LESION TRACKING IN 4D LONGITUDINAL IMAGING STUDIES

Organization Name

Inventor(s)

Florin-Cristian Ghesu of Skillman NJ (US)

Anamaria Vizitiu of Sfantu Gheorghe (RO)

LESION TRACKING IN 4D LONGITUDINAL IMAGING STUDIES - A simplified explanation of the abstract

This abstract first appeared for US patent application 18058896 titled 'LESION TRACKING IN 4D LONGITUDINAL IMAGING STUDIES

Simplified Explanation

The patent application describes systems and methods for tracking an anatomical object in medical images using machine learning techniques.

Extraction of embeddings at different scales from input medical images
Comparison of embeddings to determine the location of the anatomical object
Outputting the location of the anatomical object in the second input medical image

Potential Applications

This technology could be used in various medical imaging applications such as:

Surgical navigation systems
Radiation therapy planning
Image-guided interventions

Problems Solved

This technology helps in:

Accurate tracking of anatomical structures
Improving the efficiency of medical image analysis
Enhancing the precision of medical procedures

Benefits

The benefits of this technology include:

Increased accuracy in locating anatomical objects
Reduction in manual intervention for image analysis
Enhanced patient outcomes through precise medical interventions

Potential Commercial Applications

This technology has potential commercial applications in:

Medical device manufacturing
Healthcare software development
Research institutions in the field of medical imaging

Possible Prior Art

One possible prior art in this field is the use of traditional image processing techniques for tracking anatomical objects in medical images. However, the use of machine learning for extracting embeddings at different scales is a novel approach in this patent application.

Unanswered Questions

How does the machine learning model handle variations in anatomical structures across different patients?

The patent application does not provide details on how the machine learning model adapts to variations in anatomical structures among different patients. Further research may be needed to understand the generalizability of the model.

What is the computational cost associated with extracting embeddings at multiple scales from medical images?

The patent application does not discuss the computational resources required for extracting embeddings at different scales. Understanding the computational cost can help in assessing the feasibility of implementing this technology in real-world medical imaging systems.

Original Abstract Submitted

Systems and methods for tracking an anatomical object in medical images are provided. A first input medical image and a second input medical image each depicting an anatomical object of a patient are received. The first input medical image comprises a point of interest corresponding to a location of the anatomical object. A first set of embeddings associated with a plurality of scales is extracted from the first input medical image using a machine learning based extraction network. The plurality of scales comprises a coarse scale, one or more intermediate scales, and a fine scale. A second set of embeddings associated with the plurality of scales is extracted from the second input medical image using the machine learning based extraction network. A location of the anatomical object in the second input medical image is determined by comparing embeddings of the first set of embeddings corresponding to the point of interest with embeddings of the second set of embeddings. The location of the anatomical object in the second input medical image is output.