HOW IT WORKS
Auto-detect issues
Discover outliers (anomalies) which may have an outsized impact on data-driven conclusions and should be handled with care. Auto-correct common issues in messy healthcare data to immediately produce more reliable ML models. Read more: Improve XGBoost by Improving Data and Improve LLM by Improving Data.
Assess Data Quality
Know which subset of the data is high-quality with confidence, and evaluate the quality of different data sources.
Built with AI
Cleanlab’s AI helps you correct errors in electronic health records, insurance documents, patient communications, and medical imaging metadata/annotations. Model images together with tabular (numeric, categorical) and text information.
Train and Produce Reliable Models
Automatically train robust machine learning models using complex healthcare datasets. Deploy them with one click to make predictions and catch bad decisions in real-time.
Assessing Multiple Data Annotators
Effectively analyze data labeled by multiple annotators (clinicians), and estimate which examples require additional review and which annotators are best/worst overall.
Resources and Tutorials
Videos on using Cleanlab Studio to find and fix incorrect labels for: text data (e.g. medical reports), image data (e.g. medical reports), and tabular data (e.g. medical records).
CASE STUDY
The University of Florida Shands Hospital, through its affiliation with the UF Health Science Center remains at the forefront of medical advancements, ensuring patients have access to the latest medical knowledge and cutting-edge technology.
This research hospital used Cleanlab to build datasets for real-time AI monitoring of ICU patients. The hospital’s datasets include over 18 million depth image frames and 22 million patient face image frames extracted from videos. As it is not practical to annotate the entirety of these massive datasets, Cleanlab was leveraged to implement active learning for data annotation and to assess the quality of multiple annotators.
40M
images processed with Cleanlab
“Active learning is an important machine learning technique that involves an iterative process to choose most informative data samples to be labeled.”
From AI-Enhanced Intensive Care Unit: Revolutionizing Patient Care with Pervasive Sensing, published in March 2023.
