Abstract: A method and a system for mental index prediction are provided. The method includes the following steps. A plurality of images of a subject person are obtained. A plurality of emotion tags of the subject person in the images are analyzed. A plurality of integrated emotion tags in a plurality of predetermined time periods are calculated according to the emotion tags respectively corresponding to the images. A plurality of preferred features are determined according to the integrated emotion tags. A mental index prediction model is established according to the preferred features to predict a mental index according to the emotional index prediction model.

Abstract: A feature identifying method and an electronic device are provided. The method includes: obtaining a plurality of physiological information obtained by measuring a subject at a plurality of time points in one day; converting the plurality of physiological information into a plurality of correlation features respectively; establishing a plurality of first risk prediction models according to the plurality of correlation features, and identifying at least one first correlation feature from the plurality of correlation features according to the plurality of first risk prediction models; establishing a plurality of second risk prediction models according to the at least one first correlation feature, and identifying, according to the plurality of second risk prediction models, at least one second correlation feature capable of predicting a specific disease from the at least one first correlation feature; and outputting the at least one second correlation feature.

Abstract: A method and an electronic device of checking drug interaction are provided. The method includes: generating a first odds ratio (OR) between a first drug combination and a hospitalization event, a second OR between a second drug combination and the hospitalization event, and a third OR between a third drug combination and the hospitalization event according to a plurality of medical records; generating a first fraction corresponding to a first drug according to the second OR; generating a second fraction corresponding to a second drug according to the third OR; and outputting the first drug combination in response to the first OR being greater than a first threshold, a sum of the first fraction and the second fraction being greater than a second threshold, and a quotient of the first fraction and the second fraction being less than a third threshold.

Abstract: A method and an electronic device for selecting influence indicators by using an automatic mechanism are provided. The method includes following steps. Raw data is obtained, where the raw data includes a body-related variable and a plurality of to-be-measured indicators corresponding to the body-related variable. The body-related variable is set as a target parameter. The body-related variable and the to-be-measured indicators are input into a plurality of validation models, and the to-be-measured indicators are sorted according an output result of the validation models to obtain ranking data. Importance of the to-be-measured indicators is calculated by using a screening condition according to the ranking data, so as to select a candidate indicator from the to-be-measured indicators. An influence indicator is determined by calculating a correlation between the candidate indicator and the body-related variable.

Abstract: A model training method and an electronic device are provided. The method includes the following steps: establishing a brain age prediction model according to a training set; adjusting a parameter in the brain age prediction model according to a validation set; inputting a test set into the brain age prediction model with the adjusted parameter to obtain a plurality of first predicted brain ages; determining whether the first predicted brain ages satisfy a first specific condition; and completing training of the brain age prediction model when the first predicted brain ages satisfy the first specific condition.

Abstract: A medication risk evaluation method and a medication risk evaluation device are provided. The medication risk evaluation method includes: obtaining first medication route information related to a first medicine combination in a medication database; obtaining second medication route information related to a second medicine combination in the medication database; obtaining overlapping medication information between the first medication route information and the second medication route information; determining whether the overlapping medication information meets a noise exclusion condition; and if the overlapping medication information meets the noise exclusion condition, establishing a risk evaluation model based on other medication route information in the medication database excluding the first medication route information, where the risk evaluation model is adapted to evaluate a risk of using at least one medicine in the medication database.

Abstract: A method and a system for mental index prediction are provided. The method includes the following steps. A plurality of images of a subject person are obtained. A plurality of emotion tags of the subject person in the images are analyzed. A plurality of integrated emotion tags in a plurality of predetermined time periods are calculated according to the emotion tags respectively corresponding to the images. A plurality of preferred features are determined according to the integrated emotion tags. A mental index prediction model is established according to the preferred features to predict a mental index according to the emotional index prediction model.

Abstract: A method for calculating a high risk route of administration is provided. Multiple arrangement routes composed of every two medicines among multiple medicines included in a medical record database are listed. A risk value of each arrangement route is calculated by querying the medical record database based on a specified medication result. A risk score of each arrangement route is calculated according to the risk value, and the arrangement routes are sorted based on the risk scores. Starting from the arrangement route with the highest risk score, N arrangement routes are retrieved and a combination on N of the arrangement routes is performed to obtain multiple strung routes. The number of medicines included in each strung route matches a specified medication number.

Abstract: A method for analyzing an electrocardiography (ECG) signal is provided. The ECG signal is measured through a sensor. One or more specified waveforms in the ECG signal are detected. Multiple ECG features from each specified waveform are captured. The ECG features are input into multiple candidate models to obtain multiple candidate scores. Each candidate score is compared with a standard value to obtain an ideal score among the candidate scores. One of the candidate models corresponding to the ideal score is selected as a risk prediction model to predict disease risk through the risk prediction model.

Abstract: A training data processing method and an electronic device are provided. The method includes: obtaining medical history data including at least one first disease suffered by a user; setting a plurality of disease types according to a target disease; setting a time interval; obtaining at least one second disease in the time interval from the medical history data; performing a pre-processing operation on the second disease according to the disease types to obtain processed data; and inputting the processed data to a neural network to train the neural network.

Abstract: A model training method and an electronic device are provided. The method includes the following steps: establishing a brain age prediction model according to a training set; adjusting a parameter in the brain age prediction model according to a validation set; inputting a test set into the brain age prediction model with the adjusted parameter to obtain a plurality of first predicted brain ages; determining whether the first predicted brain ages satisfy a first specific condition; and completing training of the brain age prediction model when the first predicted brain ages satisfy the first specific condition.

Abstract: A method and an electronic device for selecting influence indicators by using an automatic mechanism are provided. The method includes following steps. Raw data is obtained, where the raw data includes a body-related variable and a plurality of to-be-measured indicators corresponding to the body-related variable. The body-related variable is set as a target parameter. The body-related variable and the to-be-measured indicators are input into a plurality of validation models, and the to-be-measured indicators are sorted according an output result of the validation models to obtain ranking data. Importance of the to-be-measured indicators is calculated by using a screening condition according to the ranking data, so as to select a candidate indicator from the to-be-measured indicators. An influence indicator is determined by calculating a correlation between the candidate indicator and the body-related variable.

Abstract: A feature identifying method and an electronic device are provided. The method includes: obtaining a plurality of physiological information obtained by measuring a subject at a plurality of time points in one day; converting the plurality of physiological information into a plurality of correlation features respectively; establishing a plurality of first risk prediction models according to the plurality of correlation features, and identifying at least one first correlation feature from the plurality of correlation features according to the plurality of first risk prediction models; establishing a plurality of second risk prediction models according to the at least one first correlation feature, and identifying, according to the plurality of second risk prediction models, at least one second correlation feature capable of predicting a specific disease from the at least one first correlation feature; and outputting the at least one second correlation feature.

Abstract: A disease suffering probability prediction method and an electronic apparatus are provided. The method includes: determining a path length; obtaining a plurality of first paths conforming to the path length from a plurality of history data of a specific disease; obtaining a plurality of second paths positively related to the specific disease from the plurality of first paths; filtering the plurality of second paths to obtain a plurality of third paths, and establishing a prediction model according to the plurality of third paths; and inputting a path to be predicted to the prediction model and outputting a probability of suffering the specific disease.

Abstract: A statistical performance evaluation method for different grouping sets includes setting a plurality of first grouping ranges of a first grouping set corresponding to a sample space, setting a plurality of second grouping ranges of a second grouping set corresponding to the sample space, generating a plurality of first probability values and a plurality of first standard deviations corresponding to the plurality of first grouping ranges at each sampling time according to the sample space, generating a plurality of second probability values and a plurality of second standard deviations corresponding to the plurality of second grouping ranges at the each sampling time according to the sample space, and generating a plurality of statistical indicators corresponding to the first grouping set and the second grouping set and outputting a statistical performance ranking result of the first grouping set and the second grouping set accordingly.