| 主权项 |
1. A method implemented on a machine having at least one processor comprising hardware, storage, and a communication platform connected to a network for generating a treatment plan for an organ disease, the method comprising:
segmenting, by the at least one processor, an organ on a first image in a first imaging modality obtained from a first imaging device, the first imaging modality providing anatomical information of the organ; mapping, by the at least one processor, the segmented organ including a boundary of the organ from the first image onto a second image in a second imaging modality obtained from a second imaging device, the second imaging modality providing functional information of the organ; determining, by the at least one processor, a boundary of the organ on the second image based on the boundary of the organ mapped from the first image by geometrically transforming the boundary of the organ from the first image to the second image; identifying, by the at least one processor, one or more defective organ regions of the organ on the second image; segmenting, by the at least one processor, the one or more defective organ regions on the second image based on intensity of the second image and the boundary of the organ to the second image; mapping, by the at least one processor, the segmented one or more defective regions from the second image onto the first image; identifying, by the at least one processor, a target object on the first image based on a combination of intensity and intensity gradient of the one or more defective regions mapped from the second image and intensity and intensity gradient of the organ on the first image, wherein the identifying the target object is based on an optimization that maximizes a combined intensity gradient of the first image in the first imaging modality and the second image in the second imaging modality and minimizes an intensity variance within the target object; identifying, by the at least one processor, anatomical structures on the first image; combining, by the at least one processor, the segmented first image and the second image into a combined image; generating, by the at least one processor, one or more treatment plans based on the combined image, the target object, and the anatomical structures; determining, by the at least one processor, remnant organ function for each of the one or more treatment plans; and displaying the remnant organ function for each of the one or more treatment plans. |
