Deep Learning Models Take Shortcuts to Make Inaccurate or Bi

Deep Learning Models Take Shortcuts to Make Inaccurate or Biased Predictions for Cancer Patients


Deep Learning Models Take Shortcuts to Make Inaccurate or Biased Predictions for Cancer Patients
Reviewed by Laura ThomsonJul 23 2021
Artificial intelligence tools and deep learning models are a powerful tool in cancer treatment. They can be used to analyze digital images of tumor biopsy samples, helping physicians quickly classify the type of cancer, predict prognosis and guide a course of treatment for the patient. However, unless these algorithms are properly calibrated, they can sometimes make inaccurate or biased predictions.
A new study led by researchers from the University of Chicago shows that deep learning models trained on large sets of cancer genetic and tissue histology data can easily identify the institution that submitted the images. The models, which use machine learning methods to "teach" themselves how to recognize certain cancer signatures, end up using the submitting site as a shortcut to predicting outcomes for the patient, lumping them together with other patients from the same location instead of relying on the biology of individual patients. This in turn may lead to bias and missed opportunities for treatment in patients from racial or ethnic minority groups who may be more likely to be represented in certain medical centers and already struggle with access to care.

Related Keywords

Alexander Pearson , Laura Thomsonjul , Nature Communications , University Of Chicago , Pearson , Assistant Professor , Cancer Genome Atlas , அலெக்சாண்டர் பியர்சன் , இயற்கை தகவல்தொடர்புகள் , பல்கலைக்கழகம் ஆஃப் சிகாகோ , பியர்சன் , உதவியாளர் ப்ரொஃபெஸர் , புற்றுநோய் மரபணு அட்லஸ் ,

© 2025 Vimarsana