Siemens Medical Solutions Usa, Inc.

Cerner Innovation, Inc.

Siemens Aktiengesellschaft

Wisconsin Alumni Research Foundation

Cerner Health Services, Inc.

Glenn Fung

R Bharat Rao

Balaji Krishnapuram

Shipeng Yu

Sriram Krishnan

Faisal Farooq

Jan DeHaan

Philippe Lambin

Joseph Marcus Overhage

Romer E Rosales

Murat Dundar

Cary Dehing-Oberije

Dirk De Ruysscher

Jinbo Bi

Farbod Rahmanian

Andreas Lubbertus Aloysius Johannes Dekker

John Haley

Bharat R Rao

Arun Krishnan

Marcos Salganicoff

Kazunori Okada

Sathyakama Sandilya

Senthil Periaswamy

Jonathan Stoeckel

Harald Steck

Toshiro Kubota

Vikram Anand

Bradly G Wouters

Phan Hong Giang

Olvi L Mangasarian

Jennifer Dy

Ignazio Palmisano

Kartik Jayasurya Komati

Mark Schmidt

Yan Yan

Steven Florian Petit

Renaud G Seigneuric

Maud H W Starmans

Dimitry S A Nuyten

Hugo J W L Aerts

Arie Van Erk

Michael Gaston Magagnin

Claudia Offermann

Kasper M Rouschop

Christoffel Theodorus Anthonius Evelo

Michel Oellers

Adrian Campbell Begg

Volkan Vural

Sebastian Philipp Brandt

Stanley Chung

Luigi Iannone

John Weis

Nicholas Drummond

Mapping of semantics in healthcare may involve accessing first transaction data in a first database, the first transaction data corresponding to a collection of a first number of fields defined for a condition using a first semantic system to store information and calculating a first distribution of information in the first transaction data. Mapping may also involve accessing second transaction data in a second database, the second transaction data corresponding to a second semantic system different than the first semantic system and the second database comprising a second number of fields using the second semantic system to store information, and calculating a second distribution of information in the second transaction data. The distributions may then be compared and a map relating the semantic systems may be generated and used to communicate between the first and second semantic systems.

Automated mapping of service codes in healthcare systems

Adaptive medical data collection for medical entities may involve managing content by receiving data indicating a context, identifying at least one application or knowledge base associated with the context, designating the identified application or knowledge base as active, and accessing the active application or knowledge base to provide information at an interface point for a medical professionals and a patient.

Adaptive medical documentation system

Adaptive medical data collection for medical entities may involve triggering an analysis of electronic records in response to information input into an Electronic Medical Record (EMR) of a patient. Determining a potential condition for the patient based on the analysis. Identifying additional information indicated as relevant to the potential condition of the patient, and generating a request for the identified additional information.

Automatic mapping of semantics in healthcare is provided. Data sets have different semantics (e.g., Gender designated with M and F in one system and Sex designated with 1 or 2 in another system). For semantic interoperability, the semantic links between the semantic systems of different healthcare entities are created (e.g., Gender=Sex and/or 1=F and 2=M) by a processor from statistics of the data itself. The distribution of variables, values, or variables and values, with or without other information and/or logic, is used to create a map from one semantic system to another. Similar distributions of other variable and/or values are likely to be for variables and/or values with the same meaning.

Mapping of semantics in healthcare may involve accessing first transaction data of a first healthcare entity in a first database, the first transaction data corresponding to a collection of a first number of fields defined for a condition using a first semantic system to store information and calculating a first distribution of information in the first transaction data. Mapping may also involve accessing second transaction data of a second healthcare entity in a second database, the second transaction data corresponding to a second semantic system different than the first semantic system and the second database comprising a second number of fields using the second semantic system to store information, the second number of fields larger than the first number of fields and calculating a second distribution of information in the second transaction data. The distributions may then be compared and a map relating the semantic systems may be generated.

Functional imaging information is used to determine a probability of residual disease given a treatment. The functional imaging information shows different characteristic levels for different regions of the tumor. The probability is output for planning use and/or used to automatically determine dose by region. Using the probability, the dose may be distributed by region so that some regions receive a greater dose than other regions. This distribution by region of dose more likely treats the tumor with a same dose, allows a lesser dose to sufficient treat the tumor, and/or allows a greater dose with a lesser or no increase in risk to normal tissue. The dose plan may account for personalized tumors as each patient may have distinct tumors. Probability of dose application accuracy may also be used, so that a combined treatment probability allows efficient dose planning.

Dose distribution modeling by region from functional imaging

A method, including receiving a data source selection from a user or software application, the data source including medical information of a plurality of patients, receiving, from the user or software application, a data pattern that is related to a concept to be explored in the data source, querying the data source to find information that approximately matches the data pattern; and receiving the information from the data source, wherein the information includes unstructured data, assigning a classification to individual parts of the information based on the part's relationship to the data pattern, and outputting the classified information to the user or software application.

Automated patient/document identification and categorization for medical data

A computer-implemented method for privacy-preserving data mining to determine cancer survival rates includes providing a random matrix B agreed to by a plurality of entities, wherein each entity i possesses a data matrix Aof cancer survival data that is not publicly available, providing a class matrix Dfor each of the data matrices A, providing a kernel K(A, B) by each of said plurality of entities to allow public computation of a full kernel, and computing a binary classifier that incorporates said public full kernel, wherein said classifier is adapted to classify a new data vector according to a sign of said classifier.

System and method for privacy preserving predictive models for lung cancer survival analysis

Knowledge-based interpretable predictive modeling is provided. Expert knowledge is used to seed training of a model by a machine. The expert knowledge may be incorporated as diagram information, which relates known causal relationships between predictive variables. A predictive model is trained. In one embodiment, the model operates even with a missing value for one or more variables by using the relationship between variables. For application, the model outputs a prediction, such as the likelihood of survival for two years of a lung cancer patient. A graphical representation of the model is also output. The graphical representation shows the variables and relationships between variables used to determine the prediction. The graphical representation is interpretable by a physician or other to assist in understanding.

Knowledge-based interpretable predictive model for survival analysis

A method for multiple-label data analysis includes: obtaining labeled data points from more than one labeler; building a classifier that maximizes a measure relating the data points, labels on the data points and a predicted output label; and assigning an output label to an input data point by using the classifier.

Learning classifiers for multiple-label data analysis

Modeling of prognosis of survivability, side-effect, or both is provided. For example, RILI is predicted using bullae information. The amount, volume or ratio of Bullae, even alone, may indicate the likelihood of complication, such as the likelihood of significant (e.g., stage 3) pneumonitis. As another example, RILI is predicted using uptake values of an imaging agent. Standardized uptake from a functional image (e.g., FDG uptake from a positron emission image), alone or in combination with other features, may indicate the likelihood of side-effect. In another example, survivability, such as two-year survivability, is predicted using blood biomarkers. The characteristics of a patient's blood may be measured and, alone or in combination with other features, may indicate the likelihood of survival. The modeling may be for survivability, side-effect, or both and may use one or more of the blood biomarker, uptake value, and bullae features.

Modeling lung cancer survival probability after or side-effects from therapy

A method, including receiving multi-labeler data that includes data points labeled by a plurality of labelers; building a model from the multi-labeler data, wherein the model includes an input variable that corresponds to the data points, a label variable that corresponds to true labels for the data points, and variables for the labels given by the labelers; and executing the model, in response to receiving new data points, to determine a level of expertise of the labelers for the new data points.

Automatic labeler assignment using a model built from multi-labeler data

A method for predicting survival rates of medical patients includes providing a set D of survival data for a plurality of medical patients, providing a regression model having an associated parameter vector &#x3b2;, providing an example xof a medical patient whose survival probability is to be classified, calculating a parameter vector {circumflex over (&#x3b2;)} that maximizes a log-likelihood function of &#x3b2; over the set of survival data, l(&#x3b2;|D), wherein the log likelihood l(&#x3b2;|D) is a strictly concave function of &#x3b2; and is a function of the scalar x&#x3b2;, calculating a weight wfor example x, calculating an updated parameter vector &#x3b2;* that maximizes a function l(&#x3b2;|D&#x222a;{(y,x,w)}), wherein data points (y,x,w) augment set D, calculating a fair log likelihood ratio &#x3bb;from {circumflex over (&#x3b2;)} and &#x3b2;* using &#x3bb;=&#x3bb;(&#x3b2;*|x)+sign(&#x3bb;({circumflex over (&#x3b2;)}|x)){l({circumflex over (&#x3b2;)}|D)&#x2212;l(&#x3b2;*|D)}, and mapping the fair log likelihood ratio &#x3bb;to a fair price y, wherein said fair price is a probability that class label yfor example xhas a value of 1.

System and method for medical predictive models using likelihood gamble pricing

A method of training a classifier for computer aided detection of digitized medical image, includes providing a plurality of bags, each bag containing a plurality of feature samples of a single region-of-interest in a medical image, where each region-of-interest has been labeled as either malignant or healthy. The training uses candidates that are spatially adjacent to each other, modeled by a 'bag', rather than each candidate by itself. A classifier is trained on the plurality of bags of feature samples, subject to the constraint that at least one point in a convex hull of each bag, corresponding to a feature sample, is correctly classified according to the label of the associated region-of-interest, rather than a large set of discrete constraints where at least one instance in each bag has to be correctly classified.

System and method for multiple instance learning for computer aided detection

The present invention provides methods and compositions for predicting patient responses to cancer treatment using hypoxia gene signatures. These methods can comprise measuring in a biological sample from a patient the levels of gene expression of a group of the genes designated herein. The present invention also provides for microarrays that can detect expression from a group of genes.

Gene signature of early hypoxia to predict patient survival

Growing community of inventors

Madison, WI, United States of America

Glenn Fung