Method for generating a 3D physical model of a patient specific anatomic feature from 2D medical images
There is provided a method for generating a 3D physical model of a patient specific anatomic feature from 2D medical images. The 2D medical images are uploaded by an end-user via a Web Application and sent to a server. The server processes the 2D medical images and automatically generates a 3D printable model of a patient specific anatomic feature from the 2D medical images using a segmentation technique. The 3D printable model is 3D printed as a 3D physical model such that it represents a 1:1 scale of the patient specific anatomic feature. The method includes the step of automatically identifying the patient specific anatomic feature.
1. A method for defining a patient specific anatomical feature from 2D medical images, the method comprising:
receiving, by a server, 2D medical images of a patient;
automatically processing, by the server, the 2D medical images using a segmentation algorithm to assign a label for each pixel of the 2D medical images;
accessing, by the server, an anatomical knowledge dataset comprising a database of labelled 2D medical images of anatomical features;
using, by the server, an anatomical feature identification algorithm to probabilistically match the assigned label for each pixel of the 2D medical images against the anatomical knowledge dataset to generate a score for each pixel; and
generating, by the server, a confidence dataset comprising the label and the score for each pixel of the 2D medical images of the patient for defining the patient specific anatomical feature.
2. The method of claim 1 , wherein the label corresponds to a tissue type selected from at least one of bone, soft tissue, blood vessel, or organs.
3. The method of claim 1 , wherein the anatomical knowledge dataset comprises a database of labelled 2D medical images of anatomical features that are labelled using a medical imaging ontology.
4. The method of claim 1 , further comprising:
using, by the server, the anatomical feature identification algorithm to classify each pixel of the 2D medical images based on the score, and generate segmentation data the defines the patient specific anatomical feature based on the classification of each pixel of the 2D medical images; and
generating, by the server, a 3D surface mesh model defining a surface of the anatomical feature using the segmentation data.
5. The method of claim 4 , further comprising causing, by the server, a physical 3D model to be printed based on the 3D surface mesh model.
6. The method of claim 4 , wherein the 3D surface mesh model defines a virtual model.
7. The method of claim 4 , wherein the 3D surface mesh model is a file.
8. The method of claim 4 , further comprising displaying, by the server, the 3D surface mesh model on a web application.
9. The method of claim 4 , further comprising selecting, by the server, appropriate textures of the 3D surface mesh model to provide optimum visualization on a web application.
10. The method of claim 4 , further comprising updating, by the server, the anatomical knowledge dataset after the generation of the 3D surface mesh model.
11. The method of claim 4 , further comprising automatically detecting, by the server, an anomaly of the patient specific anatomical feature based on the segmentation data.
12. The method of claim 1 , further comprising improving, by the server, the segmentation algorithm based on end-user feedback.
13. The method of claim 1 , further comprising determining, by the server, information on the patient specific anatomical feature such as: volume, surface area, Hounsfield Unit standard deviation or average.
14. The method of claim 1 , further comprising determining, by the server, a status of the identified anatomical feature such as healthy or pathology by employing a comparative analysis of the expected appearance of an object against the normal appearance, without any deformity or pathology, of that object.
15. The method of claim 1 , wherein the segmentation algorithm uses a machine learning model trained from the database of labelled 2D medical images of anatomical features.
16. The method of claim 1 , wherein the segmentation algorithm includes a thresholding algorithm that automatically estimates a threshold value for the patient specific anatomical feature, and wherein pixels that are within the estimated threshold value are classified as being part of the patient specific anatomical feature.
17. A computer implemented system for defining a patient specific anatomical feature from 2D medical images, the system comprising at least one processor configured to:
receive 2D medical images of a patient;
automatically process the 2D medical images using a segmentation algorithm to assign a label for each pixel of the 2D medical images;
access an anatomical knowledge dataset comprising a database of labelled 2D medical images of anatomical features;
use an anatomical feature identification algorithm to probabilistically match the assigned label for each pixel of the 2D medical images against the anatomical knowledge dataset to generate a score for each pixel; and
generate a confidence dataset comprising the label and the score for each pixel of the 2D medical images of the patient for defining the patient specific patient specific anatomical feature.
18. The system of claim 17 , wherein the label corresponds to a tissue type selected from at least one of bone, soft tissue, blood vessel, or organs.
19. The system of claim 17 , wherein the database of labelled 2D medical images of anatomical features comprises 2D medical images that are labelled using a medical imaging ontology.
20. The system of claim 17 , wherein the at least one processor is further configured to:
use the anatomical feature identification algorithm to classify each pixel of the 2D medical images based on the score, and generate segmentation data the defines the patient specific anatomical feature based on the classification of each pixel of the 2D medical images; and
generate a 3D surface mesh model defining a surface of the anatomical feature using the segmentation data.
21. The system of claim 17 , wherein the at least one processor is further configured to cause a physical 3D model to be printed based on the 3D surface mesh model.
22. The system of claim 17 , wherein the at least one processor is further configured to determine a status of the identified anatomical feature such as healthy or pathology by employing a comparative analysis of the expected appearance of an object against the normal appearance, without any deformity or pathology, of that object.
23. The system of claim 17 , wherein the segmentation algorithm uses a machine learning model trained from the database of labelled 2D medical images of anatomical features.
24. The system of claim 17 , wherein the segmentation algorithm includes a thresholding algorithm that automatically estimates a threshold value for the patient specific anatomical feature, and wherein pixels that are within the estimated threshold value are classified as being part of the patient specific anatomical feature.
25. A non-transitory computer readable media having instructions that, when executed by a processor, cause the processor to:
receive 2D medical images of a patient;
automatically process the 2D medical images using a segmentation algorithm to assign a label for each pixel of the 2D medical images;
access an anatomical knowledge dataset comprising a database of labelled 2D medical images of anatomical features;
use an anatomical feature identification algorithm to probabilistically match the assigned label for each pixel of the 2D medical images against the anatomical knowledge dataset to generate a score for each pixel; and
generate a confidence dataset comprising the label and the score for each pixel of the 2D medical images of the patient for defining the patient specific patient specific anatomical feature.
26. The non-transitory computer readable media of claim 25 , wherein the label corresponds to a tissue type selected from at least one of bone, soft tissue, blood vessel, or organs.
27. The non-transitory computer readable media of claim 25 , wherein the processor is further configured to:
use the anatomical feature identification algorithm to classify each pixel of the 2D medical images based on the score, and generate segmentation data the defines the patient specific anatomical feature based on the classification of each pixel of the 2D medical images; and
generate a 3D surface mesh model defining a surface of the anatomical feature using the segmentation data.
28. The non-transitory computer readable media of claim 25 , wherein the processor is further configured to determine a status of the identified anatomical feature such as healthy or pathology by employing a comparative analysis of the expected appearance of an object against the normal appearance, without any deformity or pathology, of that object.
29. The non-transitory computer readable media of claim 25 , wherein the segmentation algorithm uses a machine learning model trained from the database of labelled 2D medical images of anatomical features.
30. The non-transitory computer readable media of claim 25 , wherein the segmentation algorithm includes a thresholding algorithm that automatically estimates a threshold value for the patient specific anatomical feature, and wherein pixels that are within the estimated threshold value are classified as being part of the patient specific anatomical feature.