International Business Machines Corporation

Jing Mei

Guo Tong Xie

Gang Hu

Yi Qin Yu

Jing Li

Wen Sun

Shi Wan Zhao

Yuan Ni

Eryu Xia

Shiwan Zhao

Haifeng Liu

Sheng Liu

En Liang Xu

Xu Min

Weijia Shen

Shao Chun Li

Wei Shen

Jian Min Jiang

Han Li

Feng Cao

Hai Feng Liu

Yuan Zhou

Bibo Hao

Jian Wang

Li Yu

Changhua Sun

Roslyn I Hickson

Shi Jing Guo

Zhi Qiao

Yan Fei Wang

Chia Yeow Khiang

Ying Xue Li

Embodiments of the present disclosure relate to a method, system, and computer program product for domain adaptation. According to the method, a source model of a source domain is obtained, where the source model is trained to generate a label indicating a predicted category of data from the source domain. A training sample from a target domain is obtained, where the training sample comprises training data from the target domain and a true label indicating a true category of the training data from the target domain. A first label is generated for the training data by using the source model. The first label indicates a predicted category of the training data. A target model of the target domain is trained based on the training data, the true label and the first label.

Domain adaptation

The present disclosure relates to a method, system and computer program product for electronic health record (EHR) data synthetization. According to the method, an original EHR dataset X is captured. A latent space Z is generated from the original EHR dataset X, wherein dimensionality of Z is lower than that of X. A stochastic process prior module is applied to the latent space Z. Synthetic EHR dataset X' is reconstructed from the latent space Z after being applied with the stochastic process prior.

Electronic health record data synthesization

Aspects of the invention include a computer-implemented method including extracting a domain-specific object feature from a first image data, wherein the feature describes an object in the first image data. A domain-specific semantic meaning of text data is determined. The object feature is mapped to a portion of the text data, wherein the portion of the text data describes the object. A joint representation of the object and the portion of the text data is created. A second image data and a query directed towards an object in the second image data is received. An answer to the query is generated based on the joint representation.

Medical visual question answering

The disclosure provides a method for data instance processing. The method includes obtaining a set of data instances collected from a plurality of organizations. Each of the data instances includes at least one record formed in an organization that stores values of a plurality of attributes of the data instance. The method also includes dividing the set of data instances into groups, wherein data instances with conflicting values for the same attribute are divided into different groups. The method further includes subdividing data instances in each of the groups into clusters.

Cross-organization data instance matching

A method, a system and a computer program product may evaluate reduction of disease risk. Patient data of a patient may be received. A selection of a disease outcome may be received. A risk score that the patient will experience the selected disease outcome may be determined. The determining may use the patient data. Intervention options may be generated based on the patient data and by accessing a medical record data structure. An intervention effect for each of the intervention options may be determined. The intervention effect may change the risk score. The intervention effects may be compared. A recommendation of at least one of the intervention options may be provided based on the comparing of the intervention effects.

Evaluation of reduction of disease risk and treatment decision

A computer-implemented method for predicting non-communicable diseases with infectious risk factors using artificial intelligence includes detecting one or more risk factors associated with a non-communicable disease based on a graph associated with person-to-person links, generating a data structure for compactly representing the graph to compute at least one person-to-person distance, and performing a machine learning technique with regularization of the at least one person-to-person distance.

Predicting non-communicable disease with infectious risk factors using artificial intelligence

Systems, computer-implemented methods and/or computer program products that facilitate predicting personalized risks based on intrinsic factors and extrinsic factors are provided. In one example, a computer-implemented method comprises: collecting, by a system operatively coupled to a processor, intrinsic factors and extrinsic factors associated with infectious diseases; generating, by the system, a probabilistic model based on the intrinsic factors and extrinsic factors, wherein the model incorporates node characteristics into infection probability; and refining, by the system, the model through concurrently learning respective node thresholds and hidden infection network structure of the model.

Personalized risk prediction based on intrinsic and extrinsic factors

A computer implemented method identifies guidelines through use of a neural network by a remote guideline server. A client computer transmits instructions to the remote guideline server to retrieve and evaluate multiple candidate guidelines. The remote guideline server utilizes a neural network to identify a string of terms found in each of the multiple candidate guidelines that match one or more strings of terms from a model guideline; to identify a semantic concept of each of the multiple candidate guidelines that matches one or more semantic concepts from the model guideline; and to identify a structural pattern of each of the multiple candidate guidelines that matches one or more structural patterns of the model guideline. The candidate guidelines that match the model guideline are then sent from the remote guideline server to the client computer.

Remote neural network processing for guideline identification

Patient treatment may be optimized using Recurrent Neural Network (RNN) based state simulation and Reinforcement learning (RL) techniques to simulate future states and actions. A RNN state simulator and a RL action generator may be trained using patient data such as historical states and actions. The RL action generator may be optimized by applying the RNN state simulator to simulating future states and applying the RL action generator to generate recommended actions based on the simulated future states. This process may be iteratively performed until a computational convergence is reached by the RL action generator which may indicate that the RL action generator has been optimized. A patient state may be fed into the optimized RL action generator to generate an optimal recommended treatment action.

Optimizing patient treatment recommendations using reinforcement learning combined with recurrent neural network patient state simulation

A computer implemented method for processing a plurality of medical indication conditions on a computer that includes a processor communicatively coupled to a memory includes obtaining a plurality of predetermined indication conditions which relate to a plurality of parameters, and forming a plurality of conditional segments based on respective values of the plurality of parameters defined in the plurality of predetermined indication conditions. Each conditional segment of the plurality of conditional segments corresponds to a combination of different value ranges of respective parameters of the plurality of parameters. The plurality of predetermined indication conditions includes a measurement item, a reference standard, a time period, a measurement condition, and a measurement manner.

Method and apparatus for processing medical data

The present disclosure provides a method, apparatus and system for processing a case management model (CMM). According to an embodiment, there is provided a method for processing a CMM, the method includes: obtaining an existing CMM having a plurality of elements; obtaining a new CMM having at least one element; aligning an element of the new CMM to an element of the existing CMM according to match costs between the element of the new CMM and the plurality of elements of the existing CMM; and fusing the new CMM into the existing CMM based on the match cost between the aligned elements.

Case management model processing

A method and system of imputing corrupted sequential data is provided. A plurality of input data vectors of a sequential data is received. For each input data vector of the sequential data, the input data vector is corrupted. The corrupted input data vector is mapped to a staging hidden layer to create a staging vector. The input data vector is reconstructed based on the staging vector, to provide an output data vector. adjusted parameter of the staging hidden layer is iteratively trained until it is within a predetermined tolerance of a loss function. A next input data vector of the sequential data is predicted based on the staging vector. The predicted next input data vector is stored.

Missing values imputation of sequential data

A method for automatically generating a semantic mapping for a relational database RDB includes obtaining a first semantic mapping from a first RDB to an ontology of linked data; obtaining a schema mapping from the first RDB to a second RDB; and generating a second semantic mapping from the second RDB to the ontology of the linked data based on the first semantic mapping and the schema mapping.

Growing community of inventors

Beijing, China

Jing Mei