System and method to detect the presence and progression of diseases characterized by systemic changes in the state of the vasculature
This invention provides a system and method that employs multiple measurements of various, relevant tissue states of a patient to detect and predict OH and similar conditions. These multi anatomic measurements are transformed by a multivariate algorithm to outputs that convey the diagnostic and prognostic risk of the disease of interest. This novel, multiple-measurement technique avoids use of a single measurement, which is generally unlikely to adequately extract sufficient information to drive a clinically useful test in the setting of complex system disease. The system and method herein thereby allows automated monitoring of currently stable patients who, are known or suspected to have OH and/or similar internal conditions on a substantially continuous basis.
1 . A system for detecting an internal condition in a living body comprising:
an polyanatomic sensing belt having a plurality of skin contactable sensors that each respectively generate one or more types of sensor data with respect to anatomical functions, the plurality of skin contactable sensors including a plurality of electrical impedance tomography (EIT) sensors, near-infrared spectroscopy (NIRS) sensors, and electrical impedance spectroscopy (EIS) sensors, each of the skin contactable sensors being constructed and arranged to respectively provide a data stream so that polyanatomic patterns can be derived, wherein the belt defines a relative contacted skin location on the body for each of the plurality of skin contactable sensors;
a controller integrated with the belt and configured to synchronize data in the respective data streams from the plurality of skin contactable sensors and to deliver the synchronized data in a multi-anatomic data stream to a processor; and
a data analysis process running on the processor, the data analysis process correlated to a degree of risk for the internal condition based on a clinical history and bedside evaluation of a wearer of the polyanatomic sensing belt, the data analysis process processing each data point in the synchronized multi-anatomic data stream as a localized quantum particle moving in a quantum potential representative of density of tissue of the body, and provides an output indicative of a probability that the internal condition is associated with a current or future disease state based on distinctions of sensed muscle tissue data from sensed subcutaneous tissue data resulting from the processing of the synchronized multi-anatomic data stream as localized quantum particles moving in a quantum potential representative of density of tissue of the body.
2 . The system as set forth in claim 1 wherein the internal condition is related to ongoing occult hemorrhage (OH).
3 . The system as set forth in claim 1 wherein the synchronized multi-anatomic data stream further includes information from at least one of: classic medical vital signs, photoplethysmography or an ECG.
4 . The system as set forth in claim 1 wherein the data analysis process includes a machine learning processor that performs feature extractions and supervised learning and thereby generates metaclassifiers.
5 . The system as set forth in claim 4 wherein the machine learning processor defines a deep neural network.
6 . The system as set forth in claim 5 wherein the deep neural network comprises at least one of a convolutional neural network (CNN), an recurrent neural network (RNN), and an Long short-term memory (LSTM).
7 . The system as set forth in claim 1 further comprising a refining process that increases accuracy of the data stream based upon priors.
8 . The system as set forth in claim 1 , wherein the data analysis process is constructed and arranged to operate multiple analysis processes concurrently, wherein each of the analysis processes is adapted to provide optimized performance in each of multiple, discrete different risk categories, respectively.
9 . The system as set forth in claim 8 wherein the analysis process includes a clinician-selectable process adapted to provide varying sensitivity and specificity.
10 . The system as set forth in claim 1 further comprising a system output display showing a combination of analysis process results and actual multiplex measurements that are inputs to the analysis process, both plotted against time.
11 . The system as set forth in claim 1 , wherein the processor provides optimal guidance for configuring the data analysis process based on a cost of false positives and on a cost of treatment delay using a stochastic dynamic programming model using backward induction or Q-learning.
12 . A medical treatment method using the system of claim 1 , the method comprising the steps of:
applying a plurality of polyanatomic sensing belts to the body at predetermined locations, including thorax and limbs; and
operating the sensors and the processor on a substantially continuous basis to generate the output.
13 . The medical treatment method as set forth in claim 12 wherein the internal condition is related to occult hemorrhage, an infectious process, or cardiovascular dysfunction.
14 . A system for determining a probability that an internal condition in a living body is associated with a current or future disease state comprising:
a polyanatomic sensing belt having a plurality of wearable sensors that each respectively monitor one or more types of anatomical functions, the plurality of wearable sensors including a plurality of electrical impedance tomography (EIT) sensors, near-infrared spectroscopy (NIRS) sensors, and electrical impedance spectroscopy (EIS) sensors, each of the wearable sensors being constructed and arranged to respectively provide a data stream representative of a monitored anatomical function for polyanatomic pattern detection, wherein the belt defines a relative contacted skin location on the body for each of the plurality of wearable sensors;
a controller system integrated with the belt and configured to synchronize data in the respective data streams from the plurality of wearable sensors and to deliver the synchronized data in a multi-anatomic data stream for polyanatomic pattern detection; and
a processor executing a data analysis process correlated to a degree of risk for the internal condition based on a clinical history and bedside evaluation of a wearer of the polyanatomic sensing belt that:
processes each data point in the multi-anatomic data stream as a localized quantum particle moving in a quantum potential representative of tissue density;
distinguishes sensed muscle tissue data from sensed subcutaneous tissue data; and
provides an output indicative of a probability that the internal condition is associated with a current or future disease state based on the sensed muscle tissue data as distinguished from the sensed subcutaneous tissue data.
15 . The system as set forth in claim 14 wherein the processor provides optimal guidance for configuring the data analysis process based on a cost of false positives and on a cost of treatment delay using a stochastic dynamic programming model using backward induction or Q-learning.
16 . The system as set forth in claim 14 wherein the data analysis process includes a machine learning processor that performs feature extractions and supervised learning and thereby generates metaclassifiers.
17 . The system as set forth in claim 16 wherein the machine learning processor defines a deep neural network.
18 . The system as set forth in claim 17 wherein the deep neural network comprises at least one of a convolutional neural network (CNN), an recurrent neural network (RNN), and an Long short-term memory (LSTM).
19 . The system as set forth in claim 14 further comprising a refining process that increases accuracy of the data stream based upon priors.
20 . The system as set forth in claim 14 , wherein the data analysis process is constructed and arranged to operate multiple analysis processes concurrently, wherein each of the analysis processes is adapted to provide optimized performance in each of multiple, discrete different risk categories, respectively.