Methods for improved surgical planning using machine learning and devices thereof

Patent metadata

Field	Value
Publication number	US-12343085-B2
Application number	US-202017427771-A
Country	US
Kind code	B2
Filing date	Feb 4, 2020
Priority date	Feb 5, 2019
Publication date	Jul 1, 2025
Grant date	Jul 1, 2025

Abstract

Methods, non-transitory computer readable media, and surgical computing devices are illustrated that improve surgical planning using machine learning. With this technology, a machine learning model is trained based on historical case log data sets associated with patients that have undergone a surgical procedure. The machine learning model is applied to current patient data for a current patient to generate a predictor equation. The current patient data comprises anatomy data for an anatomy of the current patient. The predictor equation is optimized to generate a size, position, and orientation of an implant, and resections required to achieve the position and orientation of the implant with respect to the anatomy of the current patient, as part of a surgical plan for the current patient. The machine learning model is updated based on the current patient data and current outcome.

First claim

We claim:

1. A method for improved surgical planning, the method comprising:
training at least one neural network based on historical case log data sets comprising historical outcome data correlated with one or more of historical patient data, historical implant data, or historical healthcare professional data associated with a plurality of instances of a surgical procedure;
applying, by a processor, the at least one neural network to current patient data for a current patient to generate a predictor equation, wherein the current patient data comprises at least anatomy data for an anatomy of the current patient, and wherein the predictor equation functionally relates a size, position, and orientation of an implant to an estimated response for the anatomy of the current patient;
optimizing, by the processor, the predictor equation to generate the size, position, and orientation of the implant, and one or more resection parameters for one or more resections required to achieve the position and orientation of the implant with respect to the anatomy of the current patient as part of a surgical plan for the current patient undergoing the surgical procedure;
controlling, by the processor, actuation of a surgical tool to implement a resection of the surgical procedure according to the surgical plan, based on the size, position, and orientation of the implant; and
updating the at least one neural network based on the current patient data and current outcome data generated for the current patient following execution of the surgical procedure according to the surgical plan,
wherein the surgical procedure is an orthopedic procedure.

2. The method of claim 1 , wherein the at least one neural network comprises a plurality of input nodes and downstream nodes coupled by connections having associated weighting values.

3. The method of claim 2 , wherein each of the weighting values comprises a predictor equation coefficient.

4. The method of claim 2 , further comprising:
obtaining a sensitivity threshold value; and
applying the sensitivity threshold value to disregard one or more of the input nodes.

5. The method of claim 2 , further comprising providing input data comprising signals that correspond with the input nodes to the neural network as seeding data, wherein the input data is extracted from the historical case log data sets.

6. The method of claim 5 , further comprising altering the weighting values until the neural network is configured to provide a result that corresponds with the historical outcome data.

7. A surgical computing device comprising memory comprising programmed instructions stored thereon for improved surgical planning and one or more processors coupled to the memory and configured to execute the stored programmed instructions to:
train at least one neural network based on historical case log data sets comprising historical outcome data correlated with historical procedure data;
apply the at least one neural network to current patient data for a current patient to generate a predictor equation, wherein the current patient data comprises at least anatomy data for an anatomy of the current patient, and wherein the predictor equation functionally relates one or more parameters of an implant to an estimated response for the anatomy of the current patient;
optimize the predictor equation to generate the one or more parameters of the implant and one or more resection parameters for one or more resections required to achieve a position or orientation of the implant with respect to the anatomy of the current patient as part of a surgical plan for the current patient undergoing a surgical procedure; and
control actuation of a surgical tool to implement a resection of the surgical procedure according to the surgical plan, based on the one or more parameters of the implant,
wherein the surgical procedure is an orthopedic procedure.

8. The surgical computing device of claim 7 , wherein the historical procedure data comprises one or more of historical patient data, historical implant data, or historical healthcare professional data associated with a plurality of instances of the surgical procedure.

9. The surgical computing device of claim 7 , wherein the one or more processors are further configured to execute the stored programmed instructions to update the neural network based on the current patient data and current outcome data generated for the current patient following execution of the surgical procedure according to the surgical plan.

10. The surgical computing device of claim 7 , wherein the one or more processors are further configured to execute the stored programmed instructions to use one or more of direct Monte Carlo sampling, stochastic tunneling, or parallel tempering to optimize the predictor equation.

11. The surgical computing device of claim 7 , wherein the one or more processors are further configured to execute the stored programmed instructions to:
generate the anatomy data pre-operatively from medical image data of the anatomy of the current patient;
determine an optimized resection envelope for the current patient based on a Boolean intersection of a geometry of the implant and the anatomy data; and
instruct a patient specific knee instrumentation (PSKI) system to remove the optimized resection envelope.

12. The surgical computing device of claim 7 , wherein the one or more processors are further configured to execute the stored programmed instructions to:
generate an intra-operative algorithm comprising a plurality of recommended actions associated with the surgical plan;
evaluate a result of an execution of one of the recommended actions; and
update one or more inputs to the intra-operative algorithm based on the evaluation to alter another one of the recommended actions to be executed subsequent to the one of the recommended actions.

13. The surgical computing device of claim 12 , wherein the one or more inputs are updated based on one or more deviations to the one of the recommended actions.

14. The surgical computing device of claim 7 , wherein the one or more parameters for the implant comprise one or more of a size, a position, or an orientation of the implant.

15. A non-transitory computer readable medium having stored thereon instructions for improved surgical planning using machine learning comprising executable code that, when executed by one or more processors, causes the one or more processors to
train a neural network based on an artificial neural network and historical case log data sets comprising historical outcome data correlated with one or more of historical patient data, historical implant data, or historical healthcare professional data associated with a plurality of instances of a surgical procedure, wherein the artificial neural network comprises a plurality of input nodes and downstream nodes coupled by connections having associated weighting values;
apply the neural network to current patient data for a current patient to generate a predictor equation, wherein the current patient data comprises at least anatomy data for an anatomy of the current patient, and wherein the predictor equation functionally relates a size, position, and orientation of an implant to an estimated response for the anatomy of the current patient;
optimize the predictor equation to generate one or more of the size, the position, or the orientation of the implant, and one or more resection parameters for one or more resections required to achieve the position and orientation of the implant with respect to the anatomy of the current patient as part of a surgical plan for the current pati

Assignees

• Smith & Nephew Inc
• Smith & Nephew Asia Pacific Pte Ltd
• Smith & Nephew Orthopaedics Ag

Inventors

• Mcguan Shawn P
• Laster Scott Kennedy
• Wilkinson Zachary C
• Janna Sied W
• Farley Daniel

Classifications

• G06N3/09

  Supervised learning · CPC title

• G06N3/0499

  Feedforward networks · CPC title

• G09B5/02

  with visual presentation of the material to be studied, e.g. using film strip · CPC title

• A61F2002/30952

  using CAD-CAM techniques or NC-techniques · CPC title

• G09B19/24

  Use of tools · CPC title