Intensive care early warning system based on artificial intelligence

Abstract

The invention discloses an intensive care early warning system based on artificial intelligence, and relates to the technical field of early warning, a first communication module is arranged to collect physical sign parameters of a target patient in real time, a second communication module is arranged to periodically collect detection data of the target patient, and the detection data is processed after collection; in the processing process, binary and hexadecimal conversion is carried out successively to obtain a sequence composed of numbers and English letters, then the numbers and the English letters are processed independently, for processing of the numbers, the numbers and 4 are subjected to addition and subtraction operation, binary numbers of the numbers are converted and limited in 00, 01, 10 and 11 in sequence, and then the numbers are processed independently. According to the method, control characters are transversely selected by English letters according to an ASCII (American Standard Code for Information Interchange) comparison table to serve as substitute characters, so that the recognizable degree of data can be effectively reduced, the risk that the data are maliciously cracked and restored is reduced, and the security and confidentiality of the data in the transmission and storage process are ensured.

Description

Technical Field

[0001] The present invention relates to the field of early warning technology, and in particular to an artificial intelligence-based intensive care early warning system. Background Art

[0002] In the modern medical field, monitoring of critically ill 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 always at great risk.

[0003] When monitoring critically ill patients, comprehensive and integrated data information from multiple aspects is required. Among them, vital sign data such as heart rate, blood pressure, respiratory rate, blood oxygen saturation, body temperature, etc. can be directly measured and obtained in real time through bedside monitors and other equipment. These data provide an intuitive basis for understanding the patient's current physiological state. However, another type of key data - test data, requires the help of professional medical institutions or specific testing machines to complete the test. Test data covers test results such as blood and urine, as well as medical imaging data (chest X-rays, CT scans, MRI, etc.) and special examination results such as electrocardiograms and electroencephalograms. These test data can reveal the root causes of the patient's illness, disease progression, and potential complications from a deeper level.

[0004] With the rapid development of medical information technology, in order to achieve efficient medical collaboration and resource integration, test data often needs to be transmitted between different medical devices, department systems and telemedicine platforms. During the network transmission process, the security of test data faces severe challenges. Since test data contains a large amount of patient privacy data, such as patient identity information, disease diagnosis details, genetic test results and other sensitive content, if these data are leaked, it will bring many adverse consequences to patients.

[0005] In order to solve the above problems, the present invention proposes a solution. Summary of the invention

[0006] The purpose of the present invention is to provide an artificial intelligence-based critical care early warning system to solve the problems raised in the above background technology;

[0007] The purpose of the present invention can be achieved through the following technical solutions:

[0008] An artificial intelligence-based critical care early warning system, comprising:

[0009] A periodic collection unit, used for collecting the test data of the target patient at intervals of a collection period to obtain the test data of the target patient in the corresponding collection period;

[0010] After receiving the test data of a target patient in a collection period, the test data processing unit processes the test data according to a preset processing rule to obtain the test processing data of the target patient in the corresponding collection period;

[0011] The monitoring and early warning platform is used to periodically receive the detection and processing data of the target patient. After receiving the detection and processing data of the target patient in each collection period, the detection data is restored to obtain the detection data of the target patient in the corresponding collection period, and the detection data is stored.

[0012] Furthermore, it also includes a first communication module, which is used to collect monitoring values ​​of several vital sign parameters of the target patient in real time to generate real-time vital sign communication data of the target patient.

[0013] Furthermore, the processing rules for obtaining the detection processing data of the target patient in a collection cycle are as follows:

[0014] S11: performing binary conversion on the detection data to obtain first processed data, and taking every 16 characters in the first processed data as a group of first processed arrays from left to right to obtain a plurality of groups of first processed arrays;

[0015] According to the position of each group of first processed arrays in the first processed data, all the obtained groups of first processed arrays are marked as A1, A2, ..., Aa from left to right, where a≥1;

[0016] S12: converting the first processed array A1 into hexadecimal to obtain a second processed array B1, and marking all characters constituting the second processed array B1 as C1, C2, C3 and C4 in order from left to right;

[0017] S13: Perform a non-digit determination on the character C1. If the character C1 is a digit, character 0 is selected as a replacement identification character of the character C1, and the character C1 is matched with the first numerical interval and the second numerical interval. If the character C1 matches the first numerical interval, the result of 4-C1 is first calculated, and the two binary digits of the result are used as the replacement string of the character C1. Then, the replacement identification character and the replacement string of the character C1 are concatenated 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 binary digits of the result are used 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 used 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 a replacement identification character of the character C1, and the first letter interval and the second letter interval are matched for the character C1. If the letter C1 matches the first letter interval, the number D1 of uppercase English letters between the character C1 and the uppercase English letter C is first obtained according to the order of the 26 uppercase English letters, and the two-digit binary number of the number D1 is used as the replacement string of the character C1. Then, the replacement identification character and the replacement string of the character C1 are concatenated in the order of the replacement identification character and the replacement string to obtain a replacement sequence of the character C1.

[0021] If the letter C1 matches the second letter interval, firstly, according to the order of the 26 uppercase English letters, the number E1 of the uppercase English letters between the character C1 and the uppercase English letter C is obtained, and the two-digit binary number of the result of the number E1+2 is used as the replacement string of the character C1, and then the replacement identification character of the character C1 and the replacement string are concatenated in the order of the replacement identification character and the replacement string to obtain a replacement sequence of the character C1;

[0022] If the character C1 is an uppercase English letter C, first find the row number D1 and column number D2 corresponding to the control character C from the ASCII code comparison table, then extract the control character with the row number D1 and the column number D2-1 from the ASCII code comparison table, and use the control character as the replacement sequence for the character C1;

[0023] If the character C1 is an uppercase English letter D, first find the row number D1 and column number D2 corresponding to the control character C from the ASCII code comparison table, then extract the control character with the row number D1 and the column number D2-2 from the ASCII code comparison table, and use the control character as the replacement sequence for the character C1;

[0024] S14: Obtain replacement sequences of characters C2, C3 and C4 in sequence according to S13;

[0025] S15: in the order of characters C1, C2, C3 and C4, the replacement characters of characters C1, C2, C3 and C4 are concatenated to obtain a replacement sequence of the second processing array B1. In the concatenation process, for a character that is a number, if the character is the number 4, a space is used to replace its replacement sequence;

[0026] S16: according to S12, the second processed array B2, B3, ..., Ba of the first processed array A2, A3, ..., Aa is obtained in sequence, and according to S13 to S15, the replacement sequence of the second processed array B2, B3, ..., Ba is obtained in sequence;

[0027] The replacement number series of the second processing arrays B1, B2, ..., Ba are spliced ​​in the order of the second processing arrays B1, B2, ..., Ba to obtain the detection processing data of the target patient in the corresponding acquisition period.

[0028] Furthermore, a warning model is pre-stored in the monitoring and warning platform, and the monitoring and warning platform periodically inputs the detection data and vital sign communication data of the target patient stored in the monitoring and warning platform into the warning model, and the warning model outputs warning result data, which includes a warning signal.

[0029] Beneficial effects of the present invention:

[0030] The present invention collects the vital sign parameters of the target patient in real time by setting a first communication module, and collects the detection data of the target patient periodically by setting a second communication module, and processes the detection data after collection. In the processing process, binary and hexadecimal conversions are successively performed to obtain a sequence composed of numbers and English letters, and then the numbers and English letters are processed independently. For the processing of numbers, the numbers and 4 are added and subtracted to limit the binary number conversion of the numbers to 00, 01, 10, and 11 in turn. The English letters refer to the ASCII code comparison table to select control characters as replacement characters horizontally. In this way, the recognizability of the data can be effectively reduced, the risk of data being maliciously cracked and restored can be reduced, and the security and confidentiality of the data during transmission and storage can be guaranteed;

[0031] After this encryption process, even if the data is accidentally leaked, it is difficult to directly restore the original test data and vital sign parameters due to the specially transformed numbers and replaced English letters, which greatly increases the difficulty of data cracking. Since the data is highly confusing, its real information is deeply hidden, reducing the risk of patient privacy exposure due to data leakage and effectively protecting patients' personal medical information. BRIEF DESCRIPTION OF THE DRAWINGS

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

[0033] Figure 1 It is a system block diagram of the present invention. DETAILED DESCRIPTION

[0034] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

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

[0036] The first communication module is used to collect monitoring values ​​of several vital sign parameters of the target patient in real time, wherein the vital sign parameters include heart rate, blood pressure, respiratory rate, blood oxygen saturation, body temperature, etc.;

[0037] Generate real-time vital sign communication data of the target patient according to the monitoring values ​​of all vital sign parameters collected in real time, and transmit the vital sign communication data to the monitoring and early warning platform;

[0038] The second communication module is used to periodically collect test data of the target patient, wherein the test 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, electrolytes, etc.), coagulation function indicators, inflammatory markers (such as C-reactive protein, procalcitonin), electrocardiogram results, and electroencephalogram results;

[0039] The second communication module includes a periodic collection unit and a detection data processing unit, wherein the periodic collection unit collects detection data of the target patient at intervals of a collection period to obtain detection data of the target patient in the corresponding collection period, and transmits the detection data to the detection data processing unit;

[0040] After receiving the detection data of the target patient in each collection period, the detection data processing unit processes the detection data according to the preset processing rules to obtain the detection processing data of the target patient in the corresponding collection period, and transmits the detection processing data to the monitoring and early warning platform. The processing rules are as follows:

[0041] S11: performing binary conversion on the detection data to obtain first processed data, and taking every 16 characters in the first processed data as a group of first processed arrays from left to right to obtain a plurality of groups of first processed arrays;

[0042] According to the position of each group of first processed arrays in the first processed data, all the obtained groups of first processed arrays are marked as A1, A2, ..., Aa from left to right, where a≥1;

[0043] S12: converting the first processed array A1 into hexadecimal to obtain a second processed array B1, and marking all characters constituting the second processed array B1 as C1, C2, C3 and C4 in order from left to right;

[0044] S13: Perform a non-digit determination on the character C1. If the character C1 is a digit, character 0 is selected as a replacement identification character of the character C1, and the character C1 is matched with the first numerical interval and the second numerical interval. If the character C1 matches the first numerical interval, the result of 4-C1 is first calculated, and the two binary digits of the result are used as the replacement string of the character C1. Then, the replacement identification character and the replacement string of the character C1 are concatenated in the order of the replacement identification character and the replacement string to obtain a replacement 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 binary digits of the result are used as the replacement sequence of the character C1;

[0046] If the character C1 is 9 or 4, the four-bit binary number of the character C1 is used as the replacement sequence of the character C1;

[0047] In this 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, a string is randomly selected from the string 10 or 11 as a replacement identification character of the character C1, and the first letter interval and the second letter interval are matched for the character C1. If the letter C1 matches the first letter interval, the number D1 of uppercase English letters between the character C1 and the uppercase English letter C is first obtained according to the order of the 26 uppercase English letters, and the two-digit binary number of the number D1 is used as the replacement string of the character C1. Then, the replacement identification character and the replacement string of the character C1 are concatenated in the order of the replacement identification character and the replacement string to obtain a replacement sequence of the character C1.

[0049] If the letter C1 matches the second letter interval, firstly, according to the order of the 26 uppercase English letters, the number E1 of the uppercase English letters between the character C1 and the uppercase English letter C is obtained, and the two-digit binary number of the result of the number E1+2 is used as the replacement string of the character C1, and then the replacement identification character of the character C1 and the replacement string are concatenated in the order of the replacement identification character and the replacement string to obtain a replacement 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] indicates that the character can only be one of the uppercase English letters A or B, and the second letter interval (D, F] indicates that the character can only be one of the uppercase English letters E or F.

[0051] If the character C1 is an uppercase English letter C, first find the row number D1 and column number D2 corresponding to the control character C from the ASCII code comparison table, then extract the control character with the row number D1 and the column number D2-1 from the ASCII code comparison table, and use the control character as the replacement sequence for the character C1;

[0052] If the character C1 is an uppercase English letter D, first find the row number D1 and column number D2 corresponding to the control character C from the ASCII code comparison table, then extract the control character with the row number D1 and the column number D2-2 from the ASCII code comparison table, and use the control character as the replacement sequence for the character C1;

[0053] S14: Obtain replacement sequences of characters C2, C3 and C4 in sequence according to S13;

[0054] S15: in the order of characters C1, C2, C3 and C4, the replacement characters of characters C1, C2, C3 and C4 are spliced ​​to obtain the replacement sequence of the second processing array B1. In the splicing process, for the characters that are digits, if the character is the digit 4, a space is used to replace its replacement sequence. For example, if character C1 is character 1, C2 is 4, C3 is 7, and C4 is 8, the corresponding replacement sequence is 111, 0100, 10, 11, and the replacement sequence is 1111011;

[0055] S16: according to S12, the second processed array B2, B3, ..., Ba of the first processed array A2, A3, ..., Aa is obtained in sequence, and according to S13 to S15, the replacement sequence of the second processed array B2, B3, ..., Ba is obtained in sequence;

[0056] splicing the replacement number series 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 in the corresponding acquisition period;

[0057] The monitoring and early warning platform is used to receive and store the real-time vital sign communication data of the target patient. The monitoring and early warning platform is also used to periodically receive the detection and processing data of the target patient. After receiving the detection and processing data of the target patient in each collection period, the detection and processing data is restored to obtain the detection data of the target patient in the corresponding collection period, and the detection data is stored;

[0058] The monitoring and early warning platform pre-stores a trained early warning model for monitoring critically ill patients. The monitoring and early warning platform periodically inputs the detection data and vital sign communication data of the target patient stored in the monitoring and early warning platform into the early warning model, and the early warning model outputs early warning result data. The early warning result data includes 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 be at risk within the next P1 time. When the early warning signal is 0, it indicates that the probability of the target patient being at risk within the next P1 time is 0. In this application, P1 is a preset risk trigger duration.

[0059] When the warning signal is 1, the warning result data also includes the risk type and risk level. The risk type in this application refers to the type of risk caused by the target patient in the future P1 time. In this application, the risk level is divided into one, two, and three levels. The risk level is one, indicating that the probability of the target patient to have a risk in the future P1 time is more than 50%. The risk level is two, indicating that the probability of the target patient to have a risk in the future P1 time is more than 30%. The risk level is three, indicating that the probability of the target patient to have a risk in the future P1 time is less than or equal to 30%;

[0060] When the warning signal in the warning result data is 1, the monitoring warning platform will display the warning result data to the monitoring personnel, and use a sound alarm to remind the monitoring personnel to pay attention to the situation of the target patient and take corresponding preparatory measures;

[0061] In the description of the specification, the description with reference to the terms "one embodiment", "example", "specific example", etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representation 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 contents are merely examples and explanations of the present invention. Those skilled in the art may make various modifications or additions to the specific embodiments described or replace them in similar ways. As long as they do not deviate from the scope defined by the invention, they shall all fall within the protection scope of the present invention.

[0063] The above is a detailed description of an embodiment of the present invention, but the content is only a preferred embodiment of the present invention and cannot be considered to limit the scope of implementation of the present invention. All equivalent changes and improvements made within the scope of the present invention should still fall within the scope of the patent coverage of the present invention.

Claims

1. An artificial intelligence-based critical care early warning system, characterized in that: include: A periodic collection unit, used for collecting the test data of the target patient at intervals of a collection period to obtain the test data of the target patient in the corresponding collection period; After receiving the test data of a target patient in a collection period, the test data processing unit processes the test data according to a preset processing rule to obtain the test processing data of the target patient in the corresponding collection period; The monitoring and early warning platform is used to periodically receive the detection and processing data of the target patient. After receiving the detection and processing data of the target patient in each collection period, the detection data is restored to obtain the detection data of the target patient in the corresponding collection period, and the detection data is stored.

2. The artificial intelligence-based intensive care warning system according to claim 1, characterized in that: It also includes a first communication module, which is used to collect monitoring values ​​of several vital sign parameters of the target patient in real time to generate real-time vital sign communication data of the target patient.

3. The artificial intelligence-based intensive care warning system according to claim 1, characterized in that: The processing rules for obtaining the test processing data of the target patient in a collection cycle are as follows: S11: performing binary conversion on the detection data to obtain first processed data, and taking every 16 characters in the first processed data as a group of first processed arrays from left to right to obtain a plurality of groups of first processed arrays; According to the position of each group of first processed arrays in the first processed data, all the obtained groups of first processed arrays are marked as A1, A2, ..., Aa from left to right, where a≥1; S12: converting the first processed array A1 into hexadecimal to obtain a second processed array B1, and marking all characters constituting the second processed array B1 as C1, C2, C3 and C4 in order from left to right; S13: Perform a non-digit determination on the character C1. If the character C1 is a digit, character 0 is selected as a replacement identification character of the character C1, and the character C1 is matched with the first numerical interval and the second numerical interval. If the character C1 matches the first numerical interval, the result of 4-C1 is first calculated, and the two binary digits of the result are used as the replacement string of the character C1. Then, the replacement identification character and the replacement string of the character C1 are concatenated 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 numerical interval, the result of C1-4 is calculated, and the two binary digits of the result are used as the replacement sequence of the character C1; If the character C1 is 9 or 4, the four-bit binary number of the character C1 is used as the replacement sequence of the character C1; If the character C1 is an English character, a string is randomly selected from the string 10 or 11 as a replacement identification character of the character C1, and the first letter interval and the second letter interval are matched for the character C1. If the letter C1 matches the first letter interval, the number D1 of uppercase English letters between the character C1 and the uppercase English letter C is first obtained according to the order of the 26 uppercase English letters, and the two-digit binary number of the number D1 is used as the replacement string of the character C1. Then, the replacement identification character and the replacement string of the character C1 are concatenated in the order of the replacement identification character and the replacement string to obtain a replacement sequence of the character C1. If the letter C1 matches the second letter interval, firstly, according to the order of the 26 uppercase English letters, the number E1 of the uppercase English letters between the character C1 and the uppercase English letter C is obtained, and the two-digit binary number of the result of the number E1+2 is used as the replacement string of the character C1, and then the replacement identification character of the character C1 and the replacement string are concatenated 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 is an uppercase English letter C, first find the row number D1 and column number D2 corresponding to the control character C in the ASCII code comparison table, then extract the control character with the row number D1 and the column number D2-1 in the ASCII code comparison table, and use the control character as the replacement sequence of the character C1; if the character C1 is an uppercase English letter D, first find the row number D1 and the column number D2 corresponding to the control character C in the ASCII code comparison table, then extract the control character with the row number D1 and the column number D2-2 in the ASCII code comparison table, and use the control character as the replacement sequence of the character C1; S14: Obtain replacement sequences of characters C2, C3 and C4 in sequence according to S13; S15: in the order of characters C1, C2, C3 and C4, the replacement characters of characters C1, C2, C3 and C4 are concatenated to obtain a replacement sequence of the second processing array B1. In the concatenation process, for a character that is a number, if the character is the number 4, a space is used to replace its replacement sequence; S16: according to S12, the second processed array B2, B3, ..., Ba of the first processed array A2, A3, ..., Aa is obtained in sequence, and according to S13 to S15, the replacement sequence of the second processed array B2, B3, ..., Ba is obtained in sequence; The replacement number series of the second processing arrays B1, B2, ..., Ba are spliced ​​in the order of the second processing arrays B1, B2, ..., Ba to obtain the detection processing data of the target patient in the corresponding acquisition period.

4. The artificial intelligence-based intensive care warning system according to claim 3, characterized in that: 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 critical care early warning system according to claim 1, characterized in that: The monitoring and early warning platform is also used to receive and store real-time vital sign communication data of target patients.

6. The artificial intelligence-based intensive care warning system according to claim 1, characterized in that: The monitoring and early warning platform pre-stores an early warning model, which periodically inputs the detection data and vital sign communication data of the target patient stored in the monitoring and early warning platform into the early warning model, and the early warning model outputs early warning result data, which includes an early warning signal.

7. The artificial intelligence-based critical care early warning system according to claim 6, characterized in that: The warning signal is one of the numbers 1 or 0. When the warning signal is 1, it indicates that the target patient may be at risk within the future P1 time. When the warning signal is 0, it indicates that the probability of the target patient being at risk within the future P1 time is 0. P1 is the preset risk trigger duration.

8. The artificial intelligence-based intensive care warning system according to claim 7, characterized in that: When the warning signal is 1, the warning result data also includes the risk type and risk level.

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

Images