Method and system for dynamic referencing in shape sensing of flexible devices

The method addresses FOSS limitations by dynamically referencing shape sensing using curvature profiles to track flexible devices' shapes and positions, achieving accurate and robust tracking without external systems, suitable for medical and robotic applications.

WO2026238867A1PCT designated stage Publication Date: 2026-11-19KATHOLIEKE UNIV LEUVEN
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
PCT/EP2026/064043
Authority / Receiving Office
WO · WO
Patent Type
Applications
Current Assignee / Owner
Priority Date
2025-12-01
Filing Date
2026-05-13
Publication Date
2026-11-19

AI Technical Summary

Technical Problem

Existing fiber-optic shape sensing (FOSS) methods for flexible devices face challenges in accurately determining the shape relative to a static coordinate frame and distinguishing between twisting and bending strains, leading to errors in shape reconstruction, and often rely on harmful fluoroscopic imaging and EM sensors.

Method used

A method and system for dynamic referencing in shape sensing using fiber-optic shape sensing (FOSS) that tracks the shape and position of flexible devices over time by selecting reference regions based on curvature profiles, allowing accurate mapping between evolving reference frames without external tracking systems.

Benefits of technology

Enables accurate, real-time tracking of flexible device shapes and positions in dynamic or constrained environments, enhancing localization accuracy and robustness against motion and deformation, suitable for medical and robotic applications.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure EP2026064043_19112026_PF_FP_ABST
    Figure EP2026064043_19112026_PF_FP_ABST
Patent Text Reader

Abstract

The invention relates to shape sensing and localization of flexible devices using fiber-optic sensors. A method tracks the device's shape and position over time by selecting stable reference regions along its curvature profile. These regions, identified via minimal curvature variance, enable displacement calculation between successive shapes in global and device-local reference frames. The system includes a flexible device with embedded sensors, a data acquisition module, and a processing unit for real-time mapping. Applications include medical catheters in constrained lumens and industrial robotic arms.
Need to check novelty before this filing date? Find Prior Art

Description

Method and System for Dynamic Referencing in Shape Sensing of FlexibleDevicesField of the invention

[0001] The present invention relates to the field of shape sensing and localization of flexible devices, particularly to systems and methods for determining the timevarying shape and position of such devices based on strain and / or curvature measurements. The invention is applicable to flexible devices comprising one or more shape-sensing elements, such as optical fibers with embedded sensors including fiber Bragg gratings (FBGs). In particular, the invention may be used in medical applications, including catheter-based interventions, as well as in industrial or robotic systems requiring precise shape tracking.Background of the invention

[0002] In a variety of technical fields, including robotics, industrial automation, and medical technology, there is a growing demand for precise localization and shape sensing of flexible devices. Such devices may include cables, robotic arms, inspection tools, or medical instruments that must navigate complex and constrained environments. Accurate, real-time knowledge of the device’s shape and position is critical for enhancing performance, safety, and automation.

[0003] In the medical field, for example, minimally invasive techniques have gained prominence for treating cardiovascular diseases. These techniques involve navigating flexible tools like catheters and guidewires through the vascular system to target specific areas, without the need for large surgical incisions typical of open surgery. Fluoroscopic imaging is commonly used for guidance but exposes both patients and clinicians to ionizing radiation and provides only 2D views.

[0004] Since the vasculature is inherently three-dimensional (3D), various sensor modalities have been explored for monitoring the 3D shape of the flexible instruments during the intervention. For example, bi-plane fluoroscopy systems have been used to obtain multiple projections of the 3D shape of the instruments. Despite improved visualization, bi-plane fluoroscopy tends to produce higher radiation levels compared to monoplane fluoroscopy. Efforts to reduce exposure, such aslowering frame rates and optimizing protocols, are reported, but these reduce the quality of the visualization for the clinician. Low-dose systems filter lower energy radiation, but increase scattering, thereby reducing radiation exposure for patients while maintaining similar levels for the clinicians compared to conventional systems.

[0005] Electromagnetic (EM) tracking systems offer an alternative by precisely locating small sensors embedded in the instruments without requiring a direct line of sight. In EM-based catheter shape tracking approaches, the shape is typically approximated using parametric curves such as B-splines or Bezier curves. However, a relatively high number of EM sensors need to be embedded to achieve accurate shape estimation. This not only complicates the catheter manufacturing process, but also makes the catheter more stiff and fragile. Moreover, EM sensors can only operate within a limited tracking volume and are susceptible to EM interference in the operating room.

[0006] A promising alternative for catheter shape tracking is fiber-optic shape sensing (FOSS). FOSS allows for real- time 3D reconstruction of the catheter’s shape based on strain measurements along embedded optical fibers. Fiber-optic strain sensing can be categorized into multi-point sensing (Fiber Bragg Grating) and distributed sensing (Rayleigh scattering and Brillouin scattering). While the methods differ in how they measure strains, the same shape reconstruction process can be applied. Fiber Bragg Gratings (FBG) enable high-frequency data acquisition, making them appealing for robotic applications, but have a lower spatial resolution compared to Rayleigh scattering-based techniques. Document US2011119023 Al discloses an exemplary surgical system with a shape sensor in a shape sensing segment between two links of a kinematic chain included in the surgical system. Correspondingly, document WO2013074726 Al discloses a plurality of position and shape sensors (FBGs) distributed along the length of a medical device, allowing easy and fast determination of the flexible device’s pose and shape.

[0007] In general, FOSS offers significant advantages, such as being immune to EM interference and having an extremely thin profile, making it convenient to embed the sensor in instruments. However, FOSS faces some challenges. It onlyprovides the shape relative to the first (most proximal) measurement point along the fiber. This means that the shape is not located and oriented relative to a static coordinate frame. Additionally, strain induced by twisting of the fiber cannot be distinguished from strain induced by bending. This can lead to significant errors in the reconstructed shape. It is worth noting that all fiber-optic strain sensing methods suffer from this limitation.

[0008] Various strategies have been suggested to tackle these challenges. One approach for addressing effect of twist involves using helically wrapped fibers, to enhance sensitivity to twisting-induced strain. While the reported shape reconstruction improvements are promising, the benefits must be weighed against the higher cost associated with helically wrapped fibers and their distributed shape sensing interrogator. Other approaches either implicitly or explicitly compensate for fiber twist while providing localization of the shape through sensor fusion.

[0009] The shape sensor localization problem has been addressed by fusing FOSS with additional sensors, such as (bi-plane) fluoroscopy or EM sensors. For instance, bi-plane fluoroscopic images may be used to register the shape in the fluoroscopic frame while also estimating twisting-induced strain. Others fused shape and EM sensing to localize the shape by registering EM sensors to their corresponding points on the FBG fiber. One approach using a single tip EM sensor and the other using a single monoplane fluoroscopic image. Another approach used EM sensors and sparse monoplane fluoroscopic images to localize the shape while also compensating for twist. Document US2025 / 0032195 Al discloses an example of such system using pre- or intra-operative imaging data based on fluoroscopy and an EM sensing or sensing via impedance, ultrasound, or fiber optic sensing, based on an external reference frame.

[0010] Despite offering excellent shape localization, these methods still rely on harmful fluoroscopic imaging and / or EM sensors, which have limitations as mentioned above. Another approach exploits knowledge of the preoperative vessel geometry for shape localization. In light of the foregoing challenges, there is a need for improved techniques for determining the location and shape of an optical fiber.The present invention addresses this need by providing a method and a system for locating and shape sensing an optical fiber.Summary of the invention

[0011] It is an object of embodiments of the present invention to provide a method, system and device for locating and sensing the shape of a flexible device over time, based on acquired data indicative of deformation, such as strain and / or curvature data, thereby enabling dynamic tracking of the shape and position of the flexible device within a reference frame without the need of additional external tracking systems. Indeed, the present invention enables such tracking without the need for fluoroscopic imaging, electromagnetic (EM) sensors, or other external tracking systems, by relying instead on fiber-optic shape sensing (FOSS) technologies.

[0012] The above objective is accomplished by the solution according to the present invention.

[0013] It is an advantage of the present invention that it enables time-dependent tracking of both the shape and position of a flexible device incorporating shapesensing optical fibers, thereby allowing accurate monitoring during dynamic motion or manipulation.

[0014] Moreover, it is an advantage of the method according to the invention that it supports operation in constrained environments where the flexible device exhibits a follow-the-leader behaviour. In such scenarios, commonly encountered in medical applications like endoscopy or catheter navigation, the tip of the flexible device defines a path through a confined space, such as a lumen or vessel, and the remainder of the device passively follows this path due to spatial constraints. The invention facilitates tracking of the entire device shape under these conditions by leveraging shape-sensing data and referencing previously known positions, allowing accurate reconstruction of the device’s trajectory without requiring full-body actuation or sensing. This principle is particularly effective in applications where the device is steered at the tip, and the shaft deforms accordingly to follow the tip’s path with minimal lateral displacement.

[0015] It is an advantage of the present invention that it provides enhanced localization accuracy by identifying and tracking reference shape segments within curvature profiles over time.

[0016] It is an advantage of the present invention that it offers robustness against motion and deformation of a flexible device, allowing for reliable tracking even during insertion, manipulation, or other forms of disturbance.

[0017] It is an advantage of the present invention that it supports mapping between reference frames, such that the optical fiber’s configuration can be updated relative to a fixed global reference frame, facilitating consistent positioning and orientation tracking.

[0018] It is an advantage of the present invention that it supports model-based reconstruction of the fiber’s shape from discrete curvature measurements, enabling a smooth and continuous shape estimation.

[0019] It is an advantage of the present invention that it allows for pattern recognition and tracking, such as determining a best match between curvature profiles using error-minimization techniques like least-squares calculations.

[0020] It is an advantage of the present invention that it is compatible with multicore optical fibers and FBG-based sensing elements, allowing flexibility in the design and integration of the shape sensing system.

[0021] It is an advantage of the present invention that it enables a follow-the- leader strategy, allowing curvature profiles to propagate along the fiber’s length in a time-dependent manner.

[0022] It is an advantage of the present invention that it is scalable and integrable, making it suitable for use in various fiber-based sensing applications, including medical devices and minimally invasive instruments.

[0023] In a first aspect the invention relates to a method for locating and sensing the shape of a flexible device over time, wherein the flexible device comprises one or more shape-sensing elements, the method comprising: acquiring, using said one or more shape-sensing elements, data indicative of deformation of said flexible device; computing, based on said acquired data, a first shape of said flexible deviceexpressed relative to a first reference frame; selecting a reference region along said computed first shape of said flexible device, wherein said reference region serves to identify a displacement of said flexible device during subsequent measurements; computing, at a subsequent time step and based on data subsequently acquired via said one or more shape-sensing elements, a second shape of said flexible device, wherein said second shape being expressed in a second reference frame; estimating the presence and the location of said selected reference region along said computed second shape of said flexible device; calculating, based on said estimated location of said present reference region, a displacement of said first shape along an arc length of said second shape within said second reference frame; and mapping, based on said calculated displacement, said second shape of said flexible device from said second reference frame to said first reference frame, thereby locating and sensing the shape of said flexible device over time. It is an advantage that the method enables accurate tracking of both the shape and location of a flexible device over time, even in dynamic or constrained environments, without the need of additional external tracking system. By expressing the shape relative to evolving reference frames and mapping between them, deformation-aware tracking becomes feasible, which is particularly valuable in medical and robotic applications.

[0024] According to a specific embodiment of the present invention, the steps may be carried out continuously or periodically. This allows the tracking process to be performed continuously or periodically, enabling real-time or time-resolved monitoring of the device’s behaviour. This is essential for applications requiring constant feedback or intervention, such as minimally invasive procedures.

[0025] According to a specific embodiment of the present invention, the method further comprises a step of selecting a plurality of reference regions along said computed first shape of the flexible device, wherein each of said plurality of reference regions is configured to serve as a basis for identifying displacement of said flexible device during subsequent measurements. This embodiment has the additional advantage that it enables the selection of multiple reference regions along the shape of the flexible device, thereby increasing the robustness of the shape and position estimation. By using a plurality of reference regions rather than a singlereference region, the method reduces sensitivity to local deformations or measurement noise.

[0026] According to a particular embodiment of the present invention, the presence and location of each of said plurality of reference regions is simultaneously estimated along said computed second shape, and wherein said displacement is calculated based on a comparison between the positions of said plurality of reference regions in said first and second shapes. It is an advantage of this embodiment that it allows a simultaneous estimation of the presence and position of multiple reference regions, which further enhances the accuracy and reliability of displacement calculations. The use of multiple simultaneously tracked points provides greater confidence in global alignment, particularly in non-uniform deformation scenarios.

[0027] According to a particular embodiment of the present invention, said reference region may be selected based on a curvature profile of the flexible device computed along its arc length, the method further comprising the steps of calculating a variance between curvature profiles at subsequent time steps, and determining one or more locations where the calculated variance is minimal. According to a particular embodiment of the present invention, said reference region may be selected based on a curvature profile of the flexible device computed along its arc length, the method further comprising the steps of calculating a variance over time between curvature profiles of computed shapes computed in previous step of the method, and determining one or more locations where the calculated variance is minimal. It is an advantage that the selection of reference regions is guided by a curvature profile, and optimized by identifying regions with minimal variance over time. This enables the system to dynamically choose stable and distinctive regions as reference landmarks, improving the long-term reliability of tracking.

[0028] In some embodiments, the selection of the reference region is preceded by a step of estimating the need of selecting a new reference region suitable for estimating of the presence and the location of the reference region along the computed second shape. Selection of a new reference region may be needed in the absence of reference region (e.g. for the initial selection) or if the current reference region does not allow estimating of the presence and the location of the referenceregion along the computed second shape. Embodiments relating to step c’, described hereinafter may apply herein mutatis mutandis.

[0029] According to a specific embodiment of the present invention, said displacement is calculated as a displacement of said second shape with respect to said first shape along the arc length of said first shape within said first reference frame. This provides an alternative approach to displacement calculation by computing displacement of the second shape relative to the first shape. This flexibility allows adaptation to different frame alignment strategies or applicationspecific tracking preferences.

[0030] According to a specific embodiment of the present invention, the step of estimating the presence and the location of said reference region along said computed second shape comprises computing a least-squares error between a shape of the flexible device obtained during a subsequent measurement and a reference shape, wherein the least-squares error is determined by summing the squared differences between corresponding curvature values at respective positions along the arc length of the flexible device. According to a specific embodiment of the present invention, the step of estimating the presence and the location of said reference region along said computed second shape comprises computing a least-squares error between the computed second shape and the reference shape, wherein the leastsquares error is determined by summing the squared differences between corresponding curvature values at respective positions along the arc length of the flexible device. This advantageously enables precise matching between current and reference shapes using a least-squares error approach based on curvature values along arc length. This allows robust detection of subtle shape differences and improves the accuracy of reference region localization.

[0031] According to a particular embodiment of the present invention, said flexible device is an elongated medical instrument. This embodiment has the advantage of making the invention highly relevant for clinical applications such as catheter navigation, endoscopy, or robotic surgery, where accurate tracking of flexible instruments is critical.

[0032] According to a particular embodiment of the present invention, said flexible device is inserted into a confined space, such as a lumen, that constrains its movement and results in approximate follow-the-leader behaviour. Hence, it is an advantage that the present invention supports operation in confined spaces such as lumens, where follow-the-leader behaviour dominates. This makes the method especially suitable for intraluminal or intracorporeal navigation tasks where movement is constrained and shape tracking is challenging.

[0033] According to a specific embodiment of the present invention, the step of measuring the shape is based on one or more of the following sensing technologies: FBG-based optical fibers, Fabry -Perot or Rayleigh scattering-based optical fibers, strain gauges, or spring-based inductive sensors. Hence, the invention may be compatible with a range of sensing technologies, including FBG optical fibers, Fabry -Perot sensors, Rayleigh scattering fibers, strain gauges, and inductive sensors. This flexibility allows the method to be implemented using different hardware platforms suited to varying precision, cost, or integration needs.

[0034] According to a particular embodiment of the present invention, the location of the reference region corresponds to a position along the longitudinal axis of the flexible device, expressible as a function of the arc length parameter. This ensures a precise and curvature-aware spatial interpretation that is invariant to overall device orientation. In some embodiments, the reference region is selected in a part of the arc length of flexible device comprising one or more shape sensing elements. In some embodiments, the reference region is selected in a part of the arc length of flexible device comprising one or more shape sensing elements.

[0035] According to a particular embodiment of the present invention, the first reference frame corresponds to a fixed global reference frame, and mapping the second shape comprises updating the shape and location of the flexible device relative to the fixed global reference frame. It is an advantage of this embodiment that it supports mapping between local and global reference frames, allowing tracked shapes to be updated and interpreted in a fixed spatial context. This is essential for registering device movement within the body or a robotic workspace.

[0036] According to another aspect, the invention relates to a system for locating and sensing the shape of a flexible device over time, configured to perform the method according to any of the previous embodiments, the system comprising: an elongated tubular flexible device with sensors that allow to compute the shape of the flexible device; a data acquisition module configured to acquire the sensor data needed for computing the shape of the flexible device; and a processing unit configured to: compute, based on said acquired data, a first shape of the flexible device expressed relative to a first reference frame; select a reference region along said computed first shape of the flexible device, wherein said reference region serve to identify a displacement of the flexible device during subsequent measurements; compute, at a subsequent time step and based on data subsequently acquired via the one or more shape-sensing elements, a second shape of the flexible device, wherein said second shape being expressed in a second reference frame; estimate the presence and the location of said reference region along said computed second shape of the flexible device; calculate, based on said estimated location of said present reference region, a displacement of said first shape along the arc length of said second shape within said second reference frame; and map, based on said calculated displacement, said second shape of the flexible device from the second reference frame to the first reference frame, thereby locating and shape sensing the flexible device over time.

[0037] According to another aspect, the invention relates to a device for locating and sensing the shape of a flexible device, configured to perform the method according to any of the previous embodiments, comprising: an elongated tubular flexible device with embedded or attached sensors that allow to compute the shape of the flexible device; a processor configured to: compute a curvature profile along the length of the flexible device in a reference frame based on the sensor data; track changes in the shape of the flexible device over time by comparing curvature profiles at different time points; calculate a displacement of the flexible device relative to a reference region, thereby updating the location of the flexible device within the fixed global reference frame.

[0038] For purposes of summarizing the invention and the advantages achieved over the prior art, certain objects and advantages of the invention have beendescribed herein above. Of course, it is to be understood that not necessarily all such objects or advantages may be achieved in accordance with any particular embodiment of the invention. Thus, for example, those skilled in the art will recognize that the invention may be embodied or carried out in a manner that achieves or optimizes one advantage or group of advantages as taught herein without necessarily achieving other objects or advantages as may be taught or suggested herein.

[0039] The above and other aspects of the invention will be apparent from and elucidated with reference to the embodiment(s) described hereinafter.Brief description of the drawings

[0040] The invention will now be described further, by way of example, with reference to the accompanying drawings, wherein like reference numerals refer to like elements in the various figures.

[0041] Figure 1 provides a schematic illustration of the concept of dynamic referencing according to embodiments of the present invention.

[0042] Figure 2 provides a plot of the change of the curvature profile of the optical fiber over time with respect to the fiber arc length and the localization of the reference curvature segment, according to embodiments of the present invention.

[0043] Figure 3 illustrates the selection of a next reference curvature segment based on the variance of the curvature, according to embodiments of the present invention.

[0044] Figure 4: Experiment errors. Three experiments for each of the three mock-ups, upper limb (panel A), cervical (panel B) and Aortic (panel C) is given the progression of errors(dashed line - short dash); es(dotted line) and en(full line) w.r.t. the insertion depth, and the distribution of errors for each of the three insertion trials. The vertical dashed lines (long dash) indicate instances where a new reference shape was selected.Detailed description of illustrative embodiments

[0045] The present invention will be described with respect to particular embodiments and with reference to certain drawings but the invention is not limited thereto but only by the claims.

[0046] Furthermore, the terms first, second and the like in the description and in the claims, are used for distinguishing between similar elements and not necessarily for describing a sequence, either temporally, spatially, in ranking or in any other manner. It is to be understood that the terms so used are interchangeable under appropriate circumstances and that the embodiments of the invention described herein are capable of operation in other sequences than described or illustrated herein.

[0047] It is to be noticed that the term “comprising”, used in the claims, should not be interpreted as being restricted to the means listed thereafter; it does not exclude other elements or steps. It is thus to be interpreted as specifying the presence of the stated features, integers, steps or components as referred to, but does not preclude the presence or addition of one or more other features, integers, steps or components, or groups thereof. Thus, the scope of the expression “a device comprising means A and B” should not be limited to devices consisting only of components A and B. It means that with respect to the present invention, the only relevant components of the device are A and B.

[0048] Reference throughout this specification to “one embodiment” or “an embodiment” means that a particular feature, structure or characteristic described in connection with the embodiment is included in at least one embodiment of the present invention. Thus, appearances of the phrases “in one embodiment” or “in an embodiment” in various places throughout this specification are not necessarily all referring to the same embodiment, but may. Furthermore, the particular features, structures or characteristics may be combined in any suitable manner, as would be apparent to one of ordinary skill in the art from this disclosure, in one or more embodiments.

[0049] Similarly, it should be appreciated that in the description of exemplary embodiments of the invention, various features of the invention are sometimes grouped together in a single embodiment, figure, or description thereof for thepurpose of streamlining the disclosure and aiding in the understanding of one or more of the various inventive aspects. This method of disclosure, however, is not to be interpreted as reflecting an intention that the claimed invention requires more features than are expressly recited in each claim. Rather, as the following claims reflect, inventive aspects lie in less than all features of a single foregoing disclosed embodiment. Thus, the claims following the detailed description are hereby expressly incorporated into this detailed description, with each claim standing on its own as a separate embodiment of this invention.

[0050] Furthermore, while some embodiments described herein include some but not other features included in other embodiments, combinations of features of different embodiments are meant to be within the scope of the invention, and form different embodiments, as would be understood by those in the art. For example, in the following claims, any of the claimed embodiments can be used in any combination.

[0051] It should be noted that the use of particular terminology when describing certain features or aspects of the invention should not be taken to imply that the terminology is being re-defined herein to be restricted to include any specific characteristics of the features or aspects of the invention with which that terminology is associated.

[0052] As used herein, the term “flexible device” refers to an elongate, deformable structure that includes one or more shape-sensing elements, such as optical fibers with embedded sensors (e.g., fiber Bragg gratings). The flexible device may include, but is not limited to, catheters, guidewires, robotic arms, industrial probes, or similar structures configured to navigate complex environments. The optical fibers used for shape sensing may be integrated into the flexible device or affixed thereto.

[0053] As used herein, a “shape-sensing element” refers to a sensing element or arrangement configured to detect mechanical deformation of a flexible device, including, for example, an optical fiber capable of measuring strain and / or curvature along its length. The shape-sensing element may operate by measuring strain atdiscrete locations or continuously along the optical fiber, enabling reconstruction of the fiber's shape.

[0054] In some embodiments, a shape sensor comprises one or more fiber Bragg gratings (FBGs) embedded within, or inscribed along, the optical core of the fiber. In other embodiments, the shape sensor may employ alternative optical or non- optical sensing modalities, including but not limited to Rayleigh scattering, interferometry, or micro-electromechanical systems (MEMS).

[0055] When reference is made to an “optical core”, this may refer to a single light-guiding region within an optical fiber, configured to transmit and / or reflect light for the purpose of sensing, such as via FBGs. The optical core may be part of a single-core optical fiber or one of multiple cores in a multi-core optical fiber. Each optical core may enable independent measurement of strain at discrete locations along its length. In some embodiments, an assembly of multiple single-core optical fibers may be used, wherein each fiber comprises one optical core. In other embodiments, a multi-core optical fiber may be used, comprising a plurality of optical cores within a single cladding structure.

[0056] The optical fiber comprising one or more optical cores may be embedded within an elongated medical device, such as a catheter or guidewire. In such embodiments, the optical fiber may extend along at least a portion of the catheter to enable shape sensing and / or position estimation of the device during use.

[0057] As used herein, the terms “reference region” and “reference shape segment” are used interchangeably to refer to a portion of the flexible device — typically a known or identifiable segment — used as a spatial and / or geometric reference for determining the shape and / or position of other parts of the flexible device. The reference region may include embedded shape-sensing elements and may be fixed or dynamically selected based on the application context. This reference region serves as a baseline for shape reconstruction, allowing relative deformation or position changes along the remainder of the flexible device to be calculated with respect to it.

[0058] As used herein, when reference is made to locating and sensing the shape of a flexible device “over time” or “time-dependent locating and sensing the shapeof a flexible device”, reference may be made to the determination of the spatial position and / or shape sensing, like curvature, of the flexible device at multiple points in time, enabling tracking of the flexible device’s movement or deformation over a time sequence.

[0059] In embodiments according to the present invention in which FBGs data is employed, shape sensing is performed by interrogating the optical fiber, embedded within a flexible device, to detect wavelength shifts in light reflected by the FBGs. These wavelength shifts, also referred to herein as “wavelength data” or “data indicative of deformation of the flexible device”, correspond to localized strain experienced by the optical core. The strain data may be further processed to determine curvature and reconstruct the shape and / or location of the flexible device.

[0060] The shape of the flexible device may be computed using a shape computation system, based on the acquired strain and / or curvature data from the shape sensors, embedded within the flexible device. This system may include an algorithm or a processor configured to reconstruct the shape of the flexible device and determine its position over time. According to specific embodiments of the present invention, this computation may be performed continuously, allowing for real-time shape sensing and tracking of the flexible device’s movement or deformation.

[0061] As used herein, the phrase “along the length of the flexible device” refers to the entire longitudinal extent of the flexible device, from one end to the other, regardless of its configuration or curvature. This includes both continuous and discrete positions along the device where strain and / or curvature may be measured by one or more shape-sensing elements. The term encompasses the full length as well as any portion or segment of the flexible device, whether it is straight, curved, or deformed.

[0062] The term “arc length” refers to a distance measured along the actual path of the flexible device, particularly when the device is curved or bent. In the context of shape sensing, “arc length” may be used to identify locations along the device in its deformed or operational state. Strain and / or curvature data may be acquired atspecific intervals or continuously along this length or arc length, depending on the configuration and resolution of the sensing system.

[0063] In some embodiments, the shape of the flexible device is computed within a global reference frame, which defines a fixed coordinate system for spatially positioning the device. The initial shape may be computed based on strain and / or curvature data acquired along the length of the flexible device, with measurements taken at respective points. The global reference frame enables consistent shape tracking and comparison of the device’s position and geometry across different time points and environments.

[0064] As used herein, the term “fixed global reference frame” refers to a reference frame that remains constant and unmoving throughout the execution of the method. This frame may be associated with an external, static coordinate system — such as a physical reference in the surrounding environment (e.g., a room or imaging system) — or with a predefined, fixed location relative to the device’s environment, such as a patient’s anatomy in a medical context. The fixed global reference frame serves as a stable basis for mapping and tracking the location and shape of the flexible device over time, ensuring consistency in spatial representation relative to external elements.

[0065] As used herein, the term “follow-the-leader principle” refers to a movement strategy wherein a distal or leading portion of a flexible device defines a trajectory or shape, and one or more subsequent segments of the device are guided or controlled to follow that same trajectory. This principle allows the device to replicate a previously traversed path, with each segment sequentially assuming the position and orientation previously occupied by the leading segment. As a result, the entire flexible device conforms to the established shape profile or navigated path.

[0066] In the description provided herein, numerous specific details are set forth.However, it is understood that embodiments of the invention may be practiced without these specific details. In other instances, well-known methods, structures and techniques have not been shown in detail in order not to obscure an understanding of this description.

[0067] The present invention relates to a method and a system for timedependent localization and shape sensing using solely FOSS. Observing that a flexible and elongated instrument, such as a catheter, when sufficiently constrained, exhibits an approximate follow-the-leader motion (follow the leader assumption), any of such constrained segments of the catheter could serve as a virtual reference shape through which the catheter is moving. The proposed method dynamically identifies and tracks the more stable reference segments to enable fiber localization relative to a static reference frame for indefinitely long insertions. These will be referred to as dynamic reference shapes. Hence, embodiments of the present invention dynamically identifies new reference shapes - while the elongated instrument is inserted in a lumen - allowing localization even after the sensing section of the fiber moves beyond the original reference. It is an advantage of embodiments of the present invention that those do not require the use of preoperational images. However, if they are available, they can be leveraged to apply a twist compensation method, assuming that the fiber remains inside the preoperative vessel model.

[0068] Figure 1 shows a schematic overview of the system for locating and sensing the shape of a flexible device over time, and illustrates the concept of dynamic referencing. The system comprises an elongated, tubular flexible device, such as a catheter, which incorporates one or more shape-sensing elements. These elements may include optical fibers or other sensors capable of detecting deformation along the length of the device. The device may be operably connected to a data acquisition module that receives and processes signals from the sensors. The data acquisition module may be in communication with a processing unit, which is configured to compute the shape of the flexible device, select one or more reference regions along the computed shape, and track the displacement of these regions over time. The system is thus capable of continuously or periodically monitoring the shape and position of the flexible device as it is manipulated during use, for example, within a patient’s body.

[0069] As illustrated in Figure 1, the flexible device 6 may comprise one or more shape-sensing elements, like shape-sensing fiber comprises one or more opticalcores with shape sensors, such as fiber Bragg gratings (FBGs), arranged along a sensible section 4 of the fiber. The fiber is embedded within or integrated into a catheter or another elongated, flexible device 6 that is designed for navigation through a vessel or lumen 5. As the device 6 is advanced or manipulated within the anatomical pathway, the shape of the device 6 is continuously or periodically measured and updated, enabling real-time tracking of the device’s configuration and location within the body.

[0070] To locate and sensing the shape of the flexible device 6 over time according to embodiments of the present invention, a first step comprises the acquisition of data indicative of deformation of said flexible device using one or more shape-sensing elements. Such data may be measured using FBGs along a single-core optical fiber or multi-core fiber embedded within the flexible device. In such an embodiment, typically four cores may be used: one central core aligned with the fiber’s neutral axis and three outer cores spaced 120° apart at a small distance from the central core. FBGs may be inscribed into each core at specific intervals, creating discrete measurement points. Each FBG reflects a certain wavelength (Bragg wavelength, AB) of incoming light while allowing the others to pass through. By analysing the wavelength shifts in the FBGs of the four cores at each measurement point sialong the fiber, the curvatureand bending plane angle 0b iat this point stcan be determined.

[0071] In a following step, this set of acquired data, for example 0 ≤ i ≤ N], with N the number of inscribed FBGs, is subsequently employed to reconstruct the 3D shape of the flexible device. This reconstructed shape or curvature profile is expressed relative to a first reference frame.

[0072] In a subsequent step, to locate or track the position and shape of the flexible device over time, a reference region, also referred to as reference curvature segment 3, may be selected along said computed first shape of the flexible device. This reference region or reference curvature segment may be selected from a portion of the computed flexible device’s shape or curvature profile, wherein said reference region 3 may serve to identify a displacement of the flexible device during subsequent measurements.

[0073] Next, a step of computing, at a subsequent time step and based on data subsequently acquired via said one or more shape-sensing elements, a second shape of said flexible device may be performed. This second shape may be expressed in a second reference frame. Thereafter, a step of estimating the presence and the location of said selected reference region 3 along said computed second shape of said flexible device may be performed. Based on said estimated location of said present reference region 3, a displacement of said first shape along an arc length of said second shape within said second reference frame may be calculated.

[0074] In order to locate and sensing the shape of the flexible device over time, a step of mapping the second shape of the flexible device from the second reference frame to the first reference frame may be performed.

[0075] Figure 1 also illustrates the presence of an initial reference region 1 or previous reference regions 2, based on pre-determined criteria such as stability or low variance.

[0076] Figure 2 illustrates the method of shape computation and reference region tracking as claimed. At an initial time point, the shape of the flexible device is computed and expressed relative to a first, fixed global reference frame. A reference region is selected along this computed first shape, serving as a landmark for subsequent measurements. At a later time point, after the device has been displaced, a second shape is computed in a second, device-local reference frame. The presence and location of the previously selected reference region are estimated along the second shape, and a displacement is calculated based on the estimated location of the reference region. This displacement is used to map the second shape from the device-local reference frame back to the global reference frame, thereby updating the location and shape of the device over time.

[0077] Indeed, Figure 2 provides a plot of the change of the shape, also referred to as curvature profile of the flexible device over time with respect to the arc length s. Figure 2 also illustrates the localization of the reference region or reference curvature segment Kref23 along the arc length s of the curvature profile Kt21 within a search window around the location of the best match at the previous time step sref,t-12, and the shift sref,t24. Indeed, at each time step t, the curvature profileof the flexible device Kt21 is computed over the arc length s of the flexible device. Similarly, the curvature profile Kt-122 at the previous time step t — 1 is also measured and plotted. According to embodiments of the present invention, a reference region Kref23 corresponding to a defined segment of the flexible device, is compared against the current and / or previous fiber curvature profiles to track the displacement of the fiber segment over time. A shift sref,t24 is computed, representing the change in arc length position of the reference region 23 between time steps t-1 and t, indicative of fiber motion such as insertion or retraction. The search for the best match is carried out within a predefined window of length 2P 25, which defines the search region around the previously matched location.

[0078] The position26 denotes the arc length at which the reference region Kref23 best matched the flexible device’s curvature profile Kt-122 at time t-1. At the current time step, a least-squares error 27 is calculated over the search window, comparing Kt21 and Kref23 to find the new best match location sref,t45. This matching is based on minimizing the summed squared difference between corresponding curvature values, as further described below.

[0079] Figure 3 depicts the process of selecting the reference region based on the curvature profile of the flexible device. The curvature profile is computed along the arc length of the device, and the variance in curvature profiles at subsequent time steps is calculated. The figure highlights one or more locations along the arc length where the calculated variance is minimal, indicating stable regions that are suitable for use as reference regions. By selecting reference regions in areas of minimal curvature variance, the method enhances the robustness and accuracy of displacement estimation and shape mapping, as described in the claims. A sliding window selects the curvature segment for which the sum of variance is minimal. To enhance robustness against noise and temporal variation, a time-averaged curvature profile Kt41 may be computed over multiple previous time steps, as described below. A sliding window 42 is applied along the flexible device’s arc length to identify segments with minimal temporal variation. The variance < Jt243 of the curvature at each point along the arc length is calculated to identify stable regions for potential selection as a new reference curvature segment 23.

[0080] The invention is further illustrated by the following non-limiting embodiment, which describes a method and a system for real-time shape sensing and localization of a flexible device, such as a catheter, comprising single-core optical fibers or a multi-core fiber. FBGs may be inscribed into the fiber at specific intervals, creating discrete measurement points. Each FBG reflects a certain wavelength (Bragg wavelength, AB) of incoming light while allowing the others to pass through. By analysing the wavelength shifts in the FBGs of the four cores at each measurement point sialong the fiber, the curvatureand bending plane angle 0b iat this point stcan be determined. This set of {Ki, θb,i|0 ≤ i ≤ N}, with N the number of inscribed FBGs, is subsequently employed to reconstruct the 3D shape of the fiber.

[0081] In some embodiments, additional technical effects may be achieved by incorporating one or more supplementary steps into the method for time-dependent shape sensing and localization of an optical fiber embedded within an elongated instrument. For instance, integrating preoperative vessel geometry — such as through a filtering step, a correction algorithm, or a predictive model — may enable compensation for fiber twist, thereby enhancing the accuracy, stability, or robustness of the shape sensing and localization process. These embodiments may offer particular advantages in environments characterized by high noise levels, rapid motion, or requirements for long-term tracking. However, they are not essential to the core functionality of the method as defined in the independent claims.

[0082] In some embodiments, the step of defining an initial reference region, comprise a selection or identification of a reference shapeRCref E Rl×3with corresponding curvature profile κref∈ Rlcan be employed to locate the reconstructed shape with respect to a static frame { / ?}. While embodiments of the present invention may dynamically select new reference shapes or location tracking segments, a step of identifying an initial reference shape may be beneficial. This initial reference shape may either be based on a physical structure, or a constrained section of the optical fiber (or catheter) itself, such as derived from an image or from a reconstructed shape. The choice of initial reference shape determines the static frame { / ?} with respect to which the localization will be performed. For example, ifthe reference shape corresponds to a physical or image-based structure, localization is performed relative to the corresponding, meaningful, coordinate frame. If the initial reference shape or location tracking segment is based simply on the fiber’s measured shape, no additional sensors of physical objects are needed, but the static frame { / ?} may not be a meaningful one.

[0083] According to embodiments of the present invention, the following paragraphs describe the step of curvature tracking initialization in more detail. Based on shape-sensing data, the fiber’s curvature profile may be measured. Based on this interpolated measured initial curvature profile KtE Rlat time step t = 0, an initial reference curvature segment, also referred to as initial segment or initial location segment κrefalong the arc length s of said initial curvature profile Ktmay be identified. This initial reference curvature segment is extending from a starting arc length towards an ending arc length. To this end, κrefmay be shifted incrementally along Kt. This means that the initial reference curvature segment κrefis not fixed in one position, but is gradually adjusted or translated along the arc length s of said initial curvature profile Kt. This incremental shifting allows for a more precise alignment and tracking of the curvature profile over time, ensuring accurate initialization and monitoring of the fiber's shape. The arc length sref,tcorresponding to the shift i where the error egis minimal, is assumed as the location of κrefalong Kt. This may mathematically be expressed as

[0084] In some embodiments, the step for updating the curvature tracking may be applied using a Kalman filter (“KF”). This KF is used to track the location sref,tof the curvature segment or location tracking segment more accurately by including a measurement of the shift of the full-length curvature profile over time in addition tothis shift of the reference curvature profile. The filter state xtis the position sref,tof κrefalong the arc length of KtXj ref,t (3)

[0085] First, the shift sref,tof κrefalong the arc length s of the measured curvature profile Kt, caused by insertion or retraction of the optical fiber since the previous time step, is identified. To this end, κrefis shifted along Ktwithin a search window around the reference location sref,t-1that was identified during the previous time step. At each shift step i, the least-squares error etbetween Ktand κrefis computed. The shift sref,tof κrefis taken as the shift i where the error etis minimal:with — p < i < p. The search window length 2p + 1 can be set based on the updated rate of the shape sensing and the expected maximal insertion or retraction speed. The Kalman filter calculates the predicted state estimate xtin the prediction update:xt= xt-1+ ut, Pt= Pt-1+ σu2, (6)where the input utis the local shift sref,tof κrefalong s, as described in equation (5).

[0086] An advantage of embodiments of the present invention is that the method proposes a solution of location and shape sensing of an optical fiber which only relies on shape sensing data. In other words, there is no need for additional sensors or data tracking to derive the location of the optical fiber.

[0087] If the insertion speed of the optical fiber would be known, for example during a robotic insertion, this information could also be used to estimate ut. The symbolis the previously estimated state variance, andis the process noise (considered fixed, for the sake of simplicity). In the measurement update step, the Kalman gain Ktis computed:where σz2is the measurement noise variance, which is also considered fixed. The predicted state estimate xtis updated using the measurement ztxt= x̂t+ Kt(zt- x̂t), Pt= (1 - Kt)P̂t(8)The measurement ztis obtained by discretely shifting the mean curvature κ̄t-1, where κ̄t-1is κttime-averaged from 0 to t − 1, over the measured curvature Ktwithin the search window defined by p and finding the shift associated with the minimal least-square error ez, similar to one of the previous equations:zt= sref,t-1− argmin{ez[i]}, (10) with — p < i < p. The mean difference with equation (4) is that here the shift of the full-length curvature profile is considered instead of only a section at the location of the reference segment.The mean or time averaged curvature profile κ̄tand variance σt2of the curvature may be estimated online using the Welford’s algorithm:nt[i] = nt-1[i] + 1, (11)counting the number of curvature samples that have been observed at the specific location i along the arc length st' = s — sref,tup to time t, and Ktthe estimated squared mean curvature. The origin of κ̄tand σ2corresponds to the start of the current reference curvature segment

[0088] Under some circumstances, the current reference shape’s exploitability may be expected to end in the near future, and as such the identification of a new reference may be required. Such circumstances may be, for example, when the reference curvature arc location sref,tfalls below threshold sm, the sensing segment of the fiber is nearing the end of the reference segment. In another circumstance, the least-squares error between the reference curvature κrefand the corresponding segment of Ktexceeds threshold eM, wherein κrefis no longer representative of the curvature profile of the catheter at the reference segment. To determine a suitable new reference shape, the variance of the curvature over time is considered. The idea is that specific portions of the vessel in which the catheter is more constrained will show stable follow-the-leader motion and form good locations for defining a new reference shape. To find the most suitable reference shape, the segment with the smallest sum of variance, calculated over a sliding window of width Zrey, is selected.s̃ref,t= argmin(ςt[i]),with 0 < i < I — sref,t— lrefthe location of the new reference shape along the arc length of Kt. The length lref is chosen to capture enough curvature detail without extending too far, as the chance of including unstable sections rises with segment length. Finally, the vectors nt, Ktand Ktare shifted by:Δs̃ref,t= s̃ref,t− sref,t(16)in order to adjust the indices so that their origin coincides with the new reference segment. Note that the reference segment selection procedure is introduced here for the case of catheter insertion, but works analogously for the retraction case.

[0089] According to a specific embodiment, the step of mapping, based on said calculated displacement, said second shape of the flexible device from the second reference frame to the first reference frame, thereby locating and shape sensing the flexible device over time, may be performed in the following way. Having found the segment of Ktthat corresponds to the reference shape, the homogeneous transformation matrixRFTt(rotationRFRtand translationRFtt) mapping the reconstructed catheter shapeFCt∈ Rl×3from the local fiber frame {F} to thereference frame {R}, can be calculated. Using a Singular Value Decomposition (SVD), the rotationRFRtbetween the reference shapeRCrefand the corresponding segment ofis computed, whileRFttaccounts for the translation:RFRt= (UVT),and Cref, respectively. Whenever a new reference segment is selected,RCref is updated:where i = s̃ref,t.Note that this ensures each new reference shape is consistently expressed in the same frame {R}.

[0090] According to a specific embodiment of the present invention, the flexible device may comprise an optical fiber. The optical fiber may comprise one or more optical cores each having one or more shape sensors configured to detect strain and / or curvature along the optical fiber. The method may comprise the steps of:computing, based on acquired strain and / or curvature data, a first curvature profile of the optical fiber along its length in a first reference frame; selecting, based on the computed first curvature profile, a first reference shape segment corresponding to a segment of the curvature profile along a selection of1points, wherein the first reference shape segment serves as a location tracking segment for identifying the location of changes in the curvature profile of the optical fiber during subsequent measurements;computing, at a subsequent time step and based on subsequently acquired strain and / or curvature data, a second curvature profile of the optical fiber in a second reference frame;- tracking a location of the first reference shape segment along the second curvature profile, wherein the location corresponds to a best match between the first reference shape segment and the second curvature profile; calculating, based on the tracked location, a shift in location of the first reference shape segment within the second reference frame; andmapping, based on the calculated shift in location, the second curvature profile of the optical fiber from the second reference frame to the first reference frame, thereby locating and shape sensing the optical fiber over time.

[0091] Optionally, these steps of the method may be carried out continuously or periodically.

[0092] The method may further comprise a step of selecting a subsequent reference shape segment. This subsequent reference shape segment may be used for the calculation of a variance between subsequent curvature profiles, and the determination of a location where the calculated variance is minimal.

[0093] According to a specific embodiment of the present invention, the step of tracking a location of the reference shape segment along the second or any subsequent curvature profile may comprise a step of computing a least-squares error between a curvature profile of the optical fiber obtained during a subsequent measurement and a reference curvature profile, wherein the least-squares error is determined by summing squared differences between corresponding curvature values at respective points along the length of the fiber.

[0094] According to another aspect, the invention relates to a system for locating and shape sensing a flexible device over time, wherein the flexible device may comprise an optical fiber. The system may comprise:an optical fiber comprising one or more optical cores, each optical core comprising one or more shape sensors configured to detect strain and / or curvature along the optical fiber;a data acquisition module configured to acquire strain and / or curvature data from the optical fiber over time;a processing unit configured to: compute, based on acquired strain and / or curvature data, a first curvature profile of the optical fiber along its length in a first reference frame; select a first reference shape segment based on the first curvature profile, the first reference shape segment corresponding to a segment of the curvature profile along a selection of points, wherein the first reference shape segment serves as a location tracking segment for identifying changes in the curvature profile of the optical fiber during subsequent measurements; compute a second curvature profile at a subsequent time step based on subsequently acquired strain and / or curvature data in a second reference frame; track a location of the first reference shape segment along the second curvature profile, wherein the location corresponds to a best match between the first reference shape segment and the second curvature profile; calculate a shift in location of the first reference shape segment within the second reference frame based on the tracked location; and map the second curvature profile from the second reference frame to the first reference frame, thereby locating and shape sensing the optical fiber over time.

[0095] According to another aspect, the present invention also relates to a device configured to locate and sensing the shape of a flexible device. The flexible device may comprise one or more shape-sensing elements or sensors, such as an optical fiber, which are configured to detect and acquire data representative of the shape and / or deformation of the flexible instrument. The device may comprise a processor configured tocompute a curvature profile along the length of the optical fiber in a reference frame based on the detected strain and / or curvature data;- track changes in the shape of the optical fiber over time by comparing curvature profiles at different time points; andcalculate a displacement of the optical fiber relative to a reference shape segment, thereby updating the location of the optical fiber within the fixed global reference frame.

[0096] While the invention has been illustrated and described in detail in the drawings and foregoing description, such illustration and description are to be considered illustrative or exemplary and not restrictive. The foregoing description details certain embodiments of the invention. It will be appreciated, however, that no matter how detailed the foregoing appears in text, the invention may be practiced in many ways. The invention is not limited to the disclosed embodiments.

[0097] Other variations to the disclosed embodiments can be understood and effected by those skilled in the art in practicing the claimed invention, from a study of the drawings, the disclosure and the appended claims. In the claims, the word “comprising” does not exclude other elements or steps, and the indefinite article “a” or “an” does not exclude a plurality. A single processor or other unit may fulfil the functions of several items recited in the claims. The mere fact that certain measures are recited in mutually different dependent claims does not indicate that a combination of these measures cannot be used to advantage. A computer program may be stored / distributed on a suitable medium, such as an optical storage medium or a solid-state medium supplied together with or as part of other hardware, but may also be distributed in other forms, such as via the Internet or other wired or wireless telecommunication systems. Any reference signs in the claims should not be construed as limiting the scope.In some embodiment, the step of estimating the presence and the location of the selected reference region along the computed second shape of the flexible device comprises measuring error associated with the shift of the full-length curvature profile of the second shape over time in addition to that associated with the shift of the reference curvature profile, preferably using a Kalman filter.In some embodiments, the method of the invention comprises the steps ofa) acquiring, using the one or more shape-sensing elements, data indicative of deformation of a flexible device;b) computing based on data provided at step a, a first shape of the flexible device expressed relative to a first reference frame;c) selecting a reference region along said computed first shape of the flexible device, wherein said reference region serves to identify a displacement of the flexible device during subsequent measurements; d) acquiring, using the one or more shape-sensing elements, data indicative of deformation of the flexible device at a later time point; e) computing based on data provided at step d, a second shape of the flexible device, wherein said second shape being expressed in a second reference frame;f) estimating the presence and the location of said selected reference region along said computed second shape of said flexible device; g) calculating, based on said estimated location of said present reference region, a displacement of said first shape along an arc length of said second shape within said second reference frame; and,h) mapping, based on said calculated displacement, said second shape of the flexible device from said second reference frame to said first reference frame, thereby locating and sensing the shape of the flexible device over time.In some embodiments, the method of the invention does not comprise the steps of acquiring data. In such embodiment, the method is performed in silico on data acquired outside the scope of the method of the invention.In some embodiments, the method of the invention comprises the steps of,a) providing data indicative of deformation of a flexible device previously acquired using said one or more shape-sensing elements;b) computing based on data provided at step a, a first shape of the flexible device expressed relative to a first reference frame;c) selecting a reference region along said computed first shape of said flexible device, wherein said reference region serves to identify a displacement of the flexible device during subsequent measurements; d) providing data indicative of deformation of the flexible device at a later time point previously acquired using said one or more shape-sensing elements;e) computing based on data provided at step d, a second shape of the flexible device, wherein said second shape is expressed in a second reference frame;f) estimating the presence and the location of said selected reference region along said computed second shape of the flexible device;g) calculating, based on said estimated location of said present reference region, a displacement of said first shape along an arc length of said second shape within said second reference frame; and,h) mapping, based on said calculated displacement, said second shape of said flexible device from said second reference frame to said first reference frame, thereby locating and sensing the shape of said flexible device over time.In some embodiments, the method of the invention further comprises, after the mapping step, the step of repeating steps c to h at least one time, using the previously computed second shape of the immediately preceding or previous iteration of step e as the computed first shape. The number of repetition of step c to h may be adjusted for enabling tracking of the flexible device over the desired duration. The first refence frame is preferably kept constant over the repetitions to register the movement of the flexible device in a single reference frame.In some embodiment, upon repeating steps c to h, step c is replaced by step c’ comprising- a step c’.l of estimating the need of selecting a new reference region for estimating of the presence and the location of the reference region along the computed second shape in the following step f; and- a step c’.2 of selecting a new reference region along said computed first shape if the estimation performed at step c’.1 indicates that the reference region used in the preceding step g is not suitable for providing estimation of the presence and the location of selected reference region along the computed second shape in the following step f.In some embodiment, the method of the invention further comprises, after the mapping step, the step of repeating steps c to h at least one time, using the computed second shape of the immediately previous iteration of step e as the computed first shape and wherein step c of the first iteration is replaced in the at least one repetition (optionally in each further repetition) by step c’ comprising- a step c’.l of estimating the need of selecting a new reference region for estimating of the presence and the location of the reference region along the computed second shape in the following step f; and- a step c’.2 of selecting a new reference region along said computed first shape if the estimation performed at step c’.1 indicates that the reference region used in the preceding step g is not suitable for providing estimation of the presence and the location of selected reference region along the computed second shape in the following step f.Embodiment pertaining to the selection of the reference region described herein may apply mutatis mutandis to the selection of the new reference region. In particular a plurality of new reference regions may be used as described hereinabove.In one embodiment, the step of estimating the need of selecting a new reference is based on,- the position of the reference region used in the preceding step g with regard to that of the one or more shape-sensing elements, preferably to ensure that the reference region remains, in the data indicative of deformation of the flexible device provided or acquired in the following step d, located within the one or more shape-sensing elements; and / or,- the statistical evaluation of the error in the estimation performed at the preceding step f preferably to ensure that said error is inferior to a predetermined threshold.The step of estimating the need of changing the reference region may comprise (i) determining the position of the reference region used in the preceding step g with regard to that of the one or more shape-sensing elements and (ii) estimating if the reference region used in the preceding step g will remains, in the data indicative of deformation of the flexible device provided or acquired in the following step d, located within the one or more shape-sensing elements. The selection of a new reference region is needed in case the estimation indicate that the reference region used in the preceding step g will be located outside of the one or more shape-sensing elements, in the data indicative of deformation of the flexible device provided or acquired in the following step d.The step of estimating the need of changing the reference region may comprise (i) evaluating the error of the estimation performed at the preceding step f, preferably using a statistical evaluation method of the error, and (ii) comparing said error to a predetermined threshold. The selection of a new reference region is needed in case the error is larger or equal than said predetermined threshold.Such statistical evaluation method may include, without being limited to, determination of the least-square error.In one embodiment, the selection of a new reference region is made to select a new reference region at a position wherein variance, preferably variance over time, of the curvature in the first and / or second shapes computed at previous iterations of the steps b and / or e, is minimized.The step of selecting a new reference region may hence comprise, (i) calculating variance, preferably variance over time, of the curvature in the first and / or second computed shapes computed at previous iterations of the steps b and / or e, and (ii) determining one or more position wherein the calculated variance is minimal to select one or more new reference region therein.In some embodiments, the processing unit of the system of the invention is configured to perform steps b, c, e, f, g, h, and / or c’ of the method of the invention.In some embodiments, the processor of the device of the invention is configured to perform steps b, c, e, f, g, h, and / or c’ of the method of the invention.The method can also be applied to locating and sensing the shape of a flexible device over time within inorganic material, such as artificial tubes, for example vessel mock-ups. This can be used to collect data to build up a table or library, and / or for calibration of the device. For example, the method can be used to locating and sensing the shape of a flexible device within a vessel mock-up, for calibration. Such vessel mock-ups may be 3D vessel mock-ups, including an artery or vein mock-up, an upper limb, a mock-up of a respiratory system, a cervical and an aortic mock-up, e.g. a phantom, such as a silicone phantom.The present invention can, in general, be used in a wide range of catheter-based interventions, not only in the medical field, but also in industrial or robotic systems requiring precise shape tracking. This includes industrial inspection (cracks, corrosion, etc.) for turbines, pipelines and such, as well as automotive and aviation diagnostics for internal components, with no need to dismantling parts.The method may be, hence, a method for locating and sensing the shape of a flexible device over time in accordance with the first aspect of the present invention, wherein said flexible device is inserted into a confined inorganic space that constrains its movement and results in approximate follow-the-leader behaviour. Such artificial space includes hollows between inorganic materials such as artificial vessels, pipelines and the like, or hollows between pieces of devices such as turbines, or gaps (e.g. cracks) in continuous inorganic materials including solid rocks, walls, ground or the like.The method applied to inorganic material or structures can also be carried out sequentially, or continuously. In some embodiments, the method is carried out through holes, gaps or conduits of inorganic material or structures, acquiring the data while moving the flexible device through the inorganic material or structures and processing simultaneously said data. Thus, the method of providing new reference, and optionally estimating the need of providing said new reference, can be done simultaneously with the data acquisition.ExamplesThe present invention is further illustrated by the following examples.Experiments to validate the proposed shape localization method were performed in three anatomically inspired 3D vessel mock-ups: an upper limb, a cervical and an aortic mockup. Additionally, an analysis was conducted to demonstrate the advantage of the KF for reference segment tracking. The upper limb mock-up includes the radial, ulnar, brachial, axillary and subclavian arteries. The cervical mock-up comprises the aortic arch, common, internal, and external carotid arteries, along with selected branches. The aortic mock-up includes the aorta and its primary branches.The aortic mock-up consisted of a realistic silicone phantom derived from patient data and had an inner diameter ranging from 7 to 35 mm, while the upper limb and cervical mockups were fabricated using flexible polyurethane pneumatic tubes (Festo, Esslingen, Germany) with inner diameters ranging between 4-8 mm. The tubes were mounted using 3D-printed holders at sparse anchor points to allow partial deformation of the vessels during catheter insertion, and their layouts were modeled after patient CT data.A passive 6 French (2 mm) JR-4 catheter (RunWay, Boston Scientific, Massachusetts, US) was sensorized. A 6 DoF EM sensor (Northern Digital, Waterloo, Ontario, Canada) was attached to the tip of a 0.38 mm outer diameter nitinol-enclosed multi-core FBG shape sensing fiber (FBGS International NV, Geel, Belgium) at the location of the last FBG using heat shrink tubing. The EM sensor was solely employed as a ground-truth during validation. The fiber consisted of 4 cores, each containing 38 FBGs, spaced at a distance of 13 mm, resulting in a sensing length of 481 mm. The combined fiber and EM sensor was embedded in the 1.78 mm diameter working channel of the catheter and fixed at the handle of the catheter using a 3D printed connector, preventing relative translation of the catheter and fiber, but allowing relative rotation.While this connection aimed to minimize fiber twist, the latter cannot be entirely eliminated due to friction between the fiber and catheter lumen during usage. The catheter was manually inserted inside each of the mock-ups while localizing the catheter using the method of the invention. The proposed method was implemented in Python and C++ and operated in real time at 100 Hz.Each mock-up was placed on top of an EM field generator plate such that the EM field covered as much of the vessel as possible. To create the digital vessel geometries, an EM pointer was used to trace the outer surfaces of the mock-up tubes, generating a point cloud from which the centerlines were extracted. Knowing the tube diameters, a mesh was generated by extruding a circular cross-section along the centerline path for the upper limb and cervical mock-ups. For the aortic mock-up, a photogrammetry scan was aligned to the EM point cloud.Three error metrics were considered to evaluate localization performance. The first error, eAt, was defined as the difference between the insertion length estimated from the dynamic reference, and that from the EM sensor. The insertion length was defined as the length of the catheter segment inside of the vessel. The total catheter insertion length, AZt, at each time step was calculated by accumulating the shift of the reference shape arc length location, Sreft, over time:The ground-truth insertion length, Mgt t, was calculated by accumulating the projection of the tip EM sensor translation between time t — 1 and t onto its z-axis from the previous time step t — 1:with pemitand zem tthe position and the z-axis (longitudinal axis) of the EM sensor, respectively. The insertion length error was then= AZt— lgt tThe other two errors relate to the estimated tip position within the vessel, determined by projecting both the ground-truth EM position and the estimated tip position onto the centerline, yielding (1) the difference esin vessel centerline arc length coordinate of the respective tip projections, and (2) the distance enbetween the two tip measurements in the plane perpendicular to the centerline. To compute these errors, the registration between the static reference frame {JI} and the EM coordinate frame was required. This was obtained by rigidly aligning the estimated catheter tip path (in {5?}) to the EM-measured tip path using correspondence registration. Figure 4A-C displays the errors. Discontinuities in the error progression signal correspond to instances where the EM sensor lost tracking, preventing error calculation at those points, table 1summarizes the median, interquartile range (IQR) and maximum values of the error distributions for each of the 3D mock-up experiments.Table 1: median (med), interquartile range (IQR) and maximum (max) values of the error distributions for each of the 3D mock-up experiments.To analyze the performance of the KF for tracking the reference curvature segment, the localization errors were also computed for a more basic method (accumulating the shift computed in (5) over time). Table 2 shows the median, IQR and maximum errors for the basic method without the KF.Table 2: median (med), interquartile range (IQR) and maximum (max) values of the error distributions for each of the 3D mock-up experiments.The method demonstrated submillimeter median accuracy for tip localization along the vessel centerline (es), with a precision (IQR) below 3 mm across all mockups. Perpendicular errors enshowed greater variability. For navigation purposes, however, localization of the catheter tip along the vessel centerline is arguably more critical than perpendicular to it. This is because navigation decisions - such as when to enter a branch or stop - are typically made based on the catheter position along the vessel route, not its radial offset within the vessel. Moreover, errors perpendicular to the centerline could potentially be improved by adding radial sensing (e.g., intravascular ultrasound or optical coherence tomography) to determine vessel wall distance. Importantly, esremained stable across long insertion lengths, with no significant error growth observed even after more than 800 mm of insertion (figure 4A-C). This highlights the method’s ability to maintain consistent longitudinal localization accuracy over distances well beyond the sensing range ofthe fiber itself. Since esremains smaller than the vessel diameters, this level of accuracy may be sufficient to support navigation tasks such as branch selection during endovascular navigation.Insertion length errors eAtexhibited slightly lower precision than es, which may in part reflect limitations of the EM-based ground truth. As the EM sensor is located at the catheter tip, where follow the leader (FTL) motion may not be preserved, errors in (21) can be introduced. While esand enrely on full 3D localization and are therefore susceptible to fiber twist artifacts, eAtdepends only on curvature and is unaffected by twist. The larger errors observed in enare likely a result of twist effects, which predominantly introduce displacement in directions orthogonal to the local fiber tangent. The comparatively higher stability of essuggests that longitudinal accuracy is less susceptible to such artifacts.Among the phantoms tested, the aortic mock-up yielded somewhat larger errors, likely due to its larger diameter. In such wide vessels, the catheter is less constrained and may deviate more from the FTL assumption. Nonetheless, the method still performed surprisingly well, owing to the catheter’s tendency to adopt a stable configuration even in unconstrained geometries. Even so, the proposed localization method is expected to be particularly effective in smaller vessels.Comparing table 1 with table 2, it can be seen that using the KF enhances accuracy in curvature tracking. By combining reference segment and full-length curvature tracking, the KF is more robust against poor reference segment choices and can handle situations where the FTL assumption doesn’t hold for the complete catheter.A key strength of this method lies in its simplicity and integrability. Unlike EM-based localization systems, this approach relies solely on a single fiber-optic shape sensor. The sensor is thin, passive, immune to EM interference, and easily embedded into a wide range of interventional tools. This offers a significant practical advantage in terms of system complexity, clinical workflow integration, and compatibility with existing tools. In this test, accurate localization was maintained up to 800 mm, without requiring preoperative imaging, fixed reference shapes, or real-time fluoroscopy.

Claims

Claims1. A method for locating and sensing the shape of a flexible device over time, wherein the flexible device comprises one or more shape-sensing elements, the method comprising:a) providing data indicative of deformation of the flexible device previously acquired using said one or more shape-sensing elements;b) computing, based on said acquired data, a first shape of said flexible device expressed relative to a first reference frame;c) selecting a reference region along said computed first shape of said flexible device, wherein said reference region serves to identify a displacement of said flexible device during subsequent measurements;d) providing data indicative of deformation of the flexible device previously acquired at a later time point using said one or more shape-sensing elements;e) computing, based on data provided at step (d), a second shape of said flexible device, said second shape being expressed in a second reference frame;f) estimating the presence and the location of said selected reference region along said computed second shape of said flexible device;g) calculating, based on said estimated location of said present reference region, a displacement of said first shape along an arc length of said second shape within said second reference frame;h) mapping, based on said calculated displacement, said second shape of said flexible device from said second reference frame to said first reference frame, thereby locating and sensing the shape of said flexible device over time; andi) repeating step c) to h) at least one time using the computed second shape of the previous iteration of step e) as the computed first shape and wherein step (c) is replaced by step c’ comprising- a step c’.1 of estimating the need of selecting a new reference region for estimating the presence and the location of thereference region along the computed second shape in the following step f; and,- a step c’.2 of selecting a new reference region along said computed first shape if the estimation performed at step c’.l indicates that the reference region used in the preceding step g is not suitable for estimating the presence and location of the selected reference region along the computed second shape in the following step (f).

2. The method according to any of the previous Claims, wherein said reference region or new reference region is selected based on a curvature profile of the flexible device computed along its arc length, the method further comprising the steps of calculating a variance between curvature profiles at subsequent time steps, and determining one or more locations where the calculated variance is minimal.

3. The method according to any of the previous Claims, wherein said displacement is calculated as a displacement of said second shape with respect to said first shape along the arc length of said first shape within said first reference frame.

4. The method according to any of the previous Claims, wherein the step of estimating the presence and the location of said reference region along said computed second shape comprises computing a least-squares error between a shape of the flexible device obtained during a subsequent measurement and a reference shape, wherein the least-squares error is determined by summing the squared differences between corresponding curvature values at respective positions along the arc length of the flexible device.

5. The method according to any of the previous Claims, wherein said flexible device is an elongated medical instrument.

6. The method according to any of the previous Claims, wherein said flexible device is inserted into a confined space, such as a lumen, that constrains its movement and results in approximate follow-the-leader behaviour.

7. The method according to any of the previous Claims, wherein the one or more shape sensing element is selected from the group consisting of : FBG-based optical fibers, Fabry -Perot or Rayleigh scattering-based optical fibers, strain gauges, or spring-based inductive sensors.

8. The method according to any of the previous Claims, wherein the location of the reference region corresponds to a position along the longitudinal axis of the flexible device, expressible as a function of the arc length parameter.

9. The method according to any of the previous Claims, wherein the first reference frame corresponds to a fixed global reference frame, and mapping the second shape comprises updating the shape and location of the flexible device relative to the fixed global reference frame.

10. A system for locating and sensing the shape of a flexible device over time, configured to perform the method according to any of Claims 1 to 9, the system comprising:an elongated tubular flexible device with sensors that allow to compute the shape of the flexible device;a data acquisition module configured to acquire the sensor data needed for computing the shape of the flexible device; anda processing unit configured to:b) compute, based on said acquired data, a first shape of the flexible device expressed relative to a first reference frame;c) select a reference region along said computed first shape of the flexible device, wherein said reference region serve to identify a displacement of the flexible device during subsequent measurements; e) compute, at a subsequent time step and based on data subsequently acquired via the one or more shape-sensing elements, asecond shape of the flexible device, said second shape being expressed in a second reference frame;f) estimate the presence and the location of said reference region along said computed second shape of the flexible device;g) calculate, based on said estimated location of said present reference region, a displacement of said first shape along the arc length of said second shape within said second reference frame;h) map, based on said calculated displacement, said second shape of the flexible device from the second reference frame to the first reference frame, thereby locating and shape sensing the flexible device over time; and,i) repeat step c) to h) at least one time using the computed second shape of the previous iteration of step e) as the computed first shape wherein the processing unit is further configured to replace step (c) by step c’ comprising- a step c’.l of estimating the need of selecting a new reference region for estimating the presence and the location of the reference region along the computed second shape in the following step f; and,- a step c’.2 of selecting a new reference region along said computed first shape if the estimation performed at step c’.l indicates that the reference region used in the preceding step g is not suitable for estimating the presence and location of the selected reference region along the computed second shape in the following step (f).