Classifications

• • A—HUMAN NECESSITIES
  • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
  • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
  • A61B5/00—Measuring for diagnostic purposes; Identification of persons
  • A61B5/02—Detecting, measuring or recording pulse, heart rate, blood pressure or blood flow; Combined pulse/heart-rate/blood pressure determination; Evaluating a cardiovascular condition not otherwise provided for, e.g. using combinations of techniques provided for in this group with electrocardiography or electroauscultation; Heart catheters for measuring blood pressure
  • A61B5/0205—Simultaneously evaluating both cardiovascular conditions and different types of body conditions, e.g. heart and respiratory condition
• • A—HUMAN NECESSITIES
  • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
  • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
  • A61B5/00—Measuring for diagnostic purposes; Identification of persons
  • A61B5/02—Detecting, measuring or recording pulse, heart rate, blood pressure or blood flow; Combined pulse/heart-rate/blood pressure determination; Evaluating a cardiovascular condition not otherwise provided for, e.g. using combinations of techniques provided for in this group with electrocardiography or electroauscultation; Heart catheters for measuring blood pressure
  • A61B5/024—Detecting, measuring or recording pulse rate or heart rate
  • A61B5/0245—Detecting, measuring or recording pulse rate or heart rate by using sensing means generating electric signals, i.e. ECG signals
• • A—HUMAN NECESSITIES
  • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
  • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
  • A61B5/00—Measuring for diagnostic purposes; Identification of persons
  • A61B5/24—Detecting, measuring or recording bioelectric or biomagnetic signals of the body or parts thereof
  • A61B5/242—Detecting biomagnetic fields, e.g. magnetic fields produced by bioelectric currents
  • A61B5/243—Detecting biomagnetic fields, e.g. magnetic fields produced by bioelectric currents specially adapted for magnetocardiographic [MCG] signals
• • A—HUMAN NECESSITIES
  • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
  • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
  • A61B5/00—Measuring for diagnostic purposes; Identification of persons
  • A61B5/24—Detecting, measuring or recording bioelectric or biomagnetic signals of the body or parts thereof
  • A61B5/242—Detecting biomagnetic fields, e.g. magnetic fields produced by bioelectric currents
  • A61B5/245—Detecting biomagnetic fields, e.g. magnetic fields produced by bioelectric currents specially adapted for magnetoencephalographic [MEG] signals
• • A—HUMAN NECESSITIES
  • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
  • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
  • A61B5/00—Measuring for diagnostic purposes; Identification of persons
  • A61B5/24—Detecting, measuring or recording bioelectric or biomagnetic signals of the body or parts thereof
  • A61B5/316—Modalities, i.e. specific diagnostic methods
  • A61B5/318—Heart-related electrical modalities, e.g. electrocardiography [ECG]
  • A61B5/346—Analysis of electrocardiograms
  • A61B5/349—Detecting specific parameters of the electrocardiograph cycle
• • A—HUMAN NECESSITIES
  • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
  • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
  • A61B5/00—Measuring for diagnostic purposes; Identification of persons
  • A61B5/72—Signal processing specially adapted for physiological signals or for diagnostic purposes
  • A61B5/7235—Details of waveform analysis
  • A61B5/7264—Classification of physiological signals or data, e.g. using neural networks, statistical classifiers, expert systems or fuzzy systems
  • A61B5/7267—Classification of physiological signals or data, e.g. using neural networks, statistical classifiers, expert systems or fuzzy systems involving training the classification device
• • A—HUMAN NECESSITIES
  • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
  • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
  • A61B2562/00—Details of sensors; Constructional details of sensor housings or probes; Accessories for sensors
  • A61B2562/04—Arrangements of multiple sensors of the same type
  • A61B2562/046—Arrangements of multiple sensors of the same type in a matrix array
• • A—HUMAN NECESSITIES
  • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
  • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
  • A61B5/00—Measuring for diagnostic purposes; Identification of persons
  • A61B5/05—Detecting, measuring or recording for diagnosis by means of electric currents or magnetic fields; Measuring using microwaves or radio waves 

Definitions

• EMF electromagnetic field
• An EMF associated with a heart of an individual may be sensed and analyzed to reveal a status of a heart of the individual.
• an EMF of an individual may be used to determine whether ischemia is present in a heart of an individual and whether the ischemia is caused by coronary artery disease (CAD).
• CAD coronary artery disease
• an EMF can be used to determine the severity of the coronary disease that is present.
• an EMF that is generated by a heart of an individual is sensed using an EMF sensor and analyzed by a machine learning algorithm which is trained to determine whether ischemia and/or CAD is present in the heart of the individual.
• the machine learning algorithm is further configured to determine a severity of CAD in the individual.
• a machine learning software module is trained using EMF data and data relating to an individual including their records and data relating to the organs and/or tissue in the body of the individual.
• Abnormality data includes the presence or absence of an abnormality within an organ, tissue, body, or portion thereof, and said abnormalities are identified along with any known resulting or associated disease, disorder or condition.
• Data relating to an individual includes demographic data, medical image data, clinical data (e.g. from a health record, including an Electronic Health Record), encoded data, and encoded features, or metrics derived from an electromagnetic field
• EMF data includes EMF measurements and simulations of EMF measurements.
• the machine learning software module described herein in some embodiments, is trained on both EMF data and the corresponding abnormality data (i.e. that corresponds to an EMF data), such that the machine learning software module is able to analyze new EMF data and determine whether an abnormality is present based on training. Furthermore, the machine learning software module may determine a condition associated with the detected abnormality.
• Described herein are methods, software, systems and devices for detecting the presence of an abnormality in an organ, tissue, body, or portion thereof of a subject by analysis of the EMFs generated by the organ, tissue, body, or portion thereof.
• EMF data is used to generate one or more of a medical image, clinical data, and encoded data.
• medical image data, clinical data, encoded data and/or other features or metrics derived from the EMF data are used as inputs to train a machine learning algorithm which is configured to identify a presence of an abnormality in an individual.
• Said abnormality may be related to an organ from which an EMF is sensed or other organs or systems in an individual.
• a machine learning algorithm is configured to determine or predict an abnormality score for an individual.
• a diagnostic system configured to determine if cardiac ischemia is present in an individual, said diagnostic system comprising: an electromagnetic field sensor configured to sense an electromagnetic field measurement associated with a heart of the individual; a processor operably coupled to the electromagnetic field sensor; and a non- transitory computer-readable storage media encoded with software comprising a trained machine learning software module, wherein said software is executable by the processor and causes the processor to: receive the electromagnetic field measurement from the electromagnetic field sensor; and determine whether ischemia is present in the heart of the individual.
• the software is further configured to cause the processor to determine whether the individual has a coronary artery occlusion.
• the coronary artery occlusion is determined to be a degree of occlusion of >50%. %. In some embodiments, the coronary artery occlusion is determined to be a degree of occlusion of >50% and ⁇ 70%. In some embodiments, the coronary artery occlusion is determined to be a degree of occlusion of >70%. In some embodiments, the coronary artery occlusion is determined to be a degree of occlusion of >70%. In some embodiments, the coronary artery occlusion is determined to be a degree of occlusion of >90%. In some embodiments, the individual has at least one negative troponin value.
• the individual has a normal electrocardiogram.
• the electromagnetic field sensor comprises an optically pumped magnetometer or a superconducting quantum interference device type sensor.
• the trained machine learning software module has access to stored data comprising a plurality of electromagnetic field values sensed from a plurality of individuals within a population.
• the stored data comprises a plurality of health data values associated with the plurality of individuals.
• the trained machine learning software module has access to data used to train the trained machine learning software module.
• the data used to train the trained machine learning software module comprises heart related data.
• the heart related data comprises an electromagnetic field associated with a heart of the individual.
• the processor is further configured to translate the electromagnetic measurement to a waveform.
• the software is further configured to cause the processor to determine a therapy or other course of action for treating the individual or for providing clinical decision making support.
• the software is further configured to determine a triage pathway for a patient.
• Described herein is a diagnostic method comprising: receiving an electromagnetic field measurement associated with the heart of an individual from an electromagnetic field sensor operably coupled to a sensing device comprising a processor and a trained machine learning software module; determining, using the processor, a presence of ischemia in the heart of individual based on the electromagnetic field measurement.
• the software is further configured to cause the processor to determine whether the individual has a coronary artery occlusion.
• the coronary artery occlusion is determined to be a degree of occlusion of >50%.
• the coronary artery occlusion is determined to be a degree of occlusion of >50% and ⁇ 70%.
• the coronary artery occlusion is determined to be a degree of occlusion of >70%. In some embodiments, the coronary artery occlusion is determined to be a degree of occlusion of >70%. In some embodiments, the coronary artery occlusion is determined to be a degree of occlusion of >90%. In some embodiments, the individual has at least one negative troponin value. In some embodiments, the individual has a normal electrocardiogram. In some embodiments, the sensing device comprises a sensor array and wherein the electromagnetic field sensor is positioned within the array. In some embodiments, the electromagnetic field sensor comprises an optically pumped magnetometer or a superconducting quantum interference device type sensor.
• the method comprises accessing, by the trained machine learning software module, stored data comprising a plurality of electromagnetic field values sensed from a plurality of individuals within a population.
• the stored data comprises a plurality of health data values associated with the plurality of individuals.
• the method comprises accessing, by the trained machine learning software module, data used to train the trained machine learning software module.
• the data used to train the trained machine learning software module comprises heart related data.
• the heart related data comprises an electromagnetic field associated with a heart of the individual.
• the processor is further configured to translate the electromagnetic field measurement to a waveform.
• the software is further configured to cause the processor to determine a therapy or other course of action for treating the individual or for providing clinical decision making support.
• the software is further configured to determine a triage pathway for a patient.
• FIGS. 1A and IB show schematic examples of neural network architecture in terms of flow of data within the neural network.
• FIG. 2 shows a schematic representing an exemplary machine learning software module.
• FIG. 3 shows a schematic representation of an exemplary device for sensing an analyzing an EMF.
• FIG. 4 shows a schematic of an exemplary embodiment of a system comprising a digital processing device.
• FIGS. 5A and 5B show examples of EMF data sensed from a plurality of OPM sensors positioned within proximity to a chest of an individual.
• FIGS. 6A and 6B show examples of EMF data sensed from a plurality of SQUID sensors positioned within proximity to a chest of an individual.
• FIGS. 7A and 7B show examples of EMF data sensed from a plurality of OPM sensors positioned within proximity to a fetus.
• FIGS. 8A and 8B show examples of EMF data sensed from a plurality of SQUID sensors positioned within proximity to a fetus.
• FIG. 9 shows three spatiotemporal activation representations of the magnetic activation of a healthy adult heart.
• FIG. 10 shows a schematic representation of cardiac related EMF data received by a DNN which is configured to output a reconstruction of that EMF data X.
• FIG. 11 shows an exemplary system as described herein which further configured to include one or more devices and software as described herein.
• FIGS. 12A and 12B respectively show a negative result of a scan using an EMF sensing and analysis system as described herein, wherein FIG. 12A corresponds to a result of a first test subject and FIG. 12B corresponds to a test result of a second subject.’
• FIGS. 13A and 13B respectively show a positive result of a scan using an EMF sensing and analysis system as described herein, wherein FIG. 13A corresponds to a result of a first test subject and FIG. 13B corresponds to a test result of a second subject.
• FIG. 14 shows a schematic representation of EMF sensing and analysis scan results as compared to CA and ST results.
• Described herein are systems, devices, software, and methods for determining an abnormality in an individual by sensing and analyzing EMF data associated with the individual. More specifically, described herein are systems, devices, software, and methods for evaluating a heart of an individual for the presence of an abnormality, disease, or pre-disease state.
• an EMF is sensed by an EMF sensor and then the EMF is analyzed by a machine learning algorithm which is configured to determine whether ischemia is present in a tissue of a heart.
• the machine learning algorithm in some embodiments, is further configured to determine if a coronary artery obstruction is present indicating CAD.
• the machine learning algorithm is further configured to evaluate a degree of coronary artery occlusion or level of severity of same.
• the systems, devices, software, and methods described herein are configured to identify a diagnostic modality that should be used as an additional step in evaluating an individual who is found by the systems, devices, software, and methods to have CAD.
• the systems, devices, software, and methods described herein are configured to identify a therapy for an individual.
• Sensed EMF data sensed by one or more EMF sensors, is received and analyzed by a software module comprising a machine learning software algorithm (also referred to herein as a machine learning software module).
• a software module comprising a machine learning software algorithm (also referred to herein as a machine learning software module).
• a software module as described herein comprises computer readable and executable code.
• a software module comprises a file, a section of code, a programming object, a programming structure, or combinations thereof.
• a software module comprises a plurality of files, a plurality of sections of code, a plurality of programming objects, a plurality of programming structures, or combinations thereof.
• the one or more software modules comprise, by way of non limiting examples, a web application, a mobile application, and a standalone application.
• software modules are in one computer program or application. In other embodiments, software modules are in more than one computer program or application. In some embodiments, software modules are hosted on one machine.
• software modules are hosted on more than one machine. In further embodiments, software modules are hosted on cloud computing platforms. In some embodiments, software modules are hosted on one or more machines in one location. In other embodiments, software modules are hosted on one or more machines in more than one location.
• a machine learning software module as described herein is configured to receive data, analyze data, and generate an output.
• an output generated by a machine learning software module include an abnormality, a disease state, an imbalance, a diagnosis, a prognosis, a prediction of a change in health status, a therapy suggestion including preventative therapy.
• a machine learning software module is configured to analyze sensed EMF data and generate a hypothesis function based on said sensed data.
• a hypothesis function generated by a machine learning software module as described herein in some embodiments, is configured to determine a presence of an abnormality in an individual based on an EMF sensed from the individual and inputted into the machine learning software module.
• a hypothesis function generated by a machine learning software module as described herein is configured to determine a prognosis for an individual based on an EMF sensed from the individual and inputted into the machine learning software module.
• a hypothesis function is configured to determine a therapy suggestion for an individual based on an EMF sensed from the individual and inputted into the machine learning software module, wherein said therapy does one or more of treat an existing abnormality and prevent the onset of an abnormality.
• Analysis of EMF data by a machine learning software module comprises identification of an abnormality associated with sensed EMF data.
• a machine learning algorithm as described herein may receive EMF data sensed from an individual and analyze said data to determine that said individual suffers from congestive heart failure.
• a neural network is a type of computational system that can learn the relationships between an input data set and a target data set.
• a neural network is a software representation of a human neural system (e.g. cognitive system), intended to capture “learning” and“generalization” abilities as used by a human.
• the software module comprises a neural network comprising a convolutional neural network.
• Non limiting examples of structural components of embodiments of the machine learning software described herein include: (deep) convolutional neural networks, (deep) recurrent neural networks, (deep) dilated convolutional neural networks, (deep) fully connected neural networks, deep generative models, and (deep) (restricted) Boltzmann machines.
• a machine learning software module comprises a recurrent neural network software module.
• a recurrent neural network software module is configured to receive sequential data as an input, such as consecutive EMF measurements, and the recurrent neural network software module updates an internal state at every time step.
• a machine learning software module comprises a supervised or unsupervised learning method such as, for example, SVM, random forests, clustering
• a machine learning software module comprises a neural network comprising a CNN, RNN, dilated CNN, fully connected neural networks, deep generative models and deep restricted Boltzmann machines.
• a neural network is comprised of a series of layers termed “neurons.” Atypical neuron in a neural network is shown in FIG. 1A. As illustrated in FIG. IB, in embodiments of neural networks, there is an input layer, to which data is presented; one or more internal, or“hidden,” layers; and an output layer. A neuron may be connected to neurons in other layers via connections that have weights, which are parameters that control the strength of the connection. The number of neurons in each layer may be related to the complexity of the problem to be solved. The minimum number of neurons required in a layer may be determined by the problem complexity, and the maximum number may be limited by the ability of the neural network to generalize.
• the input neurons may receive data from data being presented and transmit that data to the first hidden layer through connections’ weights, which are modified during training.
• the first hidden layer may process the data and transmit its result to the next layer through a second set of weighted connections.
• Each subsequent layer may“pool” the results from the previous layers into more complex relationships.
• neural networks are programmed by training them with a known sample set and allowing them to modify themselves during (and after) training so as to provide a desired output such as an output value. After training, when a neural network is presented with new input data, it is configured to generalize what was“learned” during training and apply what was learned from training to the new previously unseen input data in order to generate an output associated with that input.
• a machine learning software module comprises a neural network such as a deep convolutional neural network.
• the network is constructed with any number of convolutional layers, dilated layers or fully connected layers.
• the number of convolutional layers is between 1-10 and the dilated layers between 0-10.
• the number of convolutional layers is between 1-10 and the fully connected layers between 0-10.
• FIG. 2 shows a flow chart representing the architecture of an exemplary embodiment of a machine learning software module.
• raw EMF 240 of the individual is used to extract the MFCC features 245 which are fed into the deep learning module.
• the number of blocks, and the number of layers in each block can increase or decrease, so it is not limited to the configuration portrayed in FIG 2.
• a machine learning software module as described herein is configured to undergo at least one training phase wherein the machine learning software module is trained to carry out one or more tasks including data extraction, data analysis, and output generation.
• the software application comprises a training module that trains the machine learning software module.
• the training module is configured to provide training data to the machine learning software module, said training data comprising, for example, EMF measurements and the corresponding abnormality data.
• said training data is comprised of simulated EMF data with corresponding simulated abnormality data.
• a machine learning software module utilizes automatic statistical analysis of data in order to determine which features to extract and/or analyze from an EMF measurement. In some of these embodiments, the machine learning software module determines which features to extract and/or analyze from an EMF based on the training that the machine learning software module receives.
• a machine learning software module is trained using a data set and a target in a manner that might be described as supervised learning.
• the data set is conventionally divided into a training set, a test set, and, in some cases, a validation set.
• a target is specified that contains the correct classification of each input value in the data set. For example, a set of EMF data from one or more individuals is repeatedly presented to the machine learning software module, and for each sample presented during training, the output generated by the machine learning software module is compared with the desired target. The difference between the target and the set of input samples is calculated, and the machine learning software module is modified to cause the output to more closely approximate the desired target value.
• a back-propagation algorithm is utilized to cause the output to more closely approximate the desired target value.
• the machine learning software module output will closely match the desired target for each sample in the input training set.
• new input data not used during training
• it may generate an output classification value indicating which of the categories the new sample is most likely to fall into.
• the machine learning software module is said to be able to“generalize” from its training to new, previously unseen input samples. This feature of a machine learning software module allows it to be used to classify almost any input data which has a mathematically formulatable relationship to the category to which it should be assigned.
• the machine learning software module utilizes an individual learning model.
• An individual learning model is based on the machine learning software module having trained on data from a single individual and thus, a machine learning software module that utilizes an individual learning model is configured to be used on a single individual on whose data it trained.
• the machine training software module utilizes a global training model.
• a global training model is based on the machine training software module having trained on data from multiple individuals and thus, a machine training software module that utilizes a global training model is configured to be used on multiple patients/individuals.
• the machine training software module utilizes a simulated training model.
• a simulated training model is based on the machine training software module having trained on data from simulated EMF measurements.
• a machine training software module that utilizes a simulated training model is configured to be used on multiple patients/individuals.
• the use of training models changes as the availability of EMF data changes. For instance, a simulated training model may be used if there are insufficient quantities of appropriate patient data available for training the machine training software module to a desired accuracy. This may be particularly true in the early days of implementation, as few appropriate EMF measurements with associated abnormalities may be available initially. As additional data becomes available, the training model can change to a global or individual model. In some embodiments, a mixture of training models may be used to train the machine training software module. For example, a simulated and global training model may be used, utilizing a mixture of multiple patients’ data and simulated data to meet training data requirements.
• Unsupervised learning is used, in some embodiments, to train a machine training software module to use input data such as, for example, EMF data and output, for example, a diagnosis or abnormality.
• Unsupervised learning in some embodiments, includes feature extraction which is performed by the machine learning software module on the input data. Extracted features may be used for visualization, for classification, for subsequent supervised training, and more generally for representing the input for subsequent storage or analysis. In some cases, each training case may consist of a plurality of EMF data.
• Machine learning software modules that are commonly used for unsupervised training include k-means clustering, mixtures of multinomial distributions, affinity propagation, discrete factor analysis, hidden Markov models, Boltzmann machines, restricted Boltzmann machines, autoencoders, convolutional autoencoders, recurrent neural network autoencoders, and long short-term memory autoencoders. While there are many unsupervised learning models, they all have in common that, for training, they require a training set consisting of biological sequences, without associated labels.
• a machine learning software module may include a training phase and a prediction phase.
• the training phase is typically provided with data in order to train the machine learning algorithm.
• types of data inputted into a machine learning software module for the purposes of training include medical image data, clinical data (e.g. from a health record), encoded data, encoded features, or metrics derived from an electromagnetic field.
• Data that is inputted into the machine learning software module is used, in some embodiments, to construct a hypothesis function to determine the presence of an abnormality.
• a machine learning software module is configured to determine if the outcome of the hypothesis function was achieved and based on that analysis make a determination with respect to the data upon which the hypothesis function was constructed.
• the outcome tends to either reinforce the hypothesis function with respect to the data upon which the hypothesis functions was constructed or contradict the hypothesis function with respect to the data upon which the hypothesis function was constructed.
• the machine learning algorithm will either adopts, adjusts, or abandon the hypothesis function with respect to the data upon which the hypothesis function was constructed.
• the machine learning algorithm described herein dynamically learns through the training phase what characteristics of an input (e.g. data) is most predictive in determining whether the features of a patient EMF display any abnormality.
• a machine learning software module is provided with data on which to train so that it, for example, is able to determine the most salient features of a received EMF data to operate on.
• the machine learning software modules described herein train as to how to analyze the EMF data, rather than analyzing the EMF data using pre-defmed instructions.
• the machine learning software modules described herein dynamically learn through training what characteristics of an input signal are most predictive in determining whether the features of an EMF display any abnormality.
• the machine learning software module is trained by repeatedly presenting the machine learning software module with EMF data along with, for example, abnormality data.
• abnormality data is meant to comprise data concerning the existence or non-existence of an abnormality in an organ, tissue, body, or portion thereof. Any disease, disorder or condition associated with the abnormality is included in the abnormality data if available. For example, information concerning a subject displaying symptoms of hypertension, ischemia or shortness of breath is included as abnormality data. Information concerning a subject’s lack of any irregular health condition is also included as abnormality data.
• the abnormality data may be used as additional data being used to simulate the organ, tissue, body, or portion thereof.
• more than one abnormality is included in the abnormality data.
• more than one condition, disease or disorder is included in the abnormality data.
• training begins when the machine learning software module is given EMF data and asked to determine the presence of an abnormality.
• the predicted abnormality is then compared to the true abnormality data that corresponds to the EMF data.
• An optimization technique such as gradient descent and backpropagation is used to update the weights in each layer of the machine learning software module so as to produce closer agreement between the abnormality probability predicted by the machine learning software module, and the presence of the abnormality. This process is repeated with new EMF data and abnormality data until the accuracy of the network has reached the desired level.
• the abnormality data additionally comprises the type and location of the abnormality.
• the abnormality data may indicate that an abnormality is present, and that said abnormality is an ischemia of the left ventricle of the heart.
• the abnormality data additionally comprises a known resulting or related disease, disorder or condition associated with an identified abnormality.
• the abnormality data may indicate that the subject possesses an atrial flutter and arterial coronary disease.
• training begins when the machine learning software module is given the corresponding EMF data and asked to determine the presence of a condition, disorder or disease.
• the output data is then compared to the true abnormality data that corresponds to the EMF data.
• An optimization technique is used to update the weights in each layer of the machine learning software module so as to produce closer agreement between the abnormality probability predicted by the machine learning software module, and the actual abnormality. This process is repeated with new EMF data and abnormality data until the accuracy of the network has reached the desired level.
• the machine learning module is able to analyze an EMF measurement and determine the presence of an abnormality, the type and location of said abnormality and the conditions associated with such.
• the machine learning software module receives EMF data and directly determines the abnormality probability of the subject, wherein the abnormality probability comprises the probability that the EMF measurement is associated with the abnormality of the subject.
• the machine learning software module is trained on a single continuous EMF measurement with corresponding abnormality data over a period of time. This can greatly increase the amount of training data available to train a machine learning software module. For example, in a EMF recording consisting of N continuous 10-second segments with accompanying abnormality data, one can generate at least N*N pairs of such segments to train on.
• an individual’s abnormality data is inputted by the individual of the system. In some embodiments, an individual’s abnormality data is inputted by an entity other than the individual. In some embodiments, the entity can be a healthcare provider, healthcare professional, family member or acquaintance. In additional embodiments, the entity can be the instantly described system, device or an additional system that analyzes EMF measurements and provides data pertaining to physiological abnormalities. [0051] In some embodiments, a strategy for the collection of training data is provided to ensure that the EMF measurements represent a wide range of conditions so as to provide a broad training data set for the machine learning software module. For example, a prescribed number of measurements during a set period of time may be required as a section of a training data set.
• EMF measurements taken with variations in a subject’s physical state may be included in the training data set.
• physical states include accelerated heart rate and enhanced brain signaling. Additional examples include the analysis of a subjects EMF data under the influence of medication or during the course of medical treatment.
• training data may be generated by extracting random overlapping segments of EMF measurements performed by the subject.
• training examples can be provided by measurement recordings, models or algorithms that are
• a network may be trained using 5 EMF segments extracted from a subject’s measurements, and 15,000 EMF segments taken from another subject’s recordings.
• Training data can be acquired using two different methods. The first method is to directly measure the EMF measurements over a subject’s chest. The second method involves creating an accurate electro-anatomical model of the heart. This electro- anatomical model can be used to generate EMF measurements of both healthy and diseased subjects. The measurements are acquired by applying the Biot-Savart Law. This calculates the magnetic field vector at a given point in space, caused by a specific movement of current. After the EMF measurements have been acquired or calculated, they are fed into the network with a classification label, describing both the presence and location of diseased tissue.
• a machine learning algorithm is trained using a large patient database of medical image and/or clinical data and/or encoded data from one or more EMF measurements and/or any features or metrics computed from the above said data with the corresponding ground-truth values.
• the training phase constructs a transformation function for predicting probability of an abnormality in an unknown patient’s organ, tissue, body, or portion thereof by using the medical image and/or clinical data and/or encoded data from the one or more EMF measurements and/or any features or metrics computed from the above said data of the unknown patient.
• the machine learning algorithm dynamically learns through training what characteristics of an input signal are most predictive in determining whether the features of a patient EMF data display any abnormality.
• a prediction phase uses the constructed and optimized transformation function from the training phase to predict the probability of an abnormality in an unknown patient’s organ, tissue, body, or portion thereof by using the medical image and/or clinical data and/or encoded data from the one or more EMF measurements and/or any features or metrics computed from the above said data of the unknown patient.
• the machine learning algorithm is used to determine, for example, the presence or absence of an abnormality on which the system was trained using the prediction phase.
• the system can identify the location and type of an abnormality, and present conditions associated with such abnormality. For example, an EMF measurement is taken of a subject’s brain and appropriate data derived from the EMF
• a machine learning software algorithm detects an abnormality associated with epilepsy.
• the machine learning algorithm further localizes an anatomical region associated with an abnormality such as, for example, localizing an area of the brain of an individual associated with epilepsy in the individual based on an EMF
• a subject is known to possess arterial ischemia and has EMF measurements recorded before and after treatment with a medication.
• the medical image and/or clinical data and/or encoded data from the EMF measurements and/or features and/or metrics derived from the said data are submitted for analysis to a system using the described trained machine learning algorithm in order to determine the effectiveness of the medication on abnormal blood flow using the prediction phase.
• the prediction phase uses the constructed and optimized hypothesis function from the training phase to predict the probability of an abnormality in an unknown patient’s organ, tissue, body, or portion thereof by using the medical image and/or clinical data and/or encoded data from the EMF measurements and/or any features or metrics computed from the above said data of the unknown individual.
• the machine learning software module in the prediction phase, can be used to analyze data derived from its EMF measurement independent of any system or device described herein.
• the new data recording may provide a longer signal window that required for determining the presence of a subject’s abnormality.
• the longer signal can be cut to an appropriate size, for example 10 seconds and then can be used in the prediction phase to predict the probability of an abnormality of the new patient data.
• a probability threshold can be used in conjunction with a final probability to determine whether or not a given recording matches the trained abnormality.
• the probability threshold is used to tune the sensitivity of the trained network.
• the probability threshold can be 1%, 2%, 5%, 10%, 15%, 20%, 25%, 30%, 35%, 40%, 45%, 50%, 55%, 60%, 65%, 70%, 75%, 80%, 85%, 90%, 95%, 98% or 99%.
• the probability threshold is adjusted if the accuracy, sensitivity or specificity falls below a predefined adjustment threshold.
• the adjustment threshold is used to determine the parameters of the training period. For example, if the accuracy of the probability threshold falls below the adjustment threshold, the system can extend the training period and/or require additional measurements and/or abnormality data. In some embodiments, additional measurements and/or abnormality data can be included into the training data. In some embodiments, additional measurements and/or abnormality data can be used to refine the training data set.
• a machine learning software module is typically provided with data (input) in order to train the machine learning software module as to how to analyze an EMF to determine, for example, the presence of an abnormality.
• Input data is also used by a machine learning software module to generate an output.
• An input to a machine learning algorithm as described herein is data transmitted to the machine learning algorithm by a device or a system which includes an EMF sensor.
• data that is received by a machine learning algorithm software module from an electromagnetic sensor as an input may comprise EMF data expressed in a standard unit of measurement such as, for example, Tesla.
• sensed EMF data comprises an overall or total EMF generated by a body of an individual based on numerous different currents generated by the body of the individual. That is, in some embodiments, one or more EMF sensors sense an EMF that comprises an EMF associated with entire individual and is not specific to a single organ, tissue, body, or portion thereof. Likewise, in some embodiments, an EMF that is sensed from an individual that is associated with a portion of the individual but not specific to a single organ, tissue, body, or portion thereof.
• sensed EMF data comprises an EMF that is in proximity to an individual or a portion of the body of the individual and comprises an EMF associated with a single organ, organ system, or tissue.
• one or more EMF sensors are positioned in proximity to a chest of an individual and sense an EMF associated with a heart of the individual.
• one or more EMF sensors are positioned in proximity to a head of an individual and sense an EMF associated with a brain of the individual.
• one or more EMF sensors are positioned in proximity to a chest of an individual and sense an EMF associated with a cardio-pulmonary system (i.e. the heart and lungs).
• a machine learning software module is configured to receive an encoded length of EMF data as an input and to determine the window length of the input data. For example, an input to a machine learning software module in some embodiments described herein is 100 seconds of encoded EMF data, and the machine learning software module selects a 10 second segment within the 100 second data sample for examination. In some embodiments, the input is segmented into multiple inputs, any number of which is analyzed independently.
• a device, system, or method as described herein is configured to sense and/or receive data comprising data associated with an individual.
• Data is sensed, in some embodiments, by an electromagnetic field sensor that is a component of a device, system, or method described herein.
• Data is received, in some embodiments, by transmission of data to a software algorithm as described herein by a source other than an EMF that is a component of a device, system, or method that also includes the software algorithm. That is, data, in some embodiments, is received from a source remote from the device, system, or method that includes the software algorithm.
• data that is received comprises stored data.
• data that is received comprises data that is generated by a software module.
• sensed and/or received data comprises an input to a machine learning algorithm as described herein. An input is used to train a machine learning algorithm and/or is used by the machine learning algorithm to carry out an analysis or prediction.
• Data as described herein comprises EMF data as well as other information associated with an individual.
• data used as an input for a machine learning algorithm as described herein includes a medical record (e.g. an electronic health record), a diagnosis, a lab value, a vital sign, a prognosis, an electrocardiogram, a radiology image (including ultrasound, CT scan, MRI, and X-ray), an electroencephalogram, and a pathology report.
• a medical record e.g. an electronic health record
• diagnosis e.g. an electronic health record
• a lab value e.g. an a lab value
• a vital sign e.g. an electrocardiogram
• radiology image including ultrasound, CT scan, MRI, and X-ray
• electroencephalogram e.g., a pathology report.
• two or more different types of data are combined and/or correlated by the software algorithms described herein.
• EMF data in some embodiments, is used to generate other types of data that are used by the software algorithms described herein.
• EMF data in some embodiments, is used to generate medical image data which, in some embodiments, is achieved using Magnetic Field Maps (MFM).
• EMF data is used to generate medical image data using Pseudo-Current Density (PCD) maps.
• EMF data is used to generate medical data using Spatio-Temporal Activation Graphs (STAG).
• STAG Spatio-Temporal Activation Graphs
• EMF data in some embodiments, is used to generate clinical data such as MCG, MEG and MGG measurements.
• input to a software algorithm as described herein comprises EMF data which is encoded into some other form of data and the features or metrics computed from the encoded data such as, for example, MFCC.
• input to a software algorithm as described herein is generated by a computer.
• an input to a software algorithm as described herein comprises data generated by computer simulation.
• a computer simulation generates an image or other representation of an organ or other tissue (including skin, bone, and blood).
• a computer simulation generates an image or representation of a flow of a fluid such as, for example, blood, lymph, or bile.
• a computer simulation generates an image or representation of a flow of an electric current.
• additional inputs generated by a computer simulation include a medical record (e.g.
• an electronic health record an electronic health record
• diagnosis a lab value
• a vital sign a vital sign
• prognosis an electrocardiogram
• radiology image including ultrasound, CT scan, MRI, and X-ray
• electroencephalogram an electroencephalogram
• data that is received by a machine learning algorithm software module from an electromagnetic sensor as an input may comprise EMF data that has been filtered and or modified.
• filtering comprises a removal of noise or artifact from a sensed electromagnetic field data.
• Artifact or noise may comprise, for example, ambient electromagnetic signals that are sensed together with electromagnetic data sensed from an individual.
• sensed EMF data is filtered prior to and/or after transmission of said data to a processor.
• Filtering of sensed EMF data may, for example, comprise the removal of ambient signal noise from a sensed EMF data.
• Signal noise may, for example, comprises ambient EMF data generated by, for example, electronic devices, the earth’s magnetosphere, electrical grids, or other individuals (i.e. not individuals whose EMF data is being targeted).
• sensed EMF data is converted to another form of data or signal which then undergoes a signal filtering process.
• a device or system includes a processor including software that is configured to convert sensed EMF data to another form of data or signal. The process of converting sensed EMF data to another form of data or signal typically comprises an encoding process, wherein a first form of data is converted into a second form of data or signal.
• sensed EMF data is encoded into an audio signal which undergoes a filtering process.
• sensed EMF data is encoded into an audio signal or alternatively, a signal having the morphology of an audio signal.
• sensed EMF data is encoded into an audio signal which is further processed into a Mel-Frequency Cepstrum from which one or more Mel-Frequency Cepstrum Coefficients (“MFCC”) are derived.
• Mel-Frequency Cepstrum (“MFC”) represents a short term power spectrum of a sound. It is based on a linear cosine transform of a log power spectrum on a nonlinear mel scale of frequency.
• Mel-frequency cepstral coefficients (“MFCCs”) collectively make up an MFC. These are derived from a type of cepstral representation of the audio. In MFC, frequency bands are equally spaced on the mel-scale as compared to the linearly-spaced frequency bands used in the normal cepstrum. This equally spaced frequency bands allows for better representation of audio.
• a sensed EMF signal is filtered by converting the sensed EMF data into an audio signal or a signal having the morphology of an audio signal wave, and then generating MFCCs.
• MFCCs help in identifying the components of the audio signal that are able to
• steps for filtering an audio signal derived from sensed EMF data comprise: In a first step, the audio signal is framed into short frames. In a second step, the periodogram estimate of the power spectrum for each frame is calculated. In a third step, a mel filterbank is applied to the power spectrum and sum the energy in each filter. In a fourth step, the logarithm of all the filterbank energies is determined and the DCT of the log filterbank energies is calculated. In a fifth step, only the first 20 DCT coefficients are kept, and the rest are discarded.
• the filtered data is transmitted to a machine learning algorithm for analysis.
• the algorithm described herein is capable of classifying and characterizing the physiological health of human body tissues.
• the algorithm is designed to analyze input data and determine the presence and location of diseased tissue in the organ(s) recorded by
• EMF data is sensed using a device or system.
• a device or system comprises one or more EMF sensors.
• the device or system is configured to include a machine learning software module as described herein.
• the device or system is configured to transmit a sensed EMF to a machine learning software module not included as part of the device or system.
• EMF data that is sensed using an electromagnetic sensor comprises electromagnetic data associated with a passage of a current through a cell, tissue, and/or organ of an individual, such as, for example, the heart of the individual.
• a device and/or a system comprises a digital processing device configured to run a software application as described herein.
• a digital processing device includes one or more hardware central processing units (CPUs) or general purpose graphics processing units (GPGPUs) that carry out the device’s functions.
• the digital processing device further comprises an operating system configured to perform executable instructions.
• the digital processing device is optionally connected to a computer network.
• the digital processing device is optionally connected to the Internet such that it accesses the World Wide Web.
• the digital processing device is optionally connected to a cloud computing infrastructure.
• the digital processing device is optionally connected to an intranet.
• the digital processing device is optionally connected to a data storage device.
• suitable digital processing devices include, by way of non-limiting examples, server computers, desktop computers, laptop computers, notebook computers, sub-notebook computers, netbook computers, netpad computers, handheld computers, and tablet computers.
• the digital processing device includes an operating system configured to perform executable instructions.
• the operating system is, for example, software, including programs and data, which manages the device’s hardware and provides services for execution of applications.
• suitable operating systems include FreeBSD, OpenBSD, NetBSD ® , Linux, Apple ® Mac OS X Server ® , Oracle ® Solaris ® , Windows Server ® , and Novell ® NetWare ® .
• suitable personal computer operating systems include, by way of non-limiting examples, Microsoft ® Windows ® , Apple ® Mac OS X ® , UNIX ® , and UNIX-like operating systems such as GNU/Linux ® .
• the operating system is provided by cloud computing.
• a digital processing device includes a storage and/or memory device.
• the storage and/or memory device is one or more physical apparatuses used to store data or programs on a temporary or permanent basis.
• the device is volatile memory and requires power to maintain stored information.
• the device is non-volatile memory and retains stored information when the digital processing device is not powered.
• the non-volatile memory comprises flash memory.
• the non-volatile memory comprises dynamic random-access memory (DRAM).
• the non-volatile memory comprises ferroelectric random access memory (FRAM).
• the non-volatile memory comprises phase- change random access memory (PRAM).
• the device is a storage device including, by way of non-limiting examples, CD-ROMs, DVDs, flash memory devices, magnetic disk drives, magnetic tapes, optical disk drives, and cloud computing based storage.
• the storage and/or memory device is a combination of devices such as those disclosed herein.
• the digital processing device includes a display to send visual information to a subject.
• the digital processing device includes an input device to receive information from a subject.
• the input device is a keyboard.
• the input device is a pointing device including, by way of non-limiting examples, a mouse, trackball, track pad, joystick, game controller, or stylus.
• the input device is a touch screen or a multi-touch screen.
• the input device is a microphone to capture voice or other sound input.
• the input device is a video camera or other sensor to capture motion or visual input.
• the input device is a combination of devices such as those disclosed herein.
• FIG. 3 shows a schematic representation of an exemplary device for sensing an analyzing an EMF.
• the patient’s organ emits an EMF 370 which is then acquired from the EMF sensing device 375.
• the data is then processed, filtered and analyzed by a Digital Signal Processing module 380 thereby removing noise if any and extracting important information from the data.
• the processed data is then fed into the deep learning module 385 consisting of dilated convolutional neural networks.
• the deep learning module detects ischemia and localizes to a particular region in an organ 490.
• a cloud computing environment comprises a plurality of computing processors.
• FIG. 4 shows an exemplary embodiment of a system as described herein comprising a digital processing device 401.
• the digital processing device 401 includes a software application configured to perform data analysis such as analyzing an electromagnetic field to determine a condition of a subject.
• the device 401 is configured to run the software application that comprises a machine learning software module including training of the machine learning software module as described herein.
• the digital processing device 401 includes a central processing unit (CPU, also“processor” and“computer processor” herein) 405, which can be a single core or multi -core processor, or a plurality of processors for parallel processing.
• CPU central processing unit
• the digital processing device 401 also includes either memory or a memory location 410 (e.g., random-access memory, read-only memory, flash memory), electronic storage unit 415 (e.g., hard disk), communication interface 420 (e.g., network adapter) for communicating with one or more other systems, and peripheral devices 425, such as cache.
• the memory 410, storage unit 415, interface 420 and peripheral devices 425 are configured to communicate with the CPU 405 through a communication bus (solid lines), such as a motherboard.
• the digital processing device 401 is, in some embodiments, operatively coupled to a computer network (“network”) 430 with the aid of the communication interface 420.
• the network 430 in some embodiments, comprises the Internet.
• the network 430 in some embodiments is a telecommunication and/or data network.
• the CPU 405 is configured to execute machine-readable instructions embodied in a software application or module.
• the instructions may be stored in a memory location, such as the memory 410.
• the storage unit 415 in some embodiments is configured to store files, such as subject data, e.g., subject preferences, subject programs, and subject EMF data.
• subject data e.g., subject preferences, subject programs, and subject EMF data.
• Methods as described herein can be implemented by way of machine (e.g., computer processor) executable code stored on an electronic storage location of the digital processing device 401, such as, for example, on the memory 410 or electronic storage unit 415.
• the machine executable or machine readable code can be provided in the form of a software application or software module.
• the code can be executed by the processor 405.
• the code can be retrieved from the storage unit 415 and stored on the memory 410 for ready access by the processor 405.
• the electronic storage unit 415 can be precluded, and machine-executable instructions are stored on memory 410.
• a remote device 435 is configured to communicate with the digital processing device 401, and may comprises any mobile computing device, non-limiting examples of which include a tablet computer, laptop computer, smartphone, or smartwatch.
• a remote device 435 may comprises an integrated sensor or alternatively be coupled to a sensor that is configured to sense EMF data.
• sensed EMF data is transmitted directly from an electromagnetic sensor to a processor on a computing device that is encoded with a machine learning algorithm configured to analyze the received EMF data.
• Software comprising one or more software modules as described herein may, for example, be a component of a device or system that includes one or more sensors comprising an EMF sensor.
• This sensor records the magnetic fields that are naturally emitted by certain organs during physiological activity. Such organs may include the brain, heart or liver.
• this sensor may take the form of a magnetometer, fluxgate, or a superconducting quantum interference device (SQUID) fitted to perform biomagnetic measurements on an organ of interest. More specifically, any sensor or plurality of sensors that is capable of sensing an electromagnetic field is suitable for use with the devices, systems, software, and methods described herein.
• SQUID superconducting quantum interference device
• a device comprises a sensor, such as an optically pumped magnetometer (OPM) as a measurement tool, which, in some embodiments, utilizes nonradioactive self-contained alkali metal cells coupled with a closed pumping laser and photodetector setup to measure minute magnetic fields.
• OPM optically pumped magnetometer
• the devices and systems utilize OPMs in an n x n array (or grid) or alternative geometric configuration to collect magnetic field data at n discrete locations over, for example, a portion of a body of an individual such as a chest area, which, in some embodiments, is digitized using pickup electronics.
• OPMs are typically configured to utilize nonradioactive self-contained alkali metal cells coupled with a closed pumping laser and photodetector setup to measure minute magnetic fields.
• SQUIDs superconducting quantum interference devices
• OPM sensors are significantly smaller and typically do not require the use of cryogenic cooling.
• the Earth’s magnetic field is naturally present everywhere on Earth, and the amplitude is about 50 microtesla.
• OPM performance is enhanced in at least two exemplary ways in the presence of the Earth’s ambient magnetic field.
• a reference value representing Earth’s magnetic field is used as part of a vector subtraction to isolate a signal of interest in an OPM.
• Another technique involves the use of a gradiometer for active noise cancellation for the OPM.
• a sensor array configuration as utilized in some embodiments of the devices and systems described herein, comprises a custom array configuration.
• a sensor array configuration is customized to an individual’s anatomy.
• a sensor array configuration is customized to a location on the individual which is measured, such as a chest location or a head location.
• a sensor array configuration is customized to a measurement type that a device is programmed to acquire.
• a sensor array configuration is customized to be operatively coupled with a shield and/or an arm.
• a sensor array configuration is interchangeable with a different array configuration - a user may perform with interchange.
• An array configuration in some embodiments, comprises an arc (such as a generally curved shape) having a depth and comprising a radius from about 20 cm to about 50 cm or from about 10 cm to about 60 cm.
• An array configuration, such as an arc configuration in some embodiments, comprises one or more variable inter-magnetometer distances and variable sensor densities.
• An array configuration in some embodiments, comprises a concave structure (such as a concave structure configured to wrap or form around a body region, such as a head or chest).
• One or more magnetometers is positioned on at least a portion of a surface of the concave structure.
• configuration in some embodiments, comprises one or more variable inter-magnetometer distances and variable sensor density.
• a sensor array n x n sensors In some embodiments, a sensor array is a 2D rectangular array, such as a 2 x 2 array or a 4 x 4 array. In some embodiments, a sensor array is a 2D non-rectangular array, such as a 2 x 1 array or a 4 x 1 array. In some embodiments, a sensor array is a circular array or a semicircular array, such as a 3D array of sensors positioned in an arc or concave structure. In some embodiments, a sensor array is a 2D array or a 3D array. In some embodiments, a sensor of a sensor array comprises x, y, and z coordinates.
• An array in some embodiments, comprises three sensors.
• An array in some embodiments, comprises four sensors.
• An array in some embodiments, comprises nine sensors.
• An array in some embodiments, comprises sixteen sensors.
• An array in some embodiments, comprises 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50 sensors or more.
• a sensor array comprises 8 sensors.
• a sensor array comprises 16 sensors.
• a sensor array comprises a single sensor housed in a single housing.
• a sensor array comprises a plurality of sensors housed in a single housing, such as a housing having multiple sensor configurations or changeable sensor configurations.
• a sensor array comprises a plurality of sensors housed in a plurality of housings. In some embodiments, a sensor array comprises a plurality of sensors, each sensor housed in a separate housing. In some embodiments, a first sensor and second sensor of a sensor array is different. In some embodiments, a first sensor and a second sensor of a sensor array is the same. In some embodiments, each sensor of a sensor array is unique. In some embodiments, each sensor of a sensor array is identical. In some embodiments, a subset of sensors within a sensor array is unique. In some embodiments, a subset of sensors within a sensor array is identical.
• Spatial positioning of a sensor in a sensor array is adjustable, such as by a user or automated by a controller. In some embodiments, spatial positioning of a sensor in a sensor array is fixed. In some embodiments, a number of sensors in a sensor array is selected based on an application. In some embodiments, a number of sensors in a sensor array is selected based on a type of measurement or a location of a measurement. An array, in some embodiments, comprises a single channel array or a multi-channel array. In some embodiments, increasing a number of sensors of a sensor array increases a resolution of a measurement taken by the array. In some embodiments, a sensor array of sensors is densely packed, such as substantially adjacent or proximal one another.
• An array of sensors is sparsely spaced, such as having a spacing between one another.
• a subset of sensors of a sensor array is densely packed.
• a subset of sensors of a sensor array is sparsely spaced or densely spaced.
• centerpoints of any two sensors of a densely packed subset of sensors is spaced less than about: 5, 4.5, 4, 3.5, 3, 2.5, 2, 1.5, 1, 0.5, 0.1 centimeters (cm) apart.
• centerpoints of densely packed sensors is spaced centerpoint to centerpoint from about 0.1 cm to about 2.0 cm or from about 0.1 cm to about 1.5 cm or from about 1.0 cm to about 2.0 cm.
• centerpoints of any two sensors of a sparsely packed subset of sensors is spaced more than about: 1.5, 2, 2.5, 3, 3.5, 4, 4.5, 5, 8, 10 cm apart.
• centerpoints of sparsely packed sensors is spaced centerpoint to centerpoint from about 1.5 cm to about 3 cm or from about 2 cm to about 5 cm or from about 2.5 cm to about 8 cm.
• a center point is a central location of a sensor, such as a central axis.
• a centerpoint of a circular sensor is a central point at which all other edge points are of equal distance.
• a densely packed array indicates intermagnetometer placement of less than 1.5 cm, while magnetometer placement of greater than about 1.5 cm constitutes a sparsely packed array.
• a housing is configured to house a sensor or a sensor array of sensors. In some embodiments, the housing is configured to accommodate a single configuration of sensor spacing within the housing. In some embodiments, the housing is configured to accommodate multiple configurations of sensor spacing within the housing. In some
• the housing accommodates (i) adjusting sensor spacing, such as a dense spacing or a sparse spacing, or (ii) varying a number of sensors within the array.
• a housing is a universal housing for a plurality of arrays and array configurations.
• a sensor is configured to sense a presence of or measure a parameter of a magnetic field.
• a sensor in some embodiments, comprises a sensitivity to a magnetic field of about 10 femtotesla per root Hertz (fT/v Hz)
• fT/v Hz femtotesla per root Hertz
• a sensor in some embodiments, comprises a sensitivity of from about 1 fT/v Hz to about 20 fT/v Hz
• a sensor in some embodiments, comprises a sensitivity of from about 5 fT/v Hz to about 15 fT/v Hz
• a sensor in some embodiments, comprises a sensitivity of from about 0.1 ⁇ TL/Hz to about 30 ⁇ TL/Hz.
• a sensor in some embodiments, comprises a sensitivity of from about 0.5 fT/vHz to about 12 fT/vHz
• a sensor in some embodiments, comprises a sensitivity of from about 1 fT/vHz to about 15 fT/vHz
• a sensor in some embodiments, comprises a sensitivity of about: 0.001, 0.005, 0.01, 0.05, 0.1, 0.5, 1, 2, 3, 4, 5, 6, 7, 8, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20 fT/VHz.
• a sensor does not require a cooling element, such as cryogenic cooling, to collect a measurement.
• a sensor collects a measurement over a temperature range of from about 30 degrees Fahrenheit (F) to about 110 degrees F.
• a sensor collects a measurement over a temperature range of from about 50 degrees F to about 110 degrees F.
• a sensor collects a measurement over a time period of from about 1 second to about 5 hours without a need for a cooling element.
• a sensor collects a measurement over a time period of from about 1 second to about 1 hour without a need for a cooling element.
• a sensor collect a measurement over a time period of from about 1 second to about 30 minutes without a need for a cooling element.
• a noise source in some embodiments, comprises a magnetic field strength.
• a strength of a magnetic field of a noise source is measured in units of Tesla (T).
• Noise, such as ambient noise in some embodiments, comprises a magnetic field strength of less than about 100 nanotesla (nT).
• Noise in some embodiments, comprises a magnetic field strength of less than about 1000 nT.
• Noise in some embodiments, comprises a magnetic field strength of less than about 500 nT.
• Noise in some embodiments, comprises a magnetic field strength of less that about 200 nT.
• Noise in some embodiments, comprises a magnetic field strength of less than about 120 nT.
• Noise, in some embodiments comprises a magnetic field strength of less than about 80nT.
• a noise source such as a magnetic field of the Earth, in some embodiments, comprises a magnetic field strength of about 50 microtesla (mT).
• Noise in some embodiments, comprises a magnetic field strength of from about 40mT to about 60mT.
• Noise in some embodiments, comprises a magnetic field strength of from about lOmT to about lOOmT.
• Noise in some embodiments, comprises an amplitude component, a frequency component, or a combination thereof, and, in some embodiments, comprises both sources that is direct current (DC), alternating current (AC), or a combination of the two.
• the systems, methods, devices, and software described herein are used in a number of different applications including in research and healthcare settings, wherein the systems, methods, devices, and software are used to evaluate a status of an individual and in some cases provide a diagnosis for a condition that the individual has.
• a condition may comprise both an abnormality (including a pre-disease condition) as well as a disease state.
• Exemplary types of disease evaluated by the systems, methods, devices, and software described herein include cardiac disease, neurologic disease, and gastrointestinal disease.
• devices, systems, software, and methods described herein provide a suggestion for a next diagnostic step to carry out with the individual following sensing and analyzing the EMF of the individual, such as, for example, an additional diagnostic test or modality that will assist in obtaining a diagnosis.
• diagnostic modalities suggested include imaging, blood testing, and conduction monitoring (e.g. ECG and EEG).
• devices, systems, software, and methods described herein provide a suggestion for a treatment to be provided to an individual following sensing and analyzing the EMF of the individual.
• the systems, methods, devices, and software described herein are used to evaluate an individual for cardiac disease.
• cardiac disease evaluated by the systems, methods, devices, and software described herein include CAD, arrhythmia, and congestive heart failure.
• the systems, methods, devices, and software described herein are used to evaluate an individual for CAD.
• an EMF associated with a heart of an individual is sensed and based on the sensed EMF of the individual, a status of the individual is determined with respect to CAD.
• a determination is made as to whether coronary disease is present in the individual.
• a determination is made as to a degree of severity of a CAD that is present.
• a degree of severity determined in some embodiments, comprises“severe,”“moderate,” or“mild,”
• a degree of severity in some embodiments, comprises a degree of a obstruction of one or more coronary vessels.
• an individual may be determined to have >90% obstruction of their Left Anterior Descending (LAD) artery, >80% obstruction of their LAD, >70% obstruction of their LAD, >60% obstruction of their LAD, or >50% obstruction of their LAD.
• the systems, methods, devices, and software described herein determine a presence of a pre-CAD state or that a risk of developing coronary artery exists in the individual. For example, in some embodiments, it is determined that an individual has a >90% risk of developing moderate to severe CAD, a >80% risk of developing moderate to severe CAD, a >70% risk of developing moderate to severe CAD, a >60% risk of developing moderate to severe CAD.
• the systems, methods, devices, and software described herein are used in an acute care setting to evaluate individuals with chest pain.
• individuals with left sided chest pain of unknown origin are ruled out of having CAD.
• individuals with left sided chest pain of unknown origin are ruled in for having CAD.
• an individual with a normal ECG and/or at last one normal troponin level is assessed by the systems, devices, methods, and software described herein and determined to either have CAD, not have CAD, have a high likelihood of having CAD, or have a high likelihood of not having CAD.
• a system as described herein includes at least one EMF sensor (or a plurality of EMF sensors, or a plurality of EMF sensors arranged in an array) that are positioned in proximity to the heart of an individual.
• the system further comprises shielding to shield the at least one EMF sensor from ambient EMF readings.
• the analysis process comprises the generation, by the software described herein, of a visual representation of the EMF that is then analyzed.
• a sensed EMF that shows a regular pattern without magnetic dipole dispersion represents a normal finding, an absence of a presence of CAD in the individual, or a low likelihood of a presence of CAD in the individual.
• a sensed EMF that shows an irregular pattern of magnetic pole dispersion represents an abnormal finding, a presence of CAD in the individual, or a high likelihood of a presence of CAD in the individual.
• a shift in dipole angulation or significant disorganization in the magnetic field map indicates a greater degree of vessel stenosis (i.e. greater degree of CAD).
• a suggestion for a treatment is provided.
• treatments suggested for CAD include conservative treatment (e.g. improve diet and/or exercise), cholesterol lowering treatment, vasodilating medications, rhythm modulating medications, intravascular interventions including stenting, and bypass surgery.
• the systems, methods, devices, and software described herein are used to evaluate an individual for neurological disease including abnormalities resulting from traumatic injury and stroke.
• neurological disorders evaluated by the systems, methods, devices, and software described herein include epilepsy, stroke, traumatic brain injury, traumatic spine injury, encephalitis, meningitis, tumor, Alzheimer’s disease, Parkinson’s disease, ataxia, and psychiatric disorders including schizophrenia, depression, and bipolar disease.
• the systems, methods, devices, and software described herein are used to evaluate an individual for gastrointestinal disease including any disease or disorder of any component of the gastrointestinal system including the gastrointestinal tract, the liver (including biliary system), and the pancreas.
• gastrointestinal disorders evaluated by the systems, methods, devices, and software described herein include gastrointestinal cancers (including tumors of the gastrointestinal tract, liver, and pancreas), Crohn’s disease, ulcerative colitis, irritable bowel disease, dismotility disorders, gall stones, colitis, cholangitis, liver failure, pancreatitis, and infections of the gastrointestinal system.
• FIG. 5A shows an example of EMF data sensed from a plurality of OPM sensors positioned within proximity to a chest of an individual and, therefore, within proximity to a heart of the individual.
• the EMF sensed is associated with an electric current generated by cells of the heart of the individual.
• the EMF data that is sensed is from a 58 year old male using a plurality of OPM sensors positioned in proximity to the chest of the individual.
• the waveform 500 comprises EMF data in EM units (shown on the Y-axis), said data being sensed over time in seconds (shown on the X-axis).
• waveform 500 comprises a plurality of waveforms sensed from a plurality of OPM sensors positioned in proximity to the chest of the individual.
• the plurality of waveforms are positioned at different locations relative to the chest of the individual.
• the one or more OPM sensors are positioned in a sensor array where the array is itself positioned in proximity to the chest of the individual so that each of the plurality of OPM sensors is in a different position relative to the chest of the individual
• waveform 500 comprises three separate individual waveforms 502, 504, and, 506.
• Waveforms 502, 504, and, 506 each correspond to a single PQRST complex as would be seen on a standard ECG. Therefore, each of the waveforms 502, 504, and, 506 correspond to an individual polarization and depolarization cardiac cycle or one heartbeat.
• FIG. 5B shows examples of a plurality of single EMF waveforms (as compared to the single waveform of FIG. 5A which comprises a plurality of individual waveforms).
• waveforms or tracings 508 and 510 each represent respective EMF signals sensed over the same period of time.
• EMF data is sensed from the same 58 year old male using a plurality of OPM sensors as in the example of FIG. 5A.
• FIG. 5B shows examples of a plurality of single EMF waveforms (as compared to the single waveform of FIG. 5A which comprises a plurality of individual waveforms).
• waveforms or tracings 508 and 510 each represent respective EMF signals sensed over the same period of time.
• EMF data is sensed from the same 58 year old male using a plurality of OPM sensors as in the example of FIG. 5A.
• FIG. 5B shows examples of a plurality of single EMF waveforms (as compared to the single waveform of FIG
• tracings 508 and 510 are each a tracing representing a single waveform sensed over the same period of time and are different at least in part in that each of tracings 508 and 510 correspond to an EMF signal respectively sensed from a different OPM located at a different position relative to the chest (and therefore the heart) of the individual. That is, tracing 508 corresponds to a first EMF signal sensed from a first OPM sensor and tracing 510 corresponds to a second EMF signal sensed from a second OPM sensor where each of the first and second OPM sensors are located at different positions relative to the heart of the individual.
• ECG tracing In a traditional ECG tracing, ten electrodes positioned at different locations on a body of an individual produce 12 ECG tracings, each corresponding to a different“view” of the heart.
• a particular ECG lead corresponds to a particular“view” of the heart in that each ECG tracing corresponds to a different spatial relationship between one or more ECG electrodes and the heart of the individual.
• each of tracings 508 and 510 represent a different “view” of the heart based on the position of the one or more EMF sensors relative to the chest of the individual. That is, current traveling through the heart of an individual generates a different EMF at different locations and as such is results in different appearing tracings based on the location of a sensor sensing that EMF.
• the tracings 508 and 510 each comprise a PQRST complex 512 (or, as shown, a plurality of PQRST complexes).
• the exemplary data from FIGS. 5A and 5B are provided as input to a machine learning software module described herein.
• the machine learning software module receives the input and interrelates the data in numerous ways so that input data relating to the individual (including data related to the individual received concurrently to the input as well as data related to the individual received before and subsequent to receiving the input) is associated with the sensed EMF data shown in the examples of FIGS. 5A and 5B.
• a machine learning software module as described herein correlates the age (58 years old in these examples) and gender (male in these examples) of the individual with one or more of the tracings 500, 508 and 510. Additional data relating to the individual may be provided that the machine learning software module may decide to correlate with the input as well. Non- limiting examples of the additional data comprises health records for the individual, including diagnoses, medications, lab results other EMF sensed data from the individual.
• a machine learning software module as described herein further determines how to interrelate data from this individual and how to interrelate data of other individuals so as to generate a hypothesis function which is used to identify the presence of an abnormality in the individual and/or predict the occurrence of an abnormality in the individual.
• FIG. 6A shows an example of EMF data sensed from a plurality of SQUID sensors positioned within proximity to a chest of an individual and, therefore, within proximity to a heart of the individual.
• the EMF sensed is associated with an electric current generated by cells of the heart of the individual.
• the EMF data that is sensed is from a 58 year old male using a plurality of SQUID sensors positioned in proximity to the chest of the individual.
• the waveform 600 comprises EMF data in EM units (shown on the Y-axis), said data being sensed over time in seconds (shown on the X-axis).
• waveform 600 comprises a plurality of waveforms sensed from a plurality of SQUID sensors positioned in proximity to the chest of the individual.
• the plurality of waveforms are positioned at different locations relative to the chest of the individual.
• the one or more SQUID sensors are positioned in a sensor array where the array is itself positioned in proximity to the chest of the individual so that each of the plurality of SQUID sensors is in a different position relative to the chest of the individual (although it should be understood that one or more SQUID sensors may be arranged in other configurations in other embodiments of the systems, methods, devices, and software described herein and need not always be arranged in an array).
• waveform 600 comprises three separate individual waveforms 602, 604, and, 606. Waveforms 602, 604, and, 606 each correspond to a single PQRST complex as would be seen on a standard ECG.
• each of the waveforms 602, 604, and, 606 correspond to an individual polarization and depolarization cardiac cycle or one heartbeat.
• FIG. 6B shows examples of a plurality of single EMF waveforms (as compared to the single waveform of FIG. 6A which comprises a plurality of individual waveforms).
• waveforms or tracings 608 and 610 each represent respective EMF signals sensed over the same period of time.
• EMF data is sensed from the same 58 year old male using a plurality of SQUID sensors as in the example of FIG. 6A.
• FIG. 6B shows examples of a plurality of single EMF waveforms (as compared to the single waveform of FIG. 6A which comprises a plurality of individual waveforms).
• waveforms or tracings 608 and 610 each represent respective EMF signals sensed over the same period of time.
• EMF data is sensed from the same 58 year old male using a plurality of SQUID sensors as in the example of FIG. 6A.
• tracings 608 and 610 are each a tracing representing a single waveform sensed over the same period of time and are different at least in part in that each of tracings 608 and 610 correspond to an EMF signal respectively sensed from a different SQUID located at a different position relative to the chest (and therefore the heart) of the individual. That is, tracing 608 corresponds to a first EMF signal sensed from a first SQUID sensor and tracing 610 corresponds to a second EMF signal sensed from a second SQUID sensor where each of the first and second SQUID sensors are located at different positions relative to the heart of the individual.
• ECG tracing In a traditional ECG tracing, ten electrodes positioned at different locations on a body of an individual produce 12 ECG tracings, each corresponding to a different“view” of the heart. A particular ECG lead corresponds to a particular“view” of the heart in that each ECG tracing corresponds to a different spatial relationship between one or more ECG electrodes and the heart of the individual.
• each of tracings 608 and 610 represent a different “view” of the heart based on the position of the one or more EMF sensors relative to the chest of the individual. That is, current traveling through the heart of an individual generates a different EMF at different locations and as such is results in different appearing tracings based on the location of a sensor sensing that EMF.
• the tracings 608 and 610 each comprise a PQRST complex 612 (or, as shown, a plurality of PQRST complexes).
• the exemplary data from FIGS. 6A and 6B are provided as input to a machine learning software module described herein.
• the machine learning software module receives the input and interrelates the data in numerous ways so that input data relating to the individual (including data related to the individual received concurrently to the input as well as data related to the individual received before and subsequent to receiving the input) is associated with the sensed EMF data shown in the examples of FIGS. 6A and 6B.
• a machine learning software module as described herein correlates the age (58 years old in these examples) and gender (male in these examples) of the individual with one or more of the tracings 600, 608 and 610. Additional data relating to the individual may be provided that the machine learning software module may decide to correlate with the input as well. Non limiting examples of the additional data comprises health records for the individual, including diagnoses, medications, lab results other EMF sensed data from the individual.
• a machine learning software module as described herein further determines how to interrelate data from this individual and how to interrelate data of other individuals so as to generate a hypothesis function which is used to identify the presence of an abnormality in the individual and/or predict the occurrence of an abnormality in the individual.
• FIG. 7A shows an example of EMF data sensed from a plurality of OPM sensors positioned within proximity to a heart of a fetus.
• the EMF data that is sensed is sensed from a 39 weeks and 6 days old fetus using a plurality of OPM sensors positioned in proximity to the fetus.
• the waveform 700 comprises EMF data in EM units (shown on the Y-axis), said data being sensed over time in seconds (shown on the X-axis).
• waveform 700 comprises a plurality of waveforms sensed from a plurality of OPM sensors positioned in proximity to the chest of the individual.
• the plurality of waveforms are positioned at different locations relative to the chest of the individual.
• the one or more OPM sensors are positioned in a sensor array where the array is itself positioned in proximity to the chest of the individual so that each of the plurality of OPM sensors is in a different position relative to the chest of the individual
• waveform 700 comprises three separate individual waveforms 702, 704, and, 706.
• Waveforms 702, 704, and, 706 each correspond to a single PQRST complex as would be seen on a standard ECG. Therefore, each of the waveforms 702, 704, and, 706 correspond to an individual polarization and depolarization cardiac cycle or one heartbeat.
• FIG. 7B shows examples of a plurality of single EMF waveforms (as compared to the single waveform of FIG. 7A which comprises a plurality of individual waveforms).
• waveforms or tracings 708, 710, and 714 each represent respective EMF signals sensed over the same period of time.
• EMF data is sensed from the same 39 weeks and 6 days old fetus using a plurality of OPM sensors as in the example of FIG. 7A.
• FIG. 7B shows examples of a plurality of single EMF waveforms (as compared to the single waveform of FIG. 7A which comprises a plurality of individual waveforms).
• waveforms or tracings 708, 710, and 714 each represent respective EMF signals sensed over the same period of time.
• EMF data is sensed from the same 39 weeks and 6 days old fetus using a plurality of OPM sensors as in the example of FIG. 7A.
• FIG. 7B shows examples of a plurality of
• tracings 708, 710, and 714 are each a tracing representing a single waveform sensed over the same period of time and are different at least in part in that each of tracings 708, 710, and 714 correspond to an EMF signal respectively sensed from a different OPM located at a different position relative to the chest (and therefore the heart) of the fetus. That is, tracing 708 corresponds to a first EMF signal sensed from a first OPM sensor and tracing 710 corresponds to a second EMF signal sensed from a second OPM sensor where each of the first and second OPM sensors are located at different positions relative to the heart of the fetus.
• each ECG tracing corresponds to a different“view” of the heart.
• a particular ECG lead corresponds to a particular“view” of the heart in that each ECG tracing corresponds to a different spatial relationship between one or more ECG electrodes and the heart of the fetus.
• each of tracings 708, 710, and 714 represent a different“view” of the heart based on the position of the one or more EMF sensors relative to the chest of the fetus. That is, current traveling through the heart of an fetus generates a different EMF at different locations and as such is results in different appearing tracings based on the location of a sensor sensing that EMF.
• the tracings 708, 710, and 714 each comprise a PQRST complex 712 (or, as shown, a plurality of PQRST complexes).
• the exemplary data from FIGS. 7A and 7B are provided as input to a machine learning software module described herein.
• the machine learning software module receives the input and interrelates the data in numerous ways so that input data relating to the fetus (including data related to the fetus received concurrently to the input as well as data related to the fetus received before and subsequent to receiving the input) is associated with the sensed EMF data shown in the examples of FIGS. 7 A and 7B.
• a machine learning software module as described herein correlates the of the fetus with one or more of the tracings 700, 708, 710, and 714. Additional data relating to the fetus may be provided that the machine learning software module may decide to correlate with the input as well.
• the additional data comprises health records for the fetus, including diagnoses, medications, lab results other EMF sensed data from the fetus.
• a machine learning software module as described herein further determines how to interrelate data from this individual and how to interrelate data of other fetuses so as to generate a hypothesis function which is used to identify the presence of an abnormality in the individual and/or predict the occurrence of an abnormality in the fetus.
• FIG. 8A shows an example of EMF data sensed from a plurality of SQUID sensors positioned within proximity to a heart of a fetus.
• the EMF that data is sensed is from a 39 weeks and 6 days old fetus using a plurality of SQUID sensors positioned in proximity to the fetus.
• the waveform 800 comprises EMF data in EM units (shown on the Y- axis), said data being sensed over time in seconds (shown on the X-axis).
• waveform 800 comprises a plurality of waveforms sensed from a plurality of SQUID sensors positioned in proximity to the chest of the individual.
• the plurality of waveforms are positioned at different locations relative to the chest of the individual.
• the one or more SQUID sensors are positioned in a sensor array where the array is itself positioned in proximity to the chest of the individual so that each of the plurality of SQUID sensors is in a different position relative to the chest of the individual (although it should be understood that one or more SQUID sensors may be arranged in other configurations in other embodiments of the systems, methods, devices, and software described herein and need not always be arranged in an array).
• waveform 800 comprises three separate individual waveforms 802, 804, and, 806.
• Waveforms 802, 804, and, 806 each correspond to a single PQRST complex as would be seen on a standard ECG. Therefore, each of the waveforms 802, 804, and, 806 correspond to an individual polarization and depolarization cardiac cycle or one heartbeat.
• FIG. 8B shows examples of a plurality of single EMF waveforms (as compared to the single waveform of FIG. 8A which comprises a plurality of individual waveforms).
• waveforms or tracings 808, 810, and 814 each represent respective EMF signals sensed over the same period of time.
• EMF data is sensed from the same 39 weeks and 6 days old fetus using a plurality of SQUID sensors as in the example of FIG. 8A.
• tracings 808, 810, and 814 are each a tracing
• tracings 808, 810, and 814 correspond to an EMF signal respectively sensed from a different SQUID located at a different position relative to the chest (and therefore the heart) of the fetus. That is, tracing 808 corresponds to a first EMF signal sensed from a first SQUID sensor and tracing 810 corresponds to a second EMF signal sensed from a second SQUID sensor where each of the first and second SQUID sensors are located at different positions relative to the heart of the fetus.
• each ECG tracing corresponds to a different“view” of the heart.
• a particular ECG lead corresponds to a particular“view” of the heart in that each ECG tracing corresponds to a different spatial relationship between one or more ECG electrodes and the heart of the fetus.
• each of tracings 808, 810, and 814 represent a different“view” of the heart based on the position of the one or more EMF sensors relative to the chest of the fetus. That is, current traveling through the heart of an fetus generates a different EMF at different locations and as such is results in different appearing tracings based on the location of a sensor sensing that EMF.
• the tracings 808, 810, and 814 each comprise a PQRST complex 812 (or, as shown, a plurality of PQRST complexes).
• the exemplary data from FIGS. 8A and 8B are provided as input to a machine learning software module described herein.
• the machine learning software module receives the input and interrelates the data in numerous ways so that input data relating to the fetus (including data related to the fetus received concurrently to the input as well as data related to the fetus received before and subsequent to receiving the input) is associated with the sensed EMF data shown in the examples of FIGS. 8 A and 8B.
• a machine learning software module as described herein correlates, for example, the age the fetus with one or more of the tracings 800, 808, 810, and 814.
• Additional data relating to the fetus may be provided that the machine learning software module may decide to correlate with the input as well.
• additional data comprises health records for the fetus, including diagnoses, medications, lab results other EMF sensed data from the fetus.
• a machine learning software module as described herein further determines how to interrelate data from this fetus and how to interrelate data of other fetuses so as to generate a hypothesis function which is used to identify the presence of an abnormality in the fetus and/or predict the occurrence of an abnormality in the fetus.
• FIG. 9 shows three spatiotemporal activation representations of the magnetic activation of a healthy adult heart.
• the first spatiotemporal activation representation 902 comprises a butterfly plot.
• the second spatiotemporal activation representation 904 comprises a“view” of magnetic activation of a heart from base to apex.
• representation 906 comprises a“view” of magnetic activation of a heart from left to right.
• the exemplary data from FIG. 9 is provided as input to a machine learning software module described herein.
• the machine learning software module receives the input and interrelates the data in numerous ways so that input data relating to the individual from which the data was obtained (including data related to the individual received concurrently to the input as well as data related to the individual received before and subsequent to receiving the input) is associated with other sensed EMF data.
• a machine learning software module as described herein correlates, for example, the age the individual with other data relating to the individual. Additional data relating to the individual may be provided that the machine learning software module may decide to correlate with the input as well.
• Non-limiting examples of the additional data comprises health records for the individual, including diagnoses, medications, lab results other EMF sensed data from the individual.
• a machine learning software module as described herein further determines how to interrelate data from this individual and how to interrelate data of other individuals so as to generate a hypothesis function which is used to identify the presence of an abnormality in the individual and/or predict the occurrence of an abnormality in the individual.
• FIG. 10 shows a schematic representation of cardiac related EMF data received by a DNN which is configured to output a reconstruction of that EMF data X’.
• a hypothesis function used by the DNN to verify the accuracy of a reconstruction of EMF data compares the sensed input EMF data sample to the reconstruction generated by the DNN and determines a degree of error between the reconstruction and the input EMF data.
• the degree of error between the reconstruction generated by the DNN and the input EMF data is compared to a threshold value which is generated using the received 10,000 normal EMF data samples.
• a prediction phase uses the constructed and optimized hypothesis function from the training phase to predict the probability of an abnormality in an unknown patient’s organ, tissue, body, or portion thereof by using the EMF data of the patient.
• hypothesis function H is defined as:
• H Reconstruction error(E) between the original input (X) and the reconstructed input (X’)
• Cardiovascular disease remains the leading cause of death among men and women in the United States, representing over 25% all-cause mortality. (1) Approximately 8 million
• Genetesis, Inc. has developed a novel EMF analysis system that uses a series of diagnostic algorithms to convert EMF data into dynamic images that can indicate CAD or ischemia.
• ED chest pain patients with suspected acute coronary syndrome (ACS) placed in the EDOU for further evaluation were potentially eligible for the pilot study.
• Inclusion criteria included patients > 18 years of age with low-intermediate risk chest pain defined as having no ACS diagnosis in the ED, a non-diagnostic ECG and two negative cardiac troponin T results at least 3 hours apart per ED protocol that consented to having an EMF sensing and analysis system scan.
• Exclusion criteria included patients with metallic items in the chest area, claustrophobia, non-ambulatory, in atrial fibrillation with rapid ventricular response, unable to fit into the EMF sensing and analysis system device or lie supine for 2 minutes, poor candidates for follow up (e.g. no access to a phone), prisoners, and repeat participants.
• FIG. 11 shows an exemplary system as described herein which further configured to include one or more devices and software as described herein including an EMF sensor array (or sensor plate) 1101 and an ambient EMF shield 1102.
• EMF sensor array or sensor plate
• a sensor plate containing 14 sensors was positioned over the chest area and a 90 second- scan was obtained and stored in an encrypted database and sent to a fflPAA secure cloud.
• the EMF sensing and analysis system signal quality was evaluated by an automated function of the system software, and secondarily by Genetesis personnel.
• Sensed EMF data was aggregated and processed into 3 components: averaged EMF waveforms, Equivalent Current Dipole (ECD), and pseudocurrent density maps.
• ECD Equivalent Current Dipole
• pseudocurrent density maps pseudocurrent density maps.
• ECD provides a mathematical model to measure and localize the movement of current wavefront within the myocardium at different points in the cardiac cycle. These components are analyzed by the system’s software algorithms to look for significant deviations from a referenced database of normal images.
• EMF sensing and analysis system scan was defined as current dipole deviation pattern findings correlating to ⁇ 50% stenosis and a positive analysis result of the EMF sensing and analysis system as 50% - 69% and > 70% stenosis.
• CA was used as the gold standard if both ST and CA were performed.
• Hospital data were collected using the hospital electronic medical record and included patient demographics, cardiac risk factors, cardiac co-morbidities, laboratory reports, consultant reports, diagnostic and operative reports, and discharge diagnosis.
• a history of CAD was found in 9.9% (10/101) of patients, and 5.0% (5/101) had a history of heart failure or valvular heart disease. Mean number of cardiac risk factors was two and 28.7% (29/101) had > 3 risk factors. ST were performed in 96% (97/101) of patients; 56% stress echocardiograms (SE), 17% dobutamine echocardiograms (DE), 27% persantine stress test (PST). Eighteen (17.8%) patients underwent CA.
• SE stress echocardiograms
• DE dobutamine echocardiograms
• PST persantine stress test
• a normal EMF sensing and analysis system scan resulted in a regular pattern without magnetic dipole dispersion whereas an abnormal EMF sensing and analysis system scan demonstrated an irregular pattern of magnetic pole dispersion. It was theorized that an extreme shift in dipole angulation or significant disorganization in the magnetic field map (e.g. a triple pole) would indicate a greater degree of vessel sten
• FIGS. 12A and 12B respectively show a negative result of a scan using an EMF sensing and analysis system as described herein, wherein FIG. 12A corresponds to a result of a first test subject and FIG. 12B corresponds to a test result of a second subject. More specifically, FIGS. 12A and 12B each show a regular pattern without magnetic dipole dispersion, which correspond to a negative result. The negative result in FIGS. 12A and 12B respectively indicate a negative result for the presence of CAD in the different test subjects.
• FIGS. 13A and 13B respectively show a positive result of a scan using an EMF sensing and analysis system as described herein, wherein FIG. 13A corresponds to a result of a first test subject and FIG. 13B corresponds to a test result of a second subject. More specifically, FIGS. 13A and 13B each show an irregular pattern of magnetic pole dispersion, which correspond to a positive result The positive result in FIGS. 13A and 13B respectively indicate a positive result for the presence of CAD in the different test subjects.
• FIG. 14 shows a schematic representation of EMF sensing and analysis scan results as compared to CA and ST results.
• 3 had a corresponding positive CA
• 20 had a negative ST (16) or CA (4).
• ST (1) Only 9 patients in the entire cohort were positive for ischemia: ST (1), CA (8). All 3 patients with corresponding positive CA had >70% stenosis; EMF sensing and analysis scan interpreted as >70% in 2 of these patients (Stress Echocardiogram (SE) negative in one of these patients), and EMF sensing and analysis scan 50-69% (CA finding: 75% LAD) in the third.
• SE Stress Echocardiogram
• EMF sensing and analysis scan falsely negative 4 patients had >70% stenosis on CA (one of these patients did not have a ST) and 1 had 60% stenosis also missed by SE.
• EMF sensing and analysis scan falsely positive 4 patients with non-corresponding negative CA (i.e. EMF sensing and analysis scan falsely positive)
• EMF sensing and analysis scan found 50-69% stenosis in 1 patient and >70% in 3 patients.
• the sensitivity with 95% Cl for EMF sensing and analysis scan is 33.3% [7.5%, 70.7%], specificity 78.3% [68.4%, 86.2%], PPV 13% [5.2%, 29.0%] and NPV 92.3% [88.2%, 95.1%] for >50% stenosis and 28.6% [3.7%, 71.0%], 87.1% [78.6%, 93.2%], 14.3% [4.4%, 37.6%] and 94.2% [91.0%, 96.3%] respectively for >
• EMF sensing and analysis scan is a simple, rapid, non-invasive diagnostic modality that is feasible in an OU patient population and has excellent specificity and NPV for obstructive (50-69%) and especially critical (>70%) stenosis. This compares favorably to reported pooled results of ST with imaging from multiple studies and metanalysis showing a specificity of 70-80% for identifying patients with >50% stenosis.
• testing could be completed in under 2 minutes. Incorporation into the ED workflow of chest pain evaluation at triage or further downstream in the ED workup of chest pain patients should be explored. The downstream cost benefits to patients and the hospital could be substantial.
• Limitations of this study include the small number of patients, enrollment of patients from a single site and as a convenience sample, and low prevalence of significant disease in this study cohort. Additionally, several patients were either unable to be scanned or had inadequate scans due to body habitus, i.e. some part of the body (e.g. chest, abdomen) was touching the sensor plate and causing motion artifact or sensor railing.
• Results of this pilot study of a novel diagnostic test show a resting 90-second EMF sensing and analysis scan has excellent specificity and NPV and is faster and comparable to ST in ruling out obstructive cardiac ischemia in an EDOU population of low-intermediate risk chest pain patients. Specificity and NPV improve further with physician over-read of EMF sensing and analysis scan. Due to low prevalence of disease in this study cohort, a larger study that also includes high-risk chest pain patients is needed to better assess the accuracy of EMF sensing and analysis scan in detecting cardiac ischemia.

Landscapes

• Health & Medical Sciences (AREA)
• Life Sciences & Earth Sciences (AREA)
• Engineering & Computer Science (AREA)
• Cardiology (AREA)
• Physics & Mathematics (AREA)
• Biophysics (AREA)
• Public Health (AREA)
• Veterinary Medicine (AREA)
• General Health & Medical Sciences (AREA)
• Animal Behavior & Ethology (AREA)
• Surgery (AREA)
• Molecular Biology (AREA)
• Medical Informatics (AREA)
• Pathology (AREA)
• Biomedical Technology (AREA)
• Heart & Thoracic Surgery (AREA)
• Physiology (AREA)
• Artificial Intelligence (AREA)
• Signal Processing (AREA)
• Evolutionary Computation (AREA)
• Fuzzy Systems (AREA)
• Psychiatry (AREA)
• Mathematical Physics (AREA)
• Computer Vision & Pattern Recognition (AREA)
• Pulmonology (AREA)
• Measurement And Recording Of Electrical Phenomena And Electrical Characteristics Of The Living Body (AREA)
• Measuring And Recording Apparatus For Diagnosis (AREA)

Priority Applications (4)

Applications Claiming Priority (1)

Publications (1)

Family

ID=70774412

Family Applications (1)

Country Status (4)

Cited By (4)

Citations (6)

Family Cites Families (5)

• 2018
  • 2018-11-20 CN CN201880100637.4A patent/CN113316412A/zh active Pending
  • 2018-11-20 EP EP18940759.6A patent/EP3883457A4/en active Pending
  • 2018-11-20 JP JP2021527232A patent/JP7394133B2/ja active Active
  • 2018-11-20 WO PCT/US2018/062113 patent/WO2020106284A1/en unknown

Patent Citations (6)

Non-Patent Citations (1)

Also Published As

Similar Documents

Legal Events