Method, System and Computer Program Product for Evaluating the Accuracy of Blood Glucose Monitoring Sensors/Devices

• US 20070232878A1
• Filed: 04/21/2005
• Published: 10/04/2007
• Est. Priority Date: 04/21/2004
• Status: Active Grant

• Current Assignee: University of Virginia
• Original Assignee: University of Virginia Patent Foundation (University of Virginia)
• Inventors: Gonder-Frederick, Linda, Cox, Daniel, Clarke, William, Kovatchev, Boris
• Granted Patent: US 7,815,569 B2
• Time in Patent Office: Days
• US Class Current: 600/365
• CPC Class Codes: A61B 2560/0223 of calibration, e.g. protocols for calibrating sensors
• CPC Class Codes: A61B 5/14532 for measuring glucose, e.g. by tissue impedance measurement

Abstract

Continuous Glucose Error-Grid Analysis (CG-EGA) method, system or computer program product designed for evaluation of continuous glucose sensors providing frequent BG readings. The CG-EGA estimates the precision of such sensors/devices in terms of both BG values and temporal characteristics of BG fluctuation. The CG-EPA may account for, among other things, specifics of process characterization (location, speed and direction), and for biological limitations of the observed processes (time lags associated with interstitial sensors).

Claims

1. A method for evaluating accuracy of a glucose monitoring sensor, said method comprising:

evaluating point accuracy of blood glucose (BG) values received from said monitoring sensor;

evaluating accuracy of direction and rate of BG fluctuations received from said monitoring sensor; and

combining said point accuracy and rate accuracy in an estimate of sensor precision.

2. The method of claim 1, wherein said estimate of sensor precision comprises:

performing a combination of three grids adapted specifically for hypoglycemic, euglycemic, and hyperglycemic BG ranges.

3. The method of claim 2, wherein said estimate of sensor precision includes solely a single estimate using a weighted combination of hypoglycemic, euglycemic, hyperglycemic BG ranges.

4. The method of claim 2, wherein said estimate of sensor precision includes a plurality of estimates using a weighted combination of hypoglycemic, euglycemic, hyperglycemic BG ranges.

5. The method of claim 1, wherein said method is for use with a subject, wherein said subject is a human or an animal.

6. The method of claim 1, wherein said sensor comprises a continuous glucose monitor or semi-continuous glucose monitor or combination of both.

7. The method of claim 1, wherein:

said accuracy evaluation of direction and rate of BG fluctuations includes performing rate error-grid analysis (R-EGA), and

said accuracy evaluation blood glucose (BG) values includes performing point error-grid analysis (P-EGA) that determines correct presentation of said blood glucose (BG) values.

8. The method of claim 7, wherein said estimate of sensor precision comprises:

performing a combination of three grids adapted specifically for hypoglycemic, euglycemic, and hyperglycemic BG ranges.

9. The method of claim 8, wherein said estimate of sensor precision includes solely a single estimate using a weighted combination of hypoglycemic, euglycemic, hyperglycemic BG ranges.

10. The method of claim 8, wherein said estimate of sensor precision includes a plurality of estimates using a weighted combination of hypoglycemic, euglycemic, hyperglycemic BG ranges.

11. The method of claim 7, wherein said method is for use with a subject, wherein said subject is a human or an animal.

12. The method of claim 7, wherein said sensor comprises a continuous glucose monitor or semi-continuous glucose monitor or combination of both.

13. The method of claim 7, wherein said rate error-grid analysis (R-EGA) includes:

for each pair of reference BG points, RBG(t₁) and RBG(t₂), received from a reference blood glucose monitoring device, taken at times t₁ and t₂, a reference BG rate is estimated as the ratio of the change in the BG points divided by the length of the elapsed time interval as follows;

reference BG rate of change (mg/dl/min)=(RBG(t₂)−

RBG(t₁))/(t₂−

t₁).

14. The method of claim 13, wherein said rate said error-grid analysis (R-EGA) further includes:

for each pair of sensor reported BG points, SBG(t₁) and SBG(t₂), received from said glucose monitoring sensor, taken at times t₁ and t₂, a sensor BG rate is estimated as;

sensor BG rate of change (mg/dl/min)=(SBG(t₂)−

SBG(t₁))/(t₂−

t₁).

15. The method of claim 14, wherein said rate said error-grid analysis (R-EGA) further includes:

scatter plotting said sensor BG rate of change against said reference BG rate of change.

16. The method of claim 15, wherein said rate said error-grid analysis (R-EGA) further includes dividing said scatter plot into accuracy zones.

17. The method of claim 16, wherein said accuracy zones comprise A through E accuracy zones, which have a clinical meaning similar to the clinical meaning of the conventional Error Grid Analysis (EGA).

18. The method of claim 17, wherein said clinical meanings are defined as follows:

(a) AR-zone (accurate);

(b) CR-zone (over-correction);

(c) DR-zone (failure to detect);

(d) ER-zone (erroneous reading); and

(e) BR-zone (benign errors),

wherein suffix R stands for rate.

19. The method of claim 18, wherein said rate error-grid analysis (R-EGA) further includes:

computing said R-EGA zones for each pair of said reference-sensor BG rates of change using the formulas below, wherein X is abbreviated for said reference BG rate of change and wherein Y is abbreviated for said sensor BG rate of change;

IF (Y ge X) RZONE=5,

IF (Y gt (X+1) and X ge 0-2 and X le

     1) RZONE=4,

IF (Y gt 0.5 X and X lt 0-2) RZONE=4,

IF (Y gt 2.0 X and X gt

     1) RZONE=4,

IF (Y gt (X+2) and X ge 0-1 and X lt

     1) RZONE=3,

IF (Y gt (X+2) and X ge 0-3 and X lt 0-1) RZONE=2,

IF (Y gt 0-1 and X lt 0-3) RZONE=2,

IF (Y gt 1 and X lt 0-1) RZONE=1,

IF (Y lt X) RZONE=6,

IF (Y lt (X−

1) and X ge 0-1 and X le

     2) RZONE=7,

IF (Y lt 2.0 X and X lt 0-1) RZONE=7,

IF (Y lt 0.5 X and X gt

     2) RZONE=7,

IF (Y it (X−

2) and X ge 0-1 and X le

     1) RZONE=8,

IF (Y lt (X−

2) and X gt 1 and X le

     3) RZONE=9,

IF (Y lt 1 and X gt

     3) RZONE=9, and

IF (Y lt 0-1 and X gt

     1) RZONE=10,

wherein the output of this computation is a zone, coded as follows;

1-uE 2-uD 3-uC 4-uB 5-uA 6-lA 7-lB 8-lC 9-lD 10-lE.

20. The method of claim 13, wherein said glucose monitoring sensor comprises at least one of Beckman glucose analyzes, glucose oxidase analyzer, YSI Instruments analyzer and finger stick glucose meters, or any combination thereof.

21. The method of claim 7, wherein said point error-grid analysis (P-EGA) includes:

scatter plotting said point error-grid analysis (P-EGA) for said sensor BG rate against a reference BG rate computed from the data of a reference monitoring device, wherein said scatter plot is divided into A_P, B_P, C_P, D_P, and E_P zones, wherein suffix “

P”

stands for point.

22. The method of claim 21, wherein

said B_P zone is divided into upper B_P zone and lower B_P zone;

said C_P zone is divided into upper C_P zone and lower C_P zone;

said D_P zone is divided into upper D_P zone and lower D_P zone; and

said E_P zone is divided into upper E_P zone and lower E_P zone.

23. The method of claim 22, wherein said point error-grid analysis (P-EGA) includes:

said zones are defined depending on the reference rate of BG change classified as follows;

(a) slow change range;

(b) moderate decrease range;

(c) moderate increase range;

(d) rapid decrease range; and

(e) rapid increase range.

24. The method of claim 23, wherein said classification of reference BG change comprises:

(a) said slow change range being about −

1 and about 1 mg/dl/min;

(b) moderate decrease range being about −

2 and about −

1 mg/dl/min;

(c) moderate increase range being about 1 and about 2 mg/dl/min;

(d) rapid decrease range being less than about −

2 mg/dl/min; and

(e) rapid increase range being greater than about 2 mg/dl/min.

25. The method of claim 24, wherein said point error-grid analysis (P-EGA) includes:

said zones are defined depending on the reference rate of BG change as follows;

(a) if the reference BG rate is within said slow change range, said P-EGA zones are identical to the zones of the conventional EGA;

(b) if the reference BG is falling at a rate within said moderate increase range, the upper limits of upper said A_P, B_P, and D_P zones are expanded by about 10 mg/dl,

and if the reference BG is falling faster than said rapid decrease range, the upper limits of upper A_P, B_P, and D_P zones are expanded by about 20 mg/dl; and

(c) if the reference BG is rising at a rate within said moderate increase, the lower limits of lower A_P, B_P, and D_P zones are expanded by about 10 mg/dl,

and if the reference BG is rising faster than said rapid increase range, the lower limits of lower said A_P, B_P, and D_P zones are expanded by about 20 mg/dl.

26. The method of claim 23, wherein said point error-grid analysis (P-EGA) further includes:

computing said P-EGA zones for each pair of said reference-sensor BG readings using the formulas below;

a) if the BG rate of change is in said slow change range, apply the formulas;

wherein U1 is abbreviated for said Reference BG,

wherein W1 is abbreviated for said Sensor BG,

IF (W1 ge U1) ZONE0=5,

IF (W1 ge 1.20 U1 and U1 gt

     70) ZONE0=4,

IF (W1 ge (1.03 U1+107.9) and U1 gt

     70) ZONE0=3,

IF (W1 ge 1.20 U1 and W1 gt 70 and U1 le

     70) ZONE0=2,

IF (W1 ge 180 and U1 le

     70) ZONE0=1,

IF (W1 lt U1) ZONE0=6,

IF (W1 le 0.8 U1 and U1 gt

     70) ZONE0=7,

IF (W1 le (1.4 U1-182) and U1 le

     180) ZONE0=8,

IF (U1 ge 240 and W1 gt 70 and W1 lt

     180) ZONE0=9, and

IF (U1 gt 180 and W1 le

     70) ZONE0=10;

b) if BG is falling at said moderate rate apply the formulas;

wherein U2 is abbreviated for said Reference BG,

wherein W2 is abbreviated for said Sensor BG,

IF (W2 ge U2) ZONE2=5,

IF (W2 ge (1.20 U2+10) and U2 gt

     70) ZONE2=4,

IF (W2 ge (1.03 U2+117.9) and U2 gt

     70) ZONE2=3,

IF (W2 ge (1.20 U2+10) and W2 gt 80 and U2 le

     70) ZONE2=2,

IF (W2 ge 190 and U2 le

     70) ZONE2=1,

IF (W2 lt U2) ZONE2=6,

IF (W2 le 0.8 U2 and U2 gt

     70) ZONE2=7,

IF (W2 le (1.4 U2-182) and U2 le

     180) ZONE2=8,

IF (U2 ge 240 and W2 gt 70 and W2 lt

     180) ZONE2=9, and

IF (U2 gt 180 and W2 le

     70) ZONE2=10;

c) if BG is falling at said rapid rate use the formulas;

wherein U4 is abbreviated for said Reference BG,

wherein W4 is abbreviated for said Sensor BG,

IF (W4 ge U4) ZONE4=5,

IF (W4 ge (1.20 U4+20) and U4 gt

     70) ZONE4=4,

IF (W4 ge (1.03 U4+127.9) and U4 gt

     70) ZONE4=3,

IF (W4 ge (1.20 U4+20) and W4 gt 90 and U4 le

     70) ZONE4=2,

IF (W4 ge 200 and U4 le

     70) ZONE4=1,

IF (W4 lt U4) ZONE4=6,

IF (W4 le 0.8 U4 and U4 gt

     70) ZONE4=7,

IF (W4 le (1.4 U4-182) and U4 le

     180) ZONE4=8,

IF (U4 ge 240 and W4 gt 70 and W4 lt

     180) ZONE4=9, and

IF (U4 gt 180 and W4 le

     70) ZONE4=10;

d) if BG is rising at said moderate rate use the formulas;

wherein U1 is abbreviated for said Reference BG,

wherein SENBG is abbreviated for said Sensor BG,

IF (W1 ge U1) ZONE1=5,

IF (W1 ge 1.20 U1 and U1 gt

     70) ZONE1=4,

IF (W1 ge (1.03 U1+107.9) and U1 gt

     70) ZONE1=3,

IF (W1 ge 1.20 U1 and W1 gt 70 and U1 le

     70) ZONE1=2,

IF (W1 ge 180 and U1 le

     70) ZONE1=1,

IF (W1 lt U1) ZONE1=6,

IF (W1 le (0.8 U1-10) and U1 gt

     70) ZONE1=7,

IF (W1 le (1.4 U1-192) and U1 le

     180) ZONE1=8,

IF (U1 ge 240 and W1 gt 60 and W1 lt

     170) ZONE1=9, and

IF (U1 gt 180 and W1 le

     60) ZONE1=10; and

e) if BG is rising at said rapid rate use the formulas;

wherein U3 is abbreviated for said Reference BG,

wherein W3 is abbreviated for said Sensor BG,

IF (W3 ge U3) ZONE3=5,

IF (W3 ge 1.20 U3 and U3 gt

     70) ZONE3=4,

IF (W3 ge (1.03 U3+107.9) and U3 gt

     70) ZONE3=3,

IF (W3 ge 1.20 U3 and W3 gt 70 and U3 le

     70) ZONE3=2,

IF (W3 ge 180 and U3 le

     70) ZONE3=1,

IF (W3 lt U3) ZONE3=6. U3 gt

     70) ZONE3=7,

IF (W3 le (1.4 U3-202) and U3 le

     180) ZONE3=8,

IF (U3 ge 240 and W3 gt 50 and W3 lt

     160) ZONE3=9,

IF (U3 gt 180 and W3 le

     50) ZONE3=10, and

IF (W3 le (0.8 U3-20),

wherein the output of this computation is a zone, coded as follows;

1-uE, 2-uD, 3-uC, 4-uB, 5-uA, 6-lA, 7-lB, 8-lC, 9-lD, 10-lE.

27. The method of claim 7, comprises:

splitting reference BG values received from a reference BG monitoring device into three clinically meaningful regions including hypoglycemia, euglycemia and hyperglycemia regions; and

combining results from the rate error-grid analysis (R-EGA) and point error-grid analysis (P-EGA) within each of these regions.

28. The method of claim 27, wherein:

said hypoglycemia region is defined as BG<

=70 mg/dl (3.9 mmol/l);

said euglycemia region is defined as 70 mg/dl (3.9 mmol/l)<

BG<

180 mg/dl (10 mmol/l), and

said hyperglycemia region defined as BG>

180 mg/dl (10 mmol/l).

29. The method of claim 28, further comprising:

as presented in table below;

providing a relationship of said R-EGA and P-EGA values of said sensor in a hypoglycemic range according to a table including rows of said R-EGA values and columns of P-EGA values.

30. The method of claim 29, wherein said R-EGA and P-EGA values will be considered accurate if a value falls in the A_P zone of the P-EGA column of the table and said same value falls in the A_R or B_R zones of the R-EGA rows of the table, as denoted by the white section of the table of claim 28.

31. The method of claim 29, wherein said R-EGA and P-EGA values will be considered clinically erroneous if a value falls in the

D_P zone or E_P zone of the P-EGA column of the table, or

A_P zone of the P-EGA column of the table and said same value falls in the uD or uE zones of the R-EGA rows of the table, as denoted by the shaded section of the table in claim 28.

32. The method of claim 29, wherein said R-EGA and P-EGA values will be considered benign errors if a value falls A_P zone of the P-EGA column of the table and said same value falls in the uC, lC, lD, or lE zones of the R-EGA rows of the table, as denoted by the horizontal hash section of the table in claim 28.

33. The method of claim 29, further comprising:

as presented in table below;

providing a relationship of said R-EGA and P-EGA values of said sensor in a euglycemia range according to a table including rows of said R-EGA values and columns of P-EGA values.

34. The method of claim 33, wherein said R-EGA and P-EGA values will be considered accurate if a value falls in the A_P zone or B_P zone of the P-EGA columns of the table and said same value falls in the A_R or B_R zones of the R-EGA rows of the table, as denoted by the white section of the table of claim 32.

35. The method of claim 33, wherein said R-EGA and P-EGA values will be considered clinically erroneous if a value falls in the

A_P zone or B_P zone of the P-EGA columns of the table and said same value falls in the uE or lE of the R-EGA rows of the table;

or

C_P zone of the P-EGA column of the table, as denoted by the shaded section of the table in claim 32.

36. The method of claim 33, wherein said R-EGA and P-EGA values will be considered benign errors if a value falls in the

A_P zone or B_P zone of the P-EGA columns of the table and said same value falls in the uC, lC, uD or ID of the R-EGA rows of the table, as denoted by the horizontal hash section of the table in claim 32.

37. The method of claim 33, further comprising:

as presented in table below;

providing a relationship of said R-EGA and P-EGA values of said sensor in a hyperglycemia range according to a table including rows of said R-EGA values and columns of P-EGA values.

38. The method of claim 37, wherein said R-EGA and P-EGA values will be considered accurate if a value falls in the A_P zone or B_P zone of the P-EGA columns of the table and said same value falls in the A_R or B_R zones of the R-EGA rows of the table, as denoted by the white section of the table of claim 36.

39. The method of claim 37, wherein said R-EGA and P-EGA values will be considered clinically erroneous if a value falls in the

A_P zone or B_P zone of the P-EGA columns of the table and said same value falls in the lD, uE or lE of the R-EGA rows of the table;

or

C_P zone, D_P zone or E_P zone of the P-EGA column of the table, as denoted by the shaded section of the table of claim 36.

40. The method of claim 37, wherein said R-EGA and P-EGA values will be considered benign errors if a value falls in the

A_P zone or B_P zone of the P-EGA columns of the table and said same value falls in the uC, IC or uD of the R-EGA rows of the table, as denoted by the horizontal hash section of the table of claim 36.

41. The method of claim 37, comprising:

combining the results from tables of claims 28, 32 and 36 into a single accuracy estimate of a sensor using a linear weighted combination.

42. The method of claim 41, said combining of the R-EGA and P-EGA zone as according to the following formulas:

A(Accurate Readings)=w1*A1+w2*A2+w3*A3,
B(Benign Errors)=w1*B1+w2*B2+w3*B3,
Err(Erroneous Readings)=w1*Err1+w2*Err2+w3*Err3, wherein;

A1, B1, Err1 are percentages of accurate readings, benign errors, and erroneous readings during hypoglycemia;

A2, B2, Err2 are the percentages of accurate readings, benign errors, and erroneous readings during euglycemia, and

A3, B3, Err3 are the percentages of accurate readings, benign errors, and erroneous readings during hyperglycemia, and wherein the weights w1, w2, w3 are assigned to the hypoglycemic, euglycemic, and hyperglycemic BG ranges.

43. The method of claim 42, wherein said weights are about 0.07, about 0.53, and about 0.40 for hypoglycemic, euglycemic, and hyperglycemic ranges, respectively.

44. A system for evaluating accuracy of a glucose monitoring sensor, said system comprising a microprocessor programmed to perform the following:

evaluate point accuracy of blood glucose (BG) values received from said monitoring sensor;

evaluate accuracy of direction and rate of BG fluctuations received from said monitoring sensor; and

combining said point accuracy and rate accuracy in an estimate of sensor precision.

45. The system of claim 44, wherein said estimate of sensor precision comprises:

performing a combination of three grids adapted specifically for hypoglycemic, euglycemic, and hyperglycemic BG ranges.

46. The system of claim 45, wherein said estimate of sensor precision includes solely a single estimate using a weighted combination of hypoglycemic, euglycemic, hyperglycemic BG ranges.

47. The system of claim 45, wherein said estimate of sensor precision includes a plurality of estimates using a weighted combination of hypoglycemic, euglycemic, hyperglycemic BG ranges.

48. The system of claim 44, wherein said system is for use with a subject, wherein said subject is a human or an animal.

49. The system of claim 44, wherein said sensor comprises a continuous glucose monitor or semi-continuous glucose monitor or combination of both.

50. The system of claim 44, wherein:

said accuracy evaluation of direction and rate of BG fluctuations includes performing rate error-grid analysis (R-EGA), and

said accuracy evaluation blood glucose (BG) values includes performing point error-grid analysis (P-EGA) that determines correct presentation of said blood glucose (BG) values.

51. The system of claim 50, wherein said estimate of sensor precision comprises:

performing a combination of three grids adapted specifically for hypoglycemic, euglycemic, and hyperglycemic BG ranges.

52. The system of claim 51, wherein said estimate of sensor precision includes solely a single estimate using a weighted combination of hypoglycemic, euglycemic, hyperglycemic BG ranges.

53. The system of claim 51, wherein said estimate of sensor precision includes a plurality of estimates using a weighted combination of hypoglycemic, euglycemic, hyperglycemic BG ranges.

54. The system of claim 50, wherein system is for use with a subject, wherein said subject is a human or an animal.

55. The system of claim 50, wherein said sensor comprises a continuous glucose monitor or semi-continuous glucose monitor or combination of both.

56. The system of claim 50, wherein said rate error-grid analysis (R-EGA) includes:

for each pair of reference BG points, RBG(t₁) and RBG(t₂) received from a reference_blood glucose monitoring device, taken at times t₁ and t₂, a reference BG rate is estimated as the ratio of the change in the BG points divided by the length of the elapsed time interval as follows;

reference BG rate of change (mg/dl/min)=(RBG(t₂)−

RBG(t₁))/(t₂−

t₁).

57. The system of claim 56, wherein said rate said error-grid analysis (R-EGA) further includes:

for each pair of sensor reported BG points, SBG(t₁) and SBG(t₂), ) received from said glucose monitoring sensor, taken at times t₁ and t₂, a sensor BG rate is estimated as;

sensor BG rate of change (mg/dl/min)=(SBG(t₂)−

SBG(t₁))/(t₂−

t₁).

58. The system of claim 57, wherein said rate said error-grid analysis (R-EGA) further includes:

scatter plotting said sensor BG rate of change against said reference BG rate of change.

59. The system of claim 58, wherein said rate said error-grid analysis (R-EGA) further includes dividing said scatter plot into accuracy zones.

60. The system of claim 59, wherein said accuracy zones comprise A through E accuracy zones, which have a clinical meaning similar to the clinical meaning of the conventional Error Grid Analysis (EGA).

61. The system of claim 60, wherein said clinical meanings are defined as follows:

(a) AR-zone (accurate);

(b) CR-zone (over-correction);

(c) DR-zone (failure to detect);

(d) ER-zone (erroneous reading); and

(e) BR-zone (benign errors),

wherein suffix R stands for rate.

62. The system of claim 61, wherein said rate error-grid analysis (R-EGA) further includes:

computing said R-EGA zones for each pair of said reference-sensor BG rates of change using the formulas below, wherein X is abbreviated for said reference BG rate of change and wherein Y is abbreviated for said sensor BG rate of change;

IF (Y ge X) RZONE=5,

IF (Y gt (X+1) and X ge 0-2 and X le

     1) RZONE=4,

IF (Y gt 0.5 X and X lt 0-2) RZONE=4,

IF (Y gt 2.0 X and X gt

     1) RZONE=4,

IF (Y gt (X+2) and X ge 0-1 and X lt

     1) RZONE=3,

IF (Y gt (X+2) and X ge 0-3 and X lt 0-1) RZONE=2,

IF (Y gt 0-1 and X lt 0-3) RZONE=2,

IF (Y gt 1 and X lt 0-1) RZONE=1,

IF (Y lt X) RZONE=6,

IF (Y lt (X−

1) and X ge 0-1 and X le

     2) RZONE=7,

IF (Y lt 2.0 X and X lt 0-1) RZONE=7,

IF (Y lt 0.5 X and X gt

     2) RZONE=7,

IF (Y lt (X−

2) and X ge 0-1 and X le

     1) RZONE=8,

IF (Y lt (X−

2) and X gt 1 and X le

     3) RZONE=9,

IF (Y lt 1 and X gt

     3) RZONE=9, and

IF(Y lt 0-1 and X gt

     1) RZONE=10,

wherein the output of this computation is a zone, coded as follows;

1-uE 2-uD 3-uC 4-uB 5-uA 6-lA 7-lB 8-lC 9-lD 10-lE.

63. The system of claim 56, wherein said glucose monitoring sensor comprises at least one of Beckman glucose analyzes, glucose oxidase analyzer, YSI Instruments analyzer and finger stick glucose meters, or any combination thereof.

64. The system of claim 50, wherein said point error-grid analysis (P-EGA) includes:

scatter plotting said point error-grid analysis (P-EGA) for said sensor BG rate against a reference BG rate computed from the data of a reference monitoring device, wherein said scatter plot is divided into A_P, B_P, C_P, D_P, and E_P zones, wherein suffix “

P”

stands for point.

65. The system of claim 64, wherein

said B_P zone is divided into upper B_P zone and lower B_P zone;

said C_P zone is divided into upper C_P zone and lower C_P zone;

said D_P zone is divided into upper D_P zone and lower D_P zone; and

said E_P zone is divided into upper E_P zone and lower E_P zone.

66. The system of claim 65, wherein said point error-grid analysis (P-EGA) includes:

said zones are defined depending on the reference rate of BG change classified as follows;

(a) slow change range;

(b) moderate decrease range;

(c) moderate increase range;

(d) rapid decrease range; and

(e) rapid increase range.

67. The system of claim 66, wherein said classification of reference BG change comprises:

(a) said slow change range being about −

1 and about 1 mg/dl/min;

(b) moderate decrease range being about −

2 and about −

1 mg/dl/min;

(c) moderate increase range being about 1 and about 2 mg/dl/min;

(d) rapid decrease range being less than about −

2 mg/dl/min; and

(e) rapid increase range being greater than about 2 mg/dl/min.

68. The system of claim 67, wherein said point error-grid analysis (P-EGA) includes:

said zones are defined depending on the reference rate of BG change as follows;

(a) if the reference BG rate is within said slow change range, said P-EGA zones are identical to the zones of the conventional EGA;

(b) if the reference BG is falling at a rate within said moderate increase range, the upper limits of upper said A_P, B_P, and D_P zones are expanded by about 10 mg/dl,

and if the reference BG is falling faster than said rapid decrease range, the upper limits of upper A_P, B_P, and D_P zones are expanded by about 20 mg/dl; and

(c) if the reference BG is rising at a rate within said moderate increase, the lower limits of lower A_P, B_P, and D_P zones are expanded by about 10 mg/dl,

and if the reference BG is rising faster than said rapid increase range, the lower limits of lower said A_P, B_P, and D_P zones are expanded by about 20 mg/dl.

69. The system of claim 66, wherein said point error-grid analysis (P-EGA) further includes:

computing said P-EGA zones for each pair of said reference-sensor BG readings using the formulas below;

a) if the BG rate of change is in said slow change rate, apply the formulas;

wherein U1 is abbreviated for said Reference BG,

wherein W1 is abbreviated for sad Sensor BG,

IF (W1 ge U1) ZONE0=5,

IF (W1 ge 1.20 U1 and U1 gt

     70) ZONE0=4,

IF (W1 ge (1.03 U1+107.9) and U1 gt

     70) ZONE0=3,

IF (W1 ge 1.20 U1 and W1 gt 70 and U1 le

     70) ZONE0=2,

IF (W1 ge 180 and U1 le

     70) ZONE0=1,

IF (W1 lt U1) ZONE0=6,

IF (W1 le 0.8 U1 and U1 gt

     70) ZONE0=7,

IF (W1 le (1.4 U1-182) and U1 le

     180) ZONE0=8,

IF (U1 ge 240 and W1 gt 70 and W1 lt

     180) ZONE0=9, and

IF (U1 gt 180 and W1 le

     70) ZONE0=10;

b) if BG is falling said moderate rate apply the formulas;

wherein U2 is abbreviated for said Reference BG,

wherein W2 is abbreviated for said Sensor BG,

IF (W2 ge U2) ZONE2=5,

IF (W2 ge (1.20 U2+10) and U2 gt

     70) ZONE2=4,

IF (W2 ge (1.03 U2+117.9) and U2 gt

     70) ZONE2=3,

IF (W2 ge (1.20 U2+10) and W2 gt 80 and U2 le

     70) ZONE2=2,

IF (W2 ge 190 and U2 le

     70) ZONE2=1,

IF (W2 lt U2) ZONE2=6,

IF (W2 le 0.8 U2 and U2 gt

     70) ZONE2=7,

IF (W2 le (1.4 U2-182) and U2 le

     180) ZONE2=8,

IF (U2 ge 240 and W2 gt 70 and W2 lt

     180) ZONE2=9, and

IF (U2 gt 180 and W2 le

     70) ZONE2=10;

c) if BG is falling at said rapid rate use the formulas;

wherein U4 is abbreviated for said Reference BG,

wherein W4 is abbreviated for said Sensor BG,

IF (W4 ge U4) ZONE4=5,

IF (W4 ge (1.20 U4+20) and U4 gt

     70) ZONE4=4,

IF (W4 ge (1.03 U4+127.9) and U4 gt

     70) ZONE4=3,

IF (W4 ge (1.20 U4+20) and W4 gt 90 and U4 le

     70) ZONE4=2,

IF (W4 ge 200 and U4 le

     70) ZONE4=1,

IF (W4 lt U4) ZONE4=6,

IF (W4 le 0.8 U4 and U4 gt

     70) ZONE4=7,

IF (W4 le (1.4 U4-182) and U4 le

     180) ZONE4=8,

IF (U4 ge 240 and W4 gt 70 and W4 lt

     180) ZONE4=9, and

IF (U4 gt 180 and W4 le

     70) ZONE4=10;

d) if BG is rising at said moderate rate of use the formulas;

wherein U1 is abbreviated for said Reference BG,

wherein SENBG is abbreviated for said Sensor BG,

IF (W1 ge U1) ZONE1=5,

IF (W1 ge 1.20 U1 and U1 gt

     70) ZONE1=4,

IF (W1 ge (1.03 U1+107.9) and U1 gt

     70) ZONE1=3,

IF (W1 ge 1.20 U1 and W1 gt 70 and U1 le

     70) ZONE1=2,

IF (W1 ge 180 and U1 le

     70) ZONE1=1,

IF (W1 lt U1) ZONE1=6,

IF (W1 le (0.8 U1-10) and U1 gt

     70) ZONE1=7,

IF (W1 le (1.4 U1-192) and U1 le

     180) ZONE1=8,

IF (U1 ge 240 and W1 gt 60 and W1 lt

     170) ZONE1=9, and

IF (U1 gt 180 and W1 le

     60) ZONE1=10; and

e) if BG is rising at said rapid rate use the formulas;

wherein U3 is abbreviated for said Reference BG,

wherein W3 is abbreviated for said Sensor BG,

IF (W3 ge U3) ZONE3=5,

IF (W3 ge 1.20 U3 and U3 gt

     70) ZONE3=4,

IF (W3 ge (1.03 U3+107.9) and U3 gt

     70) ZONE3=3,

IF (W3 ge 1.20 U3 and W3 gt 70 and U3 le

     70) ZONE3=2,

IF (W3 ge 180 and U3 le

     70) ZONE3=1,

IF (W3 lt U3) ZONE3=6. U3 gt

     70) ZONE3=7,

IF (W3 le (1.4 U3-202) and U3 le

     180) ZONE3=8,

IF (U3 ge 240 and W3 gt 50 and W3 lt

     160) ZONE3=9,

IF (U3 gt 180 and W3 le

     50) ZONE3=10, and

IF (W3 le (0.8 U3-20),

wherein the output of this computation is a zone, coded as follows;

1-uE, 2-uD, 3-uC, 4-uB, 5-uA, 6-lA, 7-lB, 8-lC, 9-lD, 10-lE.

70. The system of claim 50, comprises:

splitting reference BG values received from a reference BG monitoring device into three clinically meaningful regions including hypoglycemia, euglycemia and hyperglycemia regions; and

combining results from the rate error-grid analysis (R-EGA) and point error-grid analysis (P-EGA) within each of these regions.

71. The system of claim 70, wherein:

said hypoglycemia region is defined as BG<

=70 mg/dl (3.9 mmol/l);

said euglycemia region is defined as 70 mg/dl (3.9 mmol/l)<

BG<

180 mg/dl (10 mmol/l), and

said hyperglycemia region defined as BG>

180 mg/dl (10 mmol/l).

72. The system of claim 71, further comprising:

as presented in table below;

providing a relationship of said R-EGA and P-EGA values of said sensor in a hypoglycemic range according to a table including rows of said R-EGA values and columns of P-EGA values.

73. The system of claim 72, wherein said R-EGA and P-EGA values will be considered accurate if a value falls in the A_P zone of the P-EGA column of the table and said same value falls in the A_R or B_R zones of the R-EGA rows of the table, as denoted by the white section of the table of claim 69.

74. The system of claim 72, wherein said R-EGA and P-EGA values will be considered clinically erroneous if a value falls in the

D_P zone or E_P zone of the P-EGA column of the table, or

A_P zone of the P-EGA column of the table and said same value falls in the uD or uE zones of the R-EGA rows of the table, as denoted by the shaded section of the table of claim 69.

75. The system of claim 72, wherein said R-EGA and P-EGA values will be considered benign errors if a value falls A_P zone of the P-EGA column of the table and said same value falls in the uC, lC, lD, or lE zones of the R-EGA rows of the table, as denoted by the horizontal hash section of the table of claim 69.

76. The system of claim 72, further comprising:

as presented in table below;

providing a relationship of said R-EGA and P-EGA values of said sensor in a euglycemia range according to a table including rows of said R-EGA values and columns of P-EGA values.

77. The system of claim 76, wherein said R-EGA and P-EGA values will be considered accurate if a value falls in the A_P zone or B_P zone of the P-EGA columns of the table and said same value falls in the A_R or B_R zones of the R-EGA rows of the table, as denoted by the white section of the table of claim 73.

78. The system of claim 76, wherein said R-EGA and P-EGA values will be considered clinically erroneous if a value falls in the

A_P zone or B_P zone of the P-EGA columns of the table and said same value falls in the uE or lE of the R-EGA rows of the table;

or

C_P zone of the P-EGA column of the table, as denoted by the shaded section of the table of claim 73.

79. The system of claim 76, wherein said R-EGA and P-EGA values will be considered benign errors if a value falls in the

A_P zone or B_P zone of the P-EGA columns of the table and said same value falls in the uC, lC, uD or ID of the R-EGA rows of the table, as denoted by the horizontal hash section of the table of claim 73.

80. The system of claim 76, further comprising:

as presented in table below;

providing a relationship of said R-EGA and P-EGA values of said sensor in a hyperglycemia range according to a table including rows of said R-EGA values and columns of P-EGA values.

81. The system of claim 80, wherein said R-EGA and P-EGA values will be considered accurate if a value falls in the A_P zone or B_P zone of the P-EGA columns of the table and said same value falls in the A_R or B_R zones of the R-EGA rows of the table, as denoted by the white section of the table of claim 77.

82. The system of claim 80, wherein said R-EGA and P-EGA values will be considered clinically erroneous if a value falls in the

A_P zone or B_P zone of the P-EGA columns of the table and said same value falls in the lD, UE or lE of the R-EGA rows of the table;

or

C_P zone, D_P zone or E_P zone of the P-EGA column of the table, as denoted by the shaded section of the table of claim 77.

83. The system of claim 80, wherein said R-EGA and P-EGA values will be considered benign errors if a value falls in the

A_P zone or B_P zone of the P-EGA columns of the table and said same value falls in the uC, lC or uD of the R-EGA rows of the table, as denoted by the horizontal hash section of the table of claim 77.

84. The system of claim 80, comprising:

combining the results from tables of claims 69, 73 and 77 into a single accuracy estimate of a sensor using a linear weighted combination.

85. The system of claim 84, said combining of the R-EGA and P-EGA zone as according to the following formulas:

A(Accurate Readings)=w1*A1+w2*A2+w3*A3,
B(Benign Errors)=w1*B1+w2*B2+w3*B3,
Err(Erroneous Readings)=w1*Err1+w2*Err2+w3*Err3, wherein;

A1, B1, Err1 are percentages of accurate readings, benign errors, and erroneous readings during hypoglycemia;

A2, B2, Err2 are the percentages of accurate readings, benign errors, and erroneous readings during euglycemia, and

A3, B3, Err3 are the percentages of accurate readings, benign errors, and erroneous readings during hyperglycemia, and wherein the weights w1, w2, w3 are assigned to the hypoglycemic, euglycemic, and hyperglycemic BG ranges.

86. The method of claim 85, wherein said weights are about 0.07, about 0.53, and about 0.40 for hypoglycemic, euglycemic, and hyperglycemic ranges, respectively.

87. A computer program product comprising a computer usable medium having computer program logic for enabling at least one processor in a computer system to evaluate accuracy of a glucose monitoring sensor, said computer program logic comprising:

evaluating point accuracy of blood glucose (BG) values received from said monitoring sensor;

evaluating accuracy of direction and rate of BG fluctuations received from said monitoring sensor; and

combining said point accuracy and rate accuracy in an estimate of sensor precision.

88. The computer program product of claim 87, wherein said estimate of sensor precision comprises:

performing a combination of three grids adapted specifically for hypoglycemic, euglycemic, and hyperglycemic BG ranges.

89. The computer program product of claim 88, wherein said estimate of sensor precision includes solely a single estimate using a weighted combination of hypoglycemic, euglycemic, hyperglycemic BG ranges.

90. The computer program product of claim 88, wherein said estimate of sensor precision includes a plurality of estimates using a weighted combination of hypoglycemic, euglycemic, hyperglycemic BG ranges.

91. The computer program product of claim 87, wherein said computer program product is for use with a subject, wherein said subject is a human or an animal.

92. The computer program product of claim 87, wherein said sensor comprises a continuous glucose monitor or semi-continuous glucose monitor or combination of both.

93. The computer program product of claim 87, wherein:

said accuracy evaluation of direction and rate of BG fluctuations includes performing rate error-grid analysis (R-EGA), and

said accuracy evaluation blood glucose (BG) values includes performing point error-grid analysis (P-EGA) that determines correct presentation of said blood glucose (BG) values.

94. The computer program product of claim 93, wherein said estimate of sensor precision comprises:

performing a combination of three grids adapted specifically for hypoglycemic, euglycemic, and hyperglycemic BG ranges.

95. The computer program product of claim 94, wherein said estimate of sensor precision includes solely a single estimate using a weighted combination of hypoglycemic, euglycemic, hyperglycemic BG ranges.

96. The computer program product of claim 94, wherein said estimate of sensor precision includes a plurality of estimates using a weighted combination of hypoglycemic, euglycemic, hyperglycemic BG ranges.

97. The computer program product of claim 93, wherein said computer program product is for use with a subject, wherein said subject is a human or an animal.

98. The computer program product of claim 93, wherein said sensor comprises a continuous glucose monitor or semi-continuous glucose monitor or combination of both.

99. A system for evaluating accuracy of a glucose monitoring sensor, said system comprising:

said glucose monitoring sensor; and

a microprocessor programmed to perform the following;

evaluate point accuracy of blood glucose (BG) values received from said monitoring sensor;

evaluate accuracy of direction and rate of BG fluctuations received from said monitoring sensor; and

combine said point accuracy and rate accuracy in an estimate of sensor precision.

100. A system for evaluating accuracy of a glucose monitoring sensor, said system comprising:

said glucose monitoring sensor;

a reference device; and

a microprocessor programmed to perform the following;

evaluate point accuracy of blood glucose (BG) values received from said monitoring sensor;

evaluate accuracy of direction and rate of BG fluctuations received from said monitoring sensor; and

combine said point accuracy and rate accuracy in an estimate of sensor precision.