KR102622758B1 - System and method for predicating position of capsule endoscope - Google Patents

Abstract

This technology discloses a system and method for predicting the position of a capsule endoscope. A specific implementation example of this technology is to estimate the capsule position based on the pre-stored phase difference pattern for each channel due to the change in capsule position and the input digital phase difference signal, and the small size transmitted from the endoscopic capsule A stable phase difference signal can be derived from the original signal, and the capsule position can be derived in real time, thereby increasing the accuracy of lesion location and lesion diagnosis rate.

Description

Capsule endoscope position prediction system and method {SYSTEM AND METHOD FOR PREDICATING POSITION OF CAPSULE ENDOSCOPE}

The present invention relates to a capsule endoscope position prediction system and method. More specifically, the capsule position and angle are determined based on a pre-stored phase difference pattern and an input digital phase difference signal for changes in phase difference for each channel due to changes in capsule position and angle. It is about technology that allows estimation.

Capsule endoscopy is a device that performs examination of lesions in the stomach and large intestine. It measures the size of the RF signal, which is an image signal, transmitted through a certain number of antennas in the capsule and the position of the capsule using the RSSI (Received Signal Strength Indicator) method.

However, because the strength of the transmitted RF signal is small, the RSSI signal is small, and this RSSI signal is greatly influenced by the environment, reaching a limit where it is difficult to obtain a stable pattern of size and location.

Accordingly, the present applicant would like to propose a method of generating and storing a pattern for the phase difference according to the position change of the capsule based on the phase difference of the original signal transmitted from the endoscopic capsule and estimating the capsule position from the input original signal using the previously stored pattern. .

Therefore, the technical problem to be achieved by the present invention is to derive a stable phase difference from the original signal by estimating the capsule position and angle based on the pre-stored phase difference pattern for the phase difference change for each channel due to the capsule position change and the input digital phase difference signal. This is to fundamentally improve the accuracy of capsule location, accurately identify the location of the lesion, and thereby increase the lesion diagnosis rate.

The object of the present invention is not limited to the object mentioned above, and other objects and advantages of the present invention that are not mentioned can be understood through the following description and will be more clearly understood through the examples of the present invention. Additionally, it will be readily apparent that the objects and advantages of the present invention can be realized by means and combinations thereof as set forth in the claims.

The position prediction system for a capsule endoscope according to an embodiment of the present invention,

External equipment that receives an original signal containing a carrier signal of a specific frequency transmitted from a capsule endoscope through a plurality of antennas, then derives the phase difference of the received original signal and converts the derived phase difference into digital form;

a signal processing device that divides the digital phase difference signal received from the external equipment into segments of a specific length and extracts the phase difference of the carrier signal over time;

a pattern storage device that measures and stores in advance the pattern of the phase difference signal for each channel according to changes in various positions and angles of the capsule, using the phase difference signal for each channel acquired through the external equipment and the signal processing device; and

One feature is that it is provided as a position estimation device that estimates the position and angle of the capsule in real time by comparing the digital phase difference signal received through the external equipment and the signal processing device with the pattern stored in the pattern storage device.

Preferably, the external equipment is,

multiple antennas;

a BPF (Band Pass Filter) that passes the original signal in a predetermined frequency band among the original signals received through the antenna;

A phase detector (PD: Phase Detector) that detects a phase difference signal of the original signal in the predetermined frequency band; and

It may include an ADC (Analog to Digital Converter) that converts the phase difference signal of the original signal into digital form.

Preferably, the signal processing device is:

A kernel module that generates and stores a pattern of a section containing a carrier signal within a segment signal of a specific length received through the external device as a kernel; and

It may be provided as a quantization module that convolves the kernel and the phase difference signal to extract the position of the carrier signal section within the segment, calculates the average of only the phase difference of the carrier signal, and quantizes it.

Preferably, the location estimation device,

Capsule position based on the similarity between the phase difference signal for each channel at a specific time obtained through the quantization module of the signal processing device at one capsule position and angle and the phase difference pattern for each channel at a specific position and angle previously stored in the pattern storage device. and may be provided to estimate the angle.

Preferably, the location estimation device,

The phase difference pattern for each channel for a specific position and angle previously stored in the pattern storage device is input, a learning model is constructed using the position and angle as the correct answer, and the position and angle of the capsule estimated from the neural network for each input are calculated. It may be equipped to estimate the capsule position and angle using a neural network trained with the error between the correct position and angle. The control method of the position prediction system of the capsule endoscope according to an embodiment of the present invention includes the following:

Receiving a raw signal containing a carrier signal of a specific frequency transmitted from a capsule endoscope through a plurality of antennas, then deriving the phase difference of the received raw signal signal of a specific frequency and converting the derived phase difference into digital form;

Dividing a digital phase difference signal received from an external device into segments of a specific length and extracting a phase difference signal from the carrier signal over time;

Preliminarily measuring and storing the acquired phase difference signal for each channel according to changes in various positions and angles of the capsule; and

One feature includes the step of estimating the position and angle of the capsule in real time by comparing the digital phase difference signal received from the external equipment with a pre-stored pattern.

Preferably, the step of estimating the position of the capsule is,

It may be equipped to estimate the capsule position and angle based on the similarity of the phase difference signal for each channel at a specific time obtained from one capsule position and angle and the phase difference pattern for each channel at a pre-stored specific position and angle.

Preferably, the step of estimating the position of the capsule is,

The phase difference pattern for each channel for a specific position and angle previously stored in the pattern storage device is input, and a learning model is constructed using the measured position and angle as the correct answer, and the position of the capsule estimated by the neural network for each input and It may be equipped to estimate the capsule position and angle using a neural network trained with the error between the angle and the correct answer position and angle.

According to one embodiment, the capsule position and angle are estimated based on a pre-stored phase difference pattern and the input digital phase difference for each channel due to changes in capsule position and angle, thereby deriving a stable phase difference signal from a small original signal. The accuracy of capsule location can be fundamentally improved, the lesion location can be accurately identified, and the lesion diagnosis rate can be increased.

The following drawings attached to this specification illustrate preferred embodiments of the present invention, and together with the detailed description of the invention described later, serve to further understand the technical idea of the present invention. Therefore, the present invention includes the matters described in such drawings. It should not be interpreted as limited to only .
1 is a configuration diagram of a position prediction system for a capsule endoscope according to an embodiment.
Figure 2 is a detailed configuration diagram of the external equipment of Figure 1.
Figure 3 is an example of antenna installation of the external equipment of Figure 2.
Figure 4 is a detailed configuration diagram of the signal processing device of Figure 1.
Figure 5 is a diagram showing the processing algorithm of the signal processing device of Figure 4.
Figure 6 is an output waveform diagram of each part of the signal processing device of Figure 4.
Figure 7 is a diagram showing the processing algorithm of the pattern storage device of Figure 1.
Figure 8 is an example diagram showing a pattern of phase difference signal change for each channel according to capsule position change according to an embodiment.

Specific structural or functional descriptions of the embodiments are disclosed for illustrative purposes only and may be modified and implemented in various forms. Accordingly, the embodiments are not limited to the specific disclosed form, and the scope of the present specification includes changes, equivalents, or substitutes included in the technical spirit.

Terms such as first or second may be used to describe various components, but these terms should be interpreted only for the purpose of distinguishing one component from another component. For example, a first component may be named a second component, and similarly, the second component may also be named a first component.

When a component is referred to as being “connected” to another component, it should be understood that it may be directly connected or connected to the other component, but that other components may exist in between.

Singular expressions include plural expressions unless the context clearly dictates otherwise. In this specification, terms such as “comprise” or “have” are intended to designate the presence of the described features, numbers, steps, operations, components, parts, or combinations thereof, and are intended to indicate the presence of one or more other features or numbers, It should be understood that this does not exclude in advance the possibility of the presence or addition of steps, operations, components, parts, or combinations thereof.

Unless otherwise defined, all terms used herein, including technical or scientific terms, have the same meaning as commonly understood by a person of ordinary skill in the relevant technical field. Terms as defined in commonly used dictionaries should be interpreted as having meanings consistent with the meanings they have in the context of the related technology, and unless clearly defined in this specification, should not be interpreted in an idealized or overly formal sense. No.

Before explaining the specification, some terms used in the specification will be clarified. For example, in this specification, a medical image signal is a signal acquired and transmitted from a capsule for vascular endoscopy or gastrointestinal endoscopy, and the original signal is a signal that passes through a predetermined frequency band among medical images through an antenna. In addition, the original signal or medical image throughout the specification is explained as an example in RF format, but is not limited to this.

Accordingly, in one embodiment, a pattern for the phase difference according to the position change of the capsule is generated and stored based on the phase difference of the original signal transmitted from the endoscopic capsule, and the capsule position is estimated based on the previously stored pattern of the input original signal.

Hereinafter, the present invention will be described in detail by explaining preferred embodiments of the present invention with reference to the attached drawings.

FIG. 1 is a configuration diagram of an endoscope capsule position prediction system according to an embodiment, FIG. 2 is a detailed configuration diagram of the external equipment of FIG. 1, FIG. 3 is an example antenna installation diagram of the external equipment of FIG. 2, and FIG. 4 is a detailed configuration diagram of the signal processing device of FIG. 1, FIG. 5 is a diagram showing the processing algorithm of the signal processing device of FIG. 4, FIG. 6 is an output waveform diagram of each part of the signal processing device of FIG. 1, and FIG. 7 is These are example diagrams showing patterns of phase difference signal changes for each channel according to capsule position changes according to an embodiment.

Referring to FIGS. 1 to 7, the position prediction system of the capsule endoscope of one embodiment includes an external device (1), a signal processing device (2), a pattern storage device (3), and a position estimation device (4). , It is equipped with a configuration that estimates the capsule position based on the phase difference pattern for each channel according to the previously stored capsule position and angle change and the input digital phase difference signal.

Here, the external equipment (1) receives the raw signal containing the carrier signal of a specific frequency transmitted from the capsule endoscope through a plurality of antennas, then derives the phase difference of the received original signal and converts the derived phase difference into digital form. It has a configuration, and referring to FIG. 2, it may include a plurality of antennas (11), a Band Pass Filter (BPF) 12, a Phase Detector (PD) (13), and an Analog to Digital Converter (ADC) (14). there is. Here, the specific frequency is an RF frequency that can be used in a capsule endoscope, for example, the 915 MHz band.

For example, referring to FIG. 3, the external equipment 1 is manufactured in consideration of the human body size ratio, is made of acrylic to minimize attenuation of RF signals, and may be provided in a two-layer structure. In addition, the manufactured external equipment (1) has 8 ceramic antennas fixedly installed, and the RF signal received through each antenna has a range of 0 to 1.8V, and each antenna generates a 10 RF signal difference per 0.1V, The 8 RF-type raw signals from each antenna are separated into 16 signal lines using a distribution adapter and then obtained from the 16 signal lines.

In addition, the BPF (12) receives the RF-type raw signal from each antenna as input, passes the RF-type raw signal in a predetermined frequency band, and the passed raw signal is transmitted to the PD (13), and the PD (13) The phase signal obtained at the period of the phase signal having periodicity is sampled, and then the phase difference of the sampled original signal is detected. The detected phase difference is converted into digital form by the ADC 14. The phase difference converted into digital form is transmitted to the signal processing device (2). At this time, the original signal includes a carrier signal of a specific frequency.

Accordingly, the signal processing device 2 is configured to extract the phase difference of the carrier signal over time by dividing the digital phase difference signal received from the external equipment into segments of a specific length. Here, the specific length is a time unit in which the carrier frequency can be detected.

That is, the signal processing device 2 may include a kernel module 21 and a quantization module 22, as shown in FIG. 4. Accordingly, the signal processing device 2 generates and stores a pattern of a section containing a carrier signal within a segment signal of a specific length received through the external equipment 1 as a kernel, and convolves the kernel with the phase difference signal to create a segment. The position of the carrier signal section is extracted, the average of only the phase difference of the carrier signal is calculated, and quantized. The operation process of the signal processing device 2 will be described below with reference to FIG. 5. That is, the kernel module 21 generates and stores the pattern of the section containing the carrier signal within the segment signal of a specific length received through the external equipment 1 as a kernel, and the quantization module 22 generates and stores the kernel and the above. The phase difference signal is convolved to extract the position of the carrier signal section within the segment, and the average of only the phase difference of the carrier signal is obtained and quantized.

Accordingly, the phase difference of the carrier signal of the quantization module 22 is a more stable signal than the phase difference signal of the PD 13 obtained from a small-sized original signal.

Referring to FIG. 6, raw data with a certain periodicity is shown in (a), and a kernel produced from one cycle of the raw signal is shown in (b). In addition, the result of convolution of the produced kernel and the original signal is as shown in (c), and the phase difference signal combining the derived convolution result signal and the original signal is as shown in (d).

Meanwhile, the pattern storage device 3 measures the phase difference signal for each channel acquired through the external equipment 1 and the signal processing device in advance by measuring the pattern of the phase difference signal for each channel according to changes in various positions and angles of the capsule. Save.

Referring to FIG. 7, the pattern storage device 3 receives the original signal of the capsule endoscope received through the external equipment 1, acquires a phase difference signal for the original signal through the signal processing device 2, and then prints the pattern. After saving, change the position and angle of the capsule and receive the original signal from the capsule endoscope through external equipment (1). In the above-described pattern creation process, the position and angle of the capsule are sequentially changed, and then the pattern of the received capsule is created and stored.
Figure 8 is an example diagram showing a pattern of phase difference signal change for each channel according to capsule position change according to an embodiment. Referring to FIG. 8, it is a pattern showing the change in phase difference for each channel according to the capsule position and angle change stored in the pattern storage device 3. The change in phase difference for each channel according to the capsule position changes in two dimensions with a fixed angle and x and y. The pattern showing is as shown in (a), and the pattern showing the phase difference change for each channel according to the position fixation and angle change of the capsule is shown in (b).

delete

Meanwhile, the position estimation device 4 compares the digital phase difference signal received through the external equipment 1 and the signal processing device 2 with the pattern stored in the pattern storage device 3 to determine the position of the capsule in real time. The angle can be estimated.

Here, the position estimation device 4, for example, stores the phase difference signal for each channel at a specific time obtained through the quantization module 22 of the signal processing device 2 at one capsule position and angle and the pattern storage device 3. The capsule position and angle can be estimated based on the similarity of the phase difference pattern for each channel at a previously stored specific position and angle. Here, the specific time is the current time of obtaining a phase difference signal through the quantization module 22 of the signal processing device 2 at the position and angle of one capsule, and the specific position is the position and angle of the one capsule. In another example, the position estimation device 4 takes as input the phase difference pattern for each channel for a specific position and angle previously stored in the pattern storage device 3, builds a learning model with the corresponding position and angle as the correct answer, and calculates each input The capsule position and angle can be estimated using a neural network trained with the error between the position and angle of the capsule estimated by the neural network and the correct position and angle.

Accordingly, in one embodiment, the capsule position and angle are estimated based on the pre-stored phase difference pattern for each channel according to the capsule position and angle change and the input digital phase difference signal, and the small size transmitted from the endoscope capsule is estimated. A stable phase difference signal can be derived from the original signal, and the capsule position can be derived in real time, thereby increasing the accuracy of lesion location and lesion diagnosis rate.

Although the embodiments have been described with limited drawings as described above, those skilled in the art can apply various technical modifications and variations based on the above. For example, the described techniques are performed in a different order than the described method, and/or components of the described system, structure, device, circuit, etc. are combined or combined in a different form than the described method, or other components are used. Alternatively, appropriate results may be achieved even if substituted or substituted by an equivalent.

By estimating the capsule position and angle based on the pre-stored phase difference pattern for each channel due to changes in capsule position and angle and the input digital phase difference signal, the small-sized original signal transmitted from the endoscope capsule is stable. The operation accuracy and reliability of the capsule endoscope position prediction system and method, which can derive a phase contrast signal and the capsule position in real time, thereby increasing the accuracy of the lesion position and lesion diagnosis rate, and further performance efficiency It can bring about great progress in the field, can be applied in a variety of fields, and by securing the core technology of medical imaging, diagnostic monitoring in related industries can be actively utilized, and the commercialization or sales of medical imaging diagnostic systems can be facilitated. It is an invention that has industrial applicability because it not only has sufficient potential but can also be clearly implemented in reality.

1: External equipment
11: Multiple antennas
12: BPF
13: P.D.
14: ADC
2: Signal processing device
21: Kernel module
22: Quantization module
3: Pattern storage device
4: Location estimation device

Claims (8)

External equipment that receives an original signal containing a carrier signal of a specific frequency transmitted from a capsule endoscope through a plurality of antennas, then derives the phase difference of the received original signal and converts the derived phase difference into digital form;
a signal processing device that divides the digital phase difference signal received from the external equipment into segments of a specific length and extracts the phase difference of the carrier signal over time;
a pattern storage device that measures and stores in advance the pattern of the phase difference signal for each channel according to changes in various positions and angles of the capsule, using the phase difference signal for each channel acquired through the external equipment and the signal processing device; and
A capsule endoscope, characterized in that it is equipped with a position estimation device that estimates the position and angle of the capsule in real time by comparing the digital phase difference signal received through the external equipment and the signal processing device with the pattern stored in the pattern storage device. location prediction system.
The method of claim 1, wherein the external equipment is:
multiple antennas;
a BPF (Band Pass Filter) that passes the original signal in a predetermined frequency band among the original signals received through the antenna;
A phase detector (PD: Phase Detector) that detects a phase difference signal of the original signal in the predetermined frequency band; and
A position prediction system for a capsule endoscope, comprising an ADC (Analog to Digital Converter) that converts the phase difference signal of the original signal into digital form. The method of claim 1, wherein the signal processing device:
A kernel module that generates and stores a pattern of a section containing a carrier signal within a segment signal of a specific length received through the external device as a kernel; and
A position prediction system for a capsule endoscope, characterized in that it is provided with a quantization module that convolves the kernel and the phase difference signal to extract the position of the carrier signal section within the segment, calculates the average of only the phase difference of the carrier signal, and quantizes it. The method of claim 3, wherein the location estimation device,
Estimating the capsule position and angle based on the similarity between the phase difference signal for each channel at a specific time obtained through the quantization module at one capsule position and angle and the phase difference pattern for each channel at the corresponding position and angle previously stored in the pattern storage device. A position prediction system for a capsule endoscope, characterized in that it is provided to do so. The method of claim 3, wherein the location estimation device,
The phase difference pattern for each channel for a specific position and angle previously stored in the pattern storage device is input, a learning model is constructed using the position and angle as the correct answer, and the position and angle of the capsule estimated from the neural network for each input are calculated. A position prediction system for a capsule endoscope, characterized in that it is equipped to estimate the capsule position and angle using a neural network trained with the error between the correct position and angle. In the method for predicting the position of a capsule endoscope performed based on the position prediction system for the capsule endoscope of claim 1,
Receiving a raw signal containing a carrier signal of a specific frequency transmitted from a capsule endoscope through a plurality of antennas, then deriving the phase difference of the received raw signal signal of a specific frequency and converting the derived phase difference into digital form;
Dividing a digital phase difference signal received from an external device into segments of a specific length and extracting a phase difference signal from the carrier signal over time;
Preliminarily measuring and storing the acquired phase difference signal for each channel according to changes in various positions and angles of the capsule; and
An endoscopic capsule position prediction method comprising the step of estimating the position and angle of the capsule in real time by comparing the digital phase difference signal received from the external equipment with a pre-stored pattern. The method of claim 6, wherein estimating the position of the capsule comprises:
Endoscopic capsule position prediction, characterized in that it is equipped to estimate the capsule position and angle based on the similarity of the phase difference signal for each channel at a specific time obtained from one capsule position and angle and the similarity of the phase difference pattern for each channel at a pre-stored specific position and angle. method. The method of claim 6, wherein estimating the position of the capsule comprises:
A learning model is constructed using the pre-stored phase difference pattern for each channel for a specific position and angle as input, and the measured position and angle as the correct answer. For each input, the position and angle of the capsule estimated by the neural network and the correct answer position and An endoscopic capsule position prediction method, characterized in that it is equipped to estimate the capsule position and angle using a neural network trained with the error between angles.