Navigation of tubular networks
Inventors
Mintz, David S. • Ghoreyshi, Atiyeh • Jeevan, Prasanth • Xu, Yiliang • Yang, Gehua • Leotta, Matthew Joseph • Stewart, Charles V.
Assignees
Auris Health Inc • Kitware Inc
Publication Number
US-12089804-B2
Publication Date
2024-09-17
Expiration Date
2036-09-16
Interested in licensing this patent?
MTEC can help explore whether this patent might be available for licensing for your application.
Abstract
Methods and apparatuses provide improved navigation through tubular networks such as lung airways by providing improved estimation of location and orientation information of a medical instrument (e.g., an endoscope) within the tubular network. Various input data such as image data, EM data, and robot data are used by different algorithms to estimate the state of the medical instrument, and the state information is used to locate a specific site within a tubular network and/or to determine navigation information for what positions/orientations the medical instrument should travel through to arrive at the specific site. Probability distributions together with confidence values are generated corresponding to different algorithms are used to determine the medical instrument's estimated state.
Core Innovation
The patent describes methods and apparatuses for improved navigation through tubular networks, such as lung airways, by enhancing the estimation of location and orientation of a medical instrument within the network. The solution integrates multiple sources of input data—including image data, electromagnetic (EM) data, and robot data—processed by distinct algorithms to estimate the state of the instrument, including its position, orientation, depth, and branch selection. These estimations facilitate accurately locating specific sites in the tubular network and determining navigation guidance for the instrument to reach those sites.
A key aspect of the disclosed methods is the generation and use of probability distributions and confidence values produced by the different algorithms. Each algorithm provides a probability that the instrument is in a particular segment or branch, alongside a measure of confidence in the estimate considering external factors. These probabilities and confidence values are mathematically combined to form a robust, real-time estimated state for the instrument, which is then displayed to the operator on a user interface for improved navigation.
The background identifies the persistent challenge of inaccurate real-time localization of medical instruments within tubular anatomical structures, even when advanced devices and 3D models are available. Conventional navigation systems may not provide the required accuracy for instrument positioning, possibly leading to operator error and ineffective procedures. The invention addresses this issue by creating a system where various streams of data and navigation algorithms work together, quantifying both the likelihood and the reliability of instrument location and orientation estimates, thus providing more trustworthy information to clinicians during operative procedures.
Claims Coverage
The patent claims encompass three main inventive features encompassing multi-algorithm state estimation, algorithm-specific probability and confidence computations, and integration within a robotic system for navigating tubular networks.
Multi-algorithm state estimation for instrument position and orientation in tubular networks
The invention determines an estimated state for the tip of an elongated medical instrument inserted into a tubular network by: - Accessing data related to the instrument, including EM, image, and robot data - Determining probabilities for entering each opening (branch) using different algorithms - Calculating corresponding confidence values for each algorithm - Combining these by, for each opening/segment, multiplying the algorithm-specific probability by its confidence value and summing results across at least two algorithms (such as EM-based, image-based, and robot-based algorithms) - Using the sums to estimate position within segments and displaying the result on a display
Algorithm-specific probability and confidence calculation for navigation
For each algorithm (e.g., EM-based, image-based, robot-based), the method: - Determines probabilities for entering particular branches or openings in the tubular network - Computes a confidence value reflecting trust in the algorithm's estimate based on factors such as EM field distortion, registration accuracy, movement artifacts, sensor conditions, or anatomical depth - Multiplies the algorithm-specific probability by the associated confidence - Combines multiple algorithms' outputs for robust real-time estimation
Integrated robotic system with segmented model display and sensor fusion
A robotic system is described with: - A robotic manipulator and instrument driver for interacting with an elongated medical instrument within a tubular network - One or more processors and data storage containing instructions to: - Access robotic, EM, and/or image data - Compute probabilities and confidence values with different algorithms - Formally combine these (via multiplication and addition) to estimate which segment or branch contains the instrument tip - Display, on a model or in real-time, the tip’s position within the network
The claimed features center on combining probabilistic outcomes from multiple navigation algorithms, each weighted by its distinct confidence metric, to enhance the accuracy and reliability of medical instrument localization and guidance within tubular anatomical structures using an integrated robotic platform.
Stated Advantages
Improves real-time accuracy of location and orientation estimations for medical instruments within tubular networks.
Enables more reliable and convenient navigation during medical procedures by quantifying both likelihood and confidence of instrument localization.
Facilitates correction of errors in navigation through feedback from multiple algorithms, enhancing operational reliability.
Documented Applications
Navigation of medical instruments, including endoscopes and bronchoscopes, through lung airways and other tubular anatomical networks for procedures such as biopsies or surgical interventions.
Pre-operative planning and virtual endoscopy in patient-specific 3D models derived from CT scan data to prepare navigation paths prior to surgical operations.
Interested in licensing this patent?