Multidimensional stroke prevention screening method based on artificial intelligence
An artificial intelligence and stroke technology, applied in the database field, can solve problems such as not being particularly ideal, biochemical indicators difficult to reflect disease progression, and low specificity.
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Embodiment 1
[0075] Example 1: Process for establishing a database of stroke cases
[0076] Such as figure 1 Shown: The artificial intelligence multi-dimensional stroke prevention screening system uses the data of the three dimensions of stroke as the main data for database construction.
[0077] The first aspect included in the system is the doctor’s clinical consultation data, including: medical history of stroke, family history of stroke, atrial fibrillation or valvular heart disease, age, diabetes, blood lipid, blood pressure, physical exercise, BMI, smoking and alcohol abuse, etc. 10 items. The doctor’s clinical consultation data is mainly through the doctor’s interview with the screening subjects. According to the description of the above 10 stroke-related qualitative indicators by the screening subjects, the results are recorded as “yes” and “no”. The doctor inputs stroke-related data into the report and outputs the terminal device. The terminal device establishes a connection wit...
Embodiment 2
[0082] Example 2: Artificial intelligence algorithm model establishment process for stroke prevention screening
[0083] Such as figure 2 As shown, the sample data includes both qualitative data of doctors’ clinical consultation data and quantitative data of stroke biomarker results and daily monitoring indicators. The qualitative results in the sample data are 0 for “no” and 1 for “yes”. Quantitative results are calculated by inputting the original value of the data into the model as the eigenvalue of the model. According to the sample feature value may contain non-linear factors, so compare random forest, support vector machine, deep neural network and other algorithms, use Softmax through the output layer, and use Dropout to avoid model overfitting. Using Cross Validation method and Confusion Matrix as the result evaluation method and calculating accuracy rate, recall rate and F value, a deep neural network using multi-layer neural network and ADAM optimization algorithm ...
Embodiment 3
[0086] Example 3: Stroke-related data input report output device terminal device result output process
[0087] Such as image 3 As shown, the doctor’s clinical inquiry data is input, the screening object provides their own blood, and the automatic biochemical analyzer part of the terminal equipment performs 16 blood biochemical biomarker detection and analysis to obtain quantitative blood biochemical results, which are connected to the cloud service platform and combined with The daily monitoring indicators of the screening object, and according to the stroke prevention screening artificial intelligence algorithm model, finally present the report of the intelligent stroke prevention screening system, including: doctor's clinical consultation data, blood biochemical biomarkers, daily monitoring indicators, etc. The specific situation of the three dimensions, risk coefficient, result interpretation, processing opinions and other information.
PUM
Login to View More Abstract
Description
Claims
Application Information
Login to View More 


