Brian C Delaney

US Patent:

20210210200, Jul 8, 2021

Filed:

Mar 23, 2021

Appl. No.:

17/210262

Inventors:

- Burlington MA, US
Dushyant Sharma - Woburn MA, US
Daniel Paulino Almendro Barreda - London, GB
Patrick A. Naylor - Reading, GB
Paul Joseph Vozila - Arlington MA, US
Brian W. Delaney - Bolton MA, US
Neal Snider - Belmont MA, US
Joel Praveen Pinto - Aachen, DE
Miguel Enrique Ruiz - Billerica MA, US
Yi Zhang - Saarbruecken, DE
Donald E. Owen - Orlando FL, US

International Classification:

G16H 40/20
G06Q 30/02
G16H 10/60
G07C 9/00
G06F 40/35

Abstract:

A computer-implemented method, computer program product, and computing system for monitoring a plurality of encounter participants is executed on a computing device and includes obtaining encounter information of a user encounter. The encounter information is processed to: associate at least a first portion of the encounter information with at least one known encounter participant, and associate at least a second portion of the encounter information with at least one unknown encounter participant.A computer-implemented method, computer program product, and computing system for populating records via semantic frames is executed on a computing device and includes obtaining encounter information of a user encounter. The encounter information is processed to generate an encounter transcript. At least one semantic frame is generated based, at least in part, upon at least one portion of the encounter transcript, wherein the at least one semantic frame defines an abstract meaning for the at least one portion of the encounter transcript.A mixed-media ACD device is configured to monitor one or more encounter participants of a user encounter and includes a machine vision system configured to obtain machine vision encounter information concerning the user encounter. An audio recording system is configured to obtain audio encounter information concerning the user encounter, wherein the audio recording system includes a plurality of discrete audio acquisition devices.A computer-implemented method, computer program product, and computing system for tracking encounter participants is executed on a computing device and includes obtaining encounter information of a user encounter, wherein the encounter information includes machine vision encounter information obtained via one or more machine vision systems. The machine vision encounter information is processed to identify one or more humanoid shapes.

US Patent:

20200311343, Oct 1, 2020

Filed:

Nov 1, 2019

Appl. No.:

16/671576

Inventors:

- Burlington MA, US
Brian William Delaney - Bolton MA, US
Girija Yegnanarayanan - Raleigh NC, US
Radu Florian - Danbury CT, US
Martin Franz - Yorktown Heights NY, US
Scott McCarley - Yorktown Heights NY, US
John F. Pitrelli - Danbury CT, US
Imed Zitouni - White Plains NY, US
Salim E. Roukos - Scarsdale NY, US

Assignee:

Nuance Communications, Inc. - Burlington MA

International Classification:

G06F 40/20
G16H 50/70
G16H 10/60
G16H 50/20

Abstract:

Cascaded models may be applied to extract facts from a medical text. A first model may be applied to at least a portion of the medical text. The first model extracts at least one first medical fact. The at least one first medical fact is linked to at least first text in the at least a portion of the medical text. A second model may be applied to the first text. The second model extracts at least one second fact that is an attribute of the at least one first medical fact.

US Patent:

20190051374, Feb 14, 2019

Filed:

Aug 9, 2018

Appl. No.:

16/059967

Inventors:

- Burlington MA, US
Guido Remi Marcel Gallopyn - Newburyport MA, US
Brian W. Delaney - Bolton MA, US
Neal Snider - Belmont MA, US
Joel Praveen Pinto - Aachen DE, US
Miguel Enrique Ruiz - Billerica MA, US
Yi Zhang - Saarbruecken, DE

International Classification:

G16H 10/20
G16H 10/60
G06F 17/27
G06F 17/28
G06T 7/00
H04R 1/32

Abstract:

A method, computer program product, and computing system for populating medical records via semantic frames is executed on a computing device and includes obtaining encounter information of a patient encounter. The encounter information is processed to generate an encounter transcript. At least one semantic frame is generated based, at least in part, upon at least one portion of the encounter transcript, wherein the at least one semantic frame defines an abstract meaning for the at least one portion of the encounter transcript.

US Patent:

20190027235, Jan 24, 2019

Filed:

Jul 3, 2018

Appl. No.:

16/026641

Inventors:

- Burlington MA, US
Neil Barrett - Longueuil, CA
Brian Delaney - Bolton MA, US
Ravi Kumar Kondadadi - Lakeville MN, US

International Classification:

G16H 10/60
G16H 15/00
G06Q 20/14
G16H 50/30

Abstract:

A method, computer program product, and computing system for processing content concerning a plurality of patients using a CAC system to define one or more billing codes concerning a social habit status of one or more patients of the plurality of patients. The one or more billing codes concerning the social habit status of the one or more patients are provided to a user for review. Feedback is received from the user concerning the accuracy of the one or more billing codes. The feedback concerning the one or more billing codes is automatically processed to define one or more confidence scores. The CAC system is trained based, at least in part, upon the one or more confidence scores.

US Patent:

20150006199, Jan 1, 2015

Filed:

Jun 26, 2013

Appl. No.:

13/928213

Inventors:

- Burlington MA, US
Brian William Delaney - Bolton MA, US
Girija Yegnanarayanan - Raleigh NC, US
Radu Florian - Yorktown Heights NY, US
Martin Franz - Yorktown Heights NY, US
Scott McCarley - Yorktown Heights NY, US
John F. Pitrelli - Yoktown Heights NY, US
Imed Zitouni - Yorktown Heights NY, US
Salim E. Roukos - Yorktown Heights NY, US

International Classification:

G06F 19/00
G06F 17/27

US Classification:

705 3

Abstract:

Cascaded models may be applied to extract facts from a medical text. A first model may be applied to at least a portion of the medical text. The first model extracts at least one first medical fact. The at least one first medical fact is linked to at least first text in the at least a portion of the medical text. A second model may be applied to the first text. The second model extracts at least one second fact that is an attribute of the at least one first medical fact.