System, server and method for preventing suicide

• US 10,043,591 B1
• Filed: 04/10/2018
• Issued: 08/07/2018
• Est. Priority Date: 02/06/2015
• Status: Active Grant

• Current Assignee: Brain Trust Innovations I, LLC
• Original Assignee: Brain Trust Innovations I, LLC
• Inventors: LaBorde, David
• Primary Examiner(s): Blouin, Mark
• Application Number:
  US15/950,060
• Time in Patent Office: 119 Days
• Field of Search: 340 101
• CPC Class Codes: A61B 2562/08 Sensors provided with means for identification, e.g. barcodes or memory chips
• CPC Class Codes: A61B 5/165 Evaluating the state of mind, e.g. depression, anxiety
• CPC Class Codes: G06K 19/0723 the record carrier comprising an arrangement for non-contact communication, e.g. wireless communication circuits on transponder cards, non-contact smart cards or RFIDs
• CPC Class Codes: G06K 19/07762 the adhering arrangement making the record carrier wearable, e.g. having the form of a ring, watch, glove or bracelet record carriers for insertion in the human body for medical purposes A61B90/00; record carriers adapted for attachment to animals A01K11/00
• CPC Class Codes: G06K 7/10366 the interrogation device being adapted for miscellaneous applications
• CPC Class Codes: G06N 20/10 using kernel methods, e.g. support vector machines [SVM]
• CPC Class Codes: G06N 3/006 based on simulated virtual individual or collective life forms, e.g. social simulations or particle swarm optimisation [PSO]
• CPC Class Codes: G06N 3/04 Architecture, e.g. interconnection topology
• CPC Class Codes: G06N 3/0409 Adaptive resonance theory [ART] networks
• CPC Class Codes: G06N 3/044 Recurrent networks, e.g. Hopfield networks
• CPC Class Codes: G06N 3/048 Activation functions
• CPC Class Codes: G06N 3/08 Learning methods
• CPC Class Codes: G06N 3/084 Backpropagation, e.g. using gradient descent
• CPC Class Codes: G06N 3/086 using evolutionary algorithms, e.g. genetic algorithms or genetic programming
• CPC Class Codes: G06N 3/088 Non-supervised learning, e.g. competitive learning
• CPC Class Codes: G06N 5/01 Dynamic search techniques; Heuristics; Dynamic trees; Branch-and-bound
• CPC Class Codes: G06N 5/041 Abduction
• CPC Class Codes: G06N 5/046 Forward inferencing; Production systems
• CPC Class Codes: G06N 5/048 Fuzzy inferencing
• CPC Class Codes: G06N 7/01 Probabilistic graphical models, e.g. probabilistic networks
• CPC Class Codes: G06Q 10/0633 : Workflow analysis
• CPC Class Codes: G06Q 10/08 : Logistics, e.g. warehousing, loading or distribution; Inventory or stock management
• CPC Class Codes: G06Q 50/22 : Social work or social welfare, e.g. community support activities or counselling services
• CPC Class Codes: G16H 10/65 : stored on portable record carriers, e.g. on smartcards, RFID tags or CD
• CPC Class Codes: G16H 40/20 : for the management or administration of healthcare resources or facilities, e.g. managing hospital staff or surgery rooms
• CPC Class Codes: G16H 40/63 : for local operation
• CPC Class Codes: G16H 50/30 : for calculating health indices; for individual health risk assessment
• CPC Class Codes: G16H 50/70 : for mining of medical data, e.g. analysing previous cases of other patients

Abstract

A system includes a data collection engine, a plurality of items including radio-frequency identification chips, a plurality of third party data and insight sources, a plurality of interfaces, client devices, a server and method thereof for preventing suicide. The server includes trained machine learning models, business logic and attributes of a plurality of patient events. The data collection engine sends attributes of new patient events to the server. The server can predict a suicide risk of the new patient events based upon the attributes of the new patient events utilizing the trained machine learning models. Using business logic, data visualization and the trained machine learning models, the server can also make recommendations to reduce the risk of suicides.

Claims

1. A system comprising:

a plurality of radio-frequency identification (RFID) chips, wherein a first RFID chip of the plurality of RFID chips is a passive-type RFID chip, and one or more of the plurality of RFID chips include a sensor group;

a data collection engine (DCE) device communicating with the first RFID chip, wherein the DCE comprises;

a power transmission subsystem including a power source and an antenna arranged to wirelessly transmit power from the power source to the first RFID chip;

a transceiver configured to receive first data from at least one of the first RFID chip and a second RFID chip of the plurality of RFID chips while the first RFID chip is activated by the power received, the first data including identification information of the at least one of the first and second RFID chips;

a controller operatively coupled to the transceiver; and

one or more memory sources operatively coupled to the controller, the one or more memory sources including instructions for configuring the controller to generate one or more messages indicative of the identification information to be sent by the transceiver to a server device via the network connection,

wherein the first RFID chip includes an antenna for wirelessly receiving the power from the transceiver of the DCE and control logic for generating the identification information;

wherein the server device comprises;

a transceiver configured to receive the one or more messages from the DCE;

a controller operatively coupled to the transceiver; and

one or more memory sources operatively coupled to the controller, the one or more memory sources storing a trained neural network model (NNM) for generating an output value corresponding to a present event based upon one or more of the identification information and position information, wherein the output value corresponds to a suicide risk.

2. The system of claim 1, wherein in the server device:

the one or more memory sources further store a plurality of past events, each of the plurality of past events including a plurality of attributes and a quantifiable outcome; and

the controller is further configured to;

train a NNM to generate the trained NNM, wherein the training of the NNM includes;

perform pre-processing on the plurality of attributes for each of the plurality of past events to generate a plurality of input data sets;

divide the plurality of past events into a first set of training data and a second set of validation data;

iteratively perform a machine learning algorithm (MLA) to update synaptic weights of the NNM based upon the training data; and

validate the NNM based upon the second set of validation data,

wherein the MLA for updating the synoptic weights is one or more of ADALINE training, backpropagation algorithm, competitive learning, genetic algorithm training, Hopfield learning, Instar and Outstar training, the Levenberg-Marquardt algorithm (LMA), Manhanttan Update Rule Propogation, Nelder Mead Training, Particle Swarm (PSO) training, quick propagation algorithm, resilient propagation (RPROP) algorithm, scaled conjugate gradient (SCG), support vector machines, genetic programming, Bayesian statistics, decision trees, case based reasoning, information fuzzy networks, clustering, hidden Markov models, particle swarm optimization, simulated annealing.

3. A method for predicting a suicide risk associated with a new event, the method comprising:

receiving a plurality of input attributes of the new event;

performing pre-processing on the plurality of input attributes to generate an input data set;

generating an output value from a trained model based upon the input data set; and

classifying the output value into a suicide risk category to predict an outcome.

4. The method of claim 3, further comprising:

storing a plurality of past events, each of the plurality of past events including a plurality of patient attributes and a quantifiable outcome; and

training a neural network model (NNM) to generate the trained model, wherein the training of the NNM includes;

performing pre-processing on the plurality of patient attributes for each of the plurality of past events to generate a plurality of input data sets;

dividing the plurality of past events into a first set of training data and a second set of validation data;

iteratively performing a machine learning algorithm (MLA) to update synaptic weights of the NNM based upon the training data; and

validating the NNM based upon the second set of validation data,

wherein the MLA for updating the synoptic weights is one or more of ADALINE training, backpropagation algorithm, competitive learning, genetic algorithm training, Hopfield learning, Instar and Outstar training, the Levenberg-Marquardt algorithm (LMA), Manhattan Update Rule Propagation, Nelder Mead Training, Particle Swarm (PSO) training, quick propagation algorithm, resilient propagation (RPROP) algorithm, scaled conjugate gradient (SCG), support vector machines, genetic programming, Bayesian statistics, decision trees, case based reasoning, information fuzzy networks, clustering, hidden Markov models, particle swarm optimization, simulated annealing.

5. The method of claim 3, wherein one or more of the plurality of input attributes of the new event includes data from an RFID chip, a sensor, a hospital information system, a law enforcement databases, a human facial imaging device, a location device, a database of known prior suicide attempts, a socio-economic source, a census data source, a death certificate data source, news media, a patient communication data source, and relationship status or behavioral change data source.

6. The method of claim 3, wherein one or more of the plurality of input attributes of the new event is data from social media and patient facing software applications.

7. The method of claim 3, wherein one or more of the plurality of input attributes of the new event is data from:

pharmacy prescription and pharmaceutical dispending data;

drug data received from a remote client device;

healthcare transportation records;

medical claim data; and

consultation records.

8. The method of claim 3, wherein one or more of the plurality of input attributes of the new event includes community based medical center data, laboratory result data, radiographical imaging and interpretations, and vital sign observations.

9. The method of claim 4, wherein the performing of the MLA includes measuring a global error in each training iteration for the NNM by:

calculating a local error, the local error being a difference between the output value of the NNM and the quantifiable outcome;

calculating the global error by summing all of the local errors in accordance with one of;

(1) Mean Square Error (MSE) formula

10. The method of claim 3, wherein the trained model is a trained Self-Organizing Map (SOM) including a plurality of network nodes arranged in a grid or lattice and in fixed topological positions, an input layer with a plurality of input nodes representing the input attributes of the past events, wherein each of the plurality of input nodes is connected to all of the plurality of network nodes by a plurality of synaptic weights.

11. The method of claim 10, further comprising:

storing a plurality of past events, each of the plurality of past events including a plurality of patient attributes and a quantifiable outcome;

performing pre-processing on the plurality of patient attributes for each of the plurality of past events to generate a plurality of input data sets; and

training a SOM to generate the trained model, wherein the training of the SOM includes;

initializing values of the plurality of synaptic weights to random values, randomly selecting one past event and determining which of the plurality of network nodes is a best matching unit (BMU) according to a discriminant function, wherein the discriminant function is a Euclidean Distance; and

iteratively calculating a neighborhood radius associated with the BMU to determine neighboring network nodes for updating, and updating values of synoptic weights for neighboring network nodes within the calculated neighborhood radius for a fixed number of iterations to generate the trained model.

12. The method of claim 11, further comprising generating another SOM including the plurality of patient attributes to reduce dimensionality.

13. The method of claim 3, wherein the receiving a plurality of patient attributes of the new event further includes receiving one or more messages including a patient identification and location information associated with a first RFID tag and a medical professional identification and location information associated with a second RFID tag from a data collection engine (DCE).

14. The method of claim 3, wherein:

the receiving of a plurality of patient attributes of the new event further comprises receiving a plurality of new events, each including a plurality of patient attributes;

the generating of the output value and classifying the output value further includes generating a graphical image including output values for each of the plurality of new events;

the method further comprises receiving a graphical display request from a remote client device and transmitting the graphical image to the remote client device as a response;

the trained model is a trained SOM; and

the graphical image is a cluster diagram including a plurality of clusters of output values having a similar characteristic.

15. The method of claim 3, further comprising:

alerting and assigning an allocation of appropriate clinical resources to a patient based upon an optimization algorithm in accordance with the suicide risk category of the new event and attributes of available clinical resources.

16. A client device comprising:

a transceiver communicating with a server device via a connection to a network, the transceiver configured to send a request message to the server device and receive a reply message from the server device in response to the request message, the reply message including an output value generated from a trained model stored at the server device;

a controller coupled to the transceiver;

a display device coupled to the controller; and

a memory including instructions for configuring the controller to;

generate the request message; and

render a graphical display on the display device based upon the output value.

17. The client device of claim 16, wherein the reply message includes a plurality of output values, the graphical display is a cluster diagram including a plurality of clusters of similar characteristic output values of the plurality of output values.

18. The client device of claim 16, wherein the controller is further configured to:

calculate a suicide prevention readiness score (SPRS) from a trained model based upon an input data set; and

generate an information reply including a graphical display indicating the output value of trained model.

19. The client device of claim 18, wherein the input data set includes a status of healthcare provider organization personnel suicide prevention training and attributes of the suicide prevention training, and a status of a healthcare provider organization community outreach activities and attributes of the healthcare provider organization community outreach activities.

20. The client device of claim 18, wherein the graphical display further includes data visualizations, score cards and dashboard metrics describing the input data and individuals involved with the input data set.