Abstract: Systems and methods for providing health information. A method for building a health predictive model is provided based on a plurality of electronic medical records representing a plurality of electronic medical cases each associated with a diagnosis of a medical condition. Symptom-attribute-value (SAV) ontology objects are extracted from the plurality of electronic medical records. A plurality of feature vectors respectively associated with the medical cases and at least partially based on the SAV ontology objects are generated. The health predictive model can be trained based on the feature vectors by using the diagnoses as category labels of the training. The method for providing health information can include presenting a probability vector by inputting a current feature vector into the health predictive model. The current feature vector can be based on responses received via a health conversation. Advantageously, personalized and highly relevant medical advice can be provided to a user.

Abstract: A method includes estimating a value of a parameter indicative of an age or lifespan of a population of red blood cells of a subject, estimating a value of average glucose (AG) of the subject based on (i) the value of the parameter and (ii) a value indicative of an amount of glycated hemoglobin (HbA1c) of the subject, and providing information for treatment or diagnosis of a hyperglycemia condition of the subject based on the estimated value of AG.

Abstract: A method includes estimating a value of a parameter indicative of an age or lifespan of a population of red blood cells of a subject, estimating a value of average glucose (AG) of the subject based on (i) the value of the parameter and (ii) a value indicative of an amount of glycated hemoglobin (HbA1c) of the subject, and providing information for treatment or diagnosis of a hyperglycemia condition of the subject based on the estimated value of AG.

Abstract: Systems and methods for providing health information. A method for building a health predictive model is provided based on a plurality of electronic medical records representing a plurality of electronic medical cases each associated with a diagnosis of a medical condition. Symptom-attribute-value (SAV) ontology objects are extracted from the plurality of electronic medical records. A plurality of feature vectors respectively associated with the medical cases and at least partially based on the SAV ontology objects are generated. The health predictive model can be trained based on the feature vectors by using the diagnoses as category labels of the training. The method for providing health information can include presenting a probability vector by inputting a current feature vector into the health predictive model. The current feature vector can be based on responses received via a health conversation. Advantageously, personalized and highly relevant medical advice can be provided to a user.

Abstract: A method includes estimating a value of a parameter indicative of an age or lifespan of a population of red blood cells of a subject, estimating a value of average glucose (AG) of the subject based on (i) the value of the parameter and (ii) a value indicative of an amount of glycated hemoglobin (HbA1c) of the subject, and providing information for treatment or diagnosis of a hyperglycemia condition of the subject based on the estimated value of AG.