Hill-rom Services, Inc.

Cummins Inc.

Agency for Science, Technology and Research

Salesforce, Inc.

Salesforce Com, Inc.

Igloohome Pte. Ltd.

National University of Singapore

Institute of Molecular & Cell Biology

Yue Wang

Chau Chong Ye

Suresha Venkataraya

Aye Aung

Chu Hong Hoi

Shafiq Rayhan Joty

Amodh Gundlur Ramesh

Xiao-Li Xu

Graham Keith Lacy

Jack Barney Sing

Joshua A Williams

Deny D Barilea

Mark R Stepper

Catherine M Wagner

Cong Jiang

Siew Ying Koh

Nookarajesh Varma Sangadi

Hung Le

Akhilesh Deepak Gotmare

Khee Kien Ho

Weishi Wang

Tak Wei David Teo

Jerilyn Tsai

Xiaoqiu Li

Anthony Wen Jie Chow

Sharon May Ling Goh

Xiaoli Xu

Raymond Teck Ho Lee

Michael Hoksbergen

Lei Huang

Matthew David Mullin

Thomas Dollmeyer

Michael M Frondorf

Steven V McCaig

Wei T Tan

Thierry Flocard

Peter Rauznitz

Gilles Camus

Wei Soung Tan

Haishan Wang

Zhon Ye Chu

Joel Preetham Fernandes

Eugene Hong Kheng Kung

David J Brzenchek

Pratheep Kumar Mahalingam

Jean-Francois Tarsaud

Stephanie L Goerges

Aaron R Burnham

Leah Noaeill

Florian Garnero

Yin Hwee Tan

Wole C Akinyemi

Todd Slater

Hao Gong

Bernard Arnaud

Catherine Jane Pallen

Zujun Hou

Karianto Leman

Chee Keong Ng

Charles Wu

Chang Yang

Lokesh Bhatt

Eugene Kung

Kah Leong Lim

Teck Wee Chua

Nam Trung Pham

Richard Philippe Chang

Rachid Oulad Slimane

Teck Ho Raymond Lee

Shyan Huang

Pierre Deguignet

Su Ling Yeo

Jian Wen Li

Nicolas Odic

Anthony J Abbenante

Martin T Brooks

Xiaoqui Li

Cahu Chong Ye

Steven V Mccaig

Embodiments described herein provide a reinforcement learning based framework engaging pretrained language models (LMs) for program synthesis tasks. Specifically, the framework adopts a training strategy that optimizes pretrained LMs for program synthesis tasks in an actor-critic approach.

Systems and methods for program synthesis

Embodiments described herein a code generation and understanding model that builds on a Transformer-based encoder-decoder framework. The code generation and understanding model is configured to derive generic representations for programming language (PL) and natural language (NL) in code domain via pre-training on unlabeled code corpus, and then to benefit many code-related downstream tasks with fine-tuning. Apart from the denoising sequence-to-sequence objectives widely adopted for pre-training on natural language, identifier tagging and prediction pre-training objective is adopted to enable the model to better leverage the crucial token type information from PL, which specifically are the identifiers assigned by developers.

Systems and methods for code understanding and generation

A mattress overlay apparatus is provided for use with a mattress. The mattress overlay apparatus includes an overlay configured for placement atop the mattress and having a plurality of inflatable bladders, a blower, and first and second rotary plate valves pneumatically coupled to the plurality of bladders and the blower. The first and second rotary plate valves are arranged in series between the plurality of bladders and the blower. The first and second rotary plate valves and the blower are operated to provide the overlay with percussion and vibration (P&V), left and right turn assist, and microclimate management (MCM) functionality.

Rotary plate valve having seal anti-herniation structure

A visual dialogue model receives image input and text input that includes a dialogue history between the model and a current utterance by a human user. The model generates a unified contextualized representation using a transformer encoder network, in which the unified contextualized representation includes a token level encoding of the image input and text input. The model generates an encoded visual dialogue input from the unified contextualized representation using visual dialogue encoding layers. The encoded visual dialogue input includes a position level encoding and a segment type encoding. The model generates an answer prediction from the encoded visual dialogue input using a first self-attention mask associated with discriminative settings or a second self-attention mask associated with generative settings. Dense annotation fine tuning may be performed to increase accuracy of the answer prediction. The model provides the answer prediction as a response to the current utterance of the human user.

Unified vision and dialogue transformer with BERT

A nebulizer assembly for a respiratory device is provided having a housing defining a chamber. The housing also has a nebulizer port configured to receive a nebulizer to discharge atomized medication into the chamber. An outlet of a handle is coupled to the inlet of the housing. A hose is coupled to an inlet of the handle. A patient interface is coupled to the outlet of the housing. Air flows from the hose to the patient interface via the handle and the housing. The air mixes with the atomized medication within the chamber.

Respiratory therapy filter, flow control, and patient interface apparatuses, systems, and methods

A respiratory device may include an inlet port and an outlet port. A fluid pathway may extend between the inlet port and the outlet port. A filter housing may be positioned between the inlet port and the outlet port. A filter may be positioned within the filter housing. A first pressure port may be positioned on an inlet side of the filter housing. A second pressure port may be positioned on an outlet side of the filter housing. The first pressure port and the second pressure port may extend parallel to the fluid pathway. A pressure transducer may be coupled to the first pressure port and the second pressure port and configured to measure a pressure drop between the first pressure port and the second pressure port.

Apparatus and method to assess airway clearance therapy efficacy

A system is provided for monitoring a patient on an air support system. The system comprises a capacitive sensor system having a capacitive sensor positioned at the underside of an air filled mattress, and a respiration signal processing module that receives the output of the capacitive sensor system and is configured to monitor patient respiration. Respiration is monitored by: periodically sampling the capacitive sensor system output to obtain measurements of incremental body displacement; analyzing a plurality of the samples to determine one or more values for parameters of patient respiratory rate; and determining when one or more of the parameter values indicate abnormal respiration. A corresponding method and computer program are also provided.

Capacitive sensor for respiratory monitoring

An example method for assisting a patient to conduct a physical therapy session can include: presenting information about the physical therapy session on a display, the information including one or more actions to be performed by the patient; capturing one or more images of the patient as the patient performs the one or more actions; detecting, by a computing device, a safety risk related to one or more medical devices associated with the patient during the physical therapy session; and providing feedback regarding the safety risk. Other methods can include: detecting, by a computing device, using the one or more images, a level of effort from the patient during the physical therapy session; and modifying a future physical therapy session for the patient based upon the level of effort; or allowing the patient to compete with other patients during the physical therapy session based upon the level of effort.

Interactive physical therapy

A method of determining respiratory rate from a photoplethysmograph signal includes filtering the photoplethysmograph signal to create a filtered photoplethysmograph signal. The method also includes detecting peaks and valleys in the filtered photoplethysmograph signal to extract predetermined features from the filtered photoplethysmograph signal. The method also includes obtaining variation waveforms from the predetermined features. The method also includes removing outliers from each of variation waveforms. The method also includes interpolating for the outliers removed from each of the variation waveforms to acquire a frequency estimation of each of the variation waveforms. The method also includes determining a respiratory rate based on the frequency estimation of each of the variation waveforms. The method also includes determining a weighted average of the respiratory rate the variation waveforms to determine an estimated respiratory rate.

Respiration rate estimation from a photoplethysmography signal

A patient support apparatus includes a frame and a mattress positioned on the frame. A respiratory therapy device is coupled to the frame. The respiratory therapy device includes a blower having an inlet and an outlet, a patient interface, and a valve including a valve member that is rotatable through a first angular displacement in a first direction from a first position to a second position. The outlet of the blower is coupled to the patient interface so that positive pressure is provided to a patient's airway via the patient interface when the valve member is in the first position. The inlet of the blower is coupled to the patient interface so that negative pressure is provided to the patient's airway via the patient interface when the valve member is in the second position.

Combination respiratory therapy and mattress functionality system integrated into a patient bed

Selectable control valves and mattress systems incorporating the same are disclosed. In one embodiment, an inflatable mattress system includes a pump, a selectable control valve in fluid communication with the pump, and a mattress pad having a plurality of inflatable chambers that are in fluid communication with the selectable control valve. The selectable control valve includes a valve case having an inlet port and a plurality of supply ports, a stepper motor having a motor body that is coupled to the valve case, a rotating plate having a passage and coupled to the stepper motor, and a plurality of seal cartridges positioned within the valve case. The seal cartridges each include an engagement spring and a compliant seal, where the engagement spring maintains a load on the compliant seal such that the compliant seal forms a fluid tight seal with the rotating plate.

Selectable control valves and inflatable mattress systems comprising the same

A respiratory acoustic analysis system for sensing and analyzing respiratory sounds of a patient may include a High Frequency Chest Wall Oscillation (HFCWO) vest, at least one sensor coupled with the HFCWO vest, and an algorithm stored in a processor for processing sensed data from the at least one acoustic sensor to provide processed data describing the respiratory sounds of the patient, in a form that can be used by a physician or other user.

Respiratory sound analysis for lung health assessment

Mattress overlay control system with rotary valves and graphical user interface for percussion and vibration, turn assist and microclimate management

Growing community of inventors

Singapore, Singapore

Yue Wang