Generalizing Nucleus Recognition Model in Multi-source Ki67 Immunohistochemistry Stained Images via Domain-Specific Pruning

Ki67 is a significant biomarker in the diagnosis and prognosis of cancer, whose index can be evaluated by quantifying its expression in Ki67 immunohistochemistry (IHC) stained images. However, quantitative analysis on multi-source Ki67 images is yet a challenging task in practice due to cross-domain distribution differences, which result from imaging variation, staining styles and lesion types. Many recent studies have made some efforts on domain generalization (DG), whereas there are still some noteworthy limitations. Specifically in the case of Ki67 images, learning invariant representation is at the mercy of the insufficient number of domains and the cell categories mismatching in different domains. In this paper, we propose a novel method to improve DG by searching the domain-agnostic subnetwork in a domain merging scenario. Partial model parameters are iteratively pruned according to the domain gap, which is caused by the data converting from a single domain into merged domains during training. In addition, the model is optimized by fine-tuning on merged domains to eliminate the interference of class mismatching among various domains. Furthermore, an appropriate implementation is attained by applying the pruning method to different parts of the framework. Compared with known DG methods, our method yields excellent performance in multiclass nucleus recognition of Ki67 IHC images, especially in the lost category cases. Moreover, our competitive results are also evaluated on the public dataset over the state-of-the-art DG methods.

[1]  John M S Bartlett,et al.  An international multicenter study to evaluate reproducibility of automated scoring for assessment of Ki67 in breast cancer , 2018, Modern Pathology.

[2]  Michael Carbin,et al.  The Lottery Ticket Hypothesis: Finding Sparse, Trainable Neural Networks , 2018, ICLR.

[3]  Eberhard Korsching,et al.  Interlaboratory variability of Ki67 staining in breast cancer. , 2017, European Journal of Cancer.

[4]  Trevor Darrell,et al.  Adversarial Discriminative Domain Adaptation , 2017, 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[5]  Kate Saenko,et al.  Deep CORAL: Correlation Alignment for Deep Domain Adaptation , 2016, ECCV Workshops.

[6]  François Laviolette,et al.  Domain-Adversarial Training of Neural Networks , 2015, J. Mach. Learn. Res..

[7]  David Lopez-Paz,et al.  In Search of Lost Domain Generalization , 2020, ICLR.

[8]  Vladimir Vapnik,et al.  An overview of statistical learning theory , 1999, IEEE Trans. Neural Networks.

[9]  G. Klöppel,et al.  Ki67 labeling index: assessment and prognostic role in gastroenteropancreatic neuroendocrine neoplasms , 2018, Virchows Archiv.

[10]  Silvio Savarese,et al.  Generalizing to Unseen Domains via Adversarial Data Augmentation , 2018, NeurIPS.

[11]  Jian Sun,et al.  Deep Residual Learning for Image Recognition , 2015, 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[12]  N. Rajpoot,et al.  Locality Sensitive Deep Learning for Detection and Classification of Nuclei in Routine Colon Cancer Histology Images , 2016, IEEE Transactions on Medical Imaging.

[13]  H. Kuhn.  The Hungarian method for the assignment problem , 1955 .

[14]  Debashis Ghosh,et al.  Adversarial Domain Adaptation and Pseudo-Labeling for Cross-Modality Microscopy Image Quantification , 2019, MICCAI.

[15]  K. Gelmon,et al.  Ki67 in breast cancer: prognostic and predictive potential. , 2010, The Lancet. Oncology.

[16]  Jing Qin,et al.  Domain-invariant interpretable fundus image quality assessment , 2020, Medical Image Anal..

[17]  Hai Su,et al.  Efficient and robust cell detection: A structured regression approach , 2018, Medical Image Anal..

[18]  Fan Zhang,et al.  Unsupervised Domain Adaptation via Disentangled Representations: Application to Cross-Modality Liver Segmentation , 2019, MICCAI.

[19]  Yongxin Yang,et al.  Learning to Generalize: Meta-Learning for Domain Generalization , 2017, AAAI.

[20]  Lin Yang,et al.  Pixel-to-Pixel Learning With Weak Supervision for Single-Stage Nucleus Recognition in Ki67 Images , 2019, IEEE Transactions on Biomedical Engineering.

[21]  Nasir M. Rajpoot,et al.  Locality Sensitive Deep Learning for Detection and Classification of Nuclei in Routine Colon Cancer Histology Images , 2016, IEEE Trans. Medical Imaging.

[22]  Hidekata Hontani,et al.  Multi-scale Domain-adversarial Multiple-instance CNN for Cancer Subtype Classification with Unannotated Histopathological Images , 2020, 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR).

[23]  Percy Liang,et al.  Distributionally Robust Neural Networks for Group Shifts: On the Importance of Regularization for Worst-Case Generalization , 2019, ArXiv.

[24]  Harold W. Kuhn,et al.  The Hungarian method for the assignment problem , 1955, 50 Years of Integer Programming.

[25]  Liantao Li,et al.  Ki67 targeted strategies for cancer therapy , 2018, Clinical and Translational Oncology.

[26]  Zoya I. Volynskaya,et al.  Ki67 Quantitative Interpretation: Insights using Image Analysis , 2019, Journal of pathology informatics.

[27]  Shruti Tople,et al.  Domain Generalization using Causal Matching , 2020, ICML.

[28]  David Lopez-Paz,et al.  Invariant Risk Minimization , 2019, ArXiv.

[29]  Fakhri Karray,et al.  Magnification Generalization for Histopathology Image Embedding , 2021, ArXiv.

[30]  Daguang Xu,et al.  Generalizing Deep Learning for Medical Image Segmentation to Unseen Domains via Deep Stacked Transformation , 2020, IEEE Transactions on Medical Imaging.

[31]  Ioannis Mitliagkas,et al.  Generalizing to unseen domains via distribution matching , 2019 .

[32]  J. Reis-Filho,et al.  Ki67 Assessment in Breast Cancer: Are We There yet? , 2020, Journal of the National Cancer Institute.

[33]  Mitko Veta,et al.  Learning Domain-Invariant Representations of Histological Images , 2019, Front. Med..

[34]  Thomas Brox,et al.  U-Net: Convolutional Networks for Biomedical Image Segmentation , 2015, MICCAI.

[35]  Yanwu Xu,et al.  Residual-CycleGAN based Camera Adaptation for Robust Diabetic Retinopathy Screening , 2020, MICCAI.