Removing confounding information from fetal ultrasound images
This addresses data quality issues for medical AI in ultrasound, enabling use of large clinical databases, but is incremental as it adapts known techniques to a specific domain.
The paper tackled the problem of confounding text and calipers in fetal ultrasound images that bias deep learning models, developing methods to minimize these effects and validating them on standard plane classification tasks.
Confounding information in the form of text or markings embedded in medical images can severely affect the training of diagnostic deep learning algorithms. However, data collected for clinical purposes often have such markings embedded in them. In dermatology, known examples include drawings or rulers that are overrepresented in images of malignant lesions. In this paper, we encounter text and calipers placed on the images found in national databases containing fetal screening ultrasound scans, which correlate with standard planes to be predicted. In order to utilize the vast amounts of data available in these databases, we develop and validate a series of methods for minimizing the confounding effects of embedded text and calipers on deep learning algorithms designed for ultrasound, using standard plane classification as a test case.