Combination non-invasive and invasive bioparameter measuring device

A portable bioparameter-monitoring medical device usable by a patient, comprising:

• a non-invasive component structured to receive a body part of a patient and configured to generate non-invasive bioparametric readings of tissue of the body part upon insertion of the body part of the patient into the non-invasive component, the non-invasive component including at least one color image sensor configured to generate a series of color images of the tissue and to sense a magnitude of each of three colors at pixels of each color image, the non-invasive component also including a first digital processor for processing the series of color images into a signal over time reflecting a distribution of each of the three colors over time;
• an invasive component for obtaining an invasive bioparametric reading from blood of the patient, the invasive component also including a second digital processor for directing automatic transmission of the invasive bioparametric reading to the first digital processor of the non-invasive component, the invasive bioparametric readings forming entries in a column vector, Y,
• the first digital processor of the non-invasive component programmed to
• (a) (i) use a mathematical function to convert the signal to a scalar learning number and (ii) repeat step “(a)(i)”, without necessarily using the same mathematical functions, to form a learning vector that corresponds to a scalar invasive bioparameter reading entry of column vector Y;
• (b) form an n by n learning matrix, D, that is a regular matrix, by repeating step “(a)” to non-invasive readings and invasive readings enough times that the first digital processor has sufficient correlations between non-invasive bioparametric readings and invasive bioparametric readings to be able to measure the bioparameter based on a non-invasive bioparameter reading of the bioparameter at a pre-defined level of threshold acceptability;
• (c) obtain a coefficient of learning vector, C, by multiplying an inverse matrix D−1 of matrix, D by the column vector, Y;
• (d) generate a new vector, Vnew by using a mathematical function to convert a new signal to a scalar number and do so n times to form Vnew, without necessarily using the same mathematical functions, when a user uses the at least one color image sensor in the non-invasive component to generate the new signal from a new series of color images of the tissue of the body part;
• (e) use the entries of Vnew to form a regular matrix, Dnew, of n by n size and whose structure of non-zero elements is identical to learning matrix, D, and
• (f) perform a matrix vector multiplication of Dnew by coefficient of learning vector, C, to create a vector of non-invasive bioparameter measurement, R, and compare entries of R with entries of Y to find one entry of R which represents a calibrated bioparameter value for the patient.

A method of producing a portable bioparameter-monitoring medical device custom-tailored to a patient, the method comprising:

• (a) providing directly or indirectly to a patient a medical device having (i) a non-invasive component capable of generating a non-invasive bioparametric reading of the patient's bioparameter upon insertion by the patient of a body part into the non-invasive component, the non-invasive component including a first digital processor for processing digital color images of part of the body part and representing the digital images as a discrete signal over time, and having (ii) an invasive component for measuring the bioparameter from blood of the patient and obtaining an invasive bioparametric reading for the patient, the invasive component also including a second digital processor for transmitting the invasive bioparametric reading to the first digital processor of the non-invasive component, and (iii) a coupling element for maintaining the invasive component operatively engaged to the non-invasive component and allowing transmission of invasive bioparametric readings from the invasive component to the non-invasive component, the first digital processor also for calibrating the non-invasive component so that the non-invasive bioparametric readings for the patient approximate the invasive bioparametric readings for the patient for a given bioparameter under a predefined standard of approximation;
• and
• (b) calibrating the non-invasive component to the patient by (i) invasively measuring the bioparameter of the patient using the invasive component, (ii) transmitting the invasive bioparameter readings to the non-invasive component, and (iii) non-invasively measuring the bioparameter of the patient within a proximity time of the invasive measuring using mathematical functions, and performing substeps (i), (ii) and (iii) enough times that the first digital processor has sufficient correlations between non-invasive bioparametric readings and invasive bioparametric readings to be able to measure the bioparameter using a non-invasive bioparameter reading at a pre-defined level of threshold acceptability.

A method of monitoring a bioparameter, comprising:

• (a) invasively measuring the bioparameter of a patient using an invasive component of a bioparameter monitoring device and transmitting an invasive bioparameter reading to a non-invasive component of the bioparameter monitoring device, the invasive bioparameter reading to be entered in a column vector, Y;
• (b) within a proximity time of step “(a)”, non-invasively measuring the bioparameter of the patient by using one or more variable sensors in the non-invasive component of the device to generate a series of data representing a magnitude of one or more variables of tissue of a body part of the patient and by converting the magnitudes into a series of electric signals, to produce a signal over time reflecting a distribution of each of the one or more variables over time;
• (c) a digital processor of the non-invasive component (i) using a mathematical function to convert the signal to a scalar learning number and (ii) repeating step “(c)(i)”, without necessarily using the same mathematical function, to form a learning vector that corresponds to a scalar invasive bioparameter reading entry of column vector Y;
• (d) from a plurality of learning vectors, a digital processor forming an n by n learning matrix, D, that is a regular matrix, by repeating steps “(a)” through “(c)” enough times that a digital processor has sufficient correlations between non-invasive bioparametric readings and invasive bioparametric readings to be able to measure the bioparameter using a non-invasive bioparameter reading at a pre-defined level of threshold acceptability;
• (e) a digital processor obtaining a coefficient of learning vector, C, by multiplying an inverse matrix D−1 of learning matrix, D by the column vector Y;
• (f) a digital processor obtaining a new vector, Vnew by (i) non-invasively measuring the bioparameter of the patient by using the one or more variable sensors of the non-invasive component of the device to generate a series of data representing a magnitude of one or more variables of tissue of a body part of the patient and by converting the magnitudes into a series of electric signals, to produce a signal over time reflecting a distribution of each of the variables over time and by having a digital processor use a mathematical function to convert the signal to a scalar number and by (ii) repeating step “(f)(i) n times to form Vnew, without necessarily using the same mathematical functions;
• (g) using the entries of Vnew to form a regular matrix, Dnew, of n by n size and whose structure of non-zero elements is identical to a structure of non-zero elements of learning matrix, D;
• and
• (h) using a digital processor to perform a matrix vector multiplication of Dnew by coefficient of learning vector, C, to create a column vector of non-invasive bioparameter measurement, R, and comparing entries of R with entries of Y to find one entry of R which represents a calibrated bioparameter value for the patient.

A non-transitory computer-readable medium having stored thereon bioparameter monitoring software, the bioparameter monitoring software executed by one or more digital processors, the execution of the bioparameter monitoring software by the one or more digital processors performing:

• (a) (i) using a mathematical function to convert a signal generated by a non-invasive component reflecting a non-invasive bioparameter reading, to a scalar learning number and (ii) repeating step “(a)(i)”, without necessarily using the same mathematical function, to form a learning vector that corresponds to a scalar invasive bioparameter reading entry in a column vector Y, wherein entries in column vector Y represent invasive bioparameter readings of a patient;
• (b) from a plurality of learning vectors, forming an n by n learning matrix, D, that is a regular matrix, by repeating step “(a)” a pre-defined number of times, each time from newly generated non-invasive bioparameter readings and a newly generated scalar invasive bioparameter reading that is an entry for column vector, Y;
• (c) obtaining a coefficient of learning vector, C, by multiplying an inverse matrix D−1 of learning matrix, D by the column vector Y;
• (d) obtaining a new vector, Vnew by using a mathematical function to convert a further new signal to a scalar number and by (ii) repeating step “(d)(i) n times to form Vnew, without necessarily using the same mathematical functions;
• (e) using the entries of Vnew to form a regular matrix, Dnew, of n by n size and whose structure of non-zero elements is identical to a structure of non-zero elements of learning matrix, D;
• and
• (f) performing a matrix vector multiplication of Dnew by coefficient of learning vector, C, to create a column vector of non-invasive bioparameter measurement, R, and comparing entries of R with entries of Y to find one entry of R which represents a calibrated bioparameter value for the patient.

A method of monitoring a bioparameter, comprising:

• (a) invasively measuring the bioparameter of a patient using an invasive component of a bioparameter monitoring device, transmitting an invasive bioparameter reading to a non-invasive component of the bioparameter monitoring device and storing the invasive bioparameter reading in the non-invasive component;
• (b) within a proximity time of step “(a)”, one or more color image sensors in the non-invasive component of the device generating a series of color images of tissue of a body part of the patient, sensing a magnitude of each of three colors at pixels of each color image and converting the magnitudes into a series of electric signals, to produce a signal over time reflecting a distribution of each of the three colors in the color images over time, the signal representing a non-invasive measurement of the bioparameter of the patient;
• (c) one or more processors of the non-invasive component programmed to convert the signal to a scalar learning number (i) using a mathematical function to convert the signal to a scalar learning number and (ii) repeating step “(c)(i)”, without necessarily using the same mathematical function, to form a learning vector that corresponds to a scalar invasive bioparameter reading entry of a column vector Y;
• (d) from a plurality of learning vectors, the one or more processors forming an n by n learning matrix, D, that is a regular matrix, by repeating steps “(a)” through “(c)” enough times that the one or more processors have sufficient correlations between non-invasive bioparametric readings and invasive bioparametric readings to be able to measure the bioparameter using a non-invasive bioparameter reading at a pre-defined level of acceptability, each row of learning matrix, D, representing non-invasive entries that correspond to invasive entries of vector Y;
• (e) the one more digital processors obtaining a new vector, Vnew by (i) non-invasively measuring the bioparameter of the patient by using the one or more color image sensors in the non-invasive component of the device to generate a series of color images of tissue of a body part of the patient and to sense a magnitude of each of the three colors at pixels of each color image and by converting the magnitudes into a series of electric signals, to produce a signal over time reflecting a distribution of each of the three colors in the color images over time and by having the one or more digital processors use a mathematical function to convert the signal to a scalar number and by (ii) repeating step “(e)(i) n times to form Vnew, without necessarily using the same mathematical functions;
• (f) using the one or more digital processors to compare entries of Vnew with rows of learning matrix, D to find a best match involving an ith row of learning matrix, D and using the ith entry of Y as a calibrated bioparameter value for the patient.

A method of monitoring a bioparameter, comprising:

• (a) invasively measuring the bioparameter of a patient using an invasive component of a bioparameter monitoring device, transmitting an invasive bioparameter reading to a non-invasive component of the bioparameter monitoring device and storing the invasive bioparameter reading in the non-invasive component;
• (b) within a proximity time of step “(a)”, one or more variable sensors in the non-invasive component of the device generating a series of data or color images of tissue of a body part of the patient, wherein the images represent interference by the tissue with emitted light, sensing a magnitude of the emitted light at each sensor, or sensing a magnitude of light in each of three colors at pixels of each color image, and converting the magnitudes into a series of electric signals, the signal representing a non-invasive measurement of the bioparameter of the patient;
• (c) one or more processors of the non-invasive component programmed to convert the signal to a scalar learning number (i) using a mathematical function to convert the signal to a scalar learning number and (ii) repeating step “(c)(i)”, without necessarily using the same mathematical function, to form a learning vector that corresponds to a scalar invasive bioparameter reading entry of a column vector Y;
• (d) from a plurality of learning vectors, the one or more processors forming an n by n learning matrix, D, that is a regular matrix, by repeating steps “(a)” through “(c)” enough times that the one or more processors have sufficient correlations between non-invasive bioparametric readings and invasive bioparametric readings to be able to measure the bioparameter using a non-invasive bioparameter reading at a pre-defined level of acceptability;
• (e) the one or more digital processors obtaining a coefficient of learning vector, C, by multiplying an inverse matrix D−1 of learning matrix, D by the column vector Y;
• (f) the one more digital processors obtaining a new vector, Vnew by (i) non-invasively measuring the bioparameter of the patient by using the one or more variable sensors in the non-invasive component of the device to generate a series of data or color images of tissue of a body part of the patient, wherein the images represent interference by the tissue with emitted light and to sense a magnitude of emitted light at each sensor or to sense a magnitude of light in each of the three colors at pixels of each image and by converting the magnitudes into a series of electric signals over time reflecting a spatial temporal pixel color distribution and by having the one or more digital processors use a mathematical function to convert the signal to a scalar number and by (ii) repeating step “(f)(i) n times to form Vnew, without necessarily using the same mathematical functions;
• and either
• (I)(a) using the entries of Vnew to form a regular matrix, Dnew, of n by n size and whose structure of non-zero elements is identical to a structure of non-zero elements of learning matrix, D;
• and
• (I)(b) using the one or more digital processors to perform a matrix vector multiplication of Dnew by coefficient of learning vector, C, to create a column vector of non-invasive bioparameter measurement, R, and comparing entries of R with entries of Y to find one entry of R which represents a calibrated bioparameter value for the patient;
• or
• (II) using the one or more digital processors to compare entries of Vnew with rows of learning matrix, D to find a best match involving an ith row of learning matrix, D and using the ith entry of Y as a calibrated bioparameter value for the patient.