Patent Application
20090234235
The present invention relates to the field of medical devices, and, more particularly, to a system and method of tissue classification.
A large number of techniques and sensors are available today for tissue characterization, for example, to determine the presence of abnormal tissue, such as cancerous or pre-cancerous tissue. The problem of tissue characterization by applying different energy pulses and/or fields (e.g. electrical fringe field, polarizing magnetic field, RF impulses, etc.) and measuring and analyzing the respective electromagnetic or other properties of tissue (e.g. electric impedance, magnetic resonance, dielectric properties, optical characteristics, mechanical properties, etc.) has been recognized in prior art and various systems have been developed to provide a solution, for example:
U.S. Pat. No. 7,082,325 (Hashimshony et al.) entitled “Method and apparatus for examining a substance, particularly tissue, to characterize its type” discloses a method and apparatus for examining a substance volume to characterize its type, in particular, to characterize it as cancerous or non-cancerous. The method comprises applying a polarizing magnetic field through the examined substance; applying RF pulses locally to the examined substance volume such as to invoke electrical impedance (EI) responses signals corresponding to the electrical impedance of the substance, and magnetic resonance (MR) responses signals corresponding to the MR properties of the substance; detecting the EI and MR response signals; and utilizing the detected response signals for characterizing the examined substance volume type.
US Patent Application No. 2003/138,378 (Hashimshony.) entitled “Method and apparatus for examining tissue for predefined target cells, particularly cancerous cells, and a probe useful in such method and apparatus” discloses a method, apparatus and probe for examining tissue for the presence of target cells. Energy pulses are applied to the examined tissue. The changes in impedance and/or optical characteristics of the examined tissue produced by the applied energy pulses are detected and utilized for determining the presence of the target cells in the examined tissue.
US Patent Application No. 2007/179,397 (Hashimshony et al.) entitled “Method and system for examining tissue according to the dielectric properties thereof” discloses probes, systems, and methods for tissue characterization by its dielectric properties, wherein a physical feature of the probe is designed to define and delimit a tissue volume, at a tissue edge, where characterization takes place. Thus, the probe for tissue-edge characterization comprises: a first inner conductor, which comprises: proximal and distal ends, with respect to a tissue edge, along an x-axis; a first sharp edge, inherently associated with the proximal end; at least one feature, issuing from the first inner conductor, substantially at the proximal end, for forming at least one additional sharp edge, operative to enhance localized electrical fringe fields in the tissue, within a generally predefined tissue volume, at the tissue edge, the tissue volume being generally defined by physical parameters associated with the at least one feature; and a dielectric material, which encloses the conductor, in the y-z planes.
US Patent Application No. 2007/255,169 (Hashimshony et al.) entitled “Clean margin assessment tool” discloses an integrated tool, having a tissue-type sensor, for determining the tissue type at a near zone volume of a tissue surface, and a distance-measuring sensor, for determining the distance to an interface with another tissue type, for (i) confirming an existence of a clean margin of healthy tissue around a malignant tumor, which is being removed, and (ii) determining the depth of the clean margin. The integrated tool may further include a position tracking device and an incision instrument.
International Patent application WO07/83,310 (Hashimshony et al.) entitled “System and method for analysis and treatment of a body tissue” discloses a system and method for analysis and treatment of a tissue site. The system of the invention includes a probe unit containing one or more tissue sensing and monitoring probes configured to measure one or more parameters indicative of one or more states of the tissue site and one or more tissue treatment probes configured to deliver a treatment to the tissue site. A processor receives signals from the sensing and monitoring probes and determines whether the probe unit is located at the tissue site to be treated. The treatment and monitoring probes are activated in order to monitor the state of the tissue site while the treatment is being delivered to the tissue site. The processor receives signals from the sensing and monitoring probes during delivery of the treatment indicative of the state of the tissue site and determines, as the treatment is being carried out, any one or more of whether the treatment carried out so far is adequate, whether an additional treatment needs to be carried out, and whether the parameters of the treatment or the treatment targets need to be modified.
The is a need in the art to provide a system and method for automated on-line tissue classification based on acquired data indicative of tissue parameters.
In accordance with certain aspects of the present invention, there is provided a method of automated classifying a tissue, the method comprising: predefining a classification set comprising one or more classification parameters; acquiring data indicative of the tissue characteristics and sufficient for deriving values of said classification parameters; processing the acquired data and generating a value vector characterizing the tissue, said vector characterized by values of said classification parameters; matching the value vector to a plurality of predefined classification clusters in order to associate the tissue to be classified to an appropriate cluster thus giving rise to a matching cluster; transforming the value vector into a score value by use of a transformation algorithm, wherein the transformation algorithm and/or parameters thereof are selected in accordance with the matching cluster; comparing the score value with a classification criterion corresponding to said matching cluster; and providing a classification decision in accordance with results of said comparing.
In accordance with other aspects of the present invention, there is provided a system for tissue characterization comprising at least one probe unit with one or more sensors, said probe unit operatively coupled to a classification unit configured to obtain data indicative of the tissue characteristics and measured by the probe unit with the help of said sensors, said classification unit comprising a processor operatively coupled to a data repository configured to accommodate a predefined classification set comprising one or more classification parameters; wherein the processor is configured to be capable of:
In accordance with further aspects of the present invention, at least part of data indicative of tissue characteristics is acquired with the help of measuring the tissue characteristics by one or more sensors, as, by way of non-limiting examples, non-irradiative sensors and/or optical sensors.
In accordance with further aspects of the present invention the parameters in the classification set may be selected in accordance with one or more types of sensors used for measuring the tissue characteristics.
In accordance with further aspects of the present invention, the classification clusters are predefined in accordance with conclusive data and/or reference measurements, and may be grouped in accordance with respective tissue type and/or in accordance with respective tissue physiological features.
In accordance with further aspects of the present invention, the predefined classification clusters may be characterized by a certain sub-set of the classification parameters. Matching the value vector to a plurality of predefined classification clusters further comprises generating a value sub-vector characterized by values of parameters in said certain sub-set of the classification parameters; comparing said generated value sub-vector with said plurality of predefined clusters; and providing an automated decision with regard to a matching cluster.
Among advantages of certain embodiment of the present invention is facilitating automated real-time classification using affordable processing power.
In order to understand the invention and to see how it may be carried out in practice, embodiments will now be described, by way of non-limiting example only, with reference to the accompanying drawings, wherein:
a and 3b illustrate examples of classification parameters of the acquired data characterized by a broadband resonance form;
In the following detailed description, numerous specific details are set forth in order to provide a thorough understanding of the invention. However, it will be understood by those skilled in the art that the present invention may be practiced without these specific details. In other instances, well-known methods, procedures, components and circuits have not been described in detail so as not to obscure the present invention. In the drawings and description, identical reference numerals indicate those components that are common to different embodiments or configurations.
Unless specifically stated otherwise, as apparent from the following discussions, it is appreciated that throughout the specification discussions utilizing terms such as “processing”, “computing”, “calculating”, “determining”, “generating”, “matching”, “transforming”, or the like, refer to the action and/or processes of a computer or computing system, or processor or similar electronic computing device, that manipulate and/or transform data represented as physical, such as electronic, quantities within the computing system's registers and/or memories into other data, similarly represented as physical quantities within the computing system's memories, registers or other such information storage, transmission or display devices.
Embodiments of the present invention may use terms such as, processor, computer, apparatus, system, sub-system, module, unit, device (in single or plural form) for performing the operations herein. This may be specially constructed for the desired purposes, or it may comprise a general purpose computer selectively activated or reconfigured by a computer program stored in the computer. Such a computer program may be stored in a computer readable storage medium, such as, but not limited to, any type of disk including optical disks, CD-ROMs, Disk-on-Key, smart cards (e.g. SIM, chip cards, etc.), magnetic-optical disks, read-only memories (ROMs), random access memories (RAMs), electrically programmable read-only memories (EPROMs), electrically erasable and programmable read only memories (EEPROMs), magnetic or optical cards, or any other type of media suitable for storing electronic instructions capable of being conveyed via a computer system bus.
The processes/devices presented herein are not inherently related to any particular electronic component or other apparatus, unless specifically stated otherwise. Various general purpose components may be used in accordance with the teachings herein, or it may prove convenient to construct a more specialized apparatus to perform the desired method. The desired structure for a variety of these systems will appear from the description below. In addition, embodiments of the present invention are not described with reference to any particular programming language. It will be appreciated that a variety of programming languages may be used to implement the teachings of the inventions as described herein.
The term “criterion” used in this patent specification should be expansively construed to include any compound criterion, including, for example, several criteria and/or their combination.
The references cited in the background are assigned to the common assignee of the present application, and teach many principles of implementation of tissue classification that are applicable to the present invention. Therefore the full contents of these publications are incorporated by reference herein, for appropriate teachings of additional or alternative details, features and/or technical background.
Reference is made to
The probe unit and sensors thereof may be provided in different configurations, some of which are known in prior art (for example, from references cited in the background and incorporated herewith by reference). In accordance with certain embodiments of the present invention, the probe unit 11 may be configured, by way of non-limiting example, as an extracorporeal device, an intra-corporeal device, a device adapted for use on a portion of subcutaneous tissue, a device adapted for use on a portion of intra-corporeal tissue during open surgery, or otherwise. The intra-corporeal devices may be specifically configured for minimally invasive surgery and/or insertion via a trocar valve and/or for insertion via a body orifice to a body lumen (e.g. for use on a portion of an inner lumen wall for further penetrating the lumen for use on a portion of an intra-corporeal tissue outside the lumen) and/or for percutaneous insertion to a body lumen (e.g. for use on a portion of inner lumen wall for further penetrating the lumen for use on a portion of an intra-corporeal tissue outside the lumen).
According to various exemplary embodiments of the invention, the probe unit 11 is configured to obtain data indicative of one or more tissue properties as, for example, electromagnetic properties (e.g. using a non-irradiative sensor), dielectric properties, impedance, biological properties, chemical properties, optical properties (e.g. fluorescence emission and/or absorption and/or reflectance of selected wavelengths of light), MRI, energy transmission and/or reflectance (e.g. radio frequency [RF] or microwave [MW]) and temperature (e.g. via infrared thermography).
The sensors 13 may be configured, for example, as disclosed in International Published Application WO 2006/103665 entitled “Electromagnetic sensors for tissue characterization”, U.S. Pat. No. 6,813,515, entitled “Method and system for examining tissue according to the dielectric properties thereof” which are assigned to the common assignee of the present application and which is hereby incorporated by reference, or otherwise.
In some exemplary embodiments of the invention, the probe unit may include a cutting tool and/or a sampling tool.
The control unit 12 is a computer system configured to receive and transmit data and/or derivatives thereof measured by the probe unit, and to provide respective control inputs to the probe unit. The control unit typically includes inter alia a data processing utility, as well as data input and output utilities, a data storage utility, communication utility, one or more control buttons (e.g. for activating the probe unit and/or for initiating and/or measuring/classification process).
The control unit and, optionally, the probe unit are operatively coupled to a display 19 configured to display information (e.g. acquired data and/or derivatives thereof, etc.) pertinent to tissue classification, diagnosis and/or surgical planning.
In accordance with certain embodiments of the invention, the data measured by the probe unit are classified by a classification unit 14 operationally coupled to the control unit 12, to the display 19 and to the probe unit 11 (directly or via the control unit). The classification unit is configured to receive data and/or derivatives thereof from the control unit and/or the probe unit, to analyze the received data, as will be further detailed with reference to
Those versed in the art will readily appreciate that the invention is not bound by the probe unit, sensors or technique of obtaining tissue characteristics as described with reference to
Those versed in the art will also readily appreciate that the invention is not bound by the system configuration of
Those skilled in the art will also readily appreciate that the data repositories may be consolidated or divided in another manner; some of these databases may be fully or partly shared with other systems, including 3rd party equipment.
Referring to
The acquired data are processed by the processor 15. This processing comprises deriving values of a first set of predefined (and/or configurable) parameters, said values being derivatives of the measured data. The first set comprises a plurality of parameters characterizing the signal. Optionally the first set of parameters (or subset thereof) may be used for signal integrity testing. The processor 15 further derives (21) values of a second set of predefined (and/or configurable) parameters. The second set of parameters is usable for tissue classification and is referred to hereinafter as a “classification set”. In certain embodiments of the invention the classification set may be the same or may constitute a sub-set of the first set and the values may be derived as a set/subset of values of the first set. In other embodiments of the invention, one or more parameters of the classification set may be other than those comprised in the first set, and their respective values may be, for example, derived from the acquired data directly (e.g. demographic, clinical or other characteristics of a respective person and the like). The values for the classification data set are derived in accordance with certain algorithms and configurations which are accommodated in the algorithms and configurations repository 162. Typically the parameters in the first set and the classification set differ for different methods of tissue characterization and/or types of sensors and/or types of organs sensed and/or clinical purposes. The derived values of the classification set are presented by a value vector characterizing the tissue to be classified. Optionally, the parameters in the classification set may have different weights and the respective values may be processed accordingly.
a and 3b illustrate non-limiting examples of data acquired when measuring tissue by a non-irradiative electromagnetic sensor (for example such as disclosed in International Application No. WO 2006/103665). The illustrated acquired data characterize the reflection coefficient amplitude of a reflection signal (
Frequency at resonance (fo)
Amplitude at resonance (|R|dip)
Phase at resonance
Quality factor (Q)
Frequency at −3 db point (f−3db)
Frequency at +3 db point (f+3db)
Amplitude at −3 db point
Amplitude at +3 db point
Phase at −3 db point
Phase at +3 db point
The classification sets may constitute different subsets of these parameters. When measuring the dependency between the reflection parameters and the signal frequency by non-irradiative electromagnetic sensor, the measured data for various tissue types are characterized by the broadband resonator form, while the parameters in the classification set as well as values thereof may differ, for example for different sampled organs. Additionally, as was detailed with reference to
The meaningful values of parameters comprised in the classification set of parameters constitute a space of parameters. Referring back to
The clusters' configuration is predefined before the classification process and, optionally, may be reconfigured (e.g. with the help of learning algorithms) during the process. The clusters' configuration may be accommodated in the algorithms and configurations repository 162. The processor 15 generates (22) a value sub-vector representing the values of parameters in the third set. The values may be derived as a set/sub-set of values vector or directly from the obtained data. The value sub-vector represents a relatively small set of clinically useful parameter values, and may be easily evaluated and processed. The processor compares the derived values of the third set of parameters (value sub-vector) with the predefined clusters, thus matching (23) the sub-vector to an appropriate cluster, such a cluster referred to hereinafter as a “matching cluster”. As a result, the entire respective value vector and, accordingly, the tissues characterized by said vector, are associated with the matching cluster, i.e. the cluster grouping the tissues and/or tissue features which are as similar as possible to the tissue to be classified.
The matching may be provided, for example, in accordance with minimal distance between the value sub-vector and centroids of the clusters (e.g. when using algorithms described with reference to
The processor further transforms (24) the value vector into a score value, said transformation depending on the matching cluster (e.g. by a respective cluster-dependent projection to a one-dimensional space). By way of non-limiting example, for a transformation to a MANOVA generated canonical variable, the number of parameters and/or the type of parameters and/or the weights of the parameters may be cluster dependent. By way of another non-limiting example, for nearest neighbors based transformations, the number of parameters and/or the type of parameters and/or the number of neighbors; and the metric on the distance may be cluster dependent. The transformation algorithms and/or parameters thereof are predefined (and, optionally, configurable and/or trained) for each cluster. The resulting score value is compared (25) with a respective cluster-dependent threshold value. The output of this operation serves as a basis for a classification decision (26). The thresholds may be predefined for each cluster and accommodated in the reference database 161. Alternatively, the respective thresholds may result from training a neural network based on conclusive data (e.g., histological data from biopsy results, clinical data, etc.), and associated corresponding thresholds may be accommodated in the reference database 161. Among advantages of using the score values and classification threshold is facilitating suitable classification in real-time using affordable processing power. Using a trained neural network for defining classification threshold may enhance the reliability of such a decision. Those versed in the art will readily appreciate that the invention is not bound by a one-dimensional score and threshold values, and is, likewise, applicable to other dimensions of score and/or threshold values. The number of dimensions for these values depends on the desired trade-off between exact classification and available processing power.
Optionally the classification decision may be provided in a supervised manner including comparing the value vector and/or the score value with respective values associated with corresponding conclusive data accommodated in the reference database 161. The supervision operation may take the form of searching for a look-up table, defining a nearest neighbor, training a neural network, etc.
Referring to
In certain embodiments of the invention the clusters may be defined, for example, by use of a K-means algorithm. Parameters in which there is a correlation between the parameter values and major tissue type/features and/or certain tissue types/features of interest are selected. The selected parameters are first normalized, wherein normalization of each parameter is performed by first subtracting the average value of the parameter, and further dividing by the standard deviation of the parameter. Next, the normalized values pi are weighted with weights ai. Using parameters pi from a set of reference measurements, a K-means algorithm is used to find the centroids c(j) relative to piaj. The parameters pi and weights ai are adjusted until all boundaries of the clusters are as required by physiological data indications of a specific type of breast tissue, or tissue features. The operation may be provided based on conclusive data (e.g., histological data from biopsy results, clinical data, and the like), and reference measurements.
In certain embodiments of the invention the clusters may be defined, for example, by use of the parameter values of sensor readings from homogenous major tissue types and/or features. The homogenous tissues/features samples may be determined by histo-pathological analysis or otherwise. The measuring results are parameterized to a set of classification parameters. A MANOVA analysis is performed, with measurements corresponding to each tissue type assigned to a separate group. The result of the MANOVA operation generates principle/canonical variables, which are linear combinations of the classification parameters. Each group center is characterized by a specific value of the canonical variable or variables. These values of the canonical variable(s) define centroids, using the classification parameters, for each of the homogenous tissue groups selected. These centroids are used as the centroids for clustering.
By way of non-limiting example (not-illustrated in
Those versed in the art will readily appreciate that the invention is not bound by the type of sensor and/or characteristics of acquired data illustrated with reference to
Those skilled in the art will readily appreciate that the invention in some of its aspects contemplates a computer program being readable by a computer for executing the method of the invention. The invention in some of its aspects further contemplates a machine-readable memory tangibly embodying a program of instructions executable by the machine for executing the method of the invention.
It is also to be understood that the invention is not limited in its application to the details set forth in the description contained herein or illustrated in the drawings. The invention is capable of other embodiments and of being practiced and carried out in various ways. Hence, it is to be understood that the phraseology and terminology employed herein are for the purpose of description and should not be regarded as limiting. As such, those skilled in the art will appreciate that the conception, upon which this disclosure is based, may readily be utilized as a basis for designing other structures, methods, and systems for carrying out the several purposes of the present invention.
Those skilled in the art will readily appreciate that various modifications and changes can be applied to the embodiments of the invention as hereinbefore described without departing from its scope, defined in and by the appended claims.
