Patent Application
20230042032
The present disclosure relates to energy-based surgical procedures and, more particularly, to energy-based surgical systems and methods based on an artificial-intelligence learning system.
Surgical instruments are utilized to perform various treatments on tissue structures. A surgical forceps, for example, is a plier-like device which relies on mechanical action between its jaws to grasp, clamp, and constrict tissue. Energy-based surgical forceps utilize both mechanical clamping action and energy to treat tissue, such as coagulate, cauterize, and/or seal tissue.
While surgical instruments such as energy-based surgical forceps are effective at treating tissue, outcomes of using such instruments may depend on the experience of the clinician using the instruments. For example, vessel sealing is accomplished by subjecting a vessel to a specific energy profile under a specific pressure, and experience of a clinician may affect the outcome. Accordingly, there is continuing interest in improving energy-based surgical systems and methods.
The present disclosure relates to energy-based surgical systems and methods that use an artificial-intelligence learning system. Although portions of the present disclosure discuss particular types of energy-based surgical systems, aspects of the present disclosure are applicable to other types of energy-based surgical systems not expressly described herein. As used herein, the term “distal” refers to the portion that is being described which is further from a user, while the term “proximal” refers to the portion that is being described which is closer to a user. Further, to the extent consistent, any of the aspects or embodiments described herein may be used in conjunction with any or all of the other aspects or embodiments described herein.
In accordance with aspects of the disclosure, a computer implemented method for an energy-based surgical procedure includes accessing an image of tissue of a patient, accessing control parameter values of a generator configured to provide energy based on control parameters, processing the image of the tissue and the control parameter values by an artificial-intelligence learning system to provide an output relating to configuration of the control parameters, providing an indication to a clinician based on the output where the indication indicates whether to maintain the control parameter values, and providing adjusted control parameter values for the generator based on the output of the artificial-intelligence learning system, if the indication indicates not to maintain the control parameter values.
In various embodiments of the method, the output of the artificial-intelligence learning system relates to a predicted outcome of applying the energy based on the control parameter values to the tissue.
In various embodiments of the method, the computer implemented method includes providing the energy for the clinician to apply to the tissue, applying the energy to the tissue using an energy-based surgical instrument, and receiving information from the clinician on an outcome of applying the energy to the tissue.
In various embodiments of the method, the information from the clinician indicates one of: a successful outcome of applying the energy to the tissue or an unsuccessful outcome of applying the energy to the tissue. In various embodiments of the method, the information from the clinician includes at least one of an audio recording or a user-interface selection.
In various embodiments of the method, the image is an image of the tissue captured before the clinician applied the energy to the tissue, and the computer implemented method includes processing the information from the clinician to provide a tag for training the artificial-intelligence learning system, storing training data including the control parameter values and the image of the tissue captured before the clinician applied the energy to the tissue, and training the artificial-intelligence learning system based on the training data and the tag.
In various embodiments of the method, the computer implemented method includes accessing patient information, wherein the artificial-intelligence learning system is configured to provide the output relating to configuration of the control parameters based further on the patient information. In various embodiments of the method, the patient information includes at least one of patient age, moisture of the tissue, hydration of the tissue, or location of the tissue.
In various embodiments of the method, the artificial-intelligence learning system includes a convolutional neural network that processes the image of the tissue.
In various embodiments of the method, providing the adjusted control parameter values for the generator includes automatically adjusting the control parameters, and providing an indication to the clinician that the control parameters have been automatically adjusted.
In accordance with aspects of the present disclosure, an energy-based surgical system includes an image capturing device configured to capture an image of tissue of a patient, and a generator configured to provide energy based on control parameters. The generator is configured to execute instructions to perform a method including accessing the image of the tissue of the patient, accessing control parameter values of the control parameters, processing the image of the tissue and the control parameter values by an artificial-intelligence learning system to provide an output relating to configuration of the control parameters, providing an indication to a clinician based on the output where the indication indicates whether to maintain the control parameter values, and providing adjusted control parameter values for the generator based on the output of the artificial-intelligence learning system, if the indication indicates not to maintain the control parameter values.
In various embodiments of the system, the output of the artificial-intelligence learning system relates to a predicted outcome of applying the energy based on the control parameter values to the tissue.
In various embodiments of the system, the energy-based surgical system includes at least one of a user-interface or a voice recorder, and an energy-based surgical instrument. The generator provides the energy to the energy-based surgical instrument for the clinician to apply to the tissue, and the user-interface and/or the voice recorder receives information from the clinician on an outcome of applying the energy to the tissue.
In various embodiments of the system, the information from the clinician indicates one of: a successful outcome of applying the energy to the tissue or an unsuccessful outcome of applying the energy to the tissue. In various embodiments of the system, the information from the clinician includes at least one of audio recorded by the voice recorder or a selection using the user-interface.
In various embodiments of the system, the image is an image of the tissue captured before the clinician applied the energy to the tissue, and the method performed by executing the instructions in the generator includes processing the information from the clinician to provide a tag for training the artificial-intelligence learning system, storing training data including the control parameter values and the image of the tissue captured before the clinician applied the energy to the tissue, and associating and storing the training data with the tag.
In various embodiments of the system, the energy-based surgical system includes a storage the includes patient information, where the artificial-intelligence learning system is configured to provide the output relating to configuration of the control parameters based further on the patient information. In various embodiments of the system, the patient information includes at least one of patient age, moisture of the tissue, hydration of the tissue, or location of the tissue.
In various embodiments of the system, the artificial-intelligence learning system includes a convolutional neural network that processes the image of the tissue.
In various embodiments of the system, the generator, in providing the adjusted control parameter values, executes the instructions to automatically adjust the control parameters, and provide an indication to the clinician that the control parameters have been automatically adjusted.
Various aspects and features of the disclosure are described herein with reference to the drawings wherein:
The present disclosure relates to energy-based surgical systems and methods that use an artificial-intelligence learning system. Although portions of the present disclosure discuss particular types of energy-based surgical systems, aspects of the present disclosure are applicable to other types of energy-based surgical systems not expressly described herein.
As one example of an energy-based surgical procedure, tissue sealing involves heating tissue to liquefy the collagen and elastin in the tissue so that it reforms into a fused mass with significantly-reduced demarcation between the opposing tissue structures. To achieve a tissue seal without causing unwanted damage to tissue at the surgical site or collateral damage to adjacent tissue, the application of energy to tissue is controlled to control the temperature of tissue during the sealing process. To properly seal tissue, a balance should be sustained during the sealing process between sufficient heating to denature proteins and vaporize fluids and poor seal performance. In various situations, implementing such a balance to achieve a proper tissue seal may be based on clinician experience with the energy-based equipment and/or with different conditions of the patient tissue. In accordance with aspects of the present disclosure, and as described below, learning the parameters and outcomes of such clinician experience may be helpful in implementing the proper balance for achieving a successful outcome.
As detailed below, and in accordance with aspects of the present disclosure, the present disclosure involves processing data on configuration of an energy-based surgical system and processing tissue condition information, using an artificial intelligence learning system. Aspects of the present disclosure may also use clinician feedback on the application of a particular energy-based surgical system configuration to a particular tissue to train an artificial intelligence learning system.
The terms “artificial intelligence,” “data models,” or “system learning” may include, but are not limited to, neural networks, recurrent neural networks (RNN), generative adversarial networks (GAN), Bayesian Regression, Naive Bayes, nearest neighbors, least squares, means, and support vector regression, among other data science and artificial science techniques.
The term “application” may include a computer program designed to perform particular functions, tasks, or activities for the benefit of a user. Application may refer to, for example, software running locally or remotely, as a standalone program or in a web browser, or other software which would be understood by one skilled in the art to be an application. An application may run on the controller 500 or on a user device, including for example, on a mobile device, an IoT device, or a server system.
Referring now to
With reference to
The images 902 that are input to the artificial intelligence learning system 908 may show, for example, tissue bleeding and/or tissue charring, among other things. The artificial intelligence learning system 908 may be configured to identify such characteristics in the image 902. With reference to
Referring to
In various embodiments, as mentioned above, the image 902, the generator control parameter values 904, and the patient parameters 905, form training data. In various embodiments, the tag 906 relates to the training data 902-905 in a specific way. In particular, with respect to a particular application of energy to a patient based on generator control parameter values, the training data 902-905 is captured before such energy is applied to the patient, and the tag 906 is captured after such energy is applied to the patient. Thus, the training data 902-905 reflects a scenario that a clinician may face before energy is applied to a patient, and the tag 906 reflects whether the outcome of such energy application to the patient was a successful outcome or an unsuccessful outcome. In various embodiments, the clinician may “tag” whether the energy application was successful by using, for example, buttons on the instrument or on the generator or by speech. As another example, the training data may include a tag indicating seal failure weeks later in the patient after they leave hospital. Training the artificial-intelligence learning system in the manner described above results in identifying correlations between the training data 902-905 and the successful and unsuccessful outcomes of the energy-based surgical procedure indicated by the tag 906, and thereby allows the artificial intelligence learning system to learn from clinician experience.
In various embodiments, the “tagging” may be accomplished through audio capture and voice recognition. For example, the instrument or the generator may include a microphone to capture the clinician's speech indicating the outcome of the energy application. Examples of such tagging is shown in Table 1 provided below. In various embodiments, the system can display in text on the generator the words that were recorded from the clinician and that were recognized by speech recognition. In various embodiments, the generator may include indicator lights, which reflect the outcome of the energy application, as indicated by the clinician speech. In various embodiments, the tag 906 that is derived based on the voice recognition may be used as training data for supervised learning of the artificial intelligence learning system 908. It is contemplated that the training may be performed on a separate system, for example, GPU servers, simulation, etc., and the trained network would then be deployed in the energy-based surgical system. In various embodiments, the instrument or generator may include a button to cancel the previous input to correct an incorrect “tag.”
Thus, as mentioned above, the artificial-intelligence learning machine 908 determines correlations between the input data 902-905 and the successful and unsuccessful outcomes of energy-based surgical applications, and thereby learns from clinician experience through the training process. In applying a trained artificial intelligence learning system 908, the controller 500 outputs, from the artificial-intelligence learning system, an indication of whether to maintain a generator's current control parameter values based on tissue image 902, the current generator control values 904, and/or patient parameters 905. In various embodiments, the indicator may be implemented as a visual indication, such as an LED light or a display screen. In various embodiments, the generator 160 provides the visual indication. In various embodiments, the energy-based surgical instrument provides the visual indication. In various embodiments, the image data may be captured before, during, and/or after energy delivery.
Referring now to
Endoscopic forceps 100 defines a longitudinal axis “A-A” and includes a housing 120, a handle assembly 130, a rotating assembly 170, a trigger assembly 180, and an end effector assembly 10. Forceps 100 further includes a shaft 112 having a distal end 114 configured to mechanically engage end effector assembly 10 and a proximal end 116 that mechanically engages housing 120. Forceps 100 may further include a surgical cable extending therefrom and configured to connect forceps 100 to an generator 160 such that at least one of the electrically-conductive tissue treating surfaces 13, 14 of jaw members 11, 12 of end effector assembly 10 may be energized to treat tissue grasped therebetween, e.g., upon activation of activation switch 190.
With continued reference to
End effector assembly 10 is shown attached at distal end 114 of shaft 112 and includes a pair of opposing jaw members 11 and 12. Each of jaw members 11 and 12 includes an electrically-conductive tissue treating surface 13, 14, respectively, configured to grasp tissue therebetween and conduct energy therethrough to treat, e.g., seal, tissue. End effector assembly 10 is designed as a unilateral assembly, i.e., where jaw member 12 is fixed relative to shaft 112 and jaw member 11 is movable relative to shaft 112 and fixed jaw member 12. However, end effector assembly 10 may alternatively be configured as a bilateral assembly, i.e., where both jaw member 11 and jaw member 12 are movable relative to one another and to shaft 112. In some embodiments, a knife assembly (not shown) is disposed within shaft 112, and a knife channel (not shown) is defined within one or both jaw members 11, 12 to permit reciprocation of a knife blade (not shown) therethrough, e.g., upon activation of trigger 182 of trigger assembly 180.
Continuing with reference to
Referring now to
The generator 160 may be any suitable type of generator and may include a plurality of connectors to accommodate various types of energy-based surgical instruments (e.g., monopolar energy-based surgical instrument and bipolar electrosurgical instrument). The generator 160 may also be configured to operate in a variety of modes, such as ablation, cutting, coagulation, and sealing. The generator 160 may include a switching mechanism (e.g., relays) to switch the supply of RF energy among the connector ports 169 to which various energy-based surgical instruments may be connected. For example, when an energy-based surgical instrument, e.g., forceps 100 (
In various embodiments, the generator 160 may include a sensor module 166, which includes a plurality of sensors, e.g., an RF current sensor and an RF voltage sensor. Various components of the generator 160, namely, the RF output stage 162 and the RF current and voltage sensors of sensor module 166 may be disposed on a printed circuit board (PCB). The RF current sensor of sensor module 166 may be coupled to the active terminal and provides measurements of the RF current supplied by the RF output stage 162. In embodiments, the RF current sensor of sensor module 166 may be coupled to the return terminal. The RF voltage sensor of sensor module 166 is coupled to the active and return terminals and provides measurements of the RF voltage supplied by the RF output stage 162. In embodiments, the RF current and voltage sensors of sensor module 166 may be coupled to active and return leads and, which interconnect the active and return terminals and to the RF output stage 162, respectively.
The sensed voltage and current from sensor module 166 are fed to analog-to-digital converters (ADCs) 168. The ADCs 168 sample the sensed voltage and current to obtain digital samples of the voltage and current of the RF output stage 162. The digital samples are processed by the controller 500 and used to generate a control signal to control the DC/AC inverter of the RF output stage 162 and the preamplifier. The ADCs 168 communicate the digital samples to the controller 500 for further processing.
In various embodiments, the controller 500 may collect data relating to the generator 160, including time, power, and/or impedance. For example, the energy-based surgical system may include a generator 160 and an energy-based surgical instrument such as detailed above with respect to
In various embodiments, the controller 500 uses the stored settings and the indication as training data for an artificial-intelligence learning system. In various embodiments, training and machine learning may be performed by a computing device outside of the generator 160, and the result of the machine learning may be communicated to the controller 500 of generator 160. In various embodiments, the controller 500 communicates the determined generator control parameter configuration that was output from the machine learning algorithm to a computing device, e.g., of controller 500. In various embodiments, the artificial-intelligence learning system may use supervised learning, unsupervised learning, or reinforcement learning. In various embodiments, the neural network may include a temporal convolutional network, a fully connected network, or a feed forward network.
Referring to
In various embodiments, the memory 530 can be random access memory, read only memory, magnetic disk memory, solid state memory, optical disc memory, and/or another type of memory. In various embodiments, the memory 530 can be separate from the controller 500 and can communicate with the processor 520 through communication buses of a circuit board and/or through communication cables such as serial ATA cables or other types of cables. The memory 530 includes computer-readable instructions that are executable by the processor 520 to operate the controller 500. In various embodiments, the controller 500 may include a network interface 540 to communicate with other computers or a server. In embodiments, a storage device 510 may be used for storing data. In various embodiments, the controller 500 may include one or more FPGAs 550. The FPGA 550 may be used for executing various machine learning algorithms such as those provided in accordance with the disclosure, as detailed below.
The memory 530 stores suitable instructions, to be executed by the processor 520, for receiving the sensed data, e.g., sensed data from sensor module 166 via ADCs 168 (see
Referring now to
Initially, at step 802, the controller 500 may access an image of a tissue of a patient captured during an energy-based surgical procedure. In various embodiments, the image may be a video or a still image.
At step 804, the controller 500 accesses control parameter values of a generator 160, which provides energy. In various embodiments, the parameters may include time, slope, power, and/or impedance. In various embodiments, the controller 500 can also access patient parameters, which can be determined as described above in connection with
At step 806, controller 500 processes the image and the control parameters, and/or the patient parameters, by an artificial-intelligence learning system to provide an output relating to whether or not to maintain the current generator control parameter configuration. In various embodiments, the control parameters may include time, slope, power, and/or impedance.
At step 808, the controller 500 adjusts, automatically by the generator 160, the control parameters to provide adjusted energy based on the output of the artificial-intelligence learning system. In various embodiments, the controller 500 may deliver energy, by an energy-based surgical instrument configured to deliver energy to tissue using the adjusted energy. In various embodiments, the controller 500 may store settings of the energy-based surgical instrument based on the adjusted energy, in a memory of the generator 160. Thus, with the artificial intelligence learning system having been trained as detailed above, the system can determine whether to maintain a generator control parameter configuration for generating energy to treat the patient.
Referring now to
In various embodiments, the forceps 200 is a monopolar instrument. Forceps 200 is configured for use with an end effector assembly 20 that is similar to end effector assembly 10 of forceps 100 (see
A ratchet 230 may be included for selectively locking jaw members 21 and 22 of forceps 200 relative to one another at various different positions. It is envisioned that ratchet 230 may include graduations or other visual markings that enable the user to easily and quickly ascertain and control the amount of closure force desired between the jaw members 21 and 22.
With continued reference to
Similar to forceps 100 (
Referring now to
At least one additional or alternative microwave antenna probe 1112′ may be included with microwave ablation system 1100 that may have characteristics distinct from that of microwave antenna probe 1112. For example, without limitation, microwave antenna probe 1112 may be a 12-gauge probe suitable for use with energy of about 915 MHz, while microwave antenna probe 1112′ may be a 14-gauge probe suitable for use with energy of about 915 MHz. Other probe variations are contemplated within the scope of the present disclosure, for example, without limitation, a 12-gauge operable at 2450 MHz, and a 14 gauge operable at 2450 MHz. In use, the surgeon may interact with user interface 1205 of generator 1200 to preview operational characteristics of available probes 1112, 1112′, and to choose a probe for use.
Generator 1200 includes a generator module 1286 that is configured as a source of microwave energy and is disposed in operable communication with processor 1282. In embodiments, generator module 1286 is configured to provide energy of about 915 MHz. Generator module 1286 may also be configured to provide energy of about 2450 MHz (2.45 GHz.). The present disclosure contemplates embodiments wherein generator module 1286 is configure to generate a frequency other than about 915 MHz or about 2450 MHz, and embodiments wherein generator module 1286 is configured to generate variable frequency energy. Probe 1112 is operably coupled to an energy output of generator module 1286.
Generator assembly 1200 also includes user interface 1205, that may include a display 1210 such as, without limitation, a flat panel graphic LCD display, adapted to visually display at least one user interface element 1230, 1240. In embodiments, display 1210 includes touchscreen capability (not explicitly shown), e.g., the ability to receive input from an object in physical contact with the display, such as without limitation a stylus or a user's fingertip, as will be familiar to the skilled practitioner. A user interface element 1230, 1240 may have a corresponding active region, such that, by touching the screen within the active region associated with the user interface element, an input associated with the user interface element is received by the user interface 1205.
User interface 1205 may additionally or alternatively include one or more controls 1220, that may include without limitation a switch (e.g., pushbutton switch, toggle switch, slide switch) and/or a continuous actuator (e.g., rotary or linear potentiometer, rotary or linear encoder.) In embodiments, a control 1220 has a dedicated function, e.g., display contrast, power on/off, and the like. Control 1220 may also have a function which may vary in accordance with an operational mode of the ablation system 1100. A user interface element 1230 may be positioned substantially adjacently to control 1220 to indicate the function thereof. Control 1220 may also include an indicator, such as an illuminated indicator (e.g., a single- or variably-colored LED indicator).
It is contemplated that although
From the foregoing and with reference to the various figure drawings, those skilled in the art will appreciate that certain modifications can also be made to the disclosure without departing from the scope of the same. While several embodiments of the disclosure have been shown in the drawings, it is not intended that the disclosure be limited thereto, as it is intended that the disclosure be as broad in scope as the art will allow and that the specification be read likewise. Therefore, the above description should not be construed as limiting, but merely as exemplifications of particular embodiments. Those skilled in the art will envision other modifications within the scope and spirit of the claims appended hereto.
The present application is a U.S. National Stage Application filed under 35 U.S.C. § 371(a) claiming the benefit of and priority to International Patent Application No. PCT/US2020/063897, filed Dec. 9, 2020, which claims the benefit of and priority to U.S. Provisional Patent Application Ser. No. 62/952,803, filed Dec. 23, 2019, the entire disclosures of each of which being incorporated by reference herein.
| Filing Document | Filing Date | Country | Kind |
|---|---|---|---|
| PCT/US2020/063897 | 12/9/2020 | WO |
| Number | Date | Country | |
|---|---|---|---|
| 62952803 | Dec 2019 | US |
