Surgical instruments and systems incorporating machine learning based tissue identification and methods thereof
A surgical system includes an end effector assembly having first and second jaw members, a generator, and one or more machine learning applications. The first and/or second jaw member is movable relative to the other from a spaced-apart position to an approximated position for grasping tissue therebetween. The jaw members are configured to conduct energy therebetween and through tissue grasped therebetween. The generator includes an energy output configured to supply energy to the jaw members, a main controller configured to control the energy output, and sensor circuitry configured to sense impedance and/or power. The machine learning application(s) is configured to determine a type of tissue grasped between the first and second jaw members based upon the impedance and/or power sensed by the sensor circuitry.
1. A surgical system, comprising:
a surgical instrument including:
an end effector assembly having first and second jaw members, at least one of the first or second jaw members movable relative to the other from a spaced-apart position to an approximated position for grasping tissue therebetween, each of the first and second jaw members including an electrically-conductive plate, the electrically conductive plates configured to conduct energy therebetween and through tissue grasped between the first and second jaw members; and
first and second electrical leads coupled to the electrically-conductive plates of the first and second jaw members, respectively;
a generator, comprising:
an energy output configured to supply energy to the electrically-conductive plates of the first and second jaw members via the first and second electrical leads;
a main controller configured to control the energy output; and
sensor circuitry electrically coupled to the first and second electrical leads and configured to sense power during the conduction of energy between the electrically-conductive plates and through tissue grasped between the first and second jaw members; and
at least one machine learning application configured to determine a type of tissue grasped between the first and second jaw members based on at least one of an anatomical location of the tissue to be treated, patient condition information, or data obtained from an optical tissue sensor,
wherein the at least one machine learning application receives, as an input, sensed power data from the sensor circuitry,
wherein the at least one machine learning application is trained using previous procedure data, and
wherein the tissue type includes at least one of a vascular or a non-vascular tissue.
2. The system according to claim 1 , wherein the at least one machine learning application is stored within the generator.
3. The system according to claim 1 , wherein the at least one machine learning application is stored remotely from the generator.
4. The system according to claim 1 , wherein the at least one machine learning application is configured to communicate the determined tissue type to the main controller.
5. The system according to claim 4 , wherein, in response to receiving the determined tissue type, the main controller is configured to determine whether the determined tissue type is acceptable for tissue treatment or not acceptable for tissue treatment.
6. The system according to claim 5 , wherein, if the determined tissue type is determined not to be acceptable for tissue treatment, the main controller is further configured to at least one of inhibit further supply of energy to the electrically-conductive plates or output a warning.
7. The system according to claim 5 , wherein, if the determined tissue type is determined to be acceptable for tissue treatment, the main controller is further configured to control the supply of energy to the electrically-conductive plates to complete tissue treatment.
8. The system according to claim 4 , wherein the main controller is configured to determine whether the determined tissue type is acceptable for tissue treatment or not acceptable for tissue treatment during an initial stage of tissue treatment prior to permanently damaging tissue.
9. The system according to claim 8 , wherein the main controller is configured to determine whether the determined tissue type is acceptable for tissue treatment or not acceptable for tissue treatment during an initial 250 ms of tissue treatment.
10. The system according to claim 1 , wherein the at least one machine learning application is further configured to determine a type of tissue grasped between the first and second jaw members based on information relating to manual overrides.
11. The system according to claim 1 , wherein the at least one machine learning application includes at least one of a Support Vector Machine (SVM), implements Principle Component Analysis, implements a Hidden Markov Model (HMM), or employs the Monte Carlo Method.
12. The system according to claim 1 , wherein the at least one machine learning application utilizes data representative of a snapshot in time of power during the conduction of energy between the electrically-conductive plates and through tissue grasped between the first and second jaw members to determine the tissue type.
13. The system according to claim 1 , wherein the at least one machine learning application utilizes data representative of power over an elapsed time during the conduction of energy between the electrically-conductive plates and through tissue grasped between the first and second jaw members to determine the tissue type.
14. A surgical system, comprising:
a surgical instrument including:
an end effector assembly having first and second jaw members, at least one of the first or second jaw members movable relative to the other from a spaced-apart position to an approximated position for grasping tissue therebetween, each of the first and second jaw members including an electrically-conductive plate, the electrically conductive plates configured to conduct energy therebetween and through tissue grasped between the first and second jaw members; and
first and second electrical leads coupled to the electrically-conductive plates of the first and second jaw members, respectively;
a generator, comprising:
an energy output configured to supply energy to the electrically-conductive plates of the first and second jaw members via the first and second electrical leads;
a main controller configured to control the energy output; and
sensor circuitry electrically coupled to the first and second electrical leads and configured to sense power during the conduction of energy between the electrically-conductive plates and through tissue grasped between the first and second jaw members; and
at least one machine learning application configured to determine a type of tissue grasped between the first and second jaw members, based on at least one of an anatomical location of the tissue to be treated, patient condition information, or data obtained from an optical tissue sensor, and communicate the determined tissue type to the main controller,
wherein the at least one machine learning application receives, as an input, sensed power data from the sensor circuitry,
wherein the main controller is configured to determine whether the determined tissue type is acceptable for tissue treatment or not acceptable for tissue treatment during an initial stage of tissue treatment prior to permanently damaging tissue, and
the main controller is configured to determine whether the determined tissue type is acceptable for tissue treatment or not acceptable for tissue treatment during an initial 250 ms of tissue treatment.