INTELLIGENT AUTOMATIC OXYGEN THERAPY SYSTEM

Abstract

The Intelligent Automatic Oxygen Therapy System provides a device that allows the automatic and intelligent dosage of the percentage of an oxygen/air gas mixture and the flow delivered to each patient through non-invasive oxygen therapy procedures, based on the analysis of several measured variables that confirm the SpO2 value before taking any action. This device allows to measure biomedical signs with the main object of monitoring the oxygen saturation (SpO2), confirming its value with the analysis of the mentioned signs according to their concordance and interrelationship with each other. The equipment through artificial intelligence detects events that can occur due to the movement or for misplaced sensors. It also keeps and analyzes the records of the patient, evaluates the alarms in an intelligent way by correlating all acquired data; with this analysis the equipment can automatically and reliably provide the oxygen/air mixture adequate for each patient, with five operation modes. It activates the processed and valid alarms in an intelligent way and informs in a timely manner to the personal staff about possible found pathologies.

Claims

1. An intelligent automatic oxygen therapy system for a patient comprising: at least one flow actuator configured to regulate a flow of air, oxygen, or a combination thereof provided to a patient; at least one flow sensor configured to measure the flow of air, oxygen, or the combination thereof provided to said patient; an oxygen saturation sensor configured to measure oxygen saturation levels of said patient; and a processing unit configured to: receive measurements from said at least one flow sensor; receive oxygen saturation measurements from oxygen saturation sensor; analyze the received oxygen saturation measurements against at least one of an oxygen saturation predefined range or measurements of at least one vital sign of the patient; control said at least one flow actuator to adjust the flow of air, oxygen, or the combination thereof based on said analysis to maintain the oxygen saturation level of said patient within a desired range.

2. The intelligent automatic oxygen therapy system according to claim 1, wherein said at least one vital sign of the patient comprises an oxygen saturation level, a blood pressure, a volume of an organ or body of the patient, a heart rhythm, a concentration or partial pressure of carbon dioxide in respiratory gases, or a temperature.

3. The intelligent automatic oxygen therapy system according to claim 2, wherein said oxygen saturation sensor is a pulse oximetry (SpO2) sensor, said blood pressure is measured with a non-invasive blood pressure sensor, said volume of the organ or body of the patient is measured with a plethysmographic sensor, said heart rhythm is measured with an electrocardiogram (ECG) sensor, said concentration or partial pressure of carbon dioxide in respiratory gases is measured with a capnographic sensor, and said temperature is measured with a temperature sensor.

4. The intelligent automatic oxygen therapy system according to claim 1, wherein said at least one flow actuator is also controlled without said analysis.

5. The intelligent automatic oxygen therapy system according to claim 1, wherein said analysis is performed to determine at least one of sudden changes of values caused by movements of the patient, misplaced sensors, sensor failures, connection failures, or measuring errors.

6. The intelligent automatic oxygen therapy system according to claim 1, wherein said analysis is performed by an artificial intelligence level 2.

7. The intelligent automatic oxygen therapy system according to claim 5, wherein alert signals are generated when a problem with a sensor or a connection or a measuring error is detected.

8. The intelligent automatic oxygen therapy system according to claim 1, wherein said analysis comprises learning patterns based on at least one of a history of the patient, a pathology of the patient or interventions of medical staff so that said at least one flow actuator is actuated to optimize an oxygen automatic dosage.

9. The intelligent automatic oxygen therapy system according to claim 1, wherein said at least one flow actuator is coupled to a gas inlet or an oxygen inlet.

10. The intelligent automatic oxygen therapy system according to claim 1, wherein the measurements of said at least one flow sensor are displayed.

11. The intelligent automatic oxygen therapy system according to claim 1, wherein said processing unit is further configured to keep a record of at least one of a total oxygen consumption of the patient, sudden changes in measurements, or alterations in an oxygen monitoring pattern.

12. The intelligent automatic oxygen therapy system according to claim 7, wherein alert signals parameters are set by a physician.

13. The intelligent automatic oxygen therapy system according to claim 6, wherein at least one of possible patient pathologies or controlled flow ranges are suggested to a physician based on the analysis performed by said artificial intelligence level 2.

14. The intelligent automatic oxygen therapy system according to claim 1, wherein at least one of alarms, measurements received by the processing unit or analysis data are communicated to a remote device, displayed on a local device or a combination thereof.

15. The intelligent automatic oxygen therapy system according to claim 1, wherein said processing unit operates in: a manual mode where said at least one flow actuator is manually actuated without said performing analysis; an automatic mode where said at least one flow actuator is automatically actuated based on said analysis until the oxygen saturation level of said patient is within the desired range; and an intelligent mode where said at least one flow actuator is actuated to train the lungs of the patient to stop depending on the oxygen or mixture thereof based on the analysis being performed with at least one of a self-learning procedure or an artificial intelligence procedure that analyzes all information received and generated by the processing unit.

16. The intelligent automatic oxygen therapy system according to claim 15, wherein the generation of alarms is controlled in said intelligent mode to reduce false measurements.

17. The intelligent automatic oxygen therapy system according to claim 16, wherein said alarms are generated only when measurements of the patient's vital signs are outside of predetermined ranges and the measurements have been validated by the intelligent mode.

18. The intelligent automatic oxygen therapy system according to claim 16, wherein the generation of said alarms is further controlled based on a patient's reaction time according to a learning scheme of a patient's behaviour to an oxygen therapy.

19. The intelligent automatic oxygen therapy system according to claim 1, further comprising a display configured to display at least one of oxygen saturation levels or at least one vital sign of the patient; a user interface configured to allow a user to configure the system; and a communication module and an interface configured to communicate with at least one of a local device or a remote device.

20. The intelligent automatic oxygen therapy system according to claim 1, wherein said processing unit is further configured to determine a total volume of the air, oxygen, or the combination thereof delivered to the patient.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

[0030] Further features and advantages of the invention will become apparent from the following detailed description taken in conjunction with the accompanying figures showing illustrative embodiments of the invention, in which:

[0031] FIG. 1 shows an example block diagram showing the interconnection of the devices, according to the present invention.

[0032] FIG. 2 shows an example flow diagram of the solution needed to be implemented in a microprocessor-based system for an intelligent automatic oxygen therapy system, according to the present invention.

[0033] FIG. 3 shows a diagram of possible cases of behavior of a recovering patient indicating limits and control used in the analysis for the behavior of alarms, according to the present invention.

[0034] Throughout the figures, the same reference numbers, and characters, unless otherwise stated, are used to denote like elements, components, portions or features of the illustrated embodiments. The subject invention will be described in detail in conjunction with the accompanying figures, in view of the illustrative embodiments.

[0035] Other objectives, advantages, and new characteristics of the invention will become more evident with the detailed description that follows, when taken together with the drawings that are attached. The description illustrated in the drawings is an example and is not limited to the shown figures.

DETAILED DESCRIPTION OF THE INVENTION

[0036] The system (100) shown in FIG. 1 is illustrative. In some embodiments, one or more components may be optional. In some embodiments, the system may include additional components not shown in the drawing. For example, in some embodiments, the system may take other signs available in the system to include them in the calculation algorithms, may have different ways of communication or may include different forms of artificial intelligence. Besides, in some embodiments, the parts may be placed or organized in a form different than those shown in the example drawing appearing in FIG. 1.

[0037] It is performed a description of embodiments of examples of methods, systems, and apparatus that allow a device to control automatic or intelligently the oxygen therapy, by managing the oxygen flow or is mixture with air (112), based on the primary monitoring of the oxygen saturation 402 obtained through an optical sensor. This measurement is confirmed with other signs to improve reliability and effectiveness of oxygen therapy. For that purpose, the equipment has a configuration interface (114) of the parameters in some embodiments includes: oxygen saturation 402 limits, maximum control-minimum control-maximum alarm-minimum alarm-waiting time-performance time-return bands to normal values. These values are established by the medical staff according to the situation of each patient (107). Besides, in some embodiments includes the configuration of basic information of the patient (age: premature, child, adult, senior, personal data), allows to fix acceptable ranges of the parameters may be performed from the equipment (114) parameters and (109) graphics. In some embodiments it is monitored by computer (116) directly connected or using a wired or wireless communication network (117), besides, in some embodiments through an application for smartphone or portable devices. In addition, in some embodiments, it is monitored by real-time video for remote monitoring of the treating physician.

[0038] In some embodiments, three operation modes are available: manual, automatic and intelligent. Manual mode allows visualizing the measurements and the gas flow being administered to the patient; the opening of the supply valve (112) is manually performed without intervention of the control. According to the information of the SpO2 value, the system activates visual/audible alarms when leaving the safe range established in the parameters; however, in this mode no corrective action is taken. In general, oxygen manual dosage delivers to the patient an amount higher than needed, and the body takes longer to get rid of this support.

[0039] In some embodiments, in automatic mode the equipment performs the oxygen therapy control based on the calculations made by the processor (111), controlling the gas flow through solenoid valves (112) automatically at the right time and speed to compensate the need of the patient until he reaches the adequate saturation margin according to the scheduled performance band (hysteresis) (114). In turn, in some embodiments, if saturation exceeds the maximum established limit, the gas dosage decreases until saturation enters the stabilizing band (hysteresis). The alarm limits (maximum and minimum) are configured so that if there is a high decompensation that has not been quickly stabilized within the control bands, the visual/audible alarms are activated (114) to alert the medical staff. In some embodiments, the professionals based on the information provided by the system (109) and (114) may evaluate the situation and take a drastic measure to stabilize the patient, if necessary, optimizing the system alarms (111).

[0040] Regarding the control decision making (111) the received SpO2 value is analyzed (103) and data and graphics of other measured variables are confirmed (101), by verifying that the obtained values of the optical sensor (110), are commensurate with the patient's condition (107). This way, the equipment in some embodiments dismisses measures which are lost or altered by movements, light, or positioning which affect the pulse oximetry equipment, among others, according to the algorithms programmed in the equipment (111).

[0041] Besides, in some embodiments this mode analyzes the information of the sensors which is processed in several acquisition and data processing modules (101). All the waves are compared using patterns to determine the reliability of measurements; for example, the plethysmographic waveform is compared against a wave pattern through the correlation algorithms (111) and it is determined the reliability of the oxygen saturation measurement.

[0042] Within the automatic mode there are five control options, which are: Oxygen Only, Constant Flow/Variable FiO2, Variable Flow/Constant FiO2, Variable Flow/Variable FiO2, Weaning Mode.

[0043] Within these options the equipment shall operate with the parameters established according to the medical criteria to provide the most appropriate treatment according to the patient's condition, to reach the improvement of the patient and to avoid secondary effects for excessive use of oxygen.

[0044] In the intelligent mode, in some embodiments, the equipment applies self-learning algorithms and artificial intelligence (111), which analyzes all variables of the system and the historical response of the patient, allowing lowering the oxygen levels to the strictly necessary level to help to “train” the lungs to stop depending on the oxygen or its mixture. Aside from this intrinsic functionality of the equipment, in some embodiments the device allows to select mode “weaning” for newborns. In this mode the control (111) includes a gradual and programmed reduction of oxygen or of the mixture with air, so that when the medical staff deems prudent (patient out of critical condition) the equipment shall provide a lower flow to stabilize the patient with a lower amount of oxygen or its mixture until oxygen therapy is withdrawn. This complements the lungs “training”, accelerating in a controlled way the recovery of the patients in less time, and minimizing the adverse effects of excess of oxygen (for example, ROP).

[0045] Similarly, in some embodiments if the equipment through the artificial intelligence algorithms detects that the self-recovery time of the patient is increasing, it alerts the doctor to take necessary actions by readjusting the corresponding parameters. If it is not enough, in some embodiments, it allows providing guides to the doctors to analyze certain pathologies that may be causing said condition.

[0046] Besides, in some embodiments, in the intelligent mode the equipment has a storage data system (1119 which generates a historic record of the behavior of the patient, considering the time lapses and the values that were out of the established acceptable range. This historic record has two uses; first one, in some embodiments allows the doctor to manage a record of the evolution of the patient of the medical history. In second place, it allows a feed back to the control system regarding the response times of the patients and understand their behavior through artificial intelligence algorithms (111), in some embodiments mode “weaning” may be applied to some adult patients to untie them from oxygen therapy assistance.

[0047] In the intelligent mode, in some embodiments, the information of sensors (101) processed in different acquisition and data processing modules (101) is analyzed. In some embodiments, as an example, the plethysmographic waveforms are analyzed and compared to vasoconstriction and vasodilatation wave patterns through validation algorithms (111), in some embodiments, for example, we proceed in a similar way with the values obtained from capnography to generate the best control strategy.

[0048] In some embodiments, the variables comparison applies algorithms described in their essence and that include the variations applicable to the same (111). In some embodiments, the system (100) has for example, independent modules of ECG (10), oximetry (103), non-invasive blood pressure (102), capnography (105) and temperature (106) which in some embodiments are the acquisition and data processing modules (101). This module delivers the appropriately scaled and filtered signs to the graphics handler module (108) and to the processor of algorithms and learning processor (111).

[0049] In some embodiments, in the intelligent mode, the alarms activation is subject to the verification of algorithms and artificial intelligence (111), in order to avoid false detections; this implies that in the cases when the variable is out of the established ranges in some embodiments it is considered first that the oxygen therapy system detects that the alarm is real and no due to events outside of the normal operation when confirming all variables, waveforms of the system, established limits and hysteresis bands. Besides, as a second step, it considers the history of the patient to analyze whether his answer to similar events represents or not an alarm and whether he can overcome autonomously the variation occurring, so that after verifying the above it determines whether an alarm will be triggered.

[0050] In some embodiments, in this mode there are also five control options: Oxygen Only, Constant Flow/Variable FiO2, Variable Flow/Constant FiO2, Variable Flow/Variable FiO2, and Weaning Mode.

[0051] Within these alternatives the equipment works intelligently based on the analysis of the data acquired from the behavior of each patient, limits fixed by the doctors, and recovery objectives.

[0052] In some embodiments, for any of these three operation modes: Manual, Automatic, and Intelligent, the system performs a validation of numerical values and waves (101).

[0053] In some embodiments, the waves are analyzed with a pattern wave through the calculation of the correlation between them (measurement which allows to express until when these two waveforms are similar).

[0054] After filtering the waves using traditional digital techniques, they are processed for validation (111). First, with the frequency value of each wave, at least three periods of them, are taken following the criteria of Nyquist theorem, they are averaged and normalized. Second, the standard wave is scaled to the period of the input waveform to initiate the analysis. Then the waves correlation method Discrete Wavelet Transform is used to determine whether or not the input waveform is reliable and that the measurements are not incorrect due to movements of patient, sensor displacement, or wrong positioning.

[0055] In some embodiments if the waves are within the correlation limit to be validated, the dynamic time warping algorithm (DTW) is used.

[0056] In some embodiments, if the waves are reliable then the pathologies analysis is performed. For example, in the Plethysmographic Waveform it can be determined through comparison with vasoconstriction or vasodilatation wave patterns, by displaying the tendency of the patient to these patterns. Besides, it is shown the average waveform of waves validated for the analysis. Similarly, from PQRST wave of the ECG it can be determined the ST-segment displacement to determine peaks and valleys and to deliver this information for medical analysis.

[0057] Considering that the variable controlled in the system is the percentage of oxygen saturation SpO2, its validation and reliability are very important, the system (100) in some embodiments, to confirm the validity of the measurement, performs, for example, the following considerations and validations: a) verifies that the plethysmographic wave has peaks and valleys with reliable values, b) compares that the values of heart rate obtained by ECG and the one obtained by the optical sensor, are within acceptable tolerances, c) that the time between the peaks of the plethysmographic wave is similar to the time of peaks of the ECG wave, d) verifies that sudden or out of SpO2 range variations are similar in values and waves of other signs, d) verifies that the graphic of respiratory sign is within the normal patterns, e) verifies the variation of the capnographic sign, f) analysis of the variation of the SpO2 which is out of normal ranges while the other measures are adequate, g) verifies that the values of the gas flow are within the programmed parameters; h) a sudden variation of any of the measures and/or waves. In some embodiments, after all these verifications, the system validates the value of measured SpO2.

[0058] All these conditions, in some embodiments, are evaluated by control algorithms (111) to provide intelligence and reliability to the SpO2 measured value and to act appropriately in the flow control (112) of the gas/air and/or oxygen coming from a source external to the system (113). This allows the system, in some embodiments, to control for example: gas flow, wait time for system stabilization, autoregulation by changing the variables of acting time and gas flow speed, auto-adjust of performance bands (hysteresis), generation of optimized alarms for the system and the patient, evolution of the patient, improper placement or loosening of sensors, evolution prediction, behavior extrapolation, and reference diagnosis.

[0059] In some embodiments, the medical staff initially fixes the maximum and minimum values of flow and gas concentration (108); limits of oxygen saturation; and, according to the patient's condition the oxygen therapy method is established, within the five methods of the equipment: 1. Oxygen Only; 2. Constant Flow/Variable FiO; 3. Variable Flow/Constant FiO2; 4. Variable Flow/Variable FiO2; and 5. Weaning Mode. Besides, the doctor fixes the estimated self-recovery times that may be initially expected, before the equipment starts working.

[0060] In some embodiments the system registers the self-recovery times of the patient in either compensation or decompensation depending on the patient's response during a period of analysis, these values are analyzed and averaged to establish the new self-recovery time. The increases or decreases of the self-recovery times are shown on screen through trend graphs of each one for the corresponding medical analysis.

[0061] Besides, the historical data of auto self-recovery times, flow, concentrations, and evolution of oxygen saturation (SpO2) influence the implemented expert system through Fuzzy logic; thus, continuously optimizing the flow rate and concentration of the gas mixture within the oxygen therapy method, either so that the patient's oxygen saturation remains within the established parameters and/or to reduce the gas dependence.

[0062] In some embodiments, the alarms are intelligently controlled, by determining the auto self-recovery times through the expert system and the comparison between the biomedical waves of the patient with the pattern waves, performing a quantitative and qualitative analysis of data to establish the optimal times of alarms activation and to increase reliability of the system.

[0063] In some embodiments, the intelligent alarms (111), depending on the behavior of the patient, act according to the following algorithm.

[0064] The system continuously verifies the reliability of the numeric values of vital sings of the patient among sensors, such as respiration per minute of ECG and SpO2 waves. In addition, each measured waveform is analyzed with standard waves, using the analysis algorithms described above to determine their correlation and thus their reliability.

[0065] If there is a difference in numerical values greater than a set percentage and the correlation of the waveforms in all variables is low, we wait for a time according to the self-recovery time and reanalyze the data. If the conditions are maintained, a sensor check alarm (movement, incorrect placement, or displacement) is presented.

[0066] If the correlation in the waveform of a single variable is low, we wait for a time shorter than the self-recovery time and then the wave is reanalyzed. If the correlation remains low, an alarm is generated from that sensor.

[0067] Once the data is validated, the system is verifying that the oxygen saturation is within the programmed limits, acting as follows in case of deviations from them:

1) If the patient recovers within the self-recovery time, no alarm is generated.
2) If the patient does not return to the hysteresis band within the self-recovery time, and the numerical values are in range and the waveforms have an adequate correlation in the automatic and intelligent modes, the system calculates the gas mixture supply values by means of Fuzzy logic.

[0068] The calculated values of the new set point are delivered to a PID controller that regulates the opening or closing of the air and oxygen valves.

[0069] Once the controller acts, two events can occur:

(a) If the patient tends to enter the programmed hysteresis values of oxygen saturation no alarm is generated.
b) If the patient moves away from the programmed values of the hysteresis band, an alarm is generated.

[0070] FIG. 3 exemplifies the above in the case of a decrease of the SpO2 measurement below the lower limit.

[0071] In addition to this, there are the system's own alarms such as low battery, power failure or lack of gas input, among others.

[0072] In some embodiments, for example, these control strategies and algorithms allow to provide the team with the necessary expertise, giving information support to the doctor for his diagnosis and oxygen therapy strategy for each patient by recalibrating the values of the performance bands.

[0073] In some embodiments, the system may be modular being the necessary base both, pulse oximetry and oxygen or its mixture flow control.

[0074] Although the present invention has been described herein with reference to the foregoing exemplary embodiment, this embodiment does not serve to limit the scope of the present invention. Accordingly, those skilled in the art to which the present invention pertains will appreciate that various modifications are possible, without departing from the technical spirit of the present invention.