US 10,327,697 B1Granted

Digital platform to identify health conditions and therapeutic interventions using an automatic and distributed artificial intelligence system

AssigneeSpiral Physical Therapy, Inc.

InventorsSpencer Stein, Stephen Moxey, Lee Stein

Life of this patent

Estimating the term…

Loading patent…

Abstract

This disclosure is directed to method and system for automatic, distributed, computer-aided, and intelligent data collection/analytics, health monitoring, health condition identification, and patient preventive/remedial health advocacy. The system integrates (1) distributed patient health data collection devices, (2) centralized or distributed data servers running various intelligent and predictive data analytics engines for health screening, assessment, patient health condition identification, and patient preventive/remedial health advocacy, 3) specifically designed data structures including quantized health indicator vectors, patient health condition identification matrices and patient health condition vectors, (4) portal servers configured to interface with (5) distributed physician terminal devices and (6) distributed patient terminal devices for delivering health condition identification, health interventions and patient preventive/remedial health advocacy, and for monitoring and tracking patient activities. The various intelligent and predictive engines are configured to learn and extract hidden features and correlations from a large amount of data obtained from the distributed data collection devices.

Claims20

Dependent claims

A system for automatic and intelligent patient health condition identification (PHCI) and patient preventive/remedial health advocacy (PPRHA), comprising:

a data repository;
a communication interface;
and
a processing circuitry in communication with the data repository and the communication interface;
wherein the processing circuitry is configured to:
receive 3D topographical data of a target patient in a form of body mesh scan from one of a plurality of distributed 3D scanners via the communication interface;
execute a data segmentation model trained based on a first machine learning algorithm to automatically identify a predetermined set of body landmarks of the target patient and identify a set of representations corresponding to the predetermined set of body landmarks of the target patient;
determine a vertical reference line and a transverse plane from the 3D topographical data;
predefine a quantized health indicator vector space associated with health of the predetermined set of body landmarks of the target patient;
associate each of a plurality of predetermined health conditions with quantized values in the quantized health indicator vector space to generate a quantized PHCI matrix;
derive a health indicator vector in the quantized health indicator vector space based on the set of representations, the vertical reference line, and the transverse plane;
quantize the health indicator vector into the quantized health indicator vector space to obtain a quantized health indicator vector;
store the quantized health indicator vector and the 3D topographical data in the data repository;
automatically generate a patient health condition (PHC) vector comprising a plurality of component each corresponding to one of the plurality of predetermined health conditions;
execute a PPRHA model trained using a second machine learning algorithm;
and
automatically generate a PPRHA item by inputting the PHC vector into the PPRHA model.

14 more elements

A method for automatic and intelligent patient health condition identification (PHCI) and patient preventive/remedial health advocacy (PPRHA) by a processing circuitry in communication with a data repository and a communication interface, comprising:

receiving 3D topographical data in a form of body mesh scan of a target patient from one of a plurality of distributed 3D scanners via the communication interface;
executing a data segmentation model trained based on a first machine learning algorithm to automatically identify a predetermined set of body landmarks of the target patient and identify a set of representations corresponding to the predetermined set of body landmarks of the target patient;
determining a vertical reference line and a transverse plane from the 3D topographical data;
predefining a quantized health indicator vector space associated with health of the predetermined set of body landmarks of the target patient;
associating each of a plurality of predetermined health conditions with quantized values in the quantized health indicator vector space to generate a quantized PHCI matrix;
deriving a health indicator vector in the quantized health indicator vector space based on the set of representations, the vertical reference line, and the transverse plane;
quantizing the health indicator vector into the quantized health indicator vector space to obtain a quantized health indicator vector;
storing the quantized health indicator vector and the 3D topographical data in the data repository;
automatically generating a patient health condition (PHC) vector comprising a plurality of component each corresponding to one of the plurality of predetermined health conditions;
executing a PPRHA model trained using a second machine learning algorithm;
and
automatically generating a PPRHA item by inputting the PHC vector into the PPRHA model.

9 more elements

Dependencies are inferred from the claim text.

Facts

Application: 16/228,169Filed 20 Dec 2018
Granted: 25 Jun 2019
Term ends: Not estimatedThe source records don’t support an estimate.

Classification21 codes

A61B 5/1075
Measuring physical dimensions, e.g. size of the entire body or parts thereof: for measuring dimensions by non-invasive methods, e.g. for…
A61B 5/0064
Arrangements for scanning: body surface scanning
A61B 5/1071
Measuring physical dimensions, e.g. size of the entire body or parts thereof: measuring angles, e.g. using goniometers
A61B 5/1114
Local tracking of patients, e.g. in a hospital or private home: tracking parts of the body