WO2008036649A2 - System and method for diagnosing a condition of a patient - Google Patents

System and method for diagnosing a condition of a patient Download PDF

Info

Publication number
WO2008036649A2
WO2008036649A2 PCT/US2007/078754 US2007078754W WO2008036649A2 WO 2008036649 A2 WO2008036649 A2 WO 2008036649A2 US 2007078754 W US2007078754 W US 2007078754W WO 2008036649 A2 WO2008036649 A2 WO 2008036649A2
Authority
WO
WIPO (PCT)
Prior art keywords
patient
data
waveform
condition
identifying
Prior art date
Application number
PCT/US2007/078754
Other languages
French (fr)
Other versions
WO2008036649A3 (en
Inventor
Desmond Jordan
Original Assignee
The Trustees Of Columbia University In The City Of New York
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by The Trustees Of Columbia University In The City Of New York filed Critical The Trustees Of Columbia University In The City Of New York
Publication of WO2008036649A2 publication Critical patent/WO2008036649A2/en
Publication of WO2008036649A3 publication Critical patent/WO2008036649A3/en

Links

Classifications

    • AHUMAN NECESSITIES
    • A61MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61BDIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B7/00Instruments for auscultation
    • A61B7/02Stethoscopes
    • A61B7/04Electric stethoscopes
    • AHUMAN NECESSITIES
    • A61MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61BDIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00Measuring for diagnostic purposes; Identification of persons
    • A61B5/48Other medical applications
    • A61B5/4821Determining level or depth of anaesthesia

Definitions

  • the disclosed subject matter relates to a technique for diagnosing a condition of a patient.
  • Various techniques for evaluating the status of a patient can be performed based on vital signs data. For example, the evaluation of patient sound information, such as heart sounds and their associated waveforms, has always been an important part of medical diagnosis. However, professionals in the medical field have not utilized this information as a basic patient monitoring technique in lieu of technically advanced pulse and blood pressure monitors, which are now the standard of care in most ORs, Recovery rooms, or ICUs.
  • the disclosed subject matter samples data relating to physical characteristics of the patient, deriving parameters of a waveform from the data, comparing the parameters of the waveform with reference data, obtains the condition of the patient based upon the comparison, and provides an output relating to the condition of the patient.
  • the sampled data relating to the physical characteristics of the patient can include heart sounds. That data can be sampled at distinct time intervals or distinct milestones. The distinct milestones can include a resting state, a first dosage state, a high dosage state, or an anesthetized state.
  • the reference data can include a database of the physical characteristics of a population of patients, prior condition of the patient, and prior waveform of the patient.
  • Identifying parameters associated with the waveform can include identifying the heart rate of the patient, the amplitude of the waveform, the peak-to- peak spacing of the waveform, and identifying individual heart sounds, such as the Sl, S2, S3, and S4 sounds.
  • the disclosed subject matter also provides a system for diagnosing a condition of a patient.
  • the system includes an apparatus for sampling data relating to physical characteristics of the patient, a processor arranged and configured to derive parameters of a waveform from the data, to compare the parameters of the waveform with reference data, and to obtain a condition of the patient based upon the comparison, and an output device for providing the condition of the patient.
  • FIG. 1 is a diagram illustrating a method implemented in accordance with some embodiments of the disclosed subject matter
  • FIG. 2 shows the original files for one patient given resting, induced, and then high dose sevoforane;
  • FIG. 3 is the tracings of the patient in FIG. 1 once filtered for artifacts and amplified.
  • FIG. 4 is a block diagram of system in accordance with some embodiments of the disclosed subject matter. While the disclosed subject matter will now be described in detail with reference to the Figures, it is done so in connection with the illustrative embodiments.
  • Systems and methods of diagnosing a condition of a patient are provided.
  • the disclosed subject matter samples data relating to physical characteristics of the patient, derives parameters of a waveform from the data, comparing the parameters of the waveform with reference data, obtains the condition of the patient based upon the comparison, and provides an output relating to the condition of the patient.
  • FIG. 1 is a diagram illustrating a method implemented in accordance with some embodiments of the disclosed subject matter.
  • Data relating to physical characteristics of the patient are sampled at step 101.
  • the sampled data relating to the physical characteristics of the patient can include heart sounds. That data can be sampled at distinct time intervals or distinct milestones.
  • the distinct milestones can include a resting state, a first dosage state, a high dosage state, or an anesthetized state.
  • the parameters of a waveform from the data can be derived at step 102. Identifying parameters associated with the waveform can include identifying the heart rate of the patient, the amplitude of the waveform, the peak-to-peak spacing of the waveform, and identifying individual heart sounds, such as the Sl , S2, S3, and S4 sounds.
  • the parameters of the waveform can be compared with reference data at step 103.
  • the reference data can include a database of the physical characteristics of a population of patients, prior condition of the patient, and prior waveform of the patient.
  • the condition of the patient based upon the comparison can be obtained at step 104.
  • the condition of the patient can include a normal state, a bad state, or an impossible state.
  • the condition of the patient can also include changes in the
  • NY02:596797.1 3 state of the patent such as an improving state or a worsening state.
  • the condition of the patient can include states based on time, such as a previous state, an actual state, and an expected state.
  • An output relating to the condition of the patient can be provided at step 105.
  • the output can be in the form of a graphic or text indicating the condition of the patient. Additionally, the output can be a sound indicating that the condition of the patient is worsening or is in a bad or impossible state.
  • the waveform data is heart sounds.
  • Sl or First Heart Sound is made by the closing of the mitral and tricuspid valves.
  • S2 or Second Heart Sound is made by the closing of the aortic and pulmonary valves.
  • S3 or Third heart Sound is made by rapid ventricular filling when mitral and tricuspid open. This sound is typically considered normal in children and adolescents and can be heard at the Apex.
  • S4 or Fourth Heart Sound is made by the Rapid ventricular filling that occurs in atrial systole, and is heard in late diastole. An S4 can clearly be heard at Erb's point. This sound is normally found in very elderly adults but it can also be heard in young children.
  • S3 is a normal variant, while the S4 is not.
  • S3 is considered abnormal in the young adults.
  • the S4 is a normal variant after exercise and heart stress.
  • Erb's point is considered a favorable location to find a right heart S4
  • the apex of the heart is the location for the left S4.
  • the left heart S4 is considered the normal variant, while the right is not.
  • causes of a right S4 include, e.g., an augmented atrial contraction which increases filling during low ventricular compliance states, exercise and heightened cardiac output, and during positive pressure ventilation with a concordant increase in right ventricular afterload.
  • NY02:596797.1 4 recorded before and after induction, and in the intensive care unit (ICU) post surgery.
  • ICU intensive care unit
  • Erb ' s point appeared to be useful to judge heart sounds and changes, followed by the apex and then the right sternum boarder. Similarities and differences between pediatric and adult patients before and after they were anesthetized were studied. While there were differences in anesthetic technique, inhalation versus intravenous, the decreases in after-load and concurrent increase in cardiac output was found to reveal an S3, S4, augmentation or diminution in the recorded heart sounds. It was also possible the institution of positive pressure ventilation may induce auditory changes that could be captured by our recordings. Similarities and differences in adult and pediatric patients given a high dose of anesthesia were also studied.
  • the heart sounds were recorded with the stethoscope. After the sounds were recorded, they were transferred to a Windows computer via a USB infrared adapter. The Littmann Sound Analysis Software was used to determine if the recording was acceptable. If the recording was acceptable, it would be saved in the database.
  • Wave Pad a wave evaluation application
  • the application can amplify the sounds, eliminate background noise, etc. until a desired consistency was achieved.
  • the original file was used in the evaluation of intensity, amplitude and time and inter- patient correlation.
  • FIG. 2 shows the original files for one patient given resting 22, induced 24, and then high dose 26 sevoforane.
  • the heartbeat of a child is heard more clearly when recorded than for an adult.
  • the difference in the thickness of the chest wall in children and adults is a possible explanation.
  • Pediatric patients and adults have a similar pattern of heart sounds when induced with anesthesia. This appears true for inhalation inductions, propofol, and versed fentanyl inductions.
  • the appearance of the S4 occurred whether or not the patient was placed on positive pressure ventilation or not.
  • FIG. 3 is the tracings of the same patient once filtered for artifacts and amplified. While the pattern of heart sounds that occur during induction is still prominent, the clarity for clinical testing as stated above with attendings and residents can easily be done.
  • a computer program is used to execute a digital algorithm that detects first and second heart sounds, defines the systole and diastole, and characterizes the systolic murmur.
  • Heart sounds were recorded in 300 children with a cardiac murmur, using an electronic stethoscope.
  • a digital algorithm was developed for detection of first and second heart sounds. R-waves and T-waves in the electrocardiography were used as references for detection. The sound signal analysis was carried out using the short-time Fourier transform.
  • the first heart sound detection rate, with reference to the R- wave was 100% within 0.05-0.2R-R interval.
  • the second heart sound detection rate between the end of the T- wave and the 0.6R-R interval was 97%.
  • the systolic and diastolic phases of the cardiac cycle could be identified. Because of the overlap between heart sounds and murmur a systolic segment between the first and second heart sounds (20-70%) was selected for murmur analysis. The maximum intensity of the systolic murmur, its average frequency, and the mean spectral power were quantified. The frequency at the point with the highest sound intensity in the spectrum and its time from the first heart sound, the highest frequency, and frequency
  • FIG. 4 is a block diagram of system in accordance with some embodiments of the disclosed subject matter.
  • the system includes an apparatus for sampling data relating to physical characteristics of the patient, a processor arranged and configured to derive parameters of a waveform from the data, to compare the parameters of the waveform with reference data, and to obtain a condition of the patient based upon the comparison, and an output device for providing the condition of the patient.
  • the system diagnoses the condition of the patient based on the heart sounds of the patient.
  • a stethoscope at 403 can be applied to the chest of a patient at 402 such that sounds from the heart at 401 of the patient can be sampled.
  • the apparatus for sampling data relating to physical characteristics of the patient can include a stethoscope.
  • An example of a stethoscope includes the Littmann 4100 Series Electronic stethoscope.
  • the digitizer at 404 which can be connected to the stethoscope at 403, can covert the signal received by the stethoscope at 403 into a digital signal.
  • the digitizer at 404 can be connected to a processor at 405 and the digital signal can be sent to the processor at 405.
  • the processor at 405 is arranged and configured to derive parameters of a waveform from the digital signal, to compare the parameters of the waveform with reference data, and to obtain a condition of the patient based upon the comparison.
  • the processor at 405 can be implemented as a computer microchip, a stand alone computer or collection of networks computing or any device suitable for processing.
  • the disclosed subject matter can be implemented in Perl, C, or other suitable programming language.
  • Software such as the Littmann Sound Analysis Software can be used to record the heart sounds, and Wave Pad, another application, can be used to edit and evaluate the heart sounds.
  • the processor 405 may be distributed over a network in some embodiments.
  • the processor may include a client and remote server on a wired or wireless network.
  • the database at 406 is provided, which may be accessed by the processor at 405.
  • the database can store data relating to the physical characteristics
  • the data stored in the database at 406 can be used to compare the parameters of the waveform of the patient.
  • An inference monitor as described in U.S. Application No. 10/795,724, may be used to identify and/or classify abnormal patients by scanning a patient's records for relevant data, such as heart sounds data, and applying one or more inference rules.
  • the inference monitor may also scan data relating to a population of patients.
  • the inference monitor identifies and/or classifies abnormal events that occurred at or about certain milestones in the therapeutic treatment.
  • a "milestone” or “milestone event” refers herein to a distinguishable event in the course of a patient ' s treatment. It is understood that various types of milestones may be either predefined or user-defined, based on the type of treatment being offered. For instance, with regard to heart sounds, the milestones may be critical periods during surgery, such as resting (prior to anesthesia), anesthetization (at a first level of anesthesia) or highly anesthetized (at a second, higher level of anesthesia), etc.
  • the inference engine may apply the inference rule or rules to patient data that have been collected within a predefined window, such as a 20-minute window, before and/or after any one or more of the milestones.
  • the inference engine may also identify abnormal events in light of milestones, such as demographic data (e.g., the patient ' s age, gender, weight, etc.) and clinical data (e.g., the patient's vital signs, relevant allergic reactions, medications, prosthetics, preexisting medical conditions, relevant diagnoses, type of procedure suggested), and in light of the data, such as the time interval of the events (e.g., between start and stop times) and the drugs that were administered during the operation.
  • demographic data e.g., the patient 's age, gender, weight, etc.
  • clinical data e.g., the patient's vital signs, relevant allergic reactions, medications, prosthetics, preexisting medical conditions, relevant diagnoses, type of procedure suggested
  • the time interval of the events e.g., between start and stop times
  • Inferences identifying events indicative of the severity of a patient's condition include, without limitation, duration of treatment, the type of procedure, demographics, etc.
  • Various techniques may be used to identify abnormal events from a patient's data.
  • the system applies one or more inference rules that identify abnormal events based on a scoring system for inferring a patient's clinical status based on abnormal event thresholds for the scores.
  • the scoring system includes a plurality of independent scoring schemes, which include at least one scoring scheme applicable prior to a milestone, and at least one scoring scheme applicable after the milestone.
  • a single threshold, common to the plurality of scoring schemes, may be applied to infer the patient's clinical status; a plurality of thresholds, corresponding to each of the plurality of schemes, may also be applied.
  • a scoring system may comprise a first scoring scheme for identifying from patient data at least one abnormal event occurring prior to at least one milestone event, and a second scoring scheme for identifying from patient data at least one abnormal event occurring after the at least one milestone event.
  • a scoring system in accordance with some embodiment of the present invention is provided in U.S. Application No. 10/795,724, incorporated by reference hereinabove.
  • the inference engine identifies abnormal events (i.e., the severity of the patient's condition) based, at least in part, on the milestone of the patient's treatment, the occurrence of heart sounds, the amplitude of the waveform, the peak-to-peak spacing, etc.
  • the inference engine may take into account typical wave characteristics for the particular patient, and detect an improvement or worsening of their condition.
  • the inference engine considers the wave characteristics for other patients in the patient pool having similar demographic characteristics, e.g., age, etc., as well as considerations relative to a milestone event, e.g., waveform characteristics just prior to a milestone event, just following a milestone event, etc.
  • An output device can be used for providing the condition of the patient.
  • Examples of output devices include a printer at 407 or a display at 408 for providing, e.g., a visual output of the patient's condition, or a speaker at 409 for providing, e.g., an audible alert.
  • output may be incorporated into database 406, for providing additional data for the particular patient or the patient pool. The inference monitor can draw upon this data to provide improved information over time.

Abstract

Systems and methods of diagnosing a condition of a patient are provided. The disclosed subject matter accesses data relating to physical characteristics of the patient, deriving parameters of a waveform from the data, comparing the parameters of the waveform with reference data, obtains the condition of the patient based upon the comparison, and provides an output relating to the condition of the patient.

Description

SYSTEM AND METHOD FOR DIAGNOSING
A CONDITION OF A PATIENT
CROSS REFERENCE TO RELATED APPLICATIONS
This application claims priority from U.S. Provisional Application Serial No. 60/845,585, filed September 18, 2006, which is incorporated by reference in its entirety herein.
BACKGROUND
The disclosed subject matter relates to a technique for diagnosing a condition of a patient. Various techniques for evaluating the status of a patient can be performed based on vital signs data. For example, the evaluation of patient sound information, such as heart sounds and their associated waveforms, has always been an important part of medical diagnosis. However, professionals in the medical field have not utilized this information as a basic patient monitoring technique in lieu of technically advanced pulse and blood pressure monitors, which are now the standard of care in most ORs, Recovery rooms, or ICUs.
The monitoring of heart sounds with a stethoscope during an anesthetic induction, in particular, was the mainstay of clinically evaluating the well being of patients. However, under these conditions, the subjective nature in which these heart sounds are monitored, in conjunction with rapid perturbations in vital sign data, has imposed a barrier to using this technique to its full advantage.
Unfortunately, the current techniques are limited in the ability to determining the status of the patient based on the heart sounds. Such limitations arise from subjective nature in which these heart sounds are monitored. Accordingly, there exists a need for a technique for determining the status of the patient based on the use of waveforms.
SUMMARY
Systems and methods of diagnosing a condition of a patient are disclosed herein.
NY02:596797. l In an exemplary method, the disclosed subject matter samples data relating to physical characteristics of the patient, deriving parameters of a waveform from the data, comparing the parameters of the waveform with reference data, obtains the condition of the patient based upon the comparison, and provides an output relating to the condition of the patient.
The sampled data relating to the physical characteristics of the patient can include heart sounds. That data can be sampled at distinct time intervals or distinct milestones. The distinct milestones can include a resting state, a first dosage state, a high dosage state, or an anesthetized state. The reference data can include a database of the physical characteristics of a population of patients, prior condition of the patient, and prior waveform of the patient.
Identifying parameters associated with the waveform can include identifying the heart rate of the patient, the amplitude of the waveform, the peak-to- peak spacing of the waveform, and identifying individual heart sounds, such as the Sl, S2, S3, and S4 sounds.
The disclosed subject matter also provides a system for diagnosing a condition of a patient. In some embodiments, the system includes an apparatus for sampling data relating to physical characteristics of the patient, a processor arranged and configured to derive parameters of a waveform from the data, to compare the parameters of the waveform with reference data, and to obtain a condition of the patient based upon the comparison, and an output device for providing the condition of the patient.
The accompanying drawings, which are incorporated and constitute part of this disclosure, illustrate preferred embodiments of the disclosed subject matter and serve to explain its principles.
BRIEF DESCRIPTION OF DRAWINGS
FIG. 1 is a diagram illustrating a method implemented in accordance with some embodiments of the disclosed subject matter; FIG. 2 shows the original files for one patient given resting, induced, and then high dose sevoforane;
NY02:596797. I FIG. 3 is the tracings of the patient in FIG. 1 once filtered for artifacts and amplified; and
FIG. 4 is a block diagram of system in accordance with some embodiments of the disclosed subject matter. While the disclosed subject matter will now be described in detail with reference to the Figures, it is done so in connection with the illustrative embodiments.
DETAILED DESCRIPTION
Systems and methods of diagnosing a condition of a patient are provided. The disclosed subject matter samples data relating to physical characteristics of the patient, derives parameters of a waveform from the data, comparing the parameters of the waveform with reference data, obtains the condition of the patient based upon the comparison, and provides an output relating to the condition of the patient.
FIG. 1 is a diagram illustrating a method implemented in accordance with some embodiments of the disclosed subject matter. Data relating to physical characteristics of the patient are sampled at step 101. The sampled data relating to the physical characteristics of the patient can include heart sounds. That data can be sampled at distinct time intervals or distinct milestones. The distinct milestones can include a resting state, a first dosage state, a high dosage state, or an anesthetized state.
The parameters of a waveform from the data can be derived at step 102. Identifying parameters associated with the waveform can include identifying the heart rate of the patient, the amplitude of the waveform, the peak-to-peak spacing of the waveform, and identifying individual heart sounds, such as the Sl , S2, S3, and S4 sounds.
The parameters of the waveform can be compared with reference data at step 103. The reference data can include a database of the physical characteristics of a population of patients, prior condition of the patient, and prior waveform of the patient. The condition of the patient based upon the comparison can be obtained at step 104. The condition of the patient can include a normal state, a bad state, or an awful state. The condition of the patient can also include changes in the
NY02:596797.1 3 state of the patent, such as an improving state or a worsening state. Additionally, the condition of the patient can include states based on time, such as a previous state, an actual state, and an expected state.
An output relating to the condition of the patient can be provided at step 105. The output can be in the form of a graphic or text indicating the condition of the patient. Additionally, the output can be a sound indicating that the condition of the patient is worsening or is in a bad or awful state.
The techniques for sampling data, comparing the waveform will reference data, determining the condition of the patient based on the data, and providing output relating to the condition of a patient are further described in the Physiologic Inference Monitor, U.S. Application No. 10/795,724, filed March 5, 2004, and which is incorporated by reference in its entirety herein.
In some embodiments, the waveform data is heart sounds. There are four different heart sounds commonly analyzed. Sl or First Heart Sound is made by the closing of the mitral and tricuspid valves. S2 or Second Heart Sound is made by the closing of the aortic and pulmonary valves. S3 or Third heart Sound is made by rapid ventricular filling when mitral and tricuspid open. This sound is typically considered normal in children and adolescents and can be heard at the Apex. S4 or Fourth Heart Sound is made by the Rapid ventricular filling that occurs in atrial systole, and is heard in late diastole. An S4 can clearly be heard at Erb's point. This sound is normally found in very elderly adults but it can also be heard in young children.
In children an S3 is a normal variant, while the S4 is not. In addition, S3 is considered abnormal in the young adults. The S4 is a normal variant after exercise and heart stress. There are differences in the left and right heart S4 sounds. For example, Erb's point is considered a favorable location to find a right heart S4, whereas the apex of the heart is the location for the left S4. The left heart S4 is considered the normal variant, while the right is not. Causes of a right S4 include, e.g., an augmented atrial contraction which increases filling during low ventricular compliance states, exercise and heightened cardiac output, and during positive pressure ventilation with a concordant increase in right ventricular afterload.
Differences in heart sounds before and after patients were induced with anesthesia were studied. For this study, pediatric and adult patient heart sounds were
NY02:596797.1 4 recorded before and after induction, and in the intensive care unit (ICU) post surgery. Originally, five points were examined across the heart, e.g., right sternum, left 3rd intercostals, left base, apex and carotid. Only three of theses points appeared acceptable for continued study. The carotid was excluded because it lacked Sl , the right sternum boarder was excluded because S3 and S4 could not be heard and recorded clearly.
Thus, in the exemplary embodiment, Erb's point appeared to be useful to judge heart sounds and changes, followed by the apex and then the right sternum boarder. Similarities and differences between pediatric and adult patients before and after they were anesthetized were studied. While there were differences in anesthetic technique, inhalation versus intravenous, the decreases in after-load and concurrent increase in cardiac output was found to reveal an S3, S4, augmentation or diminution in the recorded heart sounds. It was also possible the institution of positive pressure ventilation may induce auditory changes that could be captured by our recordings. Similarities and differences in adult and pediatric patients given a high dose of anesthesia were also studied.
Both pediatric and adult patient's hearts with the Littmann 4100 Series Electronic stethoscope were studied at three different times; resting, anesthetized, and highly anesthetized and three major points for hearing heart sounds; the apex, erbs point, and aortic.
The heart sounds were recorded with the stethoscope. After the sounds were recorded, they were transferred to a Windows computer via a USB infrared adapter. The Littmann Sound Analysis Software was used to determine if the recording was acceptable. If the recording was acceptable, it would be saved in the database.
After the recording was saved, it was edited and catalogued in Wave Pad (a wave evaluation application). The application can amplify the sounds, eliminate background noise, etc. until a desired consistency was achieved. However, the original file was used in the evaluation of intensity, amplitude and time and inter- patient correlation.
FIG. 2 shows the original files for one patient given resting 22, induced 24, and then high dose 26 sevoforane.
NY02:596797.1 5 Analysis of the waveform provides a marker for an individual heart cycle and distinguishes Sl . Of note, a slightly higher heart rate is noticed once the patient is induced, with a wider split in the S2, and the addition of an S4. The amplitude of the heart sounds is greater and higher after induction. High dose anesthetic, results in decreased amplitude of the Sl and S2, the wide split remains, and the S4 although heard was not as prominent.
The heartbeat of a child is heard more clearly when recorded than for an adult. The difference in the thickness of the chest wall in children and adults is a possible explanation. Pediatric patients and adults have a similar pattern of heart sounds when induced with anesthesia. This appears true for inhalation inductions, propofol, and versed fentanyl inductions. The appearance of the S4 occurred whether or not the patient was placed on positive pressure ventilation or not.
FIG. 3 is the tracings of the same patient once filtered for artifacts and amplified. While the pattern of heart sounds that occur during induction is still prominent, the clarity for clinical testing as stated above with attendings and residents can easily be done.
The last topic for this preliminary evaluation of heart sounds is the approach to standardize and display results of individual wave patterns.
A computer program is used to execute a digital algorithm that detects first and second heart sounds, defines the systole and diastole, and characterizes the systolic murmur. Heart sounds were recorded in 300 children with a cardiac murmur, using an electronic stethoscope. A digital algorithm was developed for detection of first and second heart sounds. R-waves and T-waves in the electrocardiography were used as references for detection. The sound signal analysis was carried out using the short-time Fourier transform. The first heart sound detection rate, with reference to the R- wave, was 100% within 0.05-0.2R-R interval. The second heart sound detection rate between the end of the T- wave and the 0.6R-R interval was 97%. The systolic and diastolic phases of the cardiac cycle could be identified. Because of the overlap between heart sounds and murmur a systolic segment between the first and second heart sounds (20-70%) was selected for murmur analysis. The maximum intensity of the systolic murmur, its average frequency, and the mean spectral power were quantified. The frequency at the point with the highest sound intensity in the spectrum and its time from the first heart sound, the highest frequency, and frequency
NY02.596797.1 6 range were also determined. This method will serve as the foundation for computer- based detection of heart sounds and the characterization of cardiac murmurs. FIG. 4 is a block diagram of system in accordance with some embodiments of the disclosed subject matter. In a preferred embodiment, the system includes an apparatus for sampling data relating to physical characteristics of the patient, a processor arranged and configured to derive parameters of a waveform from the data, to compare the parameters of the waveform with reference data, and to obtain a condition of the patient based upon the comparison, and an output device for providing the condition of the patient. In one embodiment, the system diagnoses the condition of the patient based on the heart sounds of the patient. A stethoscope at 403 can be applied to the chest of a patient at 402 such that sounds from the heart at 401 of the patient can be sampled. The apparatus for sampling data relating to physical characteristics of the patient can include a stethoscope. An example of a stethoscope includes the Littmann 4100 Series Electronic stethoscope.
The digitizer at 404, which can be connected to the stethoscope at 403, can covert the signal received by the stethoscope at 403 into a digital signal. The digitizer at 404 can be connected to a processor at 405 and the digital signal can be sent to the processor at 405. The processor at 405 is arranged and configured to derive parameters of a waveform from the digital signal, to compare the parameters of the waveform with reference data, and to obtain a condition of the patient based upon the comparison. The processor at 405 can be implemented as a computer microchip, a stand alone computer or collection of networks computing or any device suitable for processing. The disclosed subject matter can be implemented in Perl, C, or other suitable programming language. Software, such as the Littmann Sound Analysis Software can be used to record the heart sounds, and Wave Pad, another application, can be used to edit and evaluate the heart sounds.
The processor 405 may be distributed over a network in some embodiments. For example, the processor may include a client and remote server on a wired or wireless network.
The database at 406 is provided, which may be accessed by the processor at 405. The database can store data relating to the physical characteristics
NY02:596797.1 7 of a population of patients, prior condition of the patient, and prior waveform of the patient. The data stored in the database at 406 can be used to compare the parameters of the waveform of the patient. An inference monitor, as described in U.S. Application No. 10/795,724, may be used to identify and/or classify abnormal patients by scanning a patient's records for relevant data, such as heart sounds data, and applying one or more inference rules. The inference monitor may also scan data relating to a population of patients.
In a further embodiment, the inference monitor identifies and/or classifies abnormal events that occurred at or about certain milestones in the therapeutic treatment. A "milestone" or "milestone event" refers herein to a distinguishable event in the course of a patient's treatment. It is understood that various types of milestones may be either predefined or user-defined, based on the type of treatment being offered. For instance, with regard to heart sounds, the milestones may be critical periods during surgery, such as resting (prior to anesthesia), anesthetization (at a first level of anesthesia) or highly anesthetized (at a second, higher level of anesthesia), etc. In this instance, it may be desirable for the inference engine to apply the inference rule or rules to patient data that have been collected within a predefined window, such as a 20-minute window, before and/or after any one or more of the milestones. The inference engine may also identify abnormal events in light of milestones, such as demographic data (e.g., the patient's age, gender, weight, etc.) and clinical data (e.g., the patient's vital signs, relevant allergic reactions, medications, prosthetics, preexisting medical conditions, relevant diagnoses, type of procedure suggested), and in light of the data, such as the time interval of the events (e.g., between start and stop times) and the drugs that were administered during the operation.
Inferences identifying events indicative of the severity of a patient's condition, include, without limitation, duration of treatment, the type of procedure, demographics, etc. Various techniques may be used to identify abnormal events from a patient's data. In one embodiment of the present invention, the system applies one or more inference rules that identify abnormal events based on a scoring system for inferring a patient's clinical status based on abnormal event thresholds for the scores.
NY02:596797.1 8 In another embodiment, the scoring system includes a plurality of independent scoring schemes, which include at least one scoring scheme applicable prior to a milestone, and at least one scoring scheme applicable after the milestone. A single threshold, common to the plurality of scoring schemes, may be applied to infer the patient's clinical status; a plurality of thresholds, corresponding to each of the plurality of schemes, may also be applied. By way of example, a scoring system may comprise a first scoring scheme for identifying from patient data at least one abnormal event occurring prior to at least one milestone event, and a second scoring scheme for identifying from patient data at least one abnormal event occurring after the at least one milestone event. A scoring system in accordance with some embodiment of the present invention is provided in U.S. Application No. 10/795,724, incorporated by reference hereinabove.
In some embodiments, the inference engine identifies abnormal events (i.e., the severity of the patient's condition) based, at least in part, on the milestone of the patient's treatment, the occurrence of heart sounds, the amplitude of the waveform, the peak-to-peak spacing, etc. The inference engine may take into account typical wave characteristics for the particular patient, and detect an improvement or worsening of their condition. In another embodiment, the inference engine considers the wave characteristics for other patients in the patient pool having similar demographic characteristics, e.g., age, etc., as well as considerations relative to a milestone event, e.g., waveform characteristics just prior to a milestone event, just following a milestone event, etc.
An output device can be used for providing the condition of the patient. Examples of output devices include a printer at 407 or a display at 408 for providing, e.g., a visual output of the patient's condition, or a speaker at 409 for providing, e.g., an audible alert. Alternatively, output may be incorporated into database 406, for providing additional data for the particular patient or the patient pool. The inference monitor can draw upon this data to provide improved information over time. The foregoing merely illustrates the principles of the invention.
Various modifications and alterations to the described embodiments will be apparent to those skilled in the art in view of the teachings herein. For example, it is understood that the techniques described herein are useful with other waveform data
NY02:5%797.1 9 of a patient, such as neurological or peristaltic wave patterns. It will thus be appreciated that those skilled in the art will be able to devise numerous techniques which, although not explicitly described herein, embody the principles of the invention and are thus within the spirit and scope of the invention.
NY02 596797 1 10

Claims

WE CLAIM:
1. A method of diagnosing a condition of a patient comprising:
(a) accessing data relating to physical characteristics of the patient;
(b) deriving parameters of a waveform from said data; (c) comparing said parameters of said waveform with reference data;
(d) obtaining the condition of the patient based upon said comparing; and
(e) providing an output relating to the condition of the patient.
2. The method of claim 1 , wherein said data comprises data relating to heart sounds.
3. The method of claim 1, further comprising sampling said data at distinct time intervals.
4. The method of claim 1 , further comprising sampling said data at milestones.
5. The method of claim 4, wherein said milestones are selected from the group consisting of a resting state, a first dosage state, and a second dosage state.
6. The method of claim 4, wherein said milestones are selected from the group consisting of a resting state and an anesthetized state.
7. The method of claim 1 , wherein said reference data comprises data relating to the physical characteristics of a population of patients.
8. The method of claim 1, wherein said reference data comprises data relating to the prior history of the patient.
9. The method of claim 1 , wherein said reference data comprises data relating to the prior waveform of the patient.
10. The method of claim 1, wherein identifying parameters of the waveform comprises identifying the heart rate of the patient.
1 1. The method of claim 1 , wherein identifying parameters of the waveform comprises identifying an amplitude of said waveform.
NY02:596797.1 1 1
12. The method of claim 1 , wherein identifying parameters of the waveform comprises identifying a peak-to-peak spacing.
13. The method of claim 1 , wherein identifying parameters of the waveform comprises identifying a heart sound selected from the group consisting of a Sl , S2, S3, and S4 sounds.
14. A system for diagnosing a condition of a patient comprising:
(a) an apparatus for accessing data relating to physical characteristics of the patient;
(b) a processor arranged and configured to derive parameters of a waveform from said data; compare said parameters of said waveform with reference data; and obtain a condition of the patient based upon said comparison; and
(c) an output device for providing said condition of the patient.
15. The system of claim 14, wherein said apparatus for sampling data comprises a stethoscope.
16. The system of claim 14, wherein said process is distributed over a network.
17. The system of claim 14, wherein said output device comprises a display.
18. The system of claim 14, wherein said output device comprises a speaker.
19. The system of claim 14, wherein said data comprises data relating to heart sounds.
20. The system of claim 14, wherein identifying parameters of the waveform comprises identifying the heart rate of the patient.
21. A method for inferring a patient's clinical status in the course of a treatment, comprising: (a) accessing a patient's data;
(b) deriving parameters of a waveform from the patient's data; and
(c) identifying at least one abnormal event from the patient's data by applying a scoring system for inferring a patient's clinical status,
NY02:596797.1 12 wherein the scoring system comprises a first scoring scheme for identifying from patient data at least one abnormal event occurring prior to at least one milestone event, and a second scoring scheme for identifying from patient data at least one abnormal event occurring after the at least one milestone event.
22. The method of claim 21, wherein at least one scoring scheme assigns a score to abnormal patient data that is based on the severity of the patient's condition, as reflected in the patient's data.
23. The method of claim 21 , wherein the data comprises data relating to the patient's heart sounds.
24. The method of claim 21, wherein the patient's data is collected in connection with administering anesthesia, wherein the at least one milestone event is the administration of anesthesia, and wherein the first scoring scheme is applied to data occurring prior to the administration of anesthesia, and the second scoring scheme is applied to data occurring after the administration of anesthesia.
25. The method of claim 21, wherein identification of the at least one abnormal event is based, at least in part, on at least one of: an amplitude of the waveform; peak-to-peak spacing of the waveform; and occurrence of heart sounds.
NY02:596797. l 13
PCT/US2007/078754 2006-09-18 2007-09-18 System and method for diagnosing a condition of a patient WO2008036649A2 (en)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
US84558506P 2006-09-18 2006-09-18
US60/845,585 2006-09-18

Publications (2)

Publication Number Publication Date
WO2008036649A2 true WO2008036649A2 (en) 2008-03-27
WO2008036649A3 WO2008036649A3 (en) 2008-11-27

Family

ID=39201199

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/US2007/078754 WO2008036649A2 (en) 2006-09-18 2007-09-18 System and method for diagnosing a condition of a patient

Country Status (1)

Country Link
WO (1) WO2008036649A2 (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US8870791B2 (en) 2006-03-23 2014-10-28 Michael E. Sabatino Apparatus for acquiring, processing and transmitting physiological sounds

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5687738A (en) * 1995-07-03 1997-11-18 The Regents Of The University Of Colorado Apparatus and methods for analyzing heart sounds
US6643548B1 (en) * 2000-04-06 2003-11-04 Pacesetter, Inc. Implantable cardiac stimulation device for monitoring heart sounds to detect progression and regression of heart disease and method thereof
US20040260188A1 (en) * 2003-06-17 2004-12-23 The General Hospital Corporation Automated auscultation system
US20050090755A1 (en) * 2003-10-22 2005-04-28 Guion Marie A. Analysis of auscultatory sounds using single value decomposition
US20050197865A1 (en) * 2004-03-05 2005-09-08 Desmond Jordan Physiologic inference monitor
US20060047213A1 (en) * 2002-10-21 2006-03-02 Noam Gavriely Acoustic cardiac assessment

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5687738A (en) * 1995-07-03 1997-11-18 The Regents Of The University Of Colorado Apparatus and methods for analyzing heart sounds
US6643548B1 (en) * 2000-04-06 2003-11-04 Pacesetter, Inc. Implantable cardiac stimulation device for monitoring heart sounds to detect progression and regression of heart disease and method thereof
US20060047213A1 (en) * 2002-10-21 2006-03-02 Noam Gavriely Acoustic cardiac assessment
US20040260188A1 (en) * 2003-06-17 2004-12-23 The General Hospital Corporation Automated auscultation system
US20050090755A1 (en) * 2003-10-22 2005-04-28 Guion Marie A. Analysis of auscultatory sounds using single value decomposition
US20050197865A1 (en) * 2004-03-05 2005-09-08 Desmond Jordan Physiologic inference monitor

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US8870791B2 (en) 2006-03-23 2014-10-28 Michael E. Sabatino Apparatus for acquiring, processing and transmitting physiological sounds
US8920343B2 (en) 2006-03-23 2014-12-30 Michael Edward Sabatino Apparatus for acquiring and processing of physiological auditory signals
US11357471B2 (en) 2006-03-23 2022-06-14 Michael E. Sabatino Acquiring and processing acoustic energy emitted by at least one organ in a biological system

Also Published As

Publication number Publication date
WO2008036649A3 (en) 2008-11-27

Similar Documents

Publication Publication Date Title
US7162294B2 (en) System and method for correlating sleep apnea and sudden cardiac death
US20040260188A1 (en) Automated auscultation system
US20060047213A1 (en) Acoustic cardiac assessment
JP2021524958A (en) Respiratory state management based on respiratory sounds
US20060161064A1 (en) Computer-assisted detection of systolic murmurs associated with hypertrophic cardiomyopathy
JP2010535570A (en) Automatic identification of the responsible coronary artery
US8868168B2 (en) System for cardiac condition characterization using electrophysiological signal data
JP6251035B2 (en) Operating method of n-lead ECG system
Drews et al. Error producing conditions in the intensive care unit
US20210007621A1 (en) Method to analyze cardiac rhythms using beat-to-beat display plots
Krasteva et al. Validation of arrhythmia detection library on bedside monitor data for triggering alarms in intensive care
Griffel et al. Path length entropy analysis of diastolic heart sounds
Ghassemian et al. Early detection of pediatric heart disease by automated spectral analysis of phonocardiogram
Sowan et al. Role of large clinical datasets from physiologic monitors in improving the safety of clinical alarm systems and methodological considerations: a case from Philips monitors
Naraei et al. Toward learning intracranial hypertension through physiological features: A statistical and machine learning approach
WO2008036649A2 (en) System and method for diagnosing a condition of a patient
Iskandar et al. Design and implementation electronic stethoscope on smart chair for monitoring heart rate and stress levels driver
Acharya et al. Integrated index for cardiac arrythmias diagnosis using entropies as features of heart rate variability signal
US9192315B2 (en) Patient signal analysis and characterization based on late potentials
Tedim et al. Development of a system for the automatic detection of air embolism using a precordial Doppler
Bickel et al. The Use of Pneumoperitoneum During Laparoscopic Surgery as a Model to Study Pathophysiologic Phenomena: The Correlation of Cardiac Functionality with Computerized Acoustic Indices—Preliminary Data
Nguyen et al. Individualized arrhythmia detection with ECG signals from wearable devices
El‐Segaier et al. Detection of cardiac pathology: time intervals and spectral analysis
Silvestri et al. Prolonged cardiorespiratory monitoring of children more than twelve months of age: characterization of events and approach to discontinuation
Szymanowska et al. Development of an electronic stethoscope

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 07814912

Country of ref document: EP

Kind code of ref document: A2

NENP Non-entry into the national phase in:

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 07814912

Country of ref document: EP

Kind code of ref document: A2