Predictive prognosis based on multimodal analysis

1. A method comprising:

obtaining one or more first images of a region of interest of an anatomy from an image source;

obtaining at least one of a text input, and one or more physiological signals of a patient, wherein the text input comprises information of at least one of micro-ethnicity information, an age, a race, a gender, a medical condition, a symptom, clinical history, a patient history, a medical test, medication information, and a cognitive analysis report;

automatically segmenting, through a neural network, one or more second images of at least one structure that resides within the one or more first images;

extracting one or more volumes of the at least one structure from the one or more first images of the region of interest;

determining a feature associated with the at least one structure based on the one or more volumes and one or more inputs;

rendering the feature in at least one of a two-dimensional (2D) format, a three-dimensional (3D) format, and at least one anatomical plane,

wherein the feature comprises at least one of the one or more volumes of the region of interest (ROI), a cortical thickness, an atrophy percentage, an asymmetry index score, a subfield volumetry of the region of interest, annular volume changes, a progressive supranuclear palsy (psp) index score, a magnetic resonance perfusion imaging (MRPI) score, a frontal horn width to intercaudate distance ratio (FH/CC), a medial temporal lobe atrophy (MTA) score, a global cortical atrophy (GCA) scale, identification of Intracranial bleeds, hemorrhage, microbleeds and their volume analysis, a fracture detection, a midline shift identification, a measurement of the midline shift identification and the at least one structure with respect to the midline shift identification, identifying a pathology associated with the at least one structure, classifying the pathology identified, a tissue density identification, an infarct identification, a Penumbra-core-viable tissue identification, classification and volume calculation, diffusion-weighted imaging (DWI) maps and apparent diffusion coefficient (ADC) maps of the at least one structure, perfusion maps comprising resting state functional magnetic resonance imaging (rsfMRI), an alberta stroke programme early CT score (ASPECTS) calculation, a collateral detection, a mismatch ratio calculation, an angiography labeling and/or annotation, a large vessel occlusion (LVO) detection, an Hypoperfusion index calculation, Diffusion tensor imaging (DTI) fiber tracks, neural pathway connectivity maps, correlation between a signal input, an image input and the text input, classifying the signal input, identifying a normal signal, identifying an abnormal signal, identifying a pre-ictal signal, identifying an ictal signal, extracting symptoms, and grading of condition specific effects;

transforming automatically the one or more physiological signals from a sinusoidal wave format to a quantitative format, wherein the quantitative format comprises a numerical representation of the one or more physiological signals;

predicting a prognosis based on correlation with the image input and integrated analysis of at least one spike detected, and the numerical representation of the one or more physiological signals; and

generating an analysis report based on the prognosis, wherein the analysis report comprising a snippet describing the prognosis, the one or more volumes of the at least one structure, one or more quantitative volumes, and a graphical representation of the prognosis.

2. The method of claim 1 , wherein automatically segmenting, through the neural network, the one or more second images of the at least one structure that resides within the one or more first images comprise:

performing a second quality analysis manually on the one or more second images that are segmented; and

determining whether the one or more second images, that are segmented, passes the second quality analysis.

3. The method of claim 2 , wherein determining whether the one or more second images, that are segmented, passes the second quality analysis comprises:

providing a user interface when the one or more second images that are segmented fails the second quality analysis;

manually editing and correcting at least one of boundaries and the one or more volumes of the at least one structure based on one or more inputs received; and

creating a mask for the at least one structure.

4. The method of claim 1 , wherein automatically segmenting, through the neural network, the one or more second images of the at least one structure that resides within the one or more first images comprise:

training the neural network using at least one of (a) the one or more first images, (b) the information of at least one of the micro-ethnicity information, the age, the race, the gender, the medical condition, the symptom, the clinical history, the patient history, the medical test, the medication information, and the cognitive analysis report, (c) the one or more physiological signals, (d) the one or more volumes of the at least one structure, (e) one or more reference volumes, and (f) one or more reference segmented second images.

5. The method of claim 3 , wherein manually editing and correcting at least one of the boundaries and the one or more volumes of the at least one structure based on the one or more inputs received comprises:

creating a log for the mask using the one or more inputs received;

retraining the neural network based on the log created; and

automatically segmenting, through the neural network, the one or more second images of the at least one structure in future based on the retraining provided to the neural network.

6. The method of claim 1 , wherein extracting the one or more volumes of the at least one structure from the one or more first images of the region of interest comprises:

assigning a voxel of a mask of the one or more second images, that are segmented, as a unit;

tabulating a plurality of units in the mask; and

estimating one or more quantitative volumes of the at least one structure from the plurality of units.

7. The method of claim 1 , further comprising:

recording the one or more volumes of the at least one structure in a database; and

categorizing the one or more volumes of the at least one structure in the database with respect to one or more categories of at least one of the micro-ethnicity information, an intracranial volume (ICV), the age, the race, the gender, a family history, the clinical history, the patient history, the symptom, psych analysis information, brain dominance information, food habitat information, stress information, and the medical condition.

8. The method of claim 1 , wherein extracting the one or more volumes of the at least one structure from the one or more first images of the region of interest:

extracting one or more boundaries of the at least one structure from the one or more first images; and

populating one or more voxels within the one or more boundaries of the at least one structure using one or more identifiers.

9. The method of claim 3 , wherein manually editing and correcting at least one of the boundaries, and the one or more volumes of the at least one structure based on the one or more inputs received comprises:

performing at least one of adding, and deleting one or more voxels within the boundaries of the at least one structure based on the one or more inputs received.

10. The method of claim 1 , further comprising:

detecting at least one spike within the one or more physiological signals that indicates abnormality; and

predicting the prognosis based on correlation and integrated analysis of the at least one spike detected, the text input, and the one or more volumes.

11. The method of claim 10 , further comprising:

correlating with at least one of temporal resolution and spatial resolution of the image input and detecting an abnormal region, using the neural network, in the one or more volumes based on the at least one spike detected; and

indicating the abnormal region using a different identifier.

12. The method of claim 10 , further comprising:

detecting an abnormal region, using the neural network, in the one or more physiological signals based on volumetric analysis; and

indicating the abnormal region, comprising the at least one spike, using a different identifier.

13. The method of claim 4 , wherein the one or more reference volumes range between 25th and 95th percentile, wherein the 25th and the 95th percentile are calculated by matching at least one of the age, the gender, the micro-ethnicity information, and an intracranial volume (ICV) of the patient with a normative population of individuals and then deriving the 25th and the 95th percentile references.

14. The method of claim 13 , wherein the 25th and the 95th percentile is calculated by matching the medical condition of the patient with a population of individuals having the medical condition and then deriving the 25th and the 95th percentile.

15. The method of claim 14 , further comprising:

predicting a first prognosis state of the patient based at least one of the medical condition, and first medication information of the patient at a first point of time and generating a first analysis report; and

predicting a second prognosis state of the patient based on at least one of the medical condition, and second medication information of the patient at a second point of time and generating a second analysis report.

16. The method of claim 15 , further comprising:

comparing the first prognosis state and the second prognosis state;

determining a percentage of one of a deterioration and an improvement in at least one of the one or more volumes, and one or more quantitative volumes based on comparison of the first prognosis state and the second prognosis state; and

training, the neural network, using at least one of medical condition, the first medication information, the second medication information, and the percentage of the deterioration or the improvement in at least one of the one or more volumes, and the one or more quantitative volumes at a plurality of different points of time.

17. The method of claim 16 , further comprising:

detecting a diagnosis, via the neural network, at a third point of time by comparing the first prognosis state and the second prognosis state based on the training;

performing a predictive prognosis, via the neural network, and predicting a third prognosis state of the patient at the third point of time based on the training; and

generating a third analysis report comprising a clinical analytical outcome at the third point of time.

18. A system comprising:

a server comprising a memory, and a processor communicatively coupled to the memory, the processor operable to

obtain one or more first images of a region of interest of an anatomy from an image source;

obtain at least one of a text input, and one or more physiological signals of a patient, wherein the text input comprises information of at least one of micro-ethnicity information, an age, a race, a gender, a medical condition, a symptom, clinical history, a patient history, a medical test, medication information, and a cognitive analysis report;

automatically segment, through a neural network, one or more second images of at least one structure that resides within the one or more first images;

extract one or more volumes of the at least one structure from the one or more first images of the region of interest;

determine a feature associated with the at least one structure based on the one or more volumes and one or more inputs;

render the feature in at least one of a two-dimensional (2D) format, a three-dimensional (3D) format, and at least one anatomical plane,

wherein the feature comprises at least one of the one or more volumes of the region of interests (ROI), a cortical thickness, an atrophy percentage, an asymmetry index score, a subfield volumetry of the region of interest, annular volume changes, a progressive supranuclear palsy (psp) index score, a magnetic resonance perfusion imaging (MRPI) score, a frontal horn width to intercaudate distance ratio (FH/CC), a medial temporal lobe atrophy (MTA) score, a global cortical atrophy (GCA) scale, identification of Intracranial bleeds, hemorrhage, microbleeds and their volume analysis, a fracture detection, a midline shift identification, a measurement of the midline shift identification and the at least one structure with respect to the midline shift identification, identifying a pathology associated with the at least one structure, classifying the pathology identified, a tissue density identification, an infarct identification, a Penumbra-core-viable tissue identification, classification and volume calculation, diffusion-weighted imaging (DWI) maps and apparent diffusion coefficient (ADC) maps of the at least one structure, perfusion maps comprising resting state functional magnetic resonance imaging (rsfMRI), an alberta stroke programme early CT score (ASPECTS) calculation, a collateral detection, a mismatch ratio calculation, an angiography labeling and/or annotation, a large vessel occlusion (LVO) detection, an Hypoperfusion index calculation, Diffusion tensor imaging (DTI) fiber tracks, neural pathway connectivity maps, correlation between a signal input, an image input and the text input, classifying the signal input, identifying a normal signal, identifying an abnormal signal, identifying a pre-ictal signal, identifying an ictal signal, extracting symptoms, and grading of condition specific effects;

transform automatically the one or more physiological signals from a sinusoidal wave format to a quantitative format, wherein the quantitative format comprises a numerical representation of the one or more physiological signals;

predict a prognosis based on correlation with the image input and integrated analysis of at least one spike detected, and the numerical representation of the one or more physiological signals; and

generate an analysis report based on the prognosis, wherein the analysis report comprising a snippet describing the prognosis, the one or more volumes of the at least one structure, one or more quantitative volumes, and a graphical representation of the prognosis.

19. The system of claim 18 , wherein the processor operable to detect at least one spike within the one or more physiological signals that indicates abnormality; and predict the prognosis based on correlation and integrated analysis of the at least one spike detected, the text input, and the one or more volumes.

20. A non-transitory computer storage medium storing a sequence of instructions, which when executed by a processor, causes:

obtaining one or more first images of a region of interest of an anatomy from an image source;

obtaining at least one of a text input, and one or more physiological signals of a patient, wherein the text input comprises information of at least one of micro-ethnicity information, an age, a race, a gender, a medical condition, a symptom, clinical history, a patient history, a medical test, medication information, and a cognitive analysis report;

automatically segmenting, through a neural network, one or more second images of at least one structure that resides within the one or more first images;

extracting one or more volumes of the at least one structure from the one or more first images of the region of interest;

determining a feature associated with the at least one structure based on the one or more volumes and one or more inputs;

rendering the feature in at least one of a two-dimensional (2D) format, a three-dimensional (3D) format, and at least one anatomical plane, wherein the feature comprises at least one of the one or more volumes of the region of interest (ROI), a cortical thickness, an atrophy percentage, an asymmetry index score, a subfield volumetry of the region of interest, annular volume changes, a progressive supranuclear palsy (psp) index score, a magnetic resonance perfusion imaging (MRPI) score, a frontal horn width to intercaudate distance ratio (FH/CC), a medial temporal lobe atrophy (MTA) score, a global cortical atrophy (GCA) scale, identification of Intracranial bleeds, hemorrhage, microbleeds and their volume analysis, a fracture detection, a midline shift identification, a measurement of the midline shift identification and the at least one structure with respect to the midline shift identification, identifying a pathology associated with the at least one structure, classifying the pathology identified, a tissue density identification, an infarct identification, a Penumbra-core-viable tissue identification, classification and volume calculation, diffusion-weighted imaging (DWI) maps and apparent diffusion coefficient (ADC) maps of the at least one structure, perfusion maps comprising resting state functional magnetic resonance imaging (rsfMRI), an alberta stroke programme early CT score (ASPECTS) calculation, a collateral detection, a mismatch ratio calculation, an angiography labeling and/or annotation, a large vessel occlusion (LVO) detection, an Hypoperfusion index calculation, Diffusion tensor imaging (DTI) fiber tracks, neural pathway connectivity maps, correlation between a signal input, an image input and the text input, classifying the signal input, identifying a normal signal, identifying an abnormal signal, identifying a pre-ictal signal, identifying an ictal signal, extracting symptoms, and grading of condition specific effects;

transforming automatically the one or more physiological signals from a sinusoidal wave format to a quantitative format, wherein the quantitative format comprises a numerical representation of the one or more physiological signals;

predicting a prognosis based on correlation with the image input and integrated analysis of at least one spike detected, and the numerical representation of the one or more physiological signals; and

generating an analysis report based on the prognosis, wherein the analysis report comprising a snippet describing the prognosis, the one or more volumes of the at least one structure, one or more quantitative volumes, and a graphical representation of the prognosis.