KR20220010245A - System, method, and computer program for conducting pixel-level semi-automatization of Parmar's methodology - Google Patents

Abstract

The present invention relates to a system, a method, and a computer program for conducting pixel-level semi-automatization of Parmar's methodology in units of pixels accurately and precisely, based on information inputted without professional knowledge on the Parma methodology. The information required for the Parma methodology is calibrated and derived pixel by pixel, and the Kaplan Meier survival curve is quantified as a risk rate according to the Parma's methodology.

Description

{System, method, and computer program for conducting pixel-level semi-automatization of Parmar's methodology}

When quantifying the Kaplan-Meier survival curve as a hazard ratio, Parmar's methodology is followed. Even if you are a medical technician who routinely uses the Kaplan Meier survival curve, you will need to acquire knowledge of the Parma methodology separately.

When obtaining the information required to perform the Parma methodology, technicians have used general-purpose software (eg, DigitizeIt) having a ruler or a graphical user interface (GUI). Since the calculated risk rate is changed according to this information, it affects the accuracy and precision of the risk rate. In addition, when obtaining information using a ruler or existing general-purpose software, the user selects several points on the Kaplan Meier survival curve one by one, and derives the survival probability based on the coordinate information.

According to the Parma methodology, when the Kaplan Meier survival curve is quantified as a risk rate, the present invention provides accurate (correct, accurate) information based on information that can be input by those skilled in the art using the Kaplan Meier survival curve without specialized knowledge on the Parma methodology. ) and precisely (精密, precise) is a system, method and computer program that semi-automatizes the Parma methodology in pixels.

In order to solve this, without expert knowledge of the Parma methodology, those skilled in the art using the Kaplan Meier survival curve can use the input means to display the information of the Kafla Meier survival curve 1000, pixel by pixel. By correcting the location with accuracy and precision (2000), the information necessary for the Parma methodology is derived in pixel units (3000), and the Kaplan Meier survival curve is quantified as a risk rate according to the Parma methodology. .

As in Cochrane Training [1] and meta-analysis papers [2, 3], the Kaplan Meier survival curve is quantified as risk according to the Parma methodology [4]. Tierney published a Tierney spreadsheet (“HR calculated spreadsheet”) that converts a Kaplan Meier survival curve into a hazard rate based on the Parma methodology [5], extracting survival probabilities from the Kaplan Meier survival curve and entering ask for However, before using the Tierney spreadsheet, the process of preparing such a survival probability has been difficult, requiring an understanding of the Parma methodology or the Tierney spreadsheet, even for a technician who typically uses the Kaplan Meier survival curve.

In addition, the process of preparing the survival probability to be entered into the Parma methodology or Tierney spreadsheet has problems with pixel-level accuracy and precision. For example, when obtaining survival probabilities from a Kaplan Meier survival curve, measure using a ruler on a printed survival curve, or use DigitizeIt, the Generalized Software recommended by Guyot, with a digital ruler. It is used as [6] and measured. When using this DigitizeIt general-purpose software [7], it requires the user to display coordinates accurately and precisely. However, when displaying a specific point using a mouse, it is difficult to accurately and precisely indicate the position in units of pixels. In other words, it is difficult for the user to accurately mark the same point in pixels unless the software corrects it according to the Parma methodology context. Also, even if the user intended a specific point, there is no certainty that the actually displayed point accurately represented the intended point in the context of the Parma methodology on a pixel-by-pixel basis. Thus, instead of general-purpose software, specialized software was needed to precisely and accurately calibrate the Parma methodology on a pixel-by-pixel basis.

In addition, when using Ruler or DigitizeIt, the user randomly selects several points on the Kaplan Meier survival curve to derive survival probability information. There is something I can't do. Therefore, when deriving the information necessary for the Parma methodology, the points on the Kaplan Meier survival curve are selected one by one as a computer operation by the user and information is not derived, but the points on the survival curve are packed in pixel units as information processing by computer. They need to be derived from information.

Smith, C. T. Meta-analysis of time-to-event data. 2018 [cited 2019 March, 29th]; Available from: https://training.cochrane.org/resource/meta-analysis-time-event-data. Chu, D.K., et al., Mortality and morbidity in acutely ill adults treated with liberal versus conservative oxygen therapy (IOTA): a systematic review and meta-analysis. Lancet, 2018. 391 (10131): p. 1693-1705. Roh, H.F., et al., Pulmonary resection for patients with multidrug-resistant tuberculosis based on survival outcomes: a systematic review and meta-analysis. Eur J Cardiothorac Surg, 2017. 52(4): p. 673-678. Parmar, M.K., V. Torri, and L. Stewart, Extracting summary statistics to perform meta-analyses of the published literature for survival endpoints. Stat Med, 1998. 17(24): p. 2815-34. Tierney, J.F., et al., Practical methods for incorporating summary time-to-event data into meta-analysis. Trials, 2007. 8: p. 16. Guyot, P., et al., Enhanced secondary analysis of survival data: reconstructing the data from published Kaplan-Meier survival curves. BMC Med Res Methodol, 2012. 12: p. 9. Roh, H. F. and J.M. Kim, Noninvasive pulmonary arterial pressure estimation using a logistic-based systolic model. Comput Biol Med, 2018. 95: p. 209-216. Roh, H.F., S.H. Nam, and J.M. Kim, Robot-assisted laparoscopic surgery versus conventional laparoscopic surgery in randomized controlled trials: A systematic review and meta-analysis. PLoS One, 2018. 13(1): p. e0191628. Roh, H.F., et al., Extracorporeal membrane oxygenation meta-analysis of time-to-event data in cardiopulmonary disease in adults. 2020, Protocols.io. Kleinbaum, D. G. and M. Klein, Survival Analysis: A self-Learning Text. 3rd edition ed. Statistics for Biology and Health, ed. M. Gail, et al. 2012, LLC, 233 Spring Street, New York, NY 10013, USA: Springer Science + Business Media.

Therefore, the present invention, when quantifying the Kaplan Meier survival curve as a risk rate according to the Parma methodology, on a graphical user interface (GUI), those skilled in the art using the Kaplan Meier survival curve are experts on the Parma methodology Based on information that can be input without knowledge, a method for semi-automatization of the Parma methodology in pixel units and a computer system ( Computer system integrated with a program).

In order to achieve the above object, in accordance with the Parma methodology, according to the present invention, a method and a computer system combined with a method for semi-automatization of quantification of a survival curve as a risk rate and a program according to the Parma methodology, thinking and approaching the problem in units of pixels tried to solve them.

First, in order to think and approach on a pixel-by-pixel basis, the Kaplan Meier survival curve should be recognized more accurately. Conventional rulers or general-purpose software (eg, DigitizeIt) do not correct user input values pixel by pixel according to the context of the Parma methodology. The present invention corrects the relevant input values of the Kaplan Meier survival curve in pixels when displaying information that can be input by those skilled in the art using the Kaplan Meier survival curve, so that the precision and accuracy in pixels to get

Second, when obtaining information using a ruler or existing general-purpose software, the user sparsely selects several points on the Kaplan Meier survival curve one by one to derive survival probability information. In contrast, the present invention is obtained by analyzing the pixel values of the y-axis of the survival curves of the experimental group and the control group while increasing by one pixel from the origin on the x-axis in order to reflect the information that exists in each pixel on the Kaplan Meier survival curve. .

Third, based on the expertise of the Parma methodology, instead of calculating the survival probability by displaying the position of the survival curve corresponding to each time interval using an input device on the graphic-based user interface, Kaplan Meier Based on the information that can be input by medical technicians who normally use the survival curve, information that requires expertise in the Parma methodology is derived as information processing by a computer program.

In other words, in the process of deriving information as information processing by a computer program for the Kaplan Meier survival curve recognized by correcting the data input as a computer operation by the user in pixel units, one pixel at a time from the origin on the x-axis As the y-axis pixel values of the survival curves of the experimental group and the control group are analyzed while increasing, there is no need for the user to consider the interval interval, and a specific point on the survival curve corresponding to the survival probability can be selected using the input device on the graphic-based user interface. There is an advantage in that there is no need to use the display.

In conclusion, to describe the core of the solution to the problem, claims 3, 4, and claims that can be entered by technicians who commonly use the Kaplan Meier survival curve in order not to require information that requires specialized knowledge of the Parma methodology It had to be composed of information from 5. As a solution for deriving the information required for the expertise of the Parma methodology from this information, a method of approaching the Kaplan Meier survival curve in pixel units is chosen, and there is an advantage in that the above problems are solved at the same time, and the present invention is patented. was applied for

According to the present invention, when quantifying the Kaplan Meier survival curve as a risk rate according to the Parma methodology, on a graphical user interface (GUI), those skilled in the art using the Kaplan Meier survival curve specialize in the Parma methodology A method, system, and computer program to semi-automatize the Parma methodology in pixel units accurately and precisely, based on information that can be input without any formal knowledge. , it can be helpful for meta-analysis studies [2, 3] on the risk rate. In addition, since this hazard ratio is data [4] that additionally reflects the passage of time compared to the relative risk [8], it is related to the effect of newly developed medical technology on the mortality rate. [9], it can be used when constructing big data in the healthcare industry.

1 is an exemplary diagram showing (1000) information of a Kaflameier survival curve using an input means without professional knowledge on the Parma methodology, according to an embodiment of the present invention.
2 is an exemplary diagram of correcting (2000) the position of the input information in pixel units accurately and precisely (精密, precision) according to an embodiment of the present invention;
3 is an exemplary diagram in which information necessary for the Parma methodology is derived in units of pixels (3000) based on the input information and the corrected position according to an embodiment of the present invention.
4 is a flowchart illustrating an overall flow through a large arrow for an integrated understanding of FIGS. 1, 2, and 3 according to an embodiment of the present invention;

The terms to be described later are terms defined in consideration of functions in an embodiment of the present invention, which may vary according to intentions or customs of users and operators. Therefore, the intention should be made based on the contents throughout this specification.

In the specification of the present invention, the term of "hazard ratio" [10] is defined as For other methods of calculating the risk, detailed explanations can be obtained from the Parma Methodology Papers [4] and Tierney Papers [5], so the description will be omitted herein.

In the specification of the present invention, the term "x-axis" should be understood to mean a coordinate axis that reflects the passage of time from the point of view of those skilled in the art. Also, the term “y-axis” should be understood to mean a coordinate axis reflecting the degree of survival.

Also, in this specification, terms such as x-axis and y-axis may be used to describe various components, but these components should not be limited by these terms. These terms are only used to distinguish between the components. For example, without departing from the spirit of the present invention, the x-axis may be referred to as a y-axis, and similarly, the y-axis may be referred to as an x-axis.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, a detailed description of preferred embodiments of the present invention will be described with reference to the accompanying drawings. It should be noted that like elements in the drawings are denoted by like reference numerals wherever possible. And in the accompanying drawings, some components may be exaggerated, omitted, or schematically shown. In addition, in describing the embodiments, detailed descriptions of technical contents that are widely known in the technical field to which the present invention pertains and are not directly related to the technical gist of the present invention will be omitted. This is provided to help a more general understanding of the present invention by omitting unnecessary descriptions not related to the gist of the present invention.

Hereinafter, according to an embodiment of the present invention with reference to the accompanying drawings, on a graphic-based user interface, when quantifying the Kaplan Meier survival curve as a risk rate according to the Parma methodology, those skilled in the art using the Kaplan Meier survival curve Parma methodology Based on information that can be input without specialized knowledge about The program will be described in detail.

In relation to claim 1, for an integrated understanding of FIGS. 1, 2, and 3, the overall flow is expressed through a large arrow in FIG. 4 (representative view). That is, FIG. 1 is for displaying information 1000 by means of inputting information of a Kaflameier survival curve, without specialized knowledge of the Parma methodology, as a computer operation by a user. As information processing by a computer program, based on the essential information mentioned in FIG. 1 , as in FIG. 2 , the position is accurately corrected and precisely corrected ( 2000 ) in units of pixels. 3 is an information processing by a computer program based on the input information and the corrected information, and the information required to perform the Parma methodology on behalf of the user is derived (3000) in pixel units. will be. That is, the present invention does not require the user to directly input the elements of FIG. 3, which requires expert knowledge on the Parma methodology. Instead, as information processing by a computer program, information required for the Parma methodology is derived (3000), and the Kaplan Meier survival curve is digitized as a risk rate.

Referring to FIG. 1 , as a computer operation by a user, the step of displaying information of the Kaflameier survival curve using an input means (S1000) without professional knowledge of the Parma methodology is expressed in a graphic-based user interface An origin on the x-axis (1101), a tick on the x-axis (1102), on the y-axis, displayed and entered on the Kaplan Meier survival curve image A tick on the y-axis (1103), an origin on the y-axis (1104), a Kaplan Meier survival curve for a control group (1105) ), a Kaplan Meier survival curve corresponding to the experimental group (1106), and other input, effective number alive at start of time interval (1201) of the control group (1201) , an effective number alive at start of time interval 1202 of the research group, a value for the tick on the x-axis 1203 includes information on

If censoring information of the Kaplan Meier survival curve exists, the Minimum follow-up (1204) and Maximum follow-up (1205) described in claim 4 can be entered as a part. , which is not essential. That is, the Minimum follow-up 1204 and the Maximum follow-up 1205 are information that can be filled in by a user who normally uses the Kaplan Meier survival curve, but Kaplan Meier survival curve. Since it is not always published in the original thesis that contains curves, it may be difficult to write necessarily, so these are additionally required elements.

As for the scale 1102 on the x-axis, the more distant the scale from the origin 1101 on the x-axis, the higher the overall accuracy in the process of calculating the survival probability. Able to know.

Similarly, if the scale 1103 on the y-axis is the scale farthest from the origin 1104 on the y-axis, the overall accuracy increases in the process of calculating the survival probability. can be seen intuitively. In most cases, the scale farthest from the origin 1104 on the y-axis becomes a point at which survival probability becomes 1.0, so a value for the scale 1103 on the y-axis tick on the y-axis) 1206 is typically 1.0 and does not require a write.

Hereinafter, according to an embodiment of the present invention with reference to the accompanying FIG. 2, on a graphic-based user interface that quantifies the Kaplan Meier survival curve as a risk rate, the user can display a specific part of the Kaplan Meier survival curve using an input device. A method of obtaining and correcting precisely and accurately based on the x and y values of the pixels concerned near the marked area will be described in detail.

As information processing by a computer program, for the input information, the step (S2000) of correcting the position in pixel units accurately and precisely is expressed in a graphic-based user interface. Kaplan Meier survival curve input by displaying on the image, the origin 1101 on the x-axis, the scale 1102 on the x-axis, the scale on the y-axis (1103), the origin on the y-axis (1104), Kaplan Meier corresponding to the control Survival curve (1105), Kaplan Meier survival curve (1106) corresponding to the experimental group, for each,

On the Kaplan Meier survival curve near a spot, pointed by a user (2001), according to the context of the Parma methodology, a pixel at a spot, intended by a user (2002) It consists of a step (S2010) to find and correct.

2, on the Kaplan Meier survival curve near a spot, pointed by a user (2001), according to the Parma methodology context, a location intended by the user (a spot, intended by a user) ) (2002) as an embodiment of the present invention for the step (S2010) of finding and correcting the pixel, when the user displays the intersection point, which is a specific site, on the Kaplan Meier survival curve using an input device, two line segments near the displayed site X and y values are obtained based on the RGB values of the pixels at the intersection point of . In other words, with respect to the Kaplan Meier survival curve, the positions of the black dots and lines on the white background can be determined based on the RGB values of each pixel.

The above elements related to FIG. 3 are computer manipulation by a user, even if a person of ordinary skill in the art using the Kaplan Meier survival curve does not have an expert knowledge of the Parma methodology, input using an input means or It is a stage that is difficult to express.

Fortunately, the present invention is information processing by a computer program, and according to the Parma methodology, for "x" (3001: control - blue, 3002: experimental group - purple) displayed on the survival curve with a vertical dotted line corresponding to each interval interval. , the survival probability is calculated based on the values input by the user, and the risk rate is calculated by itself.

Based on the input information and the corrected position, the step of deriving the information required for the Parma methodology in units of pixels (S3000) is the origin 1101 on the x-axis, the origin 1101 on the x-axis, which the user inputs using an input means. Scale 1102, scale 1103 on the y-axis, origin 1104 on the y-axis, effective number alive at start of time interval 1201 of the control group, initial of the research group Based on the effective number alive at start of time interval 1202, and values corresponding to a value for the tick on the x-axis 1203, The coordinates on the Kaplan Meier survival curve corresponding to the control group (3001) and the experimental group corrected in pixels to the location intended by the user (a spot, intended by a user) while increasing in pixels from the origin on the x-axis A step (S3010) of converting a value corresponding to the coordinate 3002 on the Kaplan Meier survival curve into a survival probability is performed.

Although it is a technical content on the Parma methodology, it is a content that can be helpful when implementing the present invention as a program. As shown in FIG. 3, from the origin 1101 on the x-axis to the scale 1102 on the x-axis on the line segment of go through the process of obtaining them. In the present invention, the time corresponding to one pixel is the interval interval.

As mentioned in claim 1 or 2, based on the input information and the location, based on the input information and the location, information required for the Parma methodology is derived in units of pixels, and then the Kaplan Meier survival curve is calculated according to the Parma methodology A detailed description of the step of quantifying the risk can be obtained from the Parma Methodology Papers [4] and Tierney Papers [5], so a detailed description will be omitted herein.

As mentioned above, although specific embodiments of the present invention have been described above, the spirit and scope of the present invention are not limited to these specific embodiments, and various modifications and variations are possible without changing the gist of the present invention. It will be understood by those of ordinary skill in the art to which the invention pertains.

Therefore, since the embodiments described above are provided to fully inform those of ordinary skill in the art to which the present invention belongs the scope of the invention, it should be understood that they are exemplary in all respects and not limiting, The protection scope of the present invention should be construed by the claims, and all technical ideas within the scope equivalent thereto should be construed as being included in the scope of the present invention.

According to the present invention, when quantifying the Kaplan Meier survival curve as a risk rate according to the Parma methodology, on a graphical user interface (GUI), those skilled in the art using the Kaplan Meier survival curve specialize in the Parma methodology A method, system, and computer program to semi-automatize the Parma methodology in pixel units accurately and precisely, based on information that can be input without any formal knowledge. , it can be helpful for meta-analysis studies [2, 3] on the risk rate. In addition, since this hazard ratio is data [4] that additionally reflects the passage of time compared to the relative risk [8], it is related to the effect of newly developed medical technology on the mortality rate. [9], it can be used when constructing big data in the healthcare industry.

1101: an origin on the x-axis
1102: a tick on the x-axis
1103: a tick on the y-axis
1104: an origin on the y-axis
1105: Kaplan Meier survival curve of a control group
1106: Kaplan Meier survival curve corresponding to the experimental group (a survival curve of a research group)
1201: effective number alive at start of time interval of the control group
1202: the initial effective number of the experimental group (Research group) (effective number alive at start of time interval)
1203: a value for the tick on the x-axis
1204: Minimum follow-up
1205: Maximum follow-up
1206: a value for the tick on the y-axis
2001: a spot, pointed by a user
2002: A spot, intended by a user
3001: Calibrated in pixels to the location intended by the user (a spot, intended by a user),
Coordinates on the Kaplan Meier survival curve corresponding to the control
3002: Calibrated in pixels to the location intended by the user (a spot, intended by a user),
Coordinates on the Kaplan Meier survival curve corresponding to the experimental group

Claims (7)

A computer operation by a user, comprising:
display ( 1000 ) information of Kaflameier survival curves using input means, without expert knowledge of Parma methodology;

As information processing by a computer program,
For the input information, correct (correction, accuracy) and precisely (精密, precision) the position in units of pixels (2000); and
Deriving information necessary for the Parma methodology in units of pixels based on the input information and the corrected position (3000);

A computer system combined with a program for semi-automating the Parma methodology pixel by pixel, including
A computer operation by a user, comprising:
Displaying information on the Kaflameier survival curve using an input means without professional knowledge of the Parma methodology (S1000);

As information processing by a computer program,
a step of correcting a position in a pixel unit with respect to the input information accurately and precisely (S2000); and
deriving information necessary for the Parma methodology in units of pixels based on the input information and the corrected position (S3000);

A method of semi-automating the Parma methodology on a pixel-by-pixel basis, including
3. The method of claim 2,

Without expert knowledge of the Parma methodology, the step of displaying the information of the Kaflameier survival curve using an input means (S1000),

On the Kaplan Meier survival curve image expressed in a graphical user interface (GUI), a mouse, a touch screen, a digitizer, and a stylus pen are formed with any one selected to input by displaying using an input means;
an origin on the x-axis (1101);
a tick on the x-axis (1102);
a tick on the y-axis (1103);
an origin on the y-axis (1104) ;
Kaplan Meier survival curve of a control group (1105); and
Kaplan Meier survival curve (a survival curve of a research group) corresponding to the experimental group (1106);

to enter using an input means capable of entering numbers;
effective number alive at start of time interval (1201) in the control group;
effective number alive at start of time interval 1202 in the Research group; and
a value for the tick on the x-axis 1203;

How to semi-automate the Parma methodology pixel by pixel
4. The method of claim 3,

Without expert knowledge of the Parma methodology, the step of displaying the information of the Kaflameier survival curve using an input means (S1000),

Minimum follow-up (1204); and
Maximum follow-up (1205);

A method of semi-automating the Parma methodology pixel by pixel, including information from
4. The method of claim 3,

Without expert knowledge of the Parma methodology, the step of displaying the information of the Kaflameier survival curve using an input means (S1000),

a value for the tick on the y-axis 1206;

A method of semi-automating the Parma methodology pixel by pixel, including information from
3. The method of claim 2,

With respect to the input information, the step (S2000) of correcting the position accurately and precisely (精密, precision) in units of pixels,

On the Kaplan Meier survival curve image expressed in a graphical user interface (GUI), a mouse, a touch screen, a digitizer, and a stylus pen are formed with any one selected to input by displaying using an input means;
origin 1101 on the x-axis;
scale 1102 on the x-axis;
scale 1103 on the y-axis;
origin 1104 on the y-axis;
Kaplan Meier survival curve corresponding to control (1105); and
Kaplan Meier survival curve (1106) corresponding to the experimental group;

for each,
On the Kaplan Meier survival curve near a spot, pointed by a user (2001), according to the Parma methodology context, a pixel at a spot, intended by a user (2002) finding and correcting (S2010);

A method of semi-automating the Parma methodology on a pixel-by-pixel basis, which consists of
3. The method of claim 2,

Based on the input information and the corrected position, the step of deriving information necessary for the Parma methodology in units of pixels (S3000),

input by the user using an input means;
origin (1101) on the x-axis,
scale 1102 on the x-axis,
scale on the y-axis (1103);
origin (1104) on the y-axis,
effective number alive at start of time interval (1201) in the control group;
effective number alive at start of time interval 1202 in the Research group; and
a value for the tick on the x-axis 1203;
Based on the values corresponding to

Increasing in units of pixels from the origin on the x-axis,
Calibrated in pixels to the location intended by the user (a spot, intended by a user),
Coordinates on the Kaplan Meier survival curve corresponding to control (3001), and
Coordinates (3002) on the Kaplan Meier survival curve corresponding to the experimental group,
Converting a value corresponding to a survival probability (survival probability) (S3010);

A method of semi-automating the Parma methodology on a pixel-by-pixel basis, which consists of

Images