Optimizing checkpoint locations along an insertion trajectory of a medical instrument using data analysis
Inventors
Shochat, Moran • ROTH, Ido • MOSKOVICH, Oz • Perlman, Danna • ATAROT, Gal
Assignees
Interested in licensing this patent?
MTEC can help explore whether this patent might be available for licensing for your application.
Publication Number
US-12588874-B2
Publication Date
2026-03-31
Expiration Date
Abstract
Provided are computer-implemented methods and systems for generating and/or utilizing model(s) for determining optimized checkpoint locations along a trajectory in an image-guided procedure for inserting a medical instrument to a target in a body of a patient based, inter alia, on data related to an automated medical device and/or to operation thereof.
Core Innovation
A computer-implemented method generates a checkpoint locations model for optimizing locations of a plurality of checkpoints along a non-linear trajectory in an image-guided procedure for inserting a medical instrument toward a target in a body of a patient, using an automated medical device. The method collects one or more datasets, including at least one dataset related to the automated medical device configured to steer a medical instrument toward a target in the body of a patient and/or to operation thereof, and creates a training set comprising at least a portion of the one or more datasets and one or more target parameters relating to checkpoint locations along a non-linear trajectory in one or more previous image-guided procedures.
The method trains the checkpoint locations model to predict checkpoint locations using the training set, calculates a checkpoint locations prediction error, and optimizes the checkpoint locations model using the calculated checkpoint locations prediction error. During execution of the image-guided procedure, the checkpoint locations model dynamically updates predicted checkpoint locations based on real-time feedback received from the automated medical device and procedural imaging data.
The system includes an inference module for utilizing the checkpoint locations model to optimize locations of checkpoints along a trajectory in an image-guided procedure for inserting a medical instrument toward a target. The inference module comprises a memory configured to store the one or more datasets and one or more processors configured to collect one or more datasets, create the training set, train the checkpoint locations model, calculate a checkpoint locations prediction error, and optimize the checkpoint locations model using the calculated error.
The checkpoint locations model is configured to dynamically update predicted checkpoint locations during execution based on real-time feedback and procedural imaging data, and the trajectory parameters are adjusted in real time according to the dynamically updated checkpoint locations.
Claims Coverage
The partial content includes two independent claims. Across the independent claims, there are five inventive features: data-driven training of a checkpoint locations model, optimization using checkpoint locations prediction error, dynamic updating during execution using real-time feedback and procedural imaging data, real-time adjustment of trajectory parameters, and an inference module configured to implement these functions.
Checkpoint locations model trained to predict checkpoint locations along a non-linear trajectory
Collecting one or more datasets; creating a training set with target parameters relating to checkpoint locations along a non-linear trajectory in one or more previous image-guided procedures; training the checkpoint locations model to predict checkpoint locations using the training set.
Optimizing the checkpoint locations model using checkpoint locations prediction error
Calculating a checkpoint locations prediction error; optimizing the checkpoint locations model using the calculated checkpoint locations prediction error.
Dynamically updating predicted checkpoint locations during execution using real-time feedback and procedural imaging data
Configuring the checkpoint locations model to dynamically update predicted checkpoint locations during execution of the image-guided procedure based on real-time feedback received from the automated medical device and procedural imaging data.
Adjusting trajectory parameters in real time according to dynamically updated checkpoint locations
Adjusting trajectory parameters of the automated medical device in real time according to the dynamically updated checkpoint locations.
Inference module for utilizing the checkpoint locations model and updating checkpoint locations in an image-guided procedure
Providing an inference module comprising a memory configured to store the one or more datasets and one or more processors configured to collect the one or more datasets, create the training set, train the checkpoint locations model, calculate the checkpoint locations prediction error, and optimize the checkpoint locations model using the calculated error, with the checkpoint locations model configured to dynamically update predicted checkpoint locations during execution based on real-time feedback and procedural imaging data.
The independent claims are directed to generating and optimizing a checkpoint locations model from datasets to predict checkpoint locations along a non-linear trajectory, updating predicted checkpoint locations during the image-guided procedure using real-time feedback and procedural imaging data, and adjusting automated medical device trajectory parameters in real time according to the dynamically updated checkpoint locations. The system claim implements these functions in an inference module comprising memory and processors.
Stated Advantages
Not explicitly described in patent.
Documented Applications
Not explicitly described in patent.
Interested in licensing this patent?