An artificial intelligence-based intensive care warning system

Abstract

The application discloses an intensive care early warning system based on artificial intelligence and relates to the technical field of early warning.The first communication module is arranged to collect the sign parameters of a target patient in real time, the second communication module is arranged to collect the detection data of the target patient periodically, and the detection data is processed after collection; binary conversion and hexadecimal conversion are sequentially performed in the processing process to obtain a sequence composed of numbers and English letters; the numbers and English letters are independently processed; the numbers are subjected to addition and subtraction operations with 4 to sequentially transform the binary numbers of the numbers into 00, 01, 10 and 11; the ASCII code reference table is referred to to horizontally select control characters as substitute characters; in this way, the recognizability of data can be effectively reduced, the risk that the data is maliciously cracked and restored is reduced, and the safety and confidentiality of the data in the transmission and storage process are ensured.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of early warning, in particular to an intensive care early warning system based on artificial intelligence. BACKGROUND

[0002] In the modern medical field, the monitoring work of critical patients is crucial and extremely challenging; patients in the intensive care unit (ICU) are usually in critical and complex conditions, and their lives and health are at risk at all times;

[0003] When monitoring critical patients, comprehensive and multi-faceted data information is needed; among them, physical data such as heart rate, blood pressure, respiratory rate, blood oxygen saturation, body temperature, etc. can be directly measured and obtained in real time by bedside monitors and other devices, which provides an intuitive basis for understanding the patient's current physiological state; However, another type of key data - detection data - requires the assistance of professional medical institutions or specific detection machines to complete the detection; Detection data includes test results such as blood, urine, and also includes medical imaging data (chest X-ray, CT scan, MRI, etc.) and electrocardiogram, electroencephalogram, etc. Special examination results; These detection data can reveal the patient's disease root, disease progression and potential complication risk from a deeper level;

[0004] With the rapid development of medical informatization, in order to achieve efficient medical collaboration and resource integration, detection data often needs to be transmitted between different medical devices, department systems and remote medical platforms; During network transmission, the security of detection data is severely tested; Because detection data contains a large amount of patient privacy data, such as patient identity information, disease diagnosis details, genetic test results, etc. sensitive content, if these data are leaked, it will bring a lot of adverse consequences to the patient;

[0005] In order to solve the above problems, the present application provides a solution. SUMMARY

[0006] The purpose of the present application is to provide an intensive care early warning system based on artificial intelligence, in order to solve the problems raised in the background art;

[0007] The purpose of the present application can be achieved by the following technical solutions:

[0008] An intensive care early warning system based on artificial intelligence, comprising:

[0009] A periodic acquisition unit for acquiring detection data of a target patient every interval to obtain detection data of the target patient corresponding to the acquisition period;

[0010] The detection data processing unit processes the detection data according to a preset processing rule to obtain detection processing data of the target patient in the corresponding acquisition period after receiving the detection data of the target patient in one acquisition period;

[0011] The monitoring and early warning platform periodically receives the detection processing data of the target patient, restores the detection processing data to obtain the detection data of the target patient in the corresponding acquisition period after receiving the detection processing data of the target patient in one acquisition period, and stores the detection data.

[0012] Further, the first communication module is further included to collect the monitoring values of the several vital sign parameters of the target patient to generate the real-time vital sign communication data of the target patient.

[0013] Further, the processing rule for obtaining the detection processing data of the target patient in one acquisition period is as follows:

[0014] S11: binary conversion is performed on the detection data to obtain first processing data, and each 16 characters in the first processing data are taken as a first processing array in sequence from left to right to obtain a plurality of first processing arrays;

[0015] According to the position of each first processing array in the first processing data, all the obtained first processing arrays are sequentially marked as A1, A2,..., Aa from left to right, and a≥1;

[0016] S12: hexadecimal conversion is performed on the first processing array A1 to obtain a second processing array B1, and all the characters constituting the second processing array B1 are sequentially marked as C1, C2, C3 and C4 in sequence from left to right;

[0017] S13: non-digit judgment is performed on the character C1, if the character C1 is a digit, the character 0 is selected as a replacement identification character of the character C1, and the character C1 is matched with a first numerical interval and a second numerical interval, if the character C1 matches the first numerical interval, the result of 4-C1 is calculated first, the two-bit binary number of the result is taken as a replacement string of the character C1, and then the replacement identification character and the replacement string of the character C1 are spliced in the order of the replacement identification character and the replacement string to obtain a replacement sequence of the character C1;

[0018] If the character C1 matches the second numerical interval, the result of C1-4 is calculated, and the two-bit binary number of the result is taken as the replacement sequence of the character C1;

[0019] If the character C1 is 9 or 4, the four-bit binary number of the character C1 is taken as the replacement sequence of the character C1;

[0020] If the character C1 is an English character, a string is randomly selected from the string 10 or 11 as the alternative identification character of the character C1, and the character C1 is matched with the first and second letter intervals. If the character C1 is matched with the first letter interval, the number D1 of the capital English characters between the character C1 and the capital English character C is obtained according to the order of the 26 capital English characters, and the two-bit binary number of the number D1 is taken as the alternative string of the character C1. Then, the alternative identification character and the alternative string of the character C1 are spliced according to the order of the alternative identification character and the alternative string to obtain the alternative sequence of the character C1.

[0021] If the character C1 is matched with the second letter interval, the number E1 of the capital English characters between the character C1 and the capital English character C is obtained according to the order of the 26 capital English characters, and the two-bit binary number of the number E1+2 is taken as the alternative string of the character C1. Then, the alternative identification character and the alternative string of the character C1 are spliced according to the order of the alternative identification character and the alternative string to obtain the alternative sequence of the character C1.

[0022] If the character C1 is a capital English character C, the row number D1 and the column number D2 corresponding to the control character C are found in the ASCII code table, and then the control character with the row number D1 and the column number D2-1 is extracted from the ASCII code table, which is taken as the alternative sequence of the character C1.

[0023] If the character C1 is a capital English character D, the row number D1 and the column number D2 corresponding to the control character C are found in the ASCII code table, and then the control character with the row number D1 and the column number D2-2 is extracted from the ASCII code table, which is taken as the alternative sequence of the character C1.

[0024] S14: The alternative sequences of the characters C2, C3 and C4 are obtained in sequence according to S13.

[0025] S15: The alternative characters of the characters C1, C2, C3 and C4 are spliced according to the order of the characters C1, C2, C3 and C4 to obtain the alternative sequence of the second processing array B1. In the splicing process, for the character that is a number, if the character is a number 4, a space is used to replace the alternative sequence thereof.

[0026] S16: The second processing arrays B2, B3,..., Ba of the first processing arrays A2, A3,..., Aa are obtained in sequence according to S12, and the alternative sequences of the second processing arrays B2, B3,..., Ba are obtained in sequence according to S13 to S15.

[0027] The detection processing data of the target patient corresponding to the collection period is obtained by splicing the alternative series of the second processing array B1, B2,..., Ba in the order of the second processing array B1, B2,..., Ba.

[0028] Further, the monitoring and early warning platform pre-stores a warning model, and the monitoring and early warning platform periodically inputs the detection data and the sign communication data of the target patient stored in the monitoring and early warning platform into the warning model, and outputs warning result data from the warning model, wherein the warning result data contains a warning signal.

[0029] The present application has the following beneficial effects:

[0030] The present application has the following beneficial effects:

[0031] After the encryption processing, even if the data is accidentally leaked, it is difficult to directly restore the original detection data and sign parameters from the specially transformed numbers and replaced English letters, greatly increasing the difficulty of data cracking. Since the data presents a high degree of confusion, its true information is deeply hidden, reducing the risk of patient privacy exposure caused by data leakage, and effectively protecting the patient's personal medical information. BRIEF DESCRIPTION OF DRAWINGS

[0032] The present application will be further described below in conjunction with the accompanying drawings.

[0033] Figure 1 The present application has the following beneficial effects: DETAILED DESCRIPTION

[0034] The technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor fall within the scope of the present application.

[0035] As Figure 1As shown, an artificial intelligence-based intensive care warning system comprises a first communication module, a second communication module and an intensive care warning platform;

[0036] The first communication module is configured to collect the monitoring values of a plurality of physical parameters of a target patient in real time, wherein the physical parameters include heart rate, blood pressure, respiratory rate, blood oxygen saturation, body temperature, etc.

[0037] The monitoring values of all physical parameters collected in real time are used to generate real-time physical communication data of the target patient, and the physical communication data is transmitted to the intensive care warning platform.

[0038] The second communication module is configured to periodically collect detection data of the target patient, wherein the detection data includes but is not limited to chest X-ray, CT scan, MRI image data, blood routine, biochemical indicators (such as liver and kidney function, electrolyte, etc.), coagulation function indicators, inflammatory markers (such as C-reactive protein, procalcitonin) data, electrocardiogram results, electroencephalogram results.

[0039] The second communication module comprises a periodic collection unit and a detection data processing unit, the periodic collection unit collects the detection data of the target patient every interval to obtain the detection data of the target patient in the corresponding collection period, and transmits the detection data to the detection data processing unit.

[0040] The detection data processing unit processes the detection data according to a preset processing rule to obtain the detection processing data of the target patient in the corresponding collection period after receiving the detection data of the target patient in one collection period, and transmits the detection processing data to the intensive care warning platform.

[0041] S11: binary conversion is performed on the detection data to obtain first processing data, and every 16 characters in the first processing data are taken as a first processing array in sequence from left to right to obtain a plurality of first processing arrays;

[0042] According to the position of each first processing array in the first processing data, all the obtained first processing arrays are sequentially marked as A1, A2,..., Aa from left to right, and a≥1;

[0043] S12: hexadecimal conversion is performed on the first processing array A1 to obtain a second processing array B1, and all characters constituting the second processing array B1 are sequentially marked as C1, C2, C3 and C4 in sequence from left to right.

[0044] S13: non-digit judgment is performed on the character C1, if the character C1 is a digit, at this time, the character 0 is selected as the alternative identification character of the character C1, and the first numerical interval and the second numerical interval matching are performed on the character C1, if the character C1 matches the first numerical interval, first, the result of 4-C1 is calculated, the two-bit binary number of the result is taken as the alternative string of the character C1, then the alternative identification character and the alternative string of the character C1 are spliced in the order of the alternative identification character and the alternative string to obtain the alternative sequence of the character C1;

[0045] if the character C1 matches the second numerical interval, the result of C1-4 is calculated, and the two-bit binary number of the result is taken as the alternative sequence of the character C1;

[0046] if the character C1 is 9 or 4, at this time, the four-bit binary number of the character C1 is taken as the alternative sequence of the character C1;

[0047] In the present application, the first numerical interval is [0, 4), and the second numerical interval is (4, 8];

[0048] if the character C1 is an English character, at this time, one of the strings 10 or 11 is randomly selected as the alternative identification character of the character C1, and the first letter interval and the second letter interval matching are performed on the character C1, if the letter C1 matches the first letter interval, first, according to the order of the 26 capital English letters, the number D1 of capital English letters between the character C1 and the capital English letter C is obtained, the two-bit binary number of the number D1 is taken as the alternative string of the character C1, then the alternative identification character and the alternative string of the character C1 are spliced in the order of the alternative identification character and the alternative string to obtain the alternative sequence of the character C1;

[0049] if the letter C1 matches the second letter interval, first, according to the order of the 26 capital English letters, the number E1 of capital English letters between the character C1 and the capital English letter C is obtained, the two-bit binary number of the result of E1+2 is taken as the alternative string of the character C1, then the alternative identification character and the alternative string of the character C1 are spliced in the order of the alternative identification character and the alternative string to obtain the alternative sequence of the character C1;

[0050] In the present application, the first letter interval is [A, C), and the second letter interval is (D, F], it should be noted that the first letter interval [A, C) means that the character can only be one of the capital English letters A or B, and the second letter interval (D, F] means that the character can only be one of the capital English letters E or F;

[0051] If the character C1 is a capital English letter C, first find the row number D1 and the column number D2 of the control character C in the ASCII code table, then extract the control character with the row number D1 and the column number D2-1 in the ASCII code table as the replacement sequence of the character C1;

[0052] If the character C1 is a capital English letter D, first find the row number D1 and the column number D2 of the control character C in the ASCII code table, then extract the control character with the row number D1 and the column number D2-2 in the ASCII code table as the replacement sequence of the character C1;

[0053] S14: sequentially obtain the replacement sequences of characters C2, C3 and C4 according to S13;

[0054] S15: according to the order of characters C1, C2, C3 and C4, splice the replacement characters of characters C1, C2, C3 and C4 to obtain the replacement sequence of the second processing array B1, and in the splicing process, for the character which is a number, if the character is a number 4, use a space to replace the replacement sequence thereof, for example, if the character C1 is a character 1, C2 is 4, C3 is 7, and C4 is 8, the corresponding replacement sequence is 111, 0100, 10, and 11, and the replacement sequence is 1111011;

[0055] S16: sequentially obtain the second processing arrays B2, B3,..., Ba of the first processing arrays A2, A3,..., Aa according to S12, and sequentially obtain the replacement sequences of the second processing arrays B2, B3,..., Ba according to S13 to S15;

[0056] Splice the replacement sequences of the second processing arrays B1, B2,..., Ba in the order of the second processing arrays B1, B2,..., Ba to obtain the detection processing data of the target patient corresponding to the collection period;

[0057] The monitoring and early warning platform is used for receiving and storing the real-time physical communication data of the target patient, and is also used for periodically receiving the detection processing data of the target patient, restoring the detection processing data of the target patient corresponding to each collection period after receiving the detection processing data of the target patient corresponding to each collection period, and storing the detection data;

[0058] The monitoring and early warning platform pre-stores a trained early warning model for monitoring and early warning of critical patients, periodically inputs the detection data and communication data of the target patient stored in the monitoring and early warning platform into the early warning model, and outputs early warning result data from the early warning model, wherein the early warning result data contains an early warning signal, the early warning signal can only be one of the numbers 1 or 0, when the early warning signal is 1, it indicates that the target patient may have a risk in the future P1 time, and when the early warning signal is 0, it indicates that the probability of the target patient having a risk in the future P1 time is 0, in the present application, P1 is a preset risk triggering time length;

[0059] When the early warning signal is 1, the early warning result data also contains a risk type and a risk level, in the present application, the risk type refers to the type of risk triggered by the target patient in the future P1 time, in the present application, the risk level is divided into levels one, two and three, the risk level one indicates that the probability of the target patient having a risk in the future P1 time is more than 50%, the risk level two indicates that the probability of the target patient having a risk in the future P1 time is more than 30%, and the risk level three indicates that the probability of the target patient having a risk in the future P1 time is less than or equal to 30%;

[0060] When the early warning signal in the early warning result data is 1, the monitoring and early warning platform displays the early warning result data to the monitoring personnel and uses a sound alarm to remind the monitoring personnel to pay attention to the condition of the target patient and take corresponding preparation measures;

[0061] In the description of the specification, the description of the terms "one embodiment", "example", "specific example" and the like means that the specific features, structures, materials or characteristics described in connection with the embodiment or example are contained in at least one embodiment or example of the present application. In the specification, the illustrative description of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described can be combined in any one or more embodiments or examples in a suitable manner.

[0062] The above is only an example and description of the present application, and those skilled in the art can make various modifications or supplements or use similar ways to replace the described specific embodiments, as long as they do not deviate from the scope defined by the present application, which shall belong to the protection scope of the present application.

[0063] The above has described one embodiment of the present application in detail, but the content described is only the preferred embodiment of the present application and cannot be considered as limiting the scope of the present application. Any equivalent changes and improvements made within the scope of the present application shall still belong to the patent coverage scope of the present application.

Claims

1. An artificial intelligence based intensive care warning system characterized in that, The method comprises the following steps: a periodic acquisition unit is configured to acquire detection data of a target patient to obtain detection data of the target patient in a corresponding acquisition period every interval; a detection data processing unit is configured to process the detection data of the target patient in a corresponding acquisition period according to a preset processing rule to obtain detection processing data of the target patient in the corresponding acquisition period after receiving the detection data of the target patient in the corresponding acquisition period; a monitoring and early warning platform is configured to periodically receive the detection processing data of the target patient, restore the detection processing data of the target patient in a corresponding acquisition period to obtain detection data of the target patient in the corresponding acquisition period after receiving the detection processing data of the target patient in the corresponding acquisition period, and store the detection data of the target patient in the corresponding acquisition period; the processing rule for obtaining the detection processing data of the target patient in an acquisition period is as follows: S11: binary conversion is performed on the detection data to obtain first processing data, and every 16 characters in the first processing data are taken as a first processing array in a left-to-right order to obtain a plurality of first processing arrays; all the first processing arrays obtained are sequentially marked as A1, A2,..., Aa from left to right according to the positions of each first processing array in the first processing data, and a≥1; S12: hexadecimal conversion is performed on the first processing array A1 to obtain a second processing array B1, and all the characters constituting the second processing array B1 are sequentially marked as C1, C2, C3 and C4 in a left-to-right order; S13: non-digit judgment is performed on the character C1, if the character C1 is a digit other than 9 or 4, the character 0 is selected as a replacement identification character of the character C1, and first value interval and second value interval matching are performed on the character C1, if the character C1 matches the first value interval, the result of 4-C1 is calculated first, the two-bit binary number of the result is taken as a replacement string of the character C1, and then the replacement identification character and the replacement string of the character C1 are spliced in the order of the replacement identification character and the replacement string to obtain a replacement sequence of the character C1; if the character C1 matches the second value interval, the result of C1-4 is calculated, and the two-bit binary number of the result is taken as a replacement sequence of the character C1; if the character C1 is 9 or 4, the four-bit binary number of the character C1 is taken as a replacement sequence of the character C1; S14: replacement sequences of the characters C2, C3 and C4 are sequentially obtained according to S13; S15: the replacement characters of the characters C1, C2, C3 and C4 are spliced in the order of the characters C1, C2, C3 and C4 to obtain a replacement sequence of the second processing array B1, and in the splicing process, for a character that is a digit, if the character is a digit 4, a space is used to replace the replacement sequence of the character; S16: second processing arrays B2, B3,..., Ba of the first processing arrays A2, A3,..., Aa are sequentially obtained according to S12, and replacement sequences of the second processing arrays B2, B3,..., Ba are sequentially obtained according to S13 to S15. The alternative sequence of the second processing array B1, B2, …, Ba is spliced in the order of the second processing array B1, B2, …, Ba to obtain detection processing data of the target patient corresponding to the collection period.

2. The artificial intelligence based intensive care warning system as claimed in claim 1, wherein, The first communication module is further included for collecting and generating real-time communication data of the target patient.

3. The artificial intelligence-based intensive care warning system according to claim 1, wherein, In S13, if the character C1 is an English character other than the capital English letters C and D, an alternative identification character of the character C1 is randomly selected from the string 10 or 11, and the character C1 is matched with the first letter interval and the second letter interval. If the character C1 is matched with the first letter interval, the number Z1 of capital English letters between the character C1 and the capital English letter C is obtained according to the order of the 26 capital English letters, and the two-bit binary number of the number Z1 is taken as the alternative string of the character C1. Then, the alternative identification character and the alternative string of the character C1 are spliced in the order of the alternative identification character and the alternative string to obtain the alternative sequence of the character C1. If the character C1 is matched with the second letter interval, the number E1 of capital English letters between the character C1 and the capital English letter C is obtained according to the order of the 26 capital English letters, and the two-bit binary number of the number E1+2 is taken as the alternative string of the character C1. Then, the alternative identification character and the alternative string of the character C1 are spliced in the order of the alternative identification character and the alternative string to obtain the alternative sequence of the character C1. If the character C1 is the capital English letter C, the row number D1 and the column number D2 corresponding to the capital English letter C are found in the ASCII code table, and the control character with the row number D1 and the column number D2-1 is extracted from the ASCII code table, which is taken as the alternative sequence of the character C1. If the character C1 is the capital English letter D, the row number D3 and the column number D4 corresponding to the capital English letter D are found in the ASCII code table, and the control character with the row number D3 and the column number D4-2 is extracted from the ASCII code table, which is taken as the alternative sequence of the character C1.

4. The artificial intelligence-based intensive care warning system according to claim 3, wherein, The first numerical interval is [0, 4), the second numerical interval is (4, 8], the first letter interval is [A, C), and the second letter interval is (D, F].

5. The artificial intelligence-based intensive care warning system according to claim 1, wherein, The monitoring and warning platform is also used for receiving and storing the real-time communication data of the target patient.

6. The artificial intelligence-based intensive care warning system according to claim 1, wherein, The monitoring and warning platform pre-stores a warning model, and periodically inputs the detection data and the communication data of the target patient stored in the monitoring and warning platform into the warning model to output warning result data, wherein the warning result data includes a warning signal.

7. The artificial intelligence-based intensive care warning system according to claim 6, wherein, The warning signal is one of the numbers 1 or 0. When the warning signal is 1, it indicates that the target patient may have a risk in the future P1 time. When the warning signal is 0, it indicates that the probability of the target patient having a risk in the future P1 time is 0, and P1 is a preset risk trigger duration.

8. The artificial intelligence-based intensive care warning system according to claim 7, wherein, When the early warning signal is 1, the early warning result data further contains a risk type and a risk level.

9. The artificial intelligence-based intensive care warning system according to claim 7, wherein, When the early warning signal in the early warning result data is 1, the monitoring early warning platform displays the early warning result data to a monitoring personnel and uses a sound alarm to remind the monitoring personnel to pay attention to the target patient.

Images