The Domain Shift Problem of Medical Image Segmentation and Vendor-Adaptation by Unet-GAN

Abstract

Convolutional neural network (CNN), in particular the Unet, is a powerful method for medical image segmentation. To date Unet has demonstrated state-of-art performance in many complex medical image segmentation tasks, especially under the condition when the training and testing data share the same distribution (i.e. come from the same source domain). However, in clinical practice, medical images are acquired from different vendors and centers. The performance of a U-Net trained from a particular source domain, when transferred to a different target domain (e.g. different vendor, acquisition parameter), can drop unexpectedly. Collecting a large amount of annotation from each new domain to retrain the U-Net is expensive, tedious, and practically impossible.In this work, we proposed a generic framework to address this problem, consisting of (1) an unpaired generative adversarial network (GAN) for vendoradaptation, and (2) a Unet for object segmentation. In the proposed Unet-GAN architecture, GAN learns from Unet at the feature level that is segmentationspecific. We used cardiac cine MRI as the example, with three major vendors (Philips, Siemens, and GE) as three domains, while the methodology can be extended to medical images segmentation in general. The proposed method showed significant improvement of the segmentation results across vendors. The proposed Unet-GAN provides an annotation-free solution to the cross-vendor medical image segmentation problem, potentially extending a trained deep learning model to multi-center and multi-vendor use in real clinical scenario.

A Comparative Evaluation of Computed Tomography Images for the Classification of Spirometric Severity of the Chronic Obstructive Pulmonary Disease with Deep Learning

“…We simply compared the results with and without preprocessing using a fixed model of CT imaging to understand how the presence of preprocessing affects the accuracy of the classifier. Since medical images are said to have different image characteristics depending on the vendor and imaging conditions [ 37 ], it was necessary to minimize the effect of differences in conditions caused by using images from other facilities on classification accuracy. Since COPD is a systemic inflammatory disease with various complications, it is desirable to manage COPD to maintain the long-term activity of daily living using drug therapy and rehabilitation, especially in Japan, where the population is superaged.…”

La integración de la inteligencia artificial en el abordaje clínico del paciente: enfoque en la imagen cardiaca

“…An algorithm trained on biased data (images from a local cohort or a specific vendor), might not perform well in a real-world setting, hence, the correct interpretation of a case with a pathology/phenotype outside the training data is not feasible. Transfer learning, through the combination of CNN-GAN architecture, has been used to improve the performance of DL algorithms when applied to data from alternate vendors, providing a solution to the common challenge of applying an algorithm to multicentre, multivendor data 78 . Combined with internal validation (i.e.…”

Systematic and Comprehensive Automated Ventricle Segmentation on Ventricle Images of the Elderly Patients: A Retrospective Study

“…In real-world scenarios, the thick-slice images are more easily obtained, while thin-slices images are rare, and it is more difficult for clinicians to annotate them. Moreover, the distribution of different image thicknesses can result in the domain shift problem that can confuse the deep learning models (Yan et al, 2019). Therefore, we proposed a thickness agnostic image segmentation model, which only required the annotation of thick-slice images for the model training.…”