Related Experiment Video
Updated: May 13, 2026
14:27
Identification of Disease-related Spatial Covariance Patterns using Neuroimaging Data
Published on: June 26, 2013
15.6K
al-BERT: a semi-supervised denoising technique for disease prediction.
Yun-Chien Tseng1, Chuan-Wei Kuo1, Wen-Chih Peng1
1Department of Computer Science, National Yang Ming Chiao Tung University, University Road, Hsinchiu, 30010, Taiwan.
BMC Medical Informatics and Decision Making
|May 16, 2024
Summary
A new disease prediction model, al-BERT, effectively filters irrelevant data from medical records using a semi-supervised layer and attention mechanism. This enhances disease prediction accuracy, showing a 15% improvement in recall compared to existing methods.
Area of Science:
- Artificial Intelligence in Medicine
- Natural Language Processing for Healthcare
- Machine Learning for Disease Prediction
Background:
- Medical records contain valuable patient health information but often include irrelevant data, challenging machine learning models.
- Experienced clinicians can filter extraneous information, but automated systems struggle with this 'noise'.
- The BERT framework provides a foundation for natural language understanding, but requires adaptation for noisy medical data.
Purpose of the Study:
- To develop an advanced disease prediction model, al-BERT, capable of identifying and filtering irrelevant information from electronic medical records.
- To improve the accuracy and utility of machine learning models in clinical diagnostics by refining data relevance.
- To enhance the model's ability to correlate diseases with patient data by reducing noise.
Main Methods:
- Developed al-BERT, a novel disease prediction model leveraging the BERT framework with a semi-supervised layer for data filtering.
- Implemented a focused attention mechanism within the semi-supervised learning algorithm to prioritize chronic conditions and salient features.
- Evaluated al-BERT's performance using real-world health insurance data from Taiwan's National Health Insurance.
Main Results:
- al-BERT demonstrated superior performance compared to standard BERT and BERT with basic filtering techniques.
- The model showed significant improvements in AUC-ROC, precision, recall, and overall accuracy.
- Notably, al-BERT achieved a 15% increase in recall compared to the current state-of-the-art disease prediction methods.
Conclusions:
- The attention mechanism and selection module are critical components contributing to al-BERT's effectiveness.
- The developed model successfully addresses the challenge of irrelevant data in medical records for improved disease prediction.
- al-BERT offers a more accurate and reliable tool for electronic disease diagnostics.
Related Concept Videos
01:18Steps in Outbreak Investigation
In the ever-evolving field of public health, statistical analysis serves as a cornerstone for understanding and managing disease outbreaks. By leveraging various statistical tools, health professionals can predict potential outbreaks, analyze ongoing situations, and devise effective responses to mitigate impact. For that to happen, there are a few possible stages of the analysis:
01:26Principles of Disease Surveillance
Disease surveillance is the systematic collection, analysis, and interpretation of health data essential to the planning, implementation, and evaluation of public health practice. This process integrates data dissemination to entities responsible for preventing and controlling disease, injury, and disability. Surveillance systems provide crucial information for action, helping public health authorities make informed decisions to manage and prevent outbreaks, ensure public safety, optimize...
01:26Differential Staining Technique
Differential staining is an essential microbiological technique that exploits variations in cell wall structures to classify and identify microorganisms. It facilitates the distinction of bacteria, aiding in diagnostic and research applications. Two of the most widely used differential staining methods are Gram staining and acid-fast staining, both of which rely on the chemical and structural differences in bacterial cell walls.Gram Staining TechniqueGram staining differentiates bacteria by...
01:19Key Techniques in Microbiology
Aseptic techniques prevent contamination, ensure experimental accuracy, and protect researchers and microbial cultures. These techniques are essential in clinical, industrial, and research settings where sterility is required.Maintaining Sterility in Laboratory PracticesScientists maintain sterility by sterilizing tools with heat or chemicals, disinfecting work surfaces, and handling cultures in controlled environments. Working near an open flame or within a laminar flow hood reduces the risk...
01:23Investigation of Disease Outbreaks
Multistate foodborne outbreaks pose significant public health risks and require meticulous investigation to identify sources and implement control measures. The Centers for Disease Control and Prevention (CDC) utilizes a dynamic seven-step process for these investigations, integrating data from laboratories, interviews, and environmental assessments to protect public health.Outbreak Detection: The detection of multistate outbreaks typically begins with PulseNet, the CDC's national laboratory...
01:24Automated Microbial Diagnostics
Automated diagnostic analyzers have transformed clinical microbiology by providing rapid and reliable methods for pathogen identification and antibiotic susceptibility testing. Among these systems, the Vitek 2 is widely used because it automates the traditionally labor-intensive processes of microbial identification (ID) and antibiotic susceptibility testing (AST), delivering standardized and timely results that are essential for effective patient care.Microbial Identification with ID CardsThe...