Adaptive adversarial neural networks for the analysis of lossy and domain-shifted datasets of medical images.
Manoj Kumar KanakasabapathyPrudhvi ThirumalarajuHemanth KandulaFenil DoshiAnjali Devi SivakumarDeeksha KartikRaghav GuptaRohan PooniwalaJohn A BrandaAthe M TsibrisDaniel R KuritzkesRudy Leon De WildeCharles L BormannHadi ShafieePublished in: Nature biomedical engineering (2021)
In machine learning for image-based medical diagnostics, supervised convolutional neural networks are typically trained with large and expertly annotated datasets obtained using high-resolution imaging systems. Moreover, the network's performance can degrade substantially when applied to a dataset with a different distribution. Here, we show that adversarial learning can be used to develop high-performing networks trained on unannotated medical images of varying image quality. Specifically, we used low-quality images acquired using inexpensive portable optical systems to train networks for the evaluation of human embryos, the quantification of human sperm morphology and the diagnosis of malarial infections in the blood, and show that the networks performed well across different data distributions. We also show that adversarial learning can be used with unlabelled data from unseen domain-shifted datasets to adapt pretrained supervised networks to new distributions, even when data from the original distribution are not available. Adaptive adversarial networks may expand the use of validated neural-network models for the evaluation of data collected from multiple imaging systems of varying quality without compromising the knowledge stored in the network.
Keyphrases
- neural network
- convolutional neural network
- high resolution
- deep learning
- machine learning
- big data
- healthcare
- electronic health record
- endothelial cells
- image quality
- artificial intelligence
- high speed
- optical coherence tomography
- rna seq
- computed tomography
- magnetic resonance imaging
- mass spectrometry
- magnetic resonance
- resistance training
- single cell
- data analysis
- pluripotent stem cells
- quality improvement
- ionic liquid