System for color-coding medical instrumentation and methods of use
A system including a biopsy needle device having a cannula with a distal end configured to sever a tissue sample, and a trocar disposed within the cannula having a notch configured to retain a tissue sample, wherein at least one of the cannula and the trocar is divided into at least two segments including different echogenic coatings, an ultrasound probe, and a processor configured and arranged to collect images from the ultrasound probe and color code the at least two segments based on at least one characteristic relating to different echogenic coatings, surface textures, surface contours and dimensions of the biopsy device.
1. A computer-implemented method for using machine-learning-based image analysis to colorize at least one medical device in medical imaging, the method comprising:
obtaining (i) medical imaging including a depiction of at least one medical device within an anatomy of a patient, the at least one medical device including one or more of a needle, a vacuum-assisted device, a clip or marker, an RFID clip or marker, a vascular access device, a stent, a filter, an angioplasty, a drain, or a catheter and (ii) one or more dimensions of the at least one medical device;
generating a segmentation of the at least one medical device by inputting (i) the medical imaging and (ii) the one or more dimensions of the at least one medical device into a trained machine-learning model that has been trained, based on (i) a plurality of training images of medical devices and (ii) labels of one or more portions of the medical devices, such that the trained machine-learning model is configured to identify portions of the medical device corresponding to the labels, wherein:
at least one of the portions includes a region of the at least one medical device that is not visible in the medical imaging; and
the segmentation of the not visible region is based on the input one or more dimensions of the at least one medical device;
modifying the medical imaging by applying a color-coding to the depiction of the at least one medical device based on the identified portions, the color-coding including a different color for each portion of the at least one medical device; and
causing a display to output the modified medical imaging.
2. The computer-implemented method of claim 1 , wherein the labels include surface contours of the portions, such that the trained machine-learning model is configured to determine surface contours for the portions of the medical device based on the corresponding labels.
3. The computer-implemented method of claim 1 , wherein:
each of different portions of the at least one medical device has a respective echogenicity that is different than each other; and
the labels are based on the respective echogenicities, such that the trained machine-learning model is further configured to segment the depiction of the at least one medical device in the medical imaging into portions based on relative echogenicity of the different portions.
4. The computer-implemented method of claim 1 , wherein:
the trained machine-learning model is further configured to segment the depiction of the at least one medical device in the medical imaging into the portions based on the one or more dimensions of the at least one medical device.
5. The computer-implemented method of claim 1 , wherein the medical imaging includes ultrasound medical imaging.
6. The computer-implemented method of claim 1 , wherein the color-coding includes one or more of a colorization of the at least one medical device, a coloring of a border of the at least one medical device, or a coloring of a region surrounding the at least one medical device.
7. A computer-implemented method for using machine-learning-based image analysis to colorize at least one medical device in medical imaging, the method comprising:
obtaining (i) medical imaging including a depiction of at least one medical device within an anatomy of a patient and (ii) one or more dimensions of the at least one medical device;
generating a segmentation of the at least one medical device by inputting (i) the medical imaging and (ii) the one or more dimensions of the at least one medical device into a trained machine-learning model that has been trained, based on (i) a plurality of training images of medical devices and (ii) labels of one or more portions of the medical devices, such that the trained machine-learning model is configured to identify portions of the medical device corresponding to the labels, wherein the labels correspond to one or more of surface contours of the one or more portions, foci associated with the one or more portions, or different echogenicities of the one or more portions, wherein:
at least one of the one or more portions includes a region of the at least one medical device is not visible in the medical imaging; and
the segmentation of the not visible region is based on the input one or more dimensions of the at least one medical device;
modifying the medical imaging by applying a color-coding to the depiction of the at least one medical device based on the identified portions, the color-coding including a different color for each portion of the at least one medical device; and
causing a display to output the modified medical imaging.
8. The computer-implemented method of claim 7 , wherein the medical imaging includes ultrasound medical imaging.
9. The computer-implemented method of claim 7 , wherein the color-coding includes one or more of a colorization of the at least one medical device, a coloring of a border of the at least one medical device, or a coloring of a region surrounding the at least one medical device.
10. A computer-implemented method for using machine-learning-based image analysis to colorize at least one medical device in medical imaging, the method comprising:
obtaining (i) medical imaging including a depiction of at least one medical device within an anatomy of a patient and (ii) one or more dimensions of the at least one medical device;
generating a color-coding of the at least one medical device by inputting (i) the medical imaging and (ii) the one or more dimensions of the at least one medical device into a trained machine-learning model that has been trained, based on (i) a plurality of training images of medical devices and (ii) labels of one or more portions of the medical devices, such that the trained machine-learning model is configured to identify portions of the medical device corresponding to the labels, wherein:
at least one of the one or more portions includes a region of the at least one medical device is not visible in the medical imaging; and
the color-coding of the not visible region is based on the input one or more dimensions of the at least one medical device;
modifying the medical imaging by applying a color-coding to the depiction of the at least one medical device based on the identified one or more portions, the color-coding including a different color for each portion of the at least one medical device; and
causing a display to output the modified medical imaging.
11. The computer-implemented method of claim 10 , wherein the labels include surface contours of the portions, such that the trained machine-learning model is configured to determine surface contours for the portions of the medical device based on the corresponding labels.
12. The computer-implemented method of claim 10 , wherein:
each of different portions of the at least one medical device has a respective echogenicity that is different than each other; and
the labels are based on the respective echogenicities, such that the trained machine-learning model is further configured to identify the portions based on relative echogenicity of the different portions.
13. The computer-implemented method of claim 10 , further comprising:
receiving one or more dimensions of the at least one medical device into the trained machine-learning model; and
the identifying of the one or more portions is further based on the one or more dimensions of the at least one medical device.
14. The computer-implemented method of claim 10 , wherein at least one of the one or more portions includes a region of the at least one medical device not visible in the medical imaging, the not visible region included in the at least one portion by the trained machine-learning model based on the input one or more dimensions of the medical device.