Method of characterization and differentiation of tissue

• US 8,142,368 B2
• Filed: 10/02/2008
• Issued: 03/27/2012
• Est. Priority Date: 10/02/2008
• Status: Active Grant

• Current Assignee: Artann Laboratories Inc.
• Original Assignee: Artann Laboratories Inc.
• Inventors: Egorov, Vladimir, Sarvazyan, Armen P.
• Primary Examiner(s): Hindenburg, Max
• Assistant Examiner(s): Dougherty, Sean
• Application Number:
  US12/244,235
• Publication Number:
  US 20100087756A1
• Time in Patent Office: 1,272 Days
• Field of Search: 600/306, 600/587, 606/201, 702/21, 702/33, 702/36, 702/42
• US Class Current: 600/587
• CPC Class Codes: A61B 5/0053 by applying pressure, e.g. compression, indentation, palpation, grasping, gauging
• CPC Class Codes: A61B 5/4312 Breast evaluation or disorder diagnosis A61B5/0091 takes precedence
• CPC Class Codes: A61B 5/4381 Prostate evaluation or disorder diagnosis
• CPC Class Codes: A61B 5/442 Evaluating skin mechanical properties, e.g. elasticity, hardness, texture, wrinkle assessment
• CPC Class Codes: A61B 5/444 Evaluating skin marks, e.g. mole, nevi, tumour, scar
• CPC Class Codes: A61B 5/7264 Classification of physiological signals or data, e.g. using neural networks, statistical classifiers, expert systems or fuzzy systems neural networks per se G06N3/00; expert systems per se G06N5/00
• CPC Class Codes: A61B 5/7267 involving training the classification device
• CPC Class Codes: G06F 18/24155 Bayesian classification
• CPC Class Codes: G16H 50/20 for computer-aided diagnosis, e.g. based on medical expert systems

Abstract

A novel method for soft tissue characterization includes obtaining a sequence of surface stress patterns as a function of an increasing compression force when a probe is pressed against the tissue over the location of the lesion of interest. A number of elasticity features are then calculated to characterize the tissue and the lesion located therein including strain hardening, loading curve average slope, lesion peak signal under a predetermined load, tissue heterogeneity, lesion shape and lesion mobility. At least three elasticity features are provided as an input to a statistical Bayesian classifier trained on a clinical database to calculate the probability of the lesion being benign or malignant. Additional patient-related parameters may be further provided as inputs to the classifier to increase the accuracy of differentiation between benign and malignant lesions. These parameters include a family history of cancer disease, a patient-inherited genetic factor, a history of said tissue related diseases, patient'"'"'s age, patient'"'"'s weight, and patient'"'"'s lifestyle and dietary factors. The method of the invention along with other non-invasive examinations of lesions may help in reducing the rate of biopsies, specifically breast tissue biopsies.

Claims

1. A method of soft tissue characterization and soft tissue differentiation comprising the steps of:

(a) detecting a lesion in said tissue;

(b) applying an increasing compression force over said lesion and collecting a sequence of stress patterns as a function of said increasing compression force;

(c) calculating from said sequence of stress patterns at least three elasticity features characterizing said lesion, said elasticity features selected from a group consisting of strain hardening, loading curve average slope, lesion peak signal under a predetermined load, tissue heterogeneity, lesion shape and lesion mobility;

(d) inputting said at least three calculated elasticity features into a statistical classifier to calculate probabilities of possible diagnostic outcomes, said statistical classifier created in advance using a dataset of previously collected samples with known clinical status, said samples having been collected from several patients and including data on elasticity features that are the same as said at least three calculated elasticity features; and

(e) selecting a diagnostic outcome with prevailing calculated probability for providing a diagnosis of said soft tissue.

2. The method as in claim 1, wherein step (b) further includes providing a pressure probe equipped with a pressure sensor array adapted to be placed against and compress said tissue, step (b) further including acquiring stress patterns from the surface of said tissue in real-time.

3. The method as in claim 2, wherein said step (a) further includes acquiring stress imaging data in real-time by moving said pressure probe along said tissue.

4. The method as in claim 1, wherein said tissue is selected from a group consisting of thyroid tissue, breast tissue, prostate tissue, pelvic floor tissue, brain tissue, lung tissue, liver tissue, kidney tissue, and muscle tissue.

5. The method as in claim 1, wherein said lesion peak signal under a predetermined load is calculated in step (c) while values of a maximum pressure versus said compression force against said tissue are considered as a loading curve and approximated by a second order polynomial to characterize strain hardening of said tissue.

6. The method as in claim 1, wherein said step (d) further includes said statistical classifier calculating a probability for said lesion to be either benign or malignant.

7. The method as in claim 1, wherein said step (d) further includes at least one patient-related parameter as an additional input to said statistical classifier.

8. The method as in claim 1, wherein said steps (a) through (e) are performed during surgery, said surgery providing direct access to said tissue.

9. The method as in claim 1, wherein said statistical classifier is a Bayesian classifier.

10. The method as in claim 9, wherein said statistical classifier is a naive Bayesian classifier.

23. The method of claim 7, wherein said at least one patient-related parameter is selected from a group consisting of a family history of cancer disease, a patient-inherited genetic factor, a history of said tissue related diseases, patient'"'"'s age, patient'"'"'s weight, and patient'"'"'s lifestyle and dietary factors.

11. A method of soft tissue characterization and soft tissue differentiation comprising the steps of:

(a) detecting a lesion and a lesion location in said tissue;

(b) applying an increasing compression force over said lesion location and collecting a sequence of stress patterns, said sequence of stress patterns being a function of said increasing compression force;

(c) calculating from said sequence at least three elasticity features characterizing said lesion;

(d) generating a probability distribution for possible diagnostic outcomes for said lesion by inputting said at least three elasticity features into a statistical classifier created using a clinical database of cases from several patients (i) having known diagnoses and (ii) including data on elasticity features that are the same as said at least three calculated elasticity features; and

(e) providing a diagnosis of said soft tissue based on the diagnostic outcome with prevailing calculated probability.

12. The method as in claim 11, wherein step (b) further includes providing a pressure probe equipped with a pressure sensor array adapted to be placed against and compress said tissue, step (b) further including acquiring stress patterns from the surface of said tissue in real-time.

13. The method as in claim 12, wherein said step (a) further includes acquiring stress imaging data in real-time by moving said pressure probe along said tissue.

14. The method as in claim 11, wherein said tissue is selected from a group consisting of thyroid tissue, breast tissue, prostate tissue, pelvic floor tissue, brain tissue, lung tissue, liver tissue, kidney tissue, and muscle tissue.

15. The method as in claim 11, wherein said lesion peak signal under a predetermined load is calculated in step (c) while values of a maximum pressure versus said compression force against said tissue are considered as a loading curve and approximated by a second order polynomial.

16. The method as in claim 11, wherein said step (d) further includes said statistical classifier calculating a probability for said lesion to be either benign or malignant.

17. The method as in claim 11, wherein said step (d) further includes inputting at least one patient-related parameter into said statistical classifier in order to evaluate said probability distribution for possible diagnostic outcomes for said lesion, said statistical classifier being further created using a clinical database of cases (iii) including data on a parameter that is the same as the at least one patient-related parameter.

18. The method as in claim 11, wherein said steps (a) through (e) are performed during surgery, said surgery providing direct access to said tissue.

19. The method as in claim 11, wherein said statistical classifier is a Bayesian classifier.

20. The method as in claim 19, wherein said statistical classifier is a naive Bayesian classifier.

21. The method as in claim 11, wherein said at least three elasticity features are selected from a group consisting of strain hardening, loading curve average slope, lesion peak signal under a predetermined load, tissue heterogeneity, lesion shape and lesion mobility.

22. The method of claim 17, wherein said at least one patient-related parameter is selected from a group consisting of a family history of cancer disease, a patient-inherited genetic factor, a history of said tissue related diseases, patient'"'"'s age, patient'"'"'s weight, and patient'"'"'s lifestyle and dietary factors.

24. A method of soft tissue characterization and soft tissue differentiation comprising the steps of:

(a) detecting a lesion in said tissue;

(b) compressing said tissue over said lesion with an increasing compression force and collecting a sequence of stress patterns as a function of said increasing compression force;

(c) calculating from said sequence of stress patterns at least three elasticity features characterizing said lesion;

(d) generating a probability distribution for possible diagnostic outcomes for said lesion by inputting said at least three elasticity features into a statistical classifier created using a clinical database of cases from several patients (i) having known diagnoses and (ii) including data on elasticity features that are the same as said at least three calculated elasticity features; and

(e) providing a diagnosis of said soft tissue based on said probability distribution.