Method and apparatus for identifying clinical conditions suitable for classification with machine learning models
By training and validating ML for each disease in a clinical setting, models suitable for identifying clinical diseases are identified, solving the problem of low identification accuracy in existing technologies and achieving efficient and low-risk disease identification.
CN122374837APending Publication Date: 2026-07-10SIEMENS HEALTHCARE DIAGNOSTICS INC
View PDF 0 Cites 0 Cited by
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- SIEMENS HEALTHCARE DIAGNOSTICS INC
- Filing Date
- 2024-10-17
- Publication Date
- 2026-07-10
AI Technical Summary
Technical Problem
Existing machine learning models have low success rates in disease identification in clinical settings, leading to operational burden and reduced trust, and traditional methods cannot effectively identify datasets applicable to specific clinical conditions.
Method used
By obtaining a list of clinical symptoms and a biomarker dataset, we perform ML training and validation for each symptom to determine whether the trained model meets the performance metrics and identify models suitable for recognizing clinical symptoms.
Benefits of technology
It improves the accuracy of machine learning models in clinical settings, reduces operational burden, enhances end-user trust, and provides a low-risk, high-performance ML-based clinical disease identification solution.
✦ Generated by Eureka AI based on patent content.
Smart Images
Figure CN122374837A_ABST
Abstract
In some embodiments, a method for identifying clinical conditions suitable for classification with a machine learning model includes obtaining a list of clinical conditions and obtaining a biomarker dataset for the list of clinical conditions. The method includes, for each clinical condition, (1) determining whether a prevalence of the clinical condition within the dataset is within an expected range; and (2) if the prevalence of the clinical condition within the dataset is within the expected range, (a) performing ML training and validation on the ML model; (b) determining whether the trained ML model satisfies one or more performance metrics; and (c) if the one or more performance metrics are satisfied, identifying the trained ML model as suitable for identifying the clinical condition. Numerous other embodiments are provided.
Need to check novelty before this filing date? Find Prior Art