WO2024039316A1

WO2024039316A1 - Organ sound guided radiotherapy (os-grt) system

Info

Publication number: WO2024039316A1
Application number: PCT/TR2022/050992
Authority: WO
Inventors: Fatih BİLTEKİN
Original assignee: Hacettepe Üni̇versi̇tesi̇
Priority date: 2022-08-19
Filing date: 2022-09-16
Publication date: 2024-02-22

Abstract

The invention relates to a system for the detection and tracking of tumor motion under the guidance of organ sound and the operational method of this system.

Description

ORGAN SOUND GUIDED RADIOTHERAPY (OS-GRT) SYSTEM

TECHNICAL FIELD

The invention relates to a system that involves a novel method used to determine tumor localization instantaneously under the guidance of organ sound in radiotherapy applications and additionally to detect treatment efficacy or side-effect patterns before, during, and after treatment.

STATE OF THE ART

Taking internal organ motions and especially respiratory motions into consideration has an important role in localizing and imaging a tumor during radiotherapy and correctly irradiating the tumor during treatment. Today, in order to take tumor/organ motions caused by respiration into account, the computerized tomography (CT) scans are performed in different respiratory phases by using various hardware and software during the simulation phase. Thereby, a respiratory related tumor/organ motion pattern can be identified and irradiation can be applied in defined respiratory phases or amplitude range where the tumor spend more time during respiration. This approach is defined as gating. Moreover, the traditional methods include the irradiation of the tumor in the relevant respiratory phase by performing CT simulation in a certain phase of respiration either by using the deep breath holding technique or with an active breath control system. In addition to these methods, continuous irradiation can be performed by dynamically monitoring the tumor motion with some specific algorithms and treatment devices. Furthermore, today, it has become also possible to monitor tumors in selected patient groups with the use of dynamic magnetic resonance imaging-based linear accelerators (MRI- linac).

In both the method of irradiation, gating and the dynamic tracking method in the state of the art, it is necessary to place at least three gold markers, so-called fiducial markers (at different planes and angles), into the tumor to determine the respiration-related tumor motion. Therefore, both methods require an invasive procedure and the application increases the risk of pneumothorax as well. In addition, the existing markers will likely be displaced within the tissue during treatment planning and between the treatment fractions, which constitutes a significant disadvantage for the approaches. In the deep breath hold technique and active breath control systems, a patient's respiratory functions must be above the determined threshold values and their general condition must be good enough for them to tolerate the said approaches. Today, dynamic tumor monitoring is also possible with MRI-linac. However, due to the high cost of the existing systems, it is available in a limited number of clinics and it seems not possible to integrate the MRI method into currently used modern devices and systems with a robotic gantry by means of current technology. In addition, it is currently not possible to treat patients who have MR incompatible pacemakers, cardiac defibrillators, heart valves, stents, cochlear implants, electronic drug pumps and prostheses with MRI-linac. On the other hand, MRI-linac is still a high-cost system today whose linear accelerator system involves inferior technology compared to present linear accelerators in terms of modernization. For example, with the current technology, new treatment modalities such as volumetric modulated arc therapy (VMAT) with MRI-linac are still not applicable. The maximum dose rate is lower in the present linear accelerators which may lead to prolonged treatment periods, increased uncertainty due to patient immobilization, and in some cases, the inability to obtain effective treatment plans.

BRIEF DESCRIPTION AND OBJECT OF THE INVENTION

The present invention relates to tumor localization detection under the guidance of organ sound, which meets the above-mentioned requirements, eliminates all the disadvantages, and brings some additional advantages.

The main object of the invention is to eliminate the need for dynamic radiological imaging during treatment by associating organ sound and tumor/organ motion.

The present invention relates to detecting sounds from an organ, especially from lungs, and associating the said sounds with tumor/organ motion.

Thereby, placing radio-opaque markers within the tumor and/or any extra radiological imaging is no longer required for the detection of tumor motion during treatment. In addition, tumor response and/or side effect pattern are studied via organ sound, by analyzing the differences in sounds from the relevant organ before, during, and after the treatment with the present approach.

Organ sound guided radiotherapy (OS-GRT) involves creating a correlation model between the detected organ sound, especially the lung sound, and the tumor motion, and to determine the tumor position by monitoring the tumor motion under the guidance of organ sound.

Within the scope of the invention, it is aimed to detect sounds from an organ, especially sounds from lungs, and to associate the said sounds with tumor motion during radiotherapy without requiring any radiological imaging and/or invasive procedure.

In addition to tracking tumor position, the organ function and tumor response and/or side effect pattern can be studied by analyzing acoustic sound differences (frequency values, breathing pattern/sound, etc.) coming from the related organ. Moreover, on contrary to the known methods (such as active breath control, deep breath hold, gating), the present method allows the management of not only motions caused by respiration, but also those in all anatomical regions where a difference in organ sound and tumor motion can be observed.

The invention can be employed in the diagnosis, monitoring, and treatment stages of all radiotherapy applications where the tumor/organ position changes in relation to organ sound, and organ sound can be associated with tumor/organ motion and organ function.

Within the scope of the pre-feasibility study conducted for the invention, it has been verified on the synchrony phantom specially designed for motion tracking that the sound detected by modifying a conventional stethoscope can be converted into digital data by means of a multichannel data collection software and decibel meter measurement program, the obtained data can be synchronized with the real-time respiratory cycle predefined in the system, and the tumor motion can be detected based on the frequency range of the sound in the relevant respiration phase, and most importantly, tumor tracking can be performed in real time.

The invention generally involves the process steps of placing devices or sensors with specific features (such as a sound sensor, electronic stethoscope or a special microphone system) on the body (such as thorax, neck and back region predetermined depending on the treatment area and patient position) during any imaging method (such as 4-dimensional computed tomography, fluoroscopy, dynamic MRI, ultrasonography) which is performed for treatment planning or simulation purposes during preparation for radiotherapy and allows detection of tumor/organ motion pattern in different phases of respiration; converting the acoustic organ sound received simultaneously during imaging into digital data by means of a device or different methods (by using an intermediate software), and, by the same method, associating the organ sound with the position of the tumor/organ detected by the imaging method in the relevant time period, determining the changes in the tumor/organ motion limits and positions with the organ sound based on the changes in the sound frequency range in the different repeated respiratory patterns, and creating a reference correlation model specific to the patient (organ sound-turn or/organ position), and tracking the tumor/organ by associating the reference model obtained in the imaging stage of the treatment with the real time sound data.

With the invention, it is aimed to;

• Convert organ sound into a signal and create a correlation model between the organ sound and tumor/organ motion,

• Model especially the tumor/organ motion caused by respiration by associating the organ sound with the tumor/organ motion and dynamically track the tumor/organ,

• In addition, allow plan adaptation in early period by associating tumor response or side effect profile with the organ sound during the treatment and predict treatment efficacy and side effect profiles in the early/late period after the treatment.

The advantages and solutions to the technical problems in other approaches achieved with the invention are as follows;

Problems and solutions in relation to the gating and dynamic tumor tracking methods: i) There is no need for placing fiducial or gold markers into the tumor, thereby, the risk of pneumothorax due to the procedure is eliminated, ii) The uncertainties due to the displacement of markers during treatment process are also eliminated, iii) There is no need for extra radiological imaging during radiotherapy for tumor/organ motion tracking. Thereby, the imaging related radiation dose is reduced for X-raybased approaches, iv) Today, dynamic tumor tracking seems possible with MRI-linac. However, due to the high cost of the existing systems, it is available in a limited number of clinics, and it does not seem possible to integrate the MRI method into currently used modem devices and systems that have robotic gantry by means of current technology. In addition, it is currently not possible to treat patients who have MR incompatible pacemakers, cardiac defibrillators, heart valves, stents, cochlear implants, electronic drug pumps and prostheses with MRI-linac. The present invention, on the other hand, can be integrated into all modern linear accelerators in which photon and electron beams are used; C-arm conventional linear accelerators, systems with ring gantry and robotic gantry, and proton and carbon therapy devices in which particle beams such as proton/carbon are used, by only using an intermediate software/module, and can be easily utilized for patient groups who cannot be treated with MRI-linac.

Problems and solutions in relation to the deep breath holding and active breath control methods: i) Generally, the said approaches can be applied to patients in good condition with good respiratory functions. For this reason, they appeal to a limited patient group. Since patient's respiratory control is not intervened with the present invention, it can be employed in patient groups where deep breath holding and active breath control methods cannot be used. ii) Since the invention offers solutions to all aforementioned problems and involves a technological structure that can be used in integration with the existing modern systems, it appeals to a wide range of patients and user groups in radiotherapy applications.

BRIEF DESCRIPTION OF THE FIGURES

In order to understand the embodiment of the invention together with the additional elements in the best possible manner, it shall be considered in relation to the figure which shows the implementation of the method in a case of lung cancer.

Figure 1 Organ sound guided radiotherapy (OS-GRT) approach in a case of lung cancer.

REFERENCE NUMERALS

1. Lung volume and tumor position during inhalation

2. Lung volume and tumor position during exhalation 3. At least one device/system (sound sensor, stethoscope, microphone, etc.) used for detecting organ sound

4. Detection of the signal change in lung sound within the respiratory cycle on the basis of time

5. Conversion of lung sound signal into a respiratory graph

6. Tumor

DETAILED DESCRIPTION OF THE INVENTION

In this detailed description, the organ sound guided radiotherapy (OS-GRT) according to the invention is explained in detail for a better understanding in a non-limiting manner.

The invention associates the sound coming from an organ with tumor/organ motion. Particularly, it allows the association of respiratory sounds from lungs with a tumor/organ.

The present invention has two essential components. These are;

- a sound detector (device or sound sensor) with specific features that enable the detection of organ sound (such as distinguishing different organ sounds perceived at the same time, filtering the ambient sound, and instant transmission of the sound data),

- a data processing module with a computer-based method that allows real-time tumor/organ motion to be associated with the organ sound detected during a dynamic imaging method. The method used is supported by an artificial intelligence-based mathematical modeling (using parameters such as organ sound frequency, intensity, and length) and enables that the data regarding the patient's clinic (shrinkage/growth of the tumor, change in the atelectatic region, normal tissue/organ function changes/side effects) in the early period based on the change in organ sound.

The present method (software) improves the efficiency of the treatment management by allowing treatment plan adaptation in the early period on the basis of anatomical changes (such as shrinkage/growth of the tumor, change in the atelectatic region, changes in the surrounding healthy organ functions due to treatment) with the support of an artificial intelligence-based mathematical modeling (using parameters such as organ sound frequency, intensity, length), and obtaining information in the early period after the treatment about tumor control and/or normal tissue complication probability depending on the change in organ sound.

The present invention is a new method that enables real-time tumor motion tracking with organ sound. Otherwise, if the radiotherapy device or linear accelerator is not capable of dynamic tracking, the tumor position can be detected in the desired phase of respiration with the present invention, as indicated in the gating technique.

With the present invention, sounds coming from the organ, especially from the lung, are determined and these sounds/signals are associated with the motion of the organ. Among internal organ motions, the management of the respiratory motion is still one of the most challenging issue in radiotherapy, especially for patient with a tumor located in thorax and upper abdomen. However, it is a known fact that respiratory sound is directly related with the lung motion and the respiratory sound can be easily detected with the current technology. This provides a great advantage to employ this technique for tracking lung tumors or tumors which have a similar motion patterns with respiratory motion.

A correlation model between organ sound, especially lung sound, and the tumor motion detected under the guidance of organ sound may be created by tracking the organ sound of the tumor motion in a continuous or a determined phase range (usually in the phase range in which tumor spends more time during respiration, or in a specific setting where the patients hold their breath). The relevant phase interval may vary from patient to patient. For example, for a patient in good condition, irradiation can be performed by holding a deep breath at certain intervals in this deep breath-hold zone, or irradiation can be performed in the last region of the exhalation phase in a patient whose general condition is not well. Depending on the localization of the tumor (such as at the apex of the lung or close to the diaphragm), the method to be used may also vary. Therefore, there is no standard respiratory phase interval. The optimum phase interval is defined in a patient-specific manner depending on the patient's respiratory pattern and function during imaging for radiotherapy planning. Thus, tumor/organ motion can be modeled with organ sound without the need for any imaging method during treatment. Today, lung cancer patients are among the patient groups in which tumor/organ motion causes considerable uncertainty during radiotherapy. The lung sound due to respiration can be converted into instantaneous signals by using specific devices and methods such as sound sensors or electronic stethoscope, and the respiratory pattern of the related person can be easily characterized. In the present approach, organ sound, especially lung sound, obtained using a method/system (such as a sound sensor, electronic stethoscope, or high sensitivity microphone) where organ sound can be converted into a signal, is associated with the respiratory phases for the organ/tumor motion by using an intermediate software and the tumor position is detected instantaneously in the relevant respiratory phase. Moreover, it is possible to predict the next positions of the tumor in the standard respiratory pattern. In addition to the tumor/organ position tracking, the sound differences coming from the relevant organs are analyzed before, during, and after the treatment to evaluate the organ function and tumor response and/or side effect pattern.

At the stage of obtaining organ sound, the sound detector is not necessarily a stethoscope. It is possible to detect organ sound with sound sensors or a microphone system to a certain extent. The main difference in the computer-based method is that it allows the detection of tumor position/coordinates with an organ sound, and also, with an artificial intelligence-based module, of clinical findings such as tumor response, healthy organ function change/side effect pattern in the normal tissue on the basis of organ sound changes during and after treatment in the early time period.

The conversion of sound to digital data can be accomplished using an open source software for electronic stethoscopes / a module comprising a computer-based method according to the invention) and a simple decibel meter application . By using an artificial intelligence-based module, an association can be established between treatment-related changes and the clinic of the disease according to changes in organ sound.

For the quality control of the invention, sinusoidal respiratory files with different patterns and a moving phantom (simulation platform for virtual lung motion) where such respiratory motions are simulated are used. This phantom is a standard quality control phantom used by almost all clinics (different brands and models are available). The phantom can move dynamically according to the desired breathing pattern, and thanks to the sound data obtained during this dynamic motion (here, the motor sound that moves the mechanism was taken as a reference), the respiratory phase of the phantom motion can be associated simultaneously with the changes in the sound frequency, by which means the position of the phantom can be detected. While the frequency range of the sound differs between individuals, it may also differ according to the breathing pattern of the same person (such as free breathing, deep breathing). For example, in shallow respiration, when the sound frequency value detected in any phase of respiration is in the range of 100-200 Hz, the tumor moves 1 cm in the upward direction; however, when the patient breathes deeply, the frequency value will reach 750 Hz and above at the same respiratory phase and in this case, the displacement of the tumor will be 2 cm. For these two conditions, the position of the tumor/organ in the related phase can be determined according to the patient's respiratory pattern. This process can be performed not only for two values, but also by using respiratory motions repeated several times during imaging and utilizing different sensors to be placed on different parts of the body. Thereby, data is obtained for the same phase in various scenarios or at different frequency values, and this data is associated with the tumor position during imaging.

Detected organ sound (not limited to milliseconds (ms) and can be in different units) can be converted to digital data and this data can also be associated with the patient's respiratory chart; therefore, the approach allows real-time tumor tracking.

In the light of the information above, the invention is a tumor motion detection and tracking system by means of organ sound, characterized in that it comprises;

• At least one sound detector (device or sound sensor) that allows the detection of organ sound (such as respiratory sound, heart sound),

• At least one artificial intelligence based data processing module involving a computer- based method that allows associating real-time tumor/organ motion with the organ sound detected via a dynamic imaging method.

There are preferably 2 sound detectors.

The said sound detector may be a sound device, a sound sensor, a microphone system, or a stethoscope.

The said sound detector may further be a sound detector capable of distinguishing different organ sounds detected simultaneously, filtering the ambient sound, and transmitting the sound data instantly. The invention is a computer-based method for associating organ sound with tumor motion, characterized in that it comprises the process steps of;

• Detecting organ sound by using at least one sound detector,

• Implementing a dynamic imaging method that allows the detection of tumor/organ motion patterns in different phases of the sound (respiratory sound, etc.) and converting the detected organ sound into digital data,

• Associating the converted sound data with real-time tumor/organ motion detected during imaging,

• Creating an organ sound and tumor/organ motion correlation model (organ sound- tumor/organ position) according to the changes in the frequency range of the sound in different sound (respiratory, etc.) patterns.

Another preferred embodiment of the invention is a device which associates organ sound with tumor motion by employing the aforementioned method.

By using the correlation model obtained with this method in the said data processing module, the motion and/or position of the target organ may be detected on the screen during radiotherapy without using dynamic imaging on the basis of the organ sound received from a patient.

The dynamic imaging method is preferably radiological imaging methods such as 4- dimensional computed tomography, fluoroscopy, dynamic MRI, or ultrasonography.

The said sound pattern is preferably a breathing or heartbeat pattern.

The sound mentioned in the second step of the method is preferably respiratory sound or heart sound.

Claims

1. A system for associating organ sound with tumor motion, characterized in that it comprises;

• At least one sound detector that allows the detection of the organ sound,

• At least one data processing module configured to associate real-time tumor/organ motion with the organ sound detected by using a dynamic imaging method.

2. A system according to Claim 1, characterized in that there are 2 sound detectors.

3. A system according to Claim 1, characterized in that the said sound detector is a sound device, a sound sensor, a microphone system, or a stethoscope.

4. A system according to Claim 1, characterized in that the said sound detector is a sound detector capable of distinguishing different organ sounds detected simultaneously, filtering the ambient sound, and transmitting the sound data instantly.

5. A computer-based method for associating organ sound with tumor motion, characterized in that it comprises the process steps of;

• Detecting organ sound by using at least one sound detector,

• Obtaining an image by means of a dynamic imaging system that allows the detection of tumor/organ motion patterns in different phases of the sound and converting the detected organ sound into digital data,

• Creating an organ sound and tumor/organ motion correlation model according to the changes in the frequency range of the sound in different sound patterns.

6. A system according to Claim 5, characterized in that the said dynamic imaging method is 4-dimensional computed tomography, fluoroscopy, dynamic MRI, or ultrasonography.

7. A system according to Claim 5, characterized in that the sound pattern is a respiratory pattern or heartbeat pattern.

8. A system according to Claim 5, characterized in that the sound mentioned in the second step of the method is respiratory sound or heartbeat sound.

9. A device for associating organ sound with tumor motion, characterized in that it uses the method according to Claim 5.