Systems and methods for medical image analysis
A surgical planning and assessment system is disclosed. The system may include a computing system having a processor, a data store, a patient specific planning and analysis module, and a display. The system may be configured to access a database storing a plurality of possible surgical plans. The computing system may store a target surgical plan including a plurality of patient specific inputs including at least one preoperative medical image of a spine of a target patient and analyze the target surgical plan to determine a predicted alignment of the spine of the target patient. The computing system may develop a plurality of predictive models including a predicted alignment of the spine of the target patient based on the target surgical plan and suggest at least one alternative surgical plan with respect to the target surgical plan.
1. A computer-implemented method for generating a predictive model of at least one alternative surgical plan, the method comprising:
providing a computing system comprising a processor, a data store, a display, and a patient specific planning and analysis module, the computing system being configured to access a database storing a plurality of surgical plans;
inputting a target surgical plan into a data store accessible to the computing system, the target surgical plan being based on a medical assessment of a target patient;
analyzing at least one preoperative medical image of a spine of the target patient;
developing a predictive model, using the computing system, the predictive model being based on at least one input associated with the at least one preoperative medical image;
suggesting, at least one alternative surgical plan with respect to the target surgical plan, the at least one alternative surgical plan being based on a comparative analysis of the plurality of surgical plans, the at least one preoperative medical image, and the predictive model; and
displaying the at least one alternative surgical plan on the display.
2. The computer-implemented method of claim 1 , comprising:
generating a plurality of alternative surgical plans including the at least one alternative surgical plan,
wherein each of the plurality of alternative surgical plans is ranked according to a probability of a target outcome,
wherein the at least one alternative surgical plan comprises one or more specifications of a spinal rod to be implanted to the spine of the target patient, and
wherein the at least one predictive model comprises at least one of: a generative adversarial network (GAN) algorithm, convolutional neural network (CNN) algorithm, and/or a recurrent neural network (RNN) algorithm.
3. The computer-implemented method according to claim 1 , wherein the one or more preoperative inputs comprise at least one input chosen from: lumbar lordosis (LL), preoperative thoracic kyphosis (TK), pelvic incidence (PI), pelvic tilt (PT), and sagittal vertical axis (SVA) with respect to one or more vertebrae of the target patient.
4. The computer-implemented method according to claim 3 , wherein the suggesting step is further based on at least one patient parameter, the patient parameter comprising at least one parameter chosen from: age, gender, height, weight, and body mass index (BMI).
5. The computer-implemented method according to claim 4 , comprising:
developing, iteratively, a plurality of predictive models including the predictive model, the plurality of predictive models being based on the at least one input associated with the at least one preoperative medical image and the at least one patient parameter.
6. The computer-implemented method according to claim 5 , wherein the plurality of predictive models are developed based on a machine learning analysis comparison of the plurality of surgical plans stored in the database and the target surgical plan.
7. The computer-implemented method of claim 6 , wherein the at least one alternative surgical plan is based on the plurality of predictive models and wherein the at least one alternative surgical plan comprises automatically and/or dynamically defining a suggested position of one or more implants within the spine of the target patient.
8. The computer-implemented method of claim 7 , wherein the at least one alternative surgical plan comprises determining one or more surgical gestures to be performed by a surgeon and determining one or more compensatory mechanism simulations that are specific to the target patient.
9. The computer-implemented method of claim 8 , wherein the at least one alternative surgical plan comprises developing a guide spine in an angular total and tracing at least one end point of the spine.
10. The computer-implemented method of claim 9 , wherein the at least one alternative surgical plan comprises proposing a final position of an implant.
11. A surgical planning and assessment system, the system comprising:
a computing system including a processor, a data store, a patient specific planning and analysis module, and a display; and
a database configured to store a plurality of possible surgical plans, the database being accessible by the computing system;
wherein the computing system comprises computer executable code that when executed by the processor is configured to:
store a target surgical plan in the data store, the target surgical plan comprising a plurality of patient specific inputs including at least one preoperative medical image of a spine of a target patient obtained from a medical assessment of the target patient and at least one gesture to be performed by a surgeon;
analyze the target surgical plan to determine a predicted alignment of the spine of the target patient;
develop a first predictive model of a first predicted alignment of the spine of the target patient based on the target surgical plan;
suggest at least one alternative surgical plan with respect to the target surgical plan, the at least one alternative surgical plan being based on a comparative analysis of the plurality of surgical plans to the target surgical plan and the first predictive model;
display the at least one alternative surgical plan on the display; and
develop a second predictive model of a second predicted alignment of the spine of the target patient based on the target surgical plan and the at least one alternative surgical plan.
12. The surgical planning and assessment system of claim 11 , wherein the at least one preoperative medical images of the spine of the target patient comprise at least one sagittal x-ray image and at least one frontal x-ray image.
13. The surgical planning and assessment system of claim 12 , wherein the plurality of patient specific inputs comprise at least one input chosen from: lumbar lordosis (LL), preoperative thoracic kyphosis (TK), pelvic incidence (PI), pelvic tilt (PT), and sagittal vertical axis (SVA) with respect to one or more vertebrae of the target patient.
14. The surgical planning and assessment system of claim 13 , wherein the plurality of patient specific inputs comprise at least one parameter chosen from: age, gender, height, weight, and body mass index (BMI).
15. The surgical planning and assessment system of claim 11 , wherein the at least one alternative surgical plan comprises automatically and/or dynamically defining a suggested position of one or more implants within the spine of the target patient.
16. The surgical planning and assessment system of claim 15 , wherein the at least one alternative surgical plan comprises determining one or more surgical gestures to be performed by a surgeon.
17. The surgical planning and assessment system of claim 16 , wherein the at least one alternative surgical plan comprises determining one or more compensatory mechanism simulations that are specific to the target patient.
18. The surgical planning and assessment system of claim 17 , wherein the computing system is further configured to:
continuously and iteratively suggest a plurality of alternative surgical plans including the at least one alternative surgical plan; and
continuously and iteratively develop a plurality of predictive models including the first predictive model and the second predictive model, the plurality of predictive models;
each of the predictive models comprising a unique predicted alignment of the spine of the target patient.
19. The surgical planning and assessment system of claim 18 , wherein the plurality of predictive models are developed based on a machine learning analysis comparison of the plurality of surgical plans stored in the database and the target surgical plan.
20. The surgical planning and assessment system of claim 19 , wherein at least one of the plurality of predictive models is developed by at least one of: a generative adversarial network (GAN) algorithm, convolutional neural network (CNN) algorithm, and/or a recurrent neural network (RNN) algorithm.