Method for screening and treating patients at risk of medical disorders

• US 7,194,301 B2
• Filed: 09/30/2004
• Issued: 03/20/2007
• Est. Priority Date: 10/06/2003
• Status: Expired due to Fees

• Current Assignee: Transneuronix Inc. (Medtronic Plc)
• Original Assignee: Transneuronic Incorporated
• Inventors: Jenkins, David A., Maude-Griffin, Roland
• Primary Examiner(s): Pezzuto; Robert
• Assistant Examiner(s): LEE, YUN HAENG NMN
• Application Number:
  US10/955,591
• Publication Number:
  US 20050080462A1
• Time in Patent Office: 901 Days
• Field of Search: 607/2
• US Class Current: 607/2
• CPC Class Codes: G16H 20/10 relating to drugs or medications, e.g. for ensuring correct administration to patients
• CPC Class Codes: G16H 20/60 relating to nutrition control, e.g. diets
• CPC Class Codes: G16H 20/70 relating to mental therapies, e.g. psychological therapy or autogenous training
• CPC Class Codes: G16H 50/50 for simulation or modelling of medical disorders
• CPC Class Codes: G16H 50/70 for mining of medical data, e.g. analysing previous cases of other patients

Abstract

Method for screening patients to predict which patients at risk of a medical disorder, such as morbid obesity, gastrointestinal problems, or gastroesophageal problems, will be responders, and conversely, which patients will not, to achieve a favorable outcome from therapy for that disorder. This method supports an intervention strategy for patients having weight or gastrointestinal problems that will cut health costs. It enables patients and care-givers alike to more efficiently use their time, efforts and resources by enabling an early selection of an appropriate treatment modality for a given patient. Its application also extends to other implantable medical devices and therapies using them.

Claims

1. A method for predicting weight loss outcome of application of a proposed therapy to a patient for treating an eating, gastrointestinal, or gastroesophageal disorder, comprising:

a) obtaining items of information from the patient at risk of the eating, gastrointestinal, or gastroesophageal disorder, each item of information relating to a preselected patient variable; and

b) predicting the weight loss outcome for the patient from the obtained items of information for the patient using an aggregated weight loss predictor developed from i) observed similar types of information and corresponding weight loss information obtained from an actual population of patients who previously received a similar therapy to the proposed therapy for the patient, or ii) information generated from a simulated population of patients by resampling the observed actual population information to produce pseudo-replicates;

wherein predicting the weight loss outcome comprises processing the items of information using an aggregated classification and repression tree model formed using a committee or ensemble method combining multiple predictors trained in perturbed versions of the observed similar types of information and corresponding weight loss information obtained from the actual population of patients.

2. The method of claim 1, where the proposed therapy is applied through the use of an implantable pulse generator.

3. The method of claim 1, where the proposed therapy is application of a banding type device.

4. The method of claim 1, where the proposed therapy is a bariatric type surgery.

5. The method of claim 1, where the proposed therapy is a non-surgical behavioral modification.

6. The method of claim 1, where the proposed therapy is a pharmaceutical or medicinal therapy.

7. The method of claim 1, wherein the committee or ensemble method is selected from the group consisting of a boosting algorithm, a bagging algorithm, and an arcing algorithm.

8. The method of claim 1, further comprising training the aggregated classification and regression tree model by preprocessing i) historical data comprising actual patient information of patients who previously received similar therapy to the proposed therapy for the patient, and ii) the corresponding weight loss outcomes of the patients who previously received similar therapy to learn how to predict weight loss outcomes.

9. The method of claim 8, wherein the preprocessing comprises:

reducing quantity of the historical data and corresponding patient weight loss outcomes of patients who previously received similar therapy;

reducing number of variables contained in the historical data and corresponding patient weight loss outcomes of patients who previously received similar therapy;

transforming the historical data and corresponding patient weight loss outcomes of patients who previously received similar therapy using classification and regression trees;

applying the boosting algorithm or the bagging algorithm; and

generating the classification and regression tree model to predict the weight loss outcome.

10. The method of claim 8, further comprising testing the trained aggregated classification and regression tree model through a cross-validating procedure that comprises the steps of:

1) randomly dividing the historical data and the corresponding weight loss outcomes of the patients who previously received similar therapy into a training set and a hold out test set;

2) repeatedly training the model using perturbed versions of the randomly selected training set by the committee or ensemble method selected from the group consisting of the boosting algorithm and the bagging algorithm to obtain resulting predictors;

3) applying each of the resulting predictors to the hold-out test set, and aggregating predictions by averaging and majority voting for use in screening test set subjects to determine their suitability for treatment;

4) repeating steps

     1),

     2) and

     3) at least once; and

5) averaging at least one outcome statistic of interest for screening test set subjects, and

6) using results of step

     5) to assess usefulness of the model for improving expected treatment outcomes.

11. The method of claim 8, wherein the aggregated classification and regression tree model includes a combination of many distinct trees, each estimated in a sequence of systematically or randomly perturbed version of the historical data and the corresponding weight loss outcomes of the patients who previously received similar therapy, and wherein a screening decision for the patient is then based on a combination of average predicted weight loss across distinct trees or a majority vote criterion comprising whether a majority of the distinct trees predict weight loss above a predetermined threshold level.

12. The method of claim 1, wherein the items of information include psychometric data.

13. The method of claim 12, wherein the psychometric data include answers by the patient to questions asked in a RAND Short Form 36 (SF-36) health survey.

14. The method of claim 1, wherein the items of information are selected from at least one of symptoms, demographics, tests of psychological well being, family history, eating habits, diet, exercise, and other attempted weight loss therapies.

15. The method of claim 1, wherein the items of information comprises anthropometric data.

16. The method of claim 1, wherein the items of information comprises biomarker data.

17. The method of claim 1, wherein the items of information includes body mass index.

18. The method of claim 1, wherein the items of information includes RAND scores calculated for answers made by the patient to questions asked in a RAND Short Form 36 (SF-36) health survey (version 1.0), and anthropometric data of the patient including at least one of the patient'"'"'s height, weight, sex, and age.

19. The method of claim 1, wherein the predicting comprises comparing the items of information to similar data gathered from an actual or simulated patient population for which results to an obesity therapy have been precollected or precalculated, respectively, effective to permit a classification of the patient in terms of a probable outcome to the proposed therapy, and wherein the proposed therapy is the obesity therapy.

20. The method of claim 19, wherein the processing and the comparing are at least partially performed by a computer.

21. The method of claim 19, further comprising rendering a decision as to whether or not to apply the proposed therapy based on the classification of the patient.

22. The method of claim 1, wherein the disorder is obesity, metabolic syndrome, or their related comorbidities.

23. The method of claim 22, wherein the disorder comprises at least one of the related comorbidities selected from the group consisting of high blood pressure, hypertension, high blood cholesterol, dyslipidemia, Type 2 diabetes, insulin resistance, glucose intolerance, hyperinsulinemia, coronary heart disease, angina pectoris, congestive heart failure, stroke, gallstones, cholesystitis, cholelithiasis, gastroeophageal reflux disease, gout, osteoarthritis, obstructive sleep apnea, sleep apnea complications related to pregnancy, cancer, menstrual irregularities, infertility, irregular ovulation, stress incontinence, uric acid nephrolithiasis, depression, eating disorders, distorted body image, and low self esteem.

24. The method of claim 1, wherein steps a) and b) are performed before, during, or after an animal clinical trial to demonstrate safety or efficacy of a medicinal therapy or medical device.

25. The method of claim 1, wherein the items of information comprise biomarker information selected from the group consisting of hormone information, peptide information, genomics information, and body scan information.

26. The method of claim 1, wherein the items of information comprise biomarker information selected from the group consisting of ghrelin peptide information, blood glucose measurements, and hemoglobin AIC information.

27. A method of screening or optimization of a prospective patient in advance of a medical treatment, comprising:

a) developing a predictive model for outcomes of patients who will receive the medical treatment as part of a therapy or treatment of a medical condition or disorder, using data mining techniques, and advanced regression tree analysis to build, train and test the predictive model for predicting the outcomes of patients receiving the medical treatment;

b) collecting data from the prospective patient for the medical treatment;

c) using the predictive model to predict the possible outcome of the prospective patient for the medical treatment.

28. A method of claim 27, where such screening or optimization uses scoring from a lifestyle, quality of life, mental health or mental well-being being questionnaire taken by the prospective patient.

29. The method of claim 27, wherein the medical treatment comprises the implantation of a medical device in the prospective patient.

30. The method of claim 27, wherein the medical treatment comprises applying a pharmaceutical regimen to the prospective patient.

31. The method of claim 27, wherein the medical treatment comprises applying a medicinal regimen to the prospective patient.

32. The method of claim 27, wherein steps a), b), and c) are performed before, during, or after an animal clinical trial to demonstrate safety or efficacy of a medicinal therapy or medical device.

33. A method of screening or optimization of a prospective patient in advance of the application of an obesity therapy, comprising:

a) developing a predictive model for outcomes of patients who will have the obesity therapy as part of a treatment for obesity and comorbidities thereof, using data mining techniques, and advanced regression tree analysis to build, train and test the predictive model for predicting the outcomes of patients receiving the obesity therapy;

b) collecting data from the prospective patient for the obesity therapy;

c) using the predictive model to predict the possible outcome of the prospective patient for the obesity therapy.

34. A method of claim 33, wherein the screening or optimization uses scoring from a lifestyle, quality of life, mental health or mental well-being questionnaire taken by the prospective patient.

35. The method of claim 34, wherein the questionnaire contains between about 20 and about 60 individual indications for response.

36. The method of claim 35, wherein the questionnaire is the RAND SF-36 form.

37. The method of claim 33, wherein the obesity therapy comprises a surgical procedure.

38. The method of claim 33, wherein the obesity therapy comprises installing an implantable gastric stimulator.

39. The method of claim 33, wherein the obesity therapy comprises bariatric surgery.

40. The method of claim 39, wherein the obesity therapy further comprises treatment of at least one of the comorbidities selected from the group consisting of high blood pressure, hypertension, high blood cholesterol, dyslipidemia, Type 2 diabetes, insulin resistance, glucose intolerance, hyperinsulinemia, coronary heart disease, angina pectoris, congestive heart failure, stroke, gallstones, cholesystitis, cholelithiasis, gastroeophageal reflux disease, gout, osteoarthritis, obstructive sleep apnea, sleep apnea complications related to pregnancy, cancer, menstrual irregularities, infertility, irregular ovulation, stress incontinence, uric acid nephrolithiasis, depression, eating disorders, distorted body image, and low self esteem.

41. The method of claim 33, wherein the obesity therapy comprises a pharmaceutical regimen.

42. The method of claim 33, wherein the obesity therapy comprises a medicinal regimen.

43. The method of claim 33, wherein steps a), b), and c) are performed before, during, or after an animal clinical trial to demonstrate safety or efficacy of a surgical procedure, medicinal therapy, or medical device.