that suitable mobile smartphone operating systems include, by way of non-limiting examples, Nokia® Symbian® OS, Apple® iOS®, Research In Motion® BlackBerry OS®, Google® Android®, Microsoft® Windows Phone® OS, Microsoft® Windows Mobile® OS, Linux®, and Palm® WebOS®. Those of skill in the art will also recognize that suitable media streaming device operating systems include, by way of non-limiting examples, Apple TV®, Roku®, Boxee®, Google TV®, Google Chromecast®, Amazon Fire®, and Samsung® HomeSync®. Those of skill in the art will also recognize that suitable video game console operating systems include, by way of nonlimiting examples, Sony® PS3®, Sony® PS4®, Microsoft® Xbox 360®, Microsoft Xbox One, Nintendo® Wii®, Nintendo® Wii U®, and Ouya®.Non-transitory computer readable storage medium

[0177] In some embodiments, the platforms, systems, media, and methods disclosed herein include one or more non-transitory computer readable storage media encoded with a program including instructions executable by the operating system of an optionally networked computing device. In further embodiments, a computer readable storage medium is a tangible component of a computing device. In still further embodiments, a computer readable storage medium is optionally removable from a computing device. In some embodiments, a computer readable storage medium includes, by way of non-limiting examples, CD-ROMs, DVDs, flash memory devices, solid state memory, magnetic disk drives, magnetic tape drives, optical disk drives, distributed computing systems including cloud computing systems and services, and the like. In some cases, the program and instructions are permanently, substantially permanently, semipermanently, or non-transitorily encoded on the media.Computer program

[0178] In some embodiments, the platforms, systems, media, and methods disclosed herein include at least one computer program, or use of the same. A computer program includes a sequence of instructions, executable by one or more processor(s) of the computing device’s CPU, written to perform a specified task. Computer readable instructions may be implemented as program modules, such as functions, objects, Application Programming Interfaces (APIs), computing data structures, and the like, that perform particular tasks or implement particular abstract data types. In light of the disclosure provided herein, those of skill in the art will recognize that a computer program may be written in various versions of various languages.

[0179] The functionality of the computer readable instructions may be combined or distributed as desired in various environments. In some embodiments, a computer program comprises one sequence of instructions. In some embodiments, a computer program comprises a plurality of sequences of instructions. In some embodiments, a computer program is provided from one location. In other embodiments, a computer program is provided from a plurality of locations. In various embodiments, a computer program includes one or more software modules. In various embodiments, a computer program includes, in part or in whole, one or more web applications, one or more mobile applications, one or more standalone applications, one or more web browser plug-ins, extensions, add-ins, or add-ons, or combinations thereof.Web application

[0180] In some embodiments, a computer program includes a web application. In light of the disclosure provided herein, those of skill in the art will recognize that a web application, invarious embodiments, utilizes one or more software frameworks and one or more database systems. In some embodiments, a web application is created upon a software framework such as Microsoft® .NET or Ruby on Rails (RoR). In some embodiments, a web application utilizes one or more database systems including, by way of non-limiting examples, relational, non-relational, object oriented, associative, and XML database systems. In further embodiments, suitable relational database systems include, by way of non-limiting examples, Microsoft® SQL Server, mySQL™, and Oracle®. Those of skill in the art will also recognize that a web application, in various embodiments, is written in one or more versions of one or more languages. A web application may be written in one or more markup languages, presentation definition languages, client-side scripting languages, server-side coding languages, database query languages, or combinations thereof. In some embodiments, a web application is written to some extent in a markup language such as Hypertext Markup Language (HTML), Extensible Hypertext Markup Language (XHTML), or extensible Markup Language (XML). In some embodiments, a web application is written to some extent in a presentation definition language such as Cascading Style Sheets (CSS). In some embodiments, a web application is written to some extent in a client-side scripting language such as Asynchronous Javascript and XML (AJAX), Flash® Actionscript, Javascript, or Silverlight®. In some embodiments, a web application is written to some extent in a server-side coding language such as Active Server Pages (ASP), ColdFusion®, Perl, Java™, JavaServer Pages (JSP), Hypertext Preprocessor (PHP), Python™, Ruby, Tel, Smalltalk, WebDNA®, or Groovy. In some embodiments, a web application is written to some extent in a database query language such as Structured Query Language (SQL). In some embodiments, a web application integrates enterprise server products such as IBM® Lotus Domino®. In some embodiments, a web application includes a media player element. In various further embodiments, a media player element utilizes one or more of many suitable multimedia technologies including, by way of non-limiting examples, Adobe® Flash®, HTML 5, Apple® QuickTime®, Microsoft® Silverlight®, Java™, and Unity®.

[0181] Referring to FIG. 14, in a particular embodiment, an application provision system comprises one or more databases 1400 accessed by a relational database management system (RDBMS) 1410. Suitable RDBMSs include Firebird, MySQL, PostgreSQL, SQLite, Oracle Database, Microsoft SQL Server, IBM DB2, IBM Informix, SAP Sybase, SAP Sybase, Teradata, and the like. In this embodiment, the application provision system further comprises one or more application severs 1420 (such as Java servers, .NET servers, PHP servers, and the like) and one or more web servers 1430 (such as Apache, IIS, GWS and the like). The web server(s) optionally expose one or more web services via app application programminginterfaces (APIs) 1440. Via a network, such as the Internet, the system provides browser-based and / or mobile native user interfaces.

[0182] Referring to FIG. 15, in a particular embodiment, an application provision system alternatively has a distributed, cloud-based architecture 1500 and comprises elastically load balanced, auto-scaling web server resources 1510 and application server resources 1520 as well synchronously replicated databases 1530.Mobile Application

[0183] In some embodiments, a computer program includes a mobile application provided to a mobile computing device. In some embodiments, the mobile application is provided to a mobile computing device at the time it is manufactured. In other embodiments, the mobile application is provided to a mobile computing device via the computer network described herein.

[0184] In view of the disclosure provided herein, a mobile application is created by techniques known to those of skill in the art using hardware, languages, and development environments known to the art. Those of skill in the art will recognize that mobile applications are written in several languages. Suitable programming languages include, by way of non-limiting examples, C, C++, C#, Objective-C, Java™, Javascript, Pascal, Object Pascal, Python™, Ruby, VB.NET, WML, and XHTML / HTML with or without CSS, or combinations thereof.

[0185] Suitable mobile application development environments are available from several sources. Commercially available development environments include, by way of non-limiting examples, AirplaySDK, alcheMo, Appcelerator®, Celsius, Bedrock, Flash Lite, .NET Compact Framework, Rhomobile, and WorkLight Mobile Platform. Other development environments are available without cost including, by way of non-limiting examples, Lazarus, MobiFlex, MoSync, and Phonegap. Also, mobile device manufacturers distribute software developer kits including, by way of non-limiting examples, iPhone and iPad (iOS) SDK, Android™ SDK, BlackBerry® SDK, BREW SDK, Palm® OS SDK, Symbian SDK, webOS SDK, and Windows® Mobile SDK.

[0186] Those of skill in the art will recognize that several commercial forums are available for distribution of mobile applications including, by way of non-limiting examples, Apple® App Store, Google® Play, Chrome WebStore, BlackBerry® App World, App Store for Palm devices, App Catalog for webOS, Windows® Marketplace for Mobile, Ovi Store for Nokia® devices, Samsung® Apps, and Nintendo® DSi Shop.Standalone Application

[0187] In some embodiments, a computer program includes a standalone application, which is a program that is run as an independent computer process, not an add-on to an existing process, e.g., not a plug-in. Those of skill in the art will recognize that standalone applications are often compiled. A compiler is a computer program(s) that transforms source code written in a programming language into binary object code such as assembly language or machine code. Suitable compiled programming languages include, by way of non-limiting examples, C, C++, Objective-C, COBOL, Delphi, Eiffel, Java™, Lisp, Python™, Visual Basic, and VB .NET, or combinations thereof. Compilation is often performed, at least in part, to create an executable program. In some embodiments, a computer program includes one or more executable complied applications.Web Browser Plug-in

[0188] In some embodiments, the computer program includes a web browser plug-in (e.g., extension, etc.). In computing, a plug-in is one or more software components that add specific functionality to a larger software application. Makers of software applications support plug-ins to enable third-party developers to create abilities which extend an application, to support easily adding new features, and to reduce the size of an application. When supported, plug-ins enable customizing the functionality of a software application. For example, plug-ins are commonly used in web browsers to play video, generate interactivity, scan for viruses, and display particular file types. Those of skill in the art will be familiar with several web browser plug-ins including, Adobe® Flash® Player, Microsoft® Silverlight®, and Apple® QuickTime®. In some embodiments, the toolbar comprises one or more web browser extensions, add-ins, or add-ons. In some embodiments, the toolbar comprises one or more explorer bars, tool bands, or desk bands.

[0189] In view of the disclosure provided herein, those of skill in the art will recognize that several plug-in frameworks are available that enable development of plug-ins in various programming languages, including, by way of non-limiting examples, C++, Delphi, Java™, PHP, Python™, and VB .NET, or combinations thereof.

[0190] Web browsers (also called Internet browsers) are software applications, designed for use with network-connected computing devices, for retrieving, presenting, and traversing information resources on the World Wide Web. Suitable web browsers include, by way of nonlimiting examples, Microsoft® Internet Explorer®, Mozilla® Firefox®, Google® Chrome, Apple® Safari®, Opera Software® Opera®, and KDE Konqueror. In some embodiments, the web browser is a mobile web browser. Mobile web browsers (also called microbrowsers, mini-browsers, andwireless browsers) are designed for use on mobile computing devices including, by way of nonlimiting examples, handheld computers, tablet computers, netbook computers, subnotebook computers, smartphones, music players, personal digital assistants (PDAs), and handheld video game systems. Suitable mobile web browsers include, by way of non-limiting examples, Google® Android® browser, RIM BlackBerry® Browser, Apple® Safari®, Palm® Blazer, Palm® WebOS® Browser, Mozilla® Firefox® for mobile, Microsoft® Internet Explorer® Mobile, Amazon® Kindle® Basic Web, Nokia® Browser, Opera Software® Opera® Mobile, and Sony® PSP™ browser.Software Modules

[0191] In some embodiments, the platforms, systems, media, and methods disclosed herein include software, server, and / or database modules, or use of the same. In view of the disclosure provided herein, software modules are created by techniques known to those of skill in the art using machines, software, and languages known to the art. The software modules disclosed herein are implemented in a multitude of ways. In various embodiments, a software module comprises a file, a section of code, a programming object, a programming structure, or combinations thereof. In further various embodiments, a software module comprises a plurality of files, a plurality of sections of code, a plurality of programming objects, a plurality of programming structures, or combinations thereof. In various embodiments, the one or more software modules comprise, by way of non-limiting examples, a web application, a mobile application, and a standalone application. In some embodiments, software modules are in one computer program or application. In other embodiments, software modules are in more than one computer program or application. In some embodiments, software modules are hosted on one machine. In other embodiments, software modules are hosted on more than one machine. In further embodiments, software modules are hosted on a distributed computing platform such as a cloud computing platform. In some embodiments, software modules are hosted on one or more machines in one location. In other embodiments, software modules are hosted on one or more machines in more than one location.Databases

[0192] In some embodiments, the platforms, systems, media, and methods disclosed herein include one or more databases, or use of the same. In view of the disclosure provided herein, those of skill in the art will recognize that many databases are suitable for storage and retrieval of imaging data. In various embodiments, suitable databases include, by way of non-limiting examples, relational databases, non-relational databases, object oriented databases, objectdatabases, entity-relationship model databases, associative databases, and XML databases. Further non-limiting examples include SQL, PostgreSQL, MySQL, Oracle, DB2, and Sybase. In some embodiments, a database is internet-based. In further embodiments, a database is webbased. In still further embodiments, a database is cloud computing-based. In a particular embodiment, a database is a distributed database. In other embodiments, a database is based on one or more local computer storage devices.Machine Learning

[0193] In some embodiments, machine learning algorithms are utilized to detect one or more objects of a compartment of a human subject. In some embodiments, machine learning algorithms are utilized to determine a location and a value of a metric of at least a first object of the one or more objects. In some embodiments, machine learning algorithms are utilized to identify the first object of the one or more objects as a class of a plurality of classes based on the value and the location, wherein the plurality of classes comprises an organ, a tumor, and a spacer.

[0194] In some embodiments, the machine learning algorithms herein employ one or more forms of labels including but not limited to human annotated labels and semi-supervised labels. In some embodiments, the machine learning algorithm utilizes regression modeling, wherein relationships between predictor variables and dependent variables are determined and weighted. In one embodiment, for example, the class is a dependent variable and is derived from the value and the location.

[0195] The human annotated labels can be provided by a hand-crafted heuristic. For example, the hand-crafted heuristic can comprise examining differences between tissue classes. The semisupervised labels can be determined using a clustering technique to find tissues similar to those flagged by previous human annotated labels and previous semi-supervised labels. The semisupervised labels can employ a XGBoost, a neural network, or both.

[0196] In some embodiments, the machine learning algorithms herein employ a distant supervision method, distant supervision method can create a large training set seeded by a small hand-annotated training set. The distant supervision method can comprise positive-unlabeled learning with the training set as the ‘positive’ class. The distant supervision method can employ a logistic regression model, a recurrent neural network, or both. The recurrent neural network can be advantageous for Natural Language Processing (NLP) machine learning.

[0197] Examples of machine learning algorithms can include a support vector machine (SVM), a naive Bayes classification, a random forest, a neural network, deep learning, or othersupervised learning algorithm or unsupervised learning algorithm for classification and regression. The machine learning algorithms can be trained using one or more training datasets.

[0198] In some embodiments, a machine learning algorithm is used to select catalogue images and recommend project scope. A non-limiting example of a multi -variate linear regression model algorithm is seen below: probability = Ao + Ai(Xi) + A2(X2) + As(X3) + A4(X4) + As(Xs) + Ae(Xe) + A7(X7).. .wherein Ai (Ai, A2, A3, A4, As, Ae, A7, . . .) are “weights” or coefficients found during the regression modeling; and Xi (Xi, X2, X3, X4, X5, Xe, X7, . . .) are data collected from the User. Any number of Ai and Xi variable can be included in the model. For example, in a non-limiting example Xi is the number of locations and X2 is the number of values. In some embodiments, the programming language “R” is used to run the model.

[0199] Training the ML model may include, in some cases, selecting one or more untrained data models to train using a training data set. The selected untrained data models may include any type of untrained ML models for supervised, semi-supervised, self-supervised, or unsupervised machine learning. The selected untrained data models may be specified based upon input (e.g., user input) specifying relevant parameters to use as predicted variables or other variables to use as potential explanatory variables. For example, the selected untrained data models may be specified to generate an output (e.g., a prediction) based upon the input. Conditions for training the ML model from the selected untrained data models may likewise be selected, such as limits on the ML model complexity or limits on the ML model refinement past a certain point. The ML model may be trained (e.g., via a computer system such as a server) using the training data set. In some cases, a first subset of the training data set may be selected to train the ML model. The selected untrained data models may then be trained on the first subset of training data set using appropriate ML techniques, based upon the type of ML model selected and any conditions specified for training the ML model. In some cases, due to the processing power requirements of training the ML model, the selected untrained data models may be trained using additional computing resources (e.g., cloud computing resources). Such training may continue, in some cases, until at least one aspect of the ML model is validated and meets selection criteria to be used as a predictive model.

[0200] In some cases, one or more aspects of the ML model may be validated using a second subset of the training data set (e.g., distinct from the first subset of the training data set) to determine accuracy and robustness of the ML model. Such validation may include applying the ML model to the second subset of the training data set to make predictions derived from the second subset of the training data. The ML model may then be evaluated to determine whether performance is sufficient based upon the derived predictions. The sufficiency criteria applied tothe ML model may vary depending upon the size of the training data set available for training, the performance of previous iterations of trained models, or user-specified performance requirements. If the ML model does not achieve sufficient performance, additional training may be performed. Additional training may include refinement of the ML model or retraining on a different first subset of the training dataset, after which the new ML model may again be validated and assessed. When the ML model has achieved sufficient performance, in some cases, the ML may be stored for present or future use. The ML model may be stored as sets of parameter values or weights for analysis of further input (e.g., further relevant parameters to use as further predicted variables, further explanatory variables, further user interaction data, etc.), which may also include analysis logic or indications of model validity in some instances. In some cases, a plurality of ML models may be stored for generating predictions under different sets of input data conditions. In some embodiments, the ML model may be stored in a database (e.g., associated with a server).Decision Tree and Random Forest

[0201] As described above, the machine learning model may implement a decision tree. A decision tree may be a supervised ML algorithm that can be applied to both regression and classification problems. Decision trees may mimic the decision-making process of a human brain. For example, a decision tree may grow from a root (base condition), and when it meets a condition (internal node / feature), it may split into multiple branches. The end of the branch that does not split anymore may be an outcome (leaf). A decision tree can be generated using a training data set according to the following operations: (1) Starting from a root node (the entire dataset), the algorithm may split the dataset in two branches using a decision rule or branching criterion; (2) each of these two branches may generate a new child node; (3) for each new child node, the branching process may be repeated until the dataset cannot be split any further; (4) each branching criterion may be chosen to maximize information gain (e.g., a quantification of how much a branching criterion reduces a quantification of how mixed the labels are in the children nodes). The labels may be the data or the classification that is predicted by the decision tree.

[0202] A random forest regression is an extension of the decision tree model that tends to yield more robust predictions by stretching the use of the training data partition. Whereas a decision tree may make a single pass through the data, a random forest regression may bootstrap 50% of the data (e.g., with replacement) and build many trees. Rather than using all explanatory variables as candidates for splitting, a random subset of candidate variables may be used for splitting, which may enable trees that have completely different data and differentvariables (hence the term random). The predictions from the trees, collectively referred to as the “forest,” may be then averaged together to produce the final prediction. Many trees (e.g., one hundred trees) may be included in a random forest model, with a number (e.g., 3, 6, 10, etc.) of terms sampled per split, a minimum of number (e.g., 1, 2, 4, 10, etc.) of splits per tree, and a minimum split size (e.g., 16, 32, 64, 128, 256, etc.). Random forests may be trained in a similar way as decision trees. Specifically, training a random forest may include the following operations: (1) select randomly k features from the total number of features; (2) create a decision tree from these k features using the same operations as for generating a decision tree; and (3) repeat the previous two operations until a target number of trees is created.

[0203] FIG. 16 illustrates a random forest 1600. The random forest 1600 (which may also be referred to as random forest model) is an ensemble of decision trees 1605, 1610, and 1615 with randomly selected features in each of the decision trees 1605, 1610, and 1615 so that it can provide more stable and accurate outcomes. Outcomes may be determined by majority voting in the case of a classification problem. In the example of FIG. 16, the random forest 1600, which has been trained previously by a training method, is used to decide between classifications A, B and C. For example, the random forest 1600, with only the three decision trees shown in Fig. 16, would return the classification A by majority voting.RF Classifiers

[0204] A Random Forest classifier, which generally comprises a plurality of decision trees wherein the output prediction is the mode of the predicted classifications of the individual trees, can be helpful in reducing overfitting to training data. An ensemble of decision trees can be constructed using a random subset of features at each split or decision node. The Gini criterion may be employed to choose the best partition, wherein decision nodes having the lowest calculated Gini impurity index are selected. At prediction time, a “vote” can be taken over all of the decision trees, and the majority vote (or mode of the predicted classifications) can be output as the predicted classification.

[0205] FIG. 17 is a schematic diagram illustrating a portion of an exemplary assessment model 1760 based on a Random Forest classifier. The assessment module may comprise a plurality of individual decision trees 1765, such as decision trees 1765a and 1765b, each of which can be generated independently using a random subset of features in the training data. Each decision tree may comprise one or more decision nodes such as decision nodes 1766 and 1767 shown in FIG. 17, wherein each decision node specifies a predicate condition. For example, decision node 1766 predicates the condition that, for a given dataset of an individual, the answer to ADI-R question #86 (age when abnormality is first evident) is 4 or less. Decision node 1767 predicatesthe condition that, for the given dataset, the answer to ADI-R question #52 (showing and direction attention) is 8 or less. At each decision node, a decision tree can be split based on whether the predicate condition attached to the decision node holds true, leading to prediction nodes (e.g., 1766a, 1766b, 1767a, 1767b). Each prediction node can comprise output values (‘value’ in FIG. 17) that represent “votes” for one or more of the classifications or conditions being evaluated by the assessment model. For example, in the prediction nodes shown in FIG. 17, the output values comprise votes for the individual being classified as having autism or being non-spectrum. A prediction node can lead to one or more additional decision nodes downstream (not shown in FIG. 17), each decision node leading to an additional split in the decision tree associated with corresponding prediction nodes having corresponding output values. The Gini impurity can be used as a criterion to find informative features based on which the splits in each decision tree may be constructed.

[0206] When the dataset being queried in the assessment model reaches a “leaf,” or a final prediction node with no further downstream splits, the output values of the leaf can be output as the votes for the particular decision tree. Since the Random Forest model comprises a plurality of decision trees, the final votes across all trees in the forest can be summed to yield the final votes and the corresponding classification of the subject. While only two decision trees are shown in FIG. 16, the model can comprise any number of decision trees. A large number of decision trees can help reduce overfitting of the assessment model to the training data, by reducing the variance of each individual decision tree. For example, the assessment model can comprise at least about 10 decision trees, for example at least about 100 individual decision trees or more.Computer Vision

[0207] The systems, the methods, the computer-readable media, and the techniques disclosed herein may implement one or more computer vision techniques. Computer vision is a field of artificial intelligence that uses computers to interpret and understand the visual world at least in part by processing one or more digital images from cameras and videos. In some instances, computer vision may use deep learning models (e.g., convolutional neural networks). Bounding boxes may be used in object detection techniques within computer vision. Bounding boxes may be annotation markers drawn around objects in an image. Bounding boxes, are often, although not always, may be rectangularly shaped. Bounding boxes may be applied by humans to training data sets. However, bounding boxes may also be applied to images by a trained machine learning that is trained to detect one or more different objects (e.g., humans, hands, faces, cars, etc.). In addition or in alternative to bounding boxes detection and tracking techniques may useany object detection annotation techniques, such as semantic segmentation, instance segmentation, polygon annotation, non-polygon annotation, landmarking, 3D cuboids, etc.Support Vector Machine

[0208] As also described above, the machine learning model may implement support vector machine learning techniques. In machine learning, support vector machines (SVMs) may be supervised learning models with associated learning algorithms that analyze data for classification and regression analysis. SVMs may be a robust prediction method, being based on statistical learning. SVMs may be well-suited for domains characterized by the existence of large amounts of data, noisy patterns, or the absence of general theories.

[0209] In general terms, SVMs may map input vectors into high dimensional feature space through non-linear mapping function, chosen a priori. In this high dimensional feature space, a desired separating hyperplane may be constructed. The optimal hyperplane may then be used to determine things such as class separations, regression fit, or accuracy in density estimation. More formally, a SVM constructs a hyperplane or set of hyperplanes in a high or infinitedimensional space, which can be used for classification, regression, or other tasks like outlier detection.

[0210] Support vectors may be defined as the data points that lie closest to the decision surface (or hyperplane). Support vectors may therefore be the data points that are most difficult to classify and may have direct bearing on the optimum location of the decision surface. Given a set of training examples, each marked as belonging to one of two categories, an SVM training algorithm may build a model that assigns new examples to one category or the other, making it a non-probabilistic binary linear classifier (although methods such as Platt scaling exist to use SVM in a probabilistic classification setting). SVM may map training examples to points in space so as to maximize the width of the gap between the two categories. New examples may then be mapped into that same space and predicted to belong to a category based on which side of the gap they fall. In addition to performing linear classification, SVMs can efficiently perform a non-linear classification using what is called the kernel trick, implicitly mapping their inputs into high-dimensional feature spaces.

[0211] Within a support vector machine, the dimensionally of the feature space may be large. For example, a fourth-degree polynomial mapping function may cause a 200-dimensional input space to be mapped into a 1.6 billionth dimensional feature space. The kernel trick and the Vapnik-Chervonenkis dimension may allow the SVM to thwart the “curse of dimensionality” limiting other methods and effectively derive generalizable answers from this very highdimensional feature space. Accordingly, SVMs may assist in discovering knowledge from vast amounts of input data.

[0212] Patent applications directed to support vector machines include, U.S. patent application Ser. Nos. 09 / 303,386; 09 / 303,387; 09 / 303,389; 09 / 305,345; all filed May 1, 1999; and U.S. patent application Ser. No. 09 / 568,301, filed May 9, 2000; and U.S. patent application Ser. No. 09 / 578,011, filed May 24, 2000 and also claims the benefit of U.S. Provisional Patent Application No. 60 / 161,806, filed Oct. 27, 1999; of U.S. Provisional Patent Application No. 60 / 168,703, filed Dec. 2, 1999; of U.S. Provisional Patent Application No. 60 / 184,596, filed Feb. 24, 2000; and of U.S. Provisional Patent Application Ser. No. 60 / 191,219, filed Mar. 22, 2000; all of which are herein incorporated in their entireties.Long short-term memory (LSTM)

[0213] Long short-term memory (LSTM) may be an artificial neural network used in the fields of artificial intelligence and deep learning. Unlike standard feedforward neural networks, LSTM may use feedback connections. The LSTM architecture may provide a short-term memory for a recurrent neural network (RNN). Such RNN can process not only single data points (such as images), but also entire sequences of data (such as speech or video). This characteristic may mean that LSTM networks are well-suited for processing and predicting data. The name of LSTM may refer to the analogy that a standard RNN has both “long-term memory” and “shortterm memory.” The connection weights and biases in the RNN may change once per episode of training, analogous to how physiological changes in synaptic strengths store long-term memories; the activation patterns in the network may change once per time-step, analogous to how the moment-to-moment change in electric firing patterns in the brain store short-term memories. The LSTM architecture may provide a short-term memory for an RNN that can last many (e.g., thousands) timesteps.

[0214] In some cases, a LSTM unit may comprise a cell, an input gate, an output gate, and a forget gate. The cell may remember values over arbitrary time intervals and the input gate, the output gate, and the forget gate may regulate the flow of information into and out of the cell. Forget gates may be used to decide what information to discard from a previous state by assigning a previous state, compared to a current input, a value between 0 and 1 (e.g., a (rounded) value of 1 may mean to keep the information, and a value of 0 means to discard it). The input gate may decide which pieces of new information to store in the current state, using the same system as the forget gates. The output gate may control which pieces of information in the current state to output (e.g., by assigning a value from 0 to 1 to the information, considering the previous and current states). Selectively outputting relevant information from the currentstate may allow the LSTM network to maintain useful, long-term dependencies to make predictions, both in current and future time-steps. LSTM networks may be well-suited to classifying, processing and making predictions based on time series data, since there can be lags of unknown duration between important events in a time series. LSTMs may resolve the vanishing gradient problem that can be encountered when training traditional RNNs. Relative insensitivity to gap length may be an advantage of LSTM over RNNs, hidden Markov models and other sequence learning methods in numerous applications.

[0215] In some cases, LSTMs may be used with one or more various types of neural networks (e.g., convolutional neural networks (CNNs), deep neural network (DNNs), recurrent neural networks (RNNs), etc.). In some cases, CNNs, LSTM, and DNNs are complementary in their modeling capabilities and may be combined a unified architecture. For example, in such unified architecture, CNNs may be well-suited at reducing frequency variations, LSTMs may be well- suited at temporal modeling, and DNNs may be well-suited for mapping features to a more separable space. For example, input features to a ML model using LSTM techniques in a unified architecture may include segment features for each of a plurality of segments. To process the input features for each of the plurality of segments, the segment features for the segment may be processed using one or more CNN layers to generate first features for the segment; the first features may be processed using one or more LSTM layers to generate second features for the segment; and the second features may be processed using one or more fully connected neural network layers to generate third features for the segments, where the third features may be used for classification operations. In some examples, to process the first features using the one or more LSTM layers to generate the second features, the first features may be processed using a linear layer to generate reduced features having a reduced dimension from a dimension of the first features; and the reduced features may be processed using the one or more LSTM layers to generate the second features. Short-term features having a first number of contextual frames may be generated based on the input features, where features generated using the one or more CNN layers may include long-term features having a second number of contextual frames that are more than the first number of contextual frames of the short-term features. In some cases, the one or more CNN layers, the one or more LSTM layers, and the one or more fully connected neural network layers may have been jointly trained to determine trained values of parameters of the one or more CNN layers, the one or more LSTM layers, and the one or more fully connected neural network layers. In some cases, the input features may include log-mel features having multiple dimensions. The input features may include one or more contextual frames indicating a temporal context of a signal (e.g., input data). Advantageously, implementations for such unified architecture may leverage complementary advantages associated with each of a CNN, LSTM,and DNN. For example, convolutional layers may reduce spectral variation in input, which may help the modeling of LSTM layers. Having DNN layers after LSTM layers may help reduce variation in the hidden states of the LSTM layers. Training the unified architecture jointly may provide a better overall performance. Training in the unified architecture may also remove the need to have separate CNN, LSTM and DNN architectures, which may be expensive (e.g., in computational resource, in network traffic, in financial resources, in energy consumption, etc.). By adding multi-scale information into the unified architecture, information may be captured at different time scales.Vision Transformer

[0216] A vision transformer (ViT) is a transformer-like model that handles vision processing tasks. While CNNs use convolution, a “local” operation bounded to a small neighborhood of an image, ViTs use self-attention, a “global” operation, since the ViT draws information from the whole image. This allows the ViT to capture distant semantic relevancies in an image effectively. Advantageously, ViTs may be well-suited catching long-term dependencies. In some cases, ViTs may be a competitive alternative to convolutional neural networks as ViTs may outperform the current state-of-the-art CNNs by almost four times in terms of computational efficiency and accuracy. ViTs may be well-suited to object detection, image segmentation, image classification, and action recognition. Moreover, ViTs may be applied in generative modeling and multi-model tasks, including visual grounding, visual-question answering, and visual reasoning. In some cases, ViTs may represent images as sequences, and class labels for the image are predicted, which allows models to learn image structure independently. Input images may be treated as a sequence of patches where every patch is flattened into a single vector by concatenating the channels of all pixels in a patch and then linearly projecting it to the desired input dimension. For example, a ViT architecture may include the following operations: (A) split an image into patches; (B) flatten the patches; (C) generate lower-dimensional linear embeddings from the flattened patches; (D) add positional embeddings; (E) provide the sequence as an input to a standard transformer encoder; (F) pretrain a model with image labels (e.g., fully supervised on a huge dataset); and (G) finetune on the downstream dataset for image classification. In some cases, there may be multiple blocks in a ViT encoder, with each block comprising three major processing elements: (1) Layer Norm; (2) Multi-head Attention Network; and (3) Multi-Layer Perceptrons. The Layer Norm may keep the training process on track and enable the model to adapt to the variations among the training images. The Multi-head Attention Network may be a network responsible for generating attention maps from the given embedded visual tokens. These attention maps may help the network focus on the most criticalregions in the image, such as object(s). The Multi-Layer Perceptrons may be a two-layer classification network with a Gaussian Error Linear Unit at the end. The final Multi-Layer Perceptrons block may be used as an output of the transformer. An application of softmax on this output can provide classification labels (e.g., if the application is image classification).Masked Autoencoder

[0217] Masked autoencoders (MAE) are scalable self-supervised learners for computer vision. The MAE leverages the success of autoencoders for various imaging and natural language processing tasks. Some computer vision models may be trained using supervised learning, such as using humans to look at images and created labels for the images, so that the model could learn the patterns of those labels (e.g., a human annotator would assign a class label to an image or draw bounding boxes around objects in the image). In contrast, self-supervised learning may not use any human-created labels. One technique for self-supervised image processing training using an MAE is for before an image is input into an encoder transformer, a certain set of masks are applied to the image. Due to the masks, pixels are removed from the image and therefore the model is provided an incomplete image. At a high level, the model’s task is to now learn what the full, original image looked like before the mask was applied.

[0218] In other words, MAE may include masking random patches of an input image and reconstructing the missing pixels. The MAE may be based on two core designs. First, an asymmetric encoder-decoder architecture, with an encoder that operates on the visible subset of patches (without mask tokens), along with a lightweight decoder that reconstructs the original image from the latent representation and mask tokens. Second, masking a high proportion of the input image, e.g., 75%, may yield a nontrivial and meaningful self-supervisory task. Coupling these two core designs enables training large models efficiently and effectively, thereby accelerating training (e.g., by 3* or more) and improving accuracy. MAE techniques may be scalable, enabling learning of high-capacity models that generalize well, e.g., a vanilla ViT- Huge model. As mentioned, the MAE may be effective in pre-training ViTs for natural image analysis. In some cases, the MAE uses the characteristic of redundancy of image information to observe partial images to reconstruct original images as a proxy task, and the encoder of the MAE may have the capability of deducing the content of the masked image area by aggregating context information. This contextual aggregation capability may be important in the field of image processing and analysis.Example System for Classifying Objects

[0219] FIG. 18 depicts a non-limiting example of a computer system 1800 for classifying one or more objects in an image. In this depicted example, system 1800 includes server 1810, device1820, and computing device 1830. In some embodiments, system 1800 does not include computing device 1830.

[0220] In this depicted embodiment, server 1810 further includes imaging component 1812 and classifying component 1814. In this depicted embodiment, server 1810 is configured to communicate with device 1820. In this depicted embodiment, server 1810 is configured to receive images from device 1820, such as image 1822. In this depicted embodiment, server 1810 is configured to communicate with computing device 1830. In this depicted embodiment, server 1810 is configured to provide images, such as refined image 1834, and other information, such as classifications 1836, to computing device 1830.

[0221] Device 1820 may be further configured to take one or more images. In some embodiments, device 1820 is a device capable of taking scans of a subject. In some embodiments, the scans of the subject may be of tissue of the subject. In some embodiments, the tissue of the subject may have been injected with a composition described herein. In some embodiments, device 1820 may be configured to take CT scans, MRI scans, X-rays, or transrectal ultrasounds. In some embodiments, the one or more images may comprise or may be derived from the CT scans, MRI scans, X-rays, or transrectal ultrasounds.

[0222] In this depicted embodiment, server 1810 is configured to receive image 1822, which includes an image of the tissue of a subject with a composition described herein injected into the tissue. In some embodiments, imaging component 1812 may be configured to improve the image in one or more ways, such as providing contrast to one or more regions of the image or improving the visibility of one or more objects and one or more borders of the objects to create refined image 1834. In some embodiments, classifying component 1814 may classify one or more objects of the image as described herein with one or more classifications 1836 (e.g., as a spacer, tumor, or organ) using one or more machine learning models as described herein.

[0223] In this depicted embodiment, server 1810 is further configured to provide refined image 1834 and / or classifications to computing device 1830. Computing device 1830 may further include user interface (UI) component 1832. In some embodiments, UI component 1832 includes a user interface capable of displaying one or more images such as refined image 1834 or image 1822.

[0224] Thus, after injecting a subject with a composition described herein, an image of the composition within the tissue may be obtained, refined, and analyzed, in order to provide a clear determination of where organs, tumors, and spacers may be present within the subject.

[0225] While preferred embodiments of the present disclosure have been shown and described herein, it will be obvious to those skilled in the art that such embodiments are provided by way of example only. Numerous variations, changes, and substitutions will now occur to thoseskilled in the art without departing from the disclosure. It should be understood that various alternatives to the embodiments of the disclosure described herein may be employed in practicing the disclosure.

Claims

CLAIMSWHAT IS CLAIMED IS:

1. A method of determining an object placement in a compartment of a human subject, comprising:(a) receiving an image of the compartment, wherein the image contains one or more objects of the compartment;(b) determining, using a machine learning model, a location and a value of a metric of the one or more objects of the compartment;(c) determining, using the machine learning model, a placement of a new object to be placed in the compartment based on the location and the value of the metric of the one or more objects;(d) providing an indication of the placement of the new object in the compartment.

2. The method of claim 1, wherein the placement comprises a location of the new object.

3. The method of claim 1 or 2, wherein the placement comprises a shape of the new object.

4. The method of any one of claims 1 to 3, wherein the metric comprises a proximity of the one or more objects to each other, shape of the one or more objects, size measurements of the one or more objects at one or more locations, contours of the one or more objects, volume of the one or more objects, dosimetric impact on the one or more objects.

5. The method of claim 3 or 4, wherein the shape comprises a measure of symmetry.

6. The method of claim 5, further comprising calculating the measure of symmetry by:(a) identifying a set of locations and a direction of the one or more objects;(b) assessing a measurement at the set of locations using the direction of the one or more objects;(c) calculating the measure of symmetry based at least in part on the measurement at the set of locations.

7. The method of claim 6, wherein the set of locations comprises an indication of the prostate mid-gland, an indication approximately 1 cm superior to the indication of the prostate midglad, and an indication approximately 1 cm inferior to the indication of the prostate midgland.

8. The method of claim 6 or 7, wherein the direction is a transverse slice of the new object.

9. The method of any one of claims 1 to 8, further comprising injecting a composition at the placement of the new object in the compartment.

10. The method of claim 9, wherein the composition comprises a gel spacer.

11. The method of any one of claims 1 to 10, wherein the determining the placement of the new object is based at least in part on user input.

12. The method of claim 11, wherein the user input comprises at least one of an image, a text string indicating a portion of the human subject, an indication of desired new position of the one or more objects.

13. The method of any one of claims 1 to 12, wherein the machine learning model takes as input at least one of a dose-volume histogram set, an extracted contour set, or a set of metadata.

14. The method of claim 13, wherein the set of metadata is based, at least in part, on at least one of DICOM RT struct files or RTdose files.

15. The method of claim 13 or 14, wherein the extracted contour set comprises rectal volume.

16. The method of any one of claims 4 to 15, wherein the volume of the one or more objects comprises a volume of the rectum.

17. The method of claim 16, wherein the volume of the rectum is determined through a measurement derived from the image.

18. The method of claim 17, wherein the measurement derived from the image comprises an anatomical region from the closer of rectosigmoid flexure or bottom of sacroiliac joint to the inferior extent of the ischial tuberosities.

19. The method of any one of claims 4 to 18, wherein the contours of one or more objects comprises a set of spacer contours.

20. The method of any one of claims 6 to 19, wherein the set of locations further comprises a location at a variable distance from the other locations in the set of locations, the distance based at least in part on the size of the prostate.

21. The method of any one of claims 6 to 20, wherein a mean superior-inferior length of the new object is calculated from the set of locations.

22. The method of any one of claims 1 to 21, wherein a composition is injected at the placement of the new object in the compartment.

23. The method of any one of claims 1 to 22, further comprising providing instruction for printing the indication.

24. The method of claim 23, wherein the printing comprises 3d printing.

25. A method of classifying one or more objects in an image, comprising:(a) receiving an image of a compartment of a human subject, wherein the image contains one or more obj ects of the compartment;(b) detecting, using a machine learning model, the one or more objects of the compartment;(c) determining, using the machine learning model, a location and a value of a metric of at least a first object of the one or more objects;(d) identifying, using the machine learning model, the first object of the one or more objects as a class of a plurality of classes based on the value and the location, wherein the plurality of classes comprises an organ, a tumor, and a spacer; and(e) displaying the location and the class of the first object on a user interface.

26. The method of claim 25, wherein the metric is:(a) a distance between the first object and at least a second object of the one or more objects;(b) a radiosensitivity of the first object;(c) a contrast to noise ratio of the first object;(d) a shape of the first object;(e) a volume of the first object; or(f) any combination thereof.

27. The method of claim 25 or 26, wherein the metric is the radiosensitivity of the first object.

28. The method of any one of claims 1 to 27, wherein the class of the first object is a spacer.

29. The method of any one claims 1 to 28, further comprising receiving user input from the computing device.

30. The method of claim 29, wherein the identifying is further based on the user input.

31. The method of claim 29 or 30, wherein the user input comprises:(a) the location of the first object in the image;(b) a portion of the human subject associated with the image;(c) a condition or disorder of the human subject; or(d) any combination thereof.

32. The method of any one of claims 25 to 31, wherein the identifying is further based on a comparison of the value of the metric associated with first object to a value of the metric associated with a second object of the one or more objects.

33. The method of any one of claims 25 to 32, further comprising:(a) receiving feedback from the computing device; and(b) updating the machine learning model based on the feedback.

34. The method of claim 33, wherein updating the machine learning model comprises adjusting one or more parameters of the machine learning model.

35. The method of claim 33 or 34, wherein the feedback comprises:(a) an updated class of the first object;(b) a class of a second object of the one or more objects;(c) a location of the second object of the one or more objects;(d) an updated class of the first object;(e) an updated location of the first object; or(f) any combination thereof.

36. A method of training a machine learning model comprising one or more parameters, comprising:(a) receiving an image of a compartment of a human, wherein the image contains one or more objects of the compartment;(b) determining, using the machine learning model, a location and a value of a metric of each object of the one or more objects;(c) identifying, using the machine learning model, each object of the one or more objects as a class of a plurality of classes;(d) receiving an updated class of the plurality of class for at least one object of the one or more objects; and(e) updating the one or more parameters of the machine learning model based on the updated class, thereby training the machine learning model.

37. The method of claim 36, further comprising detecting, using the machine learning model, the one or more objects.

38. The method of claim 36 or 37, wherein the metric is:(a) a distance between each object of the one or more objects;(b) a radiosensitivity of each object of the one or more objects;(c) a contrast to noise ratio of each object of the one or more objects;(d) a shape of each object of the one or more objects;(e) a volume of each object of the one or more objects; or(f) any combination thereof.

39. The method of any one of claims 36 to 38, wherein the plurality of classes comprises an organ, a tumor, or a spacer.

40. The method of any one of claims 36 to 39, wherein the updated type of the at least one object of the one or more objects indicates a spacer.

41. The method of any one of claims 36 to 40, wherein the identifying is further based on a comparison of the value of the metric associated with first object to a value of the metric associated with a second object of the one or more objects.

42. A method of displacing a gel composition, comprising:(a) detecting one or more organs and a tumor on an image of a compartment of the human subject;(b) identifying, using a machine learning model, a plurality of spaces of the compartment for administering the gel composition based on image information of the one or more organs and the tumor;(c) selecting, using the machine learning model a first space of the plurality of spaces based on a relationship between the first space and the one or more organs and the tumor; and(d) displacing the gel composition in the first space.

43. The method of claim 42, wherein the relationship is based on:(a) a distance between the first space and the one or more organs;(b) a distance between the first space and the tumor;(c) a shape of the gel composition associated with the first space;(d) a volume of the gel compositions associated with the first space or(e) any combination thereof.

44. The method of claim 42 or 43, wherein selecting on an image, using the machine learning model, the first space of the plurality of spaces is further based on a relationship between a second space and the one or more organs and the tumor.

45. The method of claim 43 or 44, further comprising, determining, using the machine learning model:(a) the distance between the first space and the one or more organs;(b) the distance between the first space and the tumor;(c) the shape of the gel composition associated with the first space;(d) the volume of the gel compositions associated with the first space;(e) the amount of contrast (signal) on the image;(f) a distance between the second space and the one or more organs;(g) a distance between the second space and the tumor;(h) a shape of the gel composition associated with the second space;(i) a volume of the gel compositions associated with the second space; or(j) any combination thereof.

46. The method of any one of claims 42 to 45, wherein the distance between the first space and the one or more organs is associated with a reduced dose of radiation applied to the one or more organs relative to another distance between another space of the plurality of spaces and the one or more organs.

47. The method of any one of claims 42 to 46, wherein the distance between the first space and the tumor is associated with a therapeutically effective dose of radiation applied to the tumor.

48. The method of claim 46 or 47, wherein the other space is the second space.

49. The method of any one of claims 42 to 48, wherein at least two spaces of the plurality of spaces overlap.

50. The method of any one of claims 42 to 49, wherein at least two spaces of the plurality of spaces do not overlap.

51. The method of claim 49 or 50, wherein the at least two spaces comprise the first space.

52. The method of any one of claims 1 to 51, further comprising displaying the first space.

53. The method of claim 52, wherein displaying the first space comprises displaying the image of the first space within the compartment on a user interface.

54. The method of claim 53, wherein the user interface is on a computing device.

55. The method of any one of claims 45 to 54, wherein the machine model determines one or more of the following based on a signal to noise ratio of one or more portions of the image:(a) the distance between the first space and the one or more organs;(b) the distance between the first space and the tumor;(c) the shape of the gel composition associated with the first space;(d) the volume of the gel compositions associated with the first space;(e) the amount of contrast (signal) on the image;(f) the distance between the second space and the one or more organs;(g) the distance between the second space and the tumor;(h) the shape of the gel composition associated with the second space; or(i) the volume of the gel compositions associated with the second space.

56. The method of any one of claims 42 to 55, wherein identifying, using a machine learning model, a plurality of spaces of the compartment comprises generating the plurality of spaces in the compartment.

57. A method of reducing a dose of radiation applied to one or more organs within a compartment of a human subject, comprising:(a) detecting the one or more organs within the compartment and a tumor in the compartment;(b) generating a plurality of beam markers based on the one or more organs within the compartment and the tumor in the compartment, wherein a beam marker of the plurality of beam markers indicates a point of administration of a dose of radiation to the human subject; and(c) selecting a first beam marker of the plurality of beam markers based on a relationship of the first beam marker to the one or more organs displaced in the compartment and the tumor in the compartment, wherein the first beam markerindicates a reduced dose of radiation applied to the one or more organs relative to a reference point of administration, and wherein the first beam marker indicates a therapeutically effective dose of radiation applied to the tumor relative to the reference point of administration.

58. The method of claim 57, further comprising detecting a spacer displaced in the compartment, wherein generating the plurality of beam markers is further based on the spacer and selecting the first beam marker of the plurality of beam markers is further based on the spacer.

59. The method of claim 57 or 58, wherein generating the plurality of beam markers based on the one or more organs displaced in said compartment and the spacer displaced in said compartment comprises generating, using a machine learning model, the plurality of beam markers based on the one or more organs displaced in said compartment and the spacer displaced in said compartment.

60. The method of any one of claims 57 to 59, wherein selecting the first beam of the plurality of beams based on the relationship of the first beam to at least one of the one or more organs and the spacer comprises wherein selecting, using a machine learning model, the first beam of the plurality of beams based on the relationship of the first beam to at least one of the one or more organs and the spacer.

61. The method of any one of claims 57 to 60, prior to detecting the one or more organs displaced in the compartment, the spacer displaced in the compartment, the tumor displaced in the compartment, or any combination thereof, displacing the spacer adjacent to the one or more organs and the tumor.

62. The method of any one of claims 57 to 61, wherein determining the shape is further based on one or more criteria associated with the plurality of beam markers.

63. The method of any one of claims 57 to 62, wherein the tumor is in need of radiation therapy.

64. The method of any one of claims 57 to 63, further comprising ranking the beam markers of the plurality of beam markers based on the one or more criteria and a plurality of relationships between the plurality of beam markers and the one or more organs, wherein the selecting is based on the ranking and the plurality of relationships comprises the relationship of the first beam.

65. The method of claim 64, wherein the one or more criteria are associated with the one or more organs.

66. The method of claim 65, wherein the one or more organs are visible in the image, and wherein the one or more criteria are based on locations of the one or more organs in the image.

67. The method of claim 66, wherein at least one of the relationship is associated with a distance from an organ of the one or more organs.

68. The method of any one of claims 65 to 67, wherein the relationship comprises:(a) a distance from one or more organs;(b) overlap of a beam marker of the plurality of beam markers with one or more organs; or(c) both.

69. The method of any one of claims 57 to 68, wherein each beam marker of the plurality of beam markers overlap.

70. The method of any one of claims 57 to 69, wherein each beam marker of the plurality of beam markers has a respective boundary.

71. The method of any one of claims 57 to 70, wherein each beam marker of the plurality of beam markers share a common location.

72. The method of any one of claims 57 to 71, further comprising displaying an indication of the first beam marker.

73. The method of claim 64, wherein the selecting the first beam marker of the plurality of beam markers is based on the first beam marker having a highest ranking.

74. The method of claim 64, wherein the first beam marker has a first topography and a first boundary, wherein the first topography and the first boundary result in less radiation applied to organs of the subject than respective topographies and respective boundaries of each other beam marker of the plurality of beam markers.

75. The method of any one of claims 57 to 74, further comprising applying the dose of radiation based on the first beam marker.

76. The method of claim 75, wherein the dose of radiation is associated with a dose radiation gradient, wherein an organ disposed within the dose radiation gradient receives a tolerable amount of radiation.

77. The method of any one of claims 57 to 76, further comprising displaying the plurality of beam markers on a user interface.

78. The method of any one of claims 57 to 77, wherein the reduced dose of radiation applied to the one or more organs is no more than 90% of the therapeutically effective dose of radiation applied to the tumor.

79. The method of any one of claims 57 to 78, wherein the reduced dose of radiation applied to the one or more organs is no more than 50% of the therapeutically effective dose of radiation applied to the tumor.

80. The method of any one of claims 57 to 79, wherein the reduced dose of radiation applied to the one or more organs is no more than 30% of the therapeutically effective dose of radiation applied to the tumor.

81. The method of any one of claims 57 to 80, wherein the point of administration is indicated by the first beam marker.

82. A method of evaluating placement of a composition, comprising:(a) detecting one or more organs and a tumor within a compartment of a human subject;(b) determining, using a machine learning model, a first space of the compartment based on the one or more organs and the tumor;(c) detecting the composition displaced in a second space of the compartment;(d) comparing the first space to the second space;(e) determining a score based on the comparing; and(f) providing the score to a user.

83. The method of claim 82, further comprising displacing the composition in the second section.

84. The method of claim 82 or 83, wherein determining the first space based on the one or more organs and the tumor is based on a relationship between the first space and the one or more organs and the tumor.

85. The method of any one of claims 82 to 84, further comprising, determining, using the machine learning model:(a) a distance between the first space and the one or more organs;(b) a distance between the first space and the tumor;(c) a shape associated with the first space;(d) a volume associated with the first space;(e) a distance between the second space and the one or more organs;(f) a distance between the second space and the tumor;(g) a shape associated with the second space;(h) a volume associated with the second space; or(i) any combination thereof.

86. The method of claim 85, wherein the relationship between the first space and the one or more organs and the tumor is based on one of more of:(a) the distance between the first space and the one or more organs;(b) the distance between the first space and the tumor;(c) the shape associated with the first space; or(d) the volume associated with the first space.

87. The method of claim 85 or 86, further comprising determining a relationship between the second space and the one or more organs and the tumor.

88. The method of claim 87, wherein determining the relationship between the second space and the one or more organs and the tumor is based on:(a) the distance between the second space and the one or more organs;(b) the distance between the second space and the tumor;(c) the shape associated with the second space;(d) the volume associated with the second space; or(e) any combination thereof.

89. The method of any one of claims 82 to 88, wherein the first space and the second space at least partially overlap.

90. The method of any one of claims 82 to 88, wherein the first space and the second space do not overlap.

91. The method of any one of claims 82 to 90, wherein the first space is selected from a plurality of spaces based on:(a) the first space being associated with a reduced dose of radiation applied to the one or more organs relative to another dose of radiation applied to the one or more organs associated with another space of the plurality of spaces;(b) the first space being associated with a therapeutically effective dose of radiation applied to the tumor; or(c) both.

92. The method of any one of claims 82 to 91, further comprising displaying one or more of the first space or the second space on a user interface.

93. The method of any one of claims 82 to 92, wherein the one or more metrics are used to determine:(a) the distance between the first space and the one or more organs;(b) the distance between the first space and the tumor;(c) the shape of the gel composition associated with the first space;(d) the volume of the gel compositions associated with the first space;(e) the distance between the second space and the one or more organs;(f) the distance between the second space and the tumor;(g) the shape of the gel composition associated with the second space;(h) the volume of the gel compositions associated with the second space; or(i) any combination thereof.

94. The method of claim 93, wherein the one or more metrics comprise:(a) a method of training a machine learning model, comprising:(i) detecting one or more organs and a tumor in a compartment in a human subject;(ii) generating a plurality of spaces of the compartment based on the one or more organs and the tumor in the compartment;(iii) determining one or more relationships between each space of the plurality of spaces and the one or more organs and the tumor based on one or more of:(1) a distance between each space of the plurality of spaces and the one or more organs;(2) a distance between each space of the plurality of spaces and the tumor;(3) a shape of each space of the plurality of spaces; or(4) a volume of each space of the plurality of spaces;(iv) receiving a first space of the plurality of spaces associated with a first relationship of the one or more relationships; and(v) adjusting the one or more relationships based on receiving the first space, thereby training the machine learning model.

95. The method of claim 94, further comprising ranking the spaces of the plurality of spaces based on the one or more relationships.

96. The method of claim 94 or 95, wherein each space of the plurality of spaces is associated with a dose of radiation applied to the one or more organs.

97. The method of claim 96, wherein the first space is associated with a reduced dose of radiation applied to the one or more organs relative to doses of radiation applied to the one or more organs associated with other spaces of the plurality of spaces.

98. The method of any one of claims 94 to 97, wherein the first space is associated with a therapeutically effective dose of radiation applied to the tumor.

99. The method of any one of claims 94 to 98, wherein training the machine learning model comprises training the machine learning model to select a space associated with a reduced dose of radiation applied to the one or more organs and a therapeutically effective dose of radiation applied to the tumor.

100. The method of any one of claims 94 to 99, wherein at least two spaces of the plurality of spaces overlap.

101. The method of any one of claims 94 to 100, wherein at least two spaces of the plurality of spaces do not overlap.

102. The method of claim 100, or 101, wherein the at least two spaces comprises the first space.

103. The method of any one of the preceding claims, wherein, the composition comprises:(a) a viscoelastic medium; and(b) a first visual additive, wherein said first visual additive comprises a metal, and wherein said metal has a particle diameter of greater than or equal to 80 micrometers (pm).

104. The method of any one of the preceding claims, wherein said metal is a precious metal.

105. The method of any one of the preceding claims, wherein said metal or said precious metal is chosen from the group consisting of gold (Au), iodine (I), gadolinium (Gd), iron (Fe), barium (Ba), calcium (Ca), magnesium (Mg), and combinations thereof.

106. The method of any one of the preceding claims, wherein said metal or precious metal is an isotope of gold (Au), iodine (I), gadolinium (Gd), iron (Fe), barium (Ba), calcium (Ca), magnesium (Mg), or combinations thereof.

107. The method of any one of the preceding claims, wherein said metal or precious metal is a powder.

108. The method of any one of the preceding claims, wherein said first visual additive has a radiographic density of between about 1.0 g / cm3 and 2.0 g / cm3.

109. The method of any one of the preceding claims, wherein half of said first visual additive is configured to disperse within a tissue within nine months.

110. The method of any one of the preceding claims, wherein said particle diameter of said metal or precious metal is between about 80 pm and about 120 pm.

111. The method of any one of the preceding claims, wherein said particle diameter of said metal or precious metal is about 100 pm.

112. The method of any one of the preceding claims, wherein said first visual additive further comprises one or more microbubbles.

113. The method of any one of the preceding claims, further comprising a second visual additive.

114. The method of any one of the preceding claims, wherein said second visual additive is different than said first visual additive.

115. The method of any one of the preceding claims, wherein said second visual additive comprises one or more microbubbles.

116. The method of any one of the preceding claims, wherein said first visual additive has a concentration within said viscoelastic medium of greater than 5 milligrams per milliliter (mg / ml).

117. The method of any one of the preceding claims, wherein said first visual additive has a concentration within said viscoelastic medium of less than 90 mg / ml.

118. The method of any one of the preceding claims, wherein said first visual additive has a concentration within said viscoelastic medium between 15 mg / ml and 30 mg / ml.

119. The method of any one of the preceding claims, wherein said metal is between about 0.5 wt% and about 9.0 wt% of said composition.

120. The method of any one of the preceding claims, wherein said metal or precious metal is between about 0.015 wt% and about 1.5 wt% of said composition.

121. The method of any one of the preceding claims, wherein said viscoelastic medium comprises a volume of about 1 milliliter (ml) to about 50 ml.

122. The method of any one of the preceding claims, wherein said composition is configured to be biodegradable.

123. The method of any one of the preceding claims, wherein said composition is configured to be present on an imaging modality for at least 9 months.

124. The method of any one of the preceding claims, wherein said composition is configured to not substantially migrate prior to or during imaging.

125. The method of any one of the preceding claims, wherein said first visual additive is configured to not substantially migrate prior to or during imaging.

126. The method of any one of the preceding claims, wherein said composition is configured to be disposed within a subject.

127. The method of claim 126, wherein said subject is in need of radiography.

128. The method of any one of the preceding claims, wherein said composition is configured to be disposed through injection.

129. The method of any one of the preceding claims, wherein said composition is configured to be disposed subcutaneously or subepidermally.

130. The method of any one of the preceding claims, wherein said composition is configured to be disposed within a compartment.

131. The method of claim 130, wherein said compartment comprises one or more of, a fat tissue, a muscle tissue, and organ tissue, or a combination thereof.

132. The method of any one of the preceding claims, wherein said composition is configured to be imaged on one or more modalities.

133. The method of claim 132, wherein said one or more modalities comprise X-Ray, MRI, CT, CBCT, ultrasound, PET, SPECT or a combination thereof.

134. The method of any one of the preceding claims, wherein said imaging comprises real-time imaging.

135. The method of any one of the preceding claims, wherein said composition is configured to be imaged within 30 min, within 90 min, within 4 hours, within 8 hours, or within 4 days of disposition.

136. The method of any one of the preceding claims, wherein said viscoelastic medium comprises hyaluronic acid, polyethylene glycol, or dextranomers.

137. The method of any one of the preceding claims, wherein said viscoelastic medium comprises hyaluronic acid, polyethylene glycol, or dextranomers at a concentration between about 5 mg / ml to about 100 mg / ml.

138. The method of any one of the preceding claims, wherein said viscoelastic medium comprises gel particles at a size range of about 0.08 mm to about 5 mm.

139. The method of any one of the preceding claims, wherein said viscoelastic medium comprises non-animal stabilized hyaluronic acid (“NASHA”).

140. The method of any one of the preceding claims, wherein said viscoelastic medium expands within said compartment to less than 10% of an original disposition volume.

141. The method of any one of the preceding claims, wherein said viscoelastic medium is injected one time every six months.

142. The method of any one of the preceding claims, wherein said viscoelastic medium is completely resorbed within 20 months.

143. The method of any one of the preceding claims, wherein said viscoelastic medium is completely resorbed within 16 months.

144. The method of any one of the preceding claims, wherein said viscoelastic medium is completely resorbed within 12 months.

145. A non-transitory computer-readable medium comprising executable instructions that, when executed by one or more processors, cause the one or more processors to perform the method according to any one of claims 1 to 144.

146. A computer system comprising: a memory comprising executable instructions; and at least one processor configured to execute the instructions, wherein when the at least one processor executes the instructions, the at least one processor causes the system to perform method according to any one of claims 1 to 145.

Images