System and method for performing probabilistic classification and decision support using multidimensional medical image databases
View Patent ↗A system and method for providing decision support to a physician during a medical examination is disclosed. Data is received from a sensor representing a particular medical measurement. The received data includes image data. The received data and context data is analyzed with respect to one or more sets of training models. Probability values for the particular medical measurement and other measurements to be taken are derived based on the analysis and based on identified classes. The received image data is compared with training images. Distance values are determined between the received image data and the training images, and the training images are associated with the identified classes. Absolute value feature sensitivity scores are derived for the particular medical measurement and other measurements to be taken based on the analysis. The probability values, distance values and absolute value feature sensitivity scores are outputted to the user.
1. A method for assigning probability classification values to a set of identified classes based on a set of measurements taken during a medical procedure of a patient in order to provide decision support for rendering a medical diagnosis, the method comprising:
sensing one or more medical measurements using a sensor and receiving data from the sensor representing the one or more medical measurements;
analyzing the received data by applying decision rules and training models derived from knowledgebase data and prior physician input and using induction algorithms to learn probabilistic models;
calculating probability values for the identified classes based on the analysis;
determining sensitivity values for the one or more medical measurements based on the analysis; and
rendering a medical diagnosis for the patient using the probability values for each identified class and the sensitivity measurements for the one or more measurements.
2. The method of claim 1 wherein the identified classes include classifications for one or more diseases.
3. The method of claim 2 wherein the one or more disease is Dilated Cardiomyopathy (DCM).
4. The method of claim 1 wherein the identified classes include a non-diseased classification.
5. The method of claim 4 wherein the non-diseased classification is non-Dilated Cardiomyopathy (non-DCM).
6. The method of claim 1 wherein the medical sensor is an ultrasound transducer.
7. The method of claim 1 wherein the medical procedure is an echocardiogram examination.
8. The method of claim 1 wherein the identified classes are linearly separable classes.
9. The method of claim 8 wherein the induction algorithms use non-parametric discriminant analysis.
10. The method of claim 9 wherein generative modeling is used in reduced discriminative space to obtain likelihood maps for each class.
11. The method of claim 9 wherein generative modeling is used in reduced discriminative space to obtain likelihood maps for each class.
12. The method of claim 1 wherein the identified classes have nonlinear class boundaries.
13. The method of claim 12 wherein the induction algorithms use kernel discriminant analysis.
14. The method of claim 13 wherein a Support Vector Model (SVM) is used to perform the kernel discriminant analysis.
15. The method of claim 1 wherein the probability values are provided in real time.
16. A method for assigning probability classification values to a set of identified classes based on a set of measurements taken during a medical procedure of a patient in order to provide decision support for rendering a medical diagnosis, the method comprising:
sensing one or more medical measurements using a sensor and receiving data from the sensor representing the one or more medical measurements;
analyzing the received data by applying decision rules and training models derived from knowledgebase data and prior physician input;
calculating probability values for the identified classes based on the analysis;
determining sensitivity values for the one or more medical measurements based on the analysis; and
rendering a medical diagnosis for the patient using the probability values for each identified class and the sensitivity measurements for the one or more measurements,
wherein the identified classes include classifications for one or more diseases, and the one or more diseases include Dilated Cardiomyopathy (DCM).
17. A method for assigning probability classification values to a set of identified classes based on a set of measurements taken during a medical procedure of a patient in order to provide decision support for rendering a medical diagnosis, the method comprising:
sensing one or more medical measurements using a sensor and receiving data from the sensor representing the one or more medical measurements;
analyzing the received data by applying decision rules and training models derived from knowledgebase data and prior physician input;
calculating probability values for the identified classes based on the analysis;
determining sensitivity values for the one or more medical measurements based on the analysis; and
rendering a medical diagnosis for the patient using the probability values for each identified class and the sensitivity measurements for the one or more measurements,
wherein the identified classes include a non-diseased classification, and wherein the non-diseased classification includes non-Dilated Cardiomyopathy (non-DCM).
18. A method for assigning probability classification values to a set of identified classes based on a set of measurements taken during a medical procedure of a patient in order to provide decision support for rendering a medical diagnosis, the method comprising:
sensing one or more medical measurements using a medical sensor and receiving data from the medical sensor representing the one or more medical measurements, wherein the medical sensor is an ultrasound transducer;
analyzing the received data by applying decision rules and training models derived from knowledgebase data and prior physician input;
calculating probability values for the identified classes based on the analysis;
determining sensitivity values for the one or more medical measurements based on the analysis; and
rendering a medical diagnosis for the patient using the probability values for each identified class and the sensitivity measurements for the one or more measurements.
19. A method for assigning probability classification values to a set of identified classes based on a set of measurements taken during a medical procedure of a patient in order to provide decision support for rendering a medical diagnosis, the method comprising:
sensing one or more medical measurements using a sensor and receiving data from the sensor representing the one or more medical measurements;
analyzing the received data by applying decision rules and training models derived from knowledgebase data and prior physician input;
calculating probability values for the identified classes based on the analysis;
determining sensitivity values for the one or more medical measurements based on the analysis; and
rendering a medical diagnosis for the patient using the probability values for each identified class and the sensitivity measurements for the one or more measurements, wherein the medical procedure is an echocardiogram examination.