Comprehensive analysis of lung cancer pathology images to discover tumor shape and boundary features that predict survival outcome

Pathology images capture tumor histomorphological details in high resolution. However, manual detection and characterization of tumor regions in pathology images is labor intensive and subjective. Using a deep convolutional neural network (CNN), we developed an automated tumor region recognition system for lung cancer pathology images. From the identified tumor regions, we extracted 22 well-defined shape and boundary features and found that 15 of them were significantly associated with patient survival outcome in lung adenocarcinoma patients from the National Lung Screening Trial. A tumor region shape-based prognostic model was developed and validated in an independent patient cohort (n = 389). The predicted high-risk group had significantly worse survival than the low-risk group (p value = 0.0029). Predicted risk group serves as an independent prognostic factor (high-risk vs. low-risk, hazard ratio = 2.25, 95% CI 1.34–3.77, p value = 0.0022) after adjusting for age, gender, smoking status, and stage. This study provides new insights into the relationship between tumor shape and patient prognosis.

[1]  N. Otsu.  A threshold selection method from gray level histograms , 1979 .

[2]  W. Kannel,et al.  Risk factor analysis. , 1984, Progress in cardiovascular diseases.

[3]  T. Starzl,et al.  Clinicopathologic factors affecting patient survival and tumor recurrence after orthotopic liver transplantation for hepatocellular carcinoma. , 1991, Transplantation proceedings.

[4]  Luc Vincent,et al.  Morphological grayscale reconstruction in image analysis: applications and efficient algorithms , 1993, IEEE Trans. Image Process..

[5]  P. Grambsch,et al.  A Package for Survival Analysis in S , 1994 .

[6]  Rangaraj M. Rangayyan,et al.  Measures of acutance and shape for classification of breast tumors , 1997, IEEE Transactions on Medical Imaging.

[7]  P. Demant,et al.  Genetic analysis of three-dimensional shape of mouse lung tumors reveals eight lung tumor shape-determining (Ltsd) loci that are associated with tumor heterogeneity and symmetry. , 2003, Cancer research.

[8]  Farrokh Dehdashti,et al.  Improved prognostic value of 18F-FDG PET using a simple visual analysis of tumor characteristics in patients with cervical cancer. , 2003, Journal of nuclear medicine : official publication, Society of Nuclear Medicine.

[9]  Y. Nishiwaki,et al.  Histopathologic prognostic factors in resected colorectal lung metastases. , 2005, The Annals of thoracic surgery.

[10]  E. Feuer,et al.  SEER Cancer Statistics Review, 1975-2003 , 2006 .

[11]  E. Feuer,et al.  SEER Cancer Statistics Review, 1975-2003, National Cancer Institute. Bethesda, MD, , 2006 .

[12]  T. Yoshimine,et al.  Scoring radiologic characteristics to predict proliferative potential in meningiomas , 2006, Brain Tumor Pathology.

[13]  P. Mahadevan,et al.  An overview , 2007, Journal of Biosciences.

[14]  J. Crowley,et al.  The IASLC Lung Cancer Staging Project: Proposals for the Revision of the TNM Stage Groupings in the Forthcoming (Seventh) Edition of the TNM Classification of Malignant Tumours , 2007, Journal of thoracic oncology : official publication of the International Association for the Study of Lung Cancer.

[15]  Mikhail Teverovskiy,et al.  Multifeature Prostate Cancer Diagnosis and Gleason Grading of Histological Images , 2007, IEEE Transactions on Medical Imaging.

[16]  Q. Cui,et al.  Identification of high-quality cancer prognostic markers and metastasis network modules , 2010, Nature communications.

[17]  J. Klijanienko,et al.  Prognostic significance of tumor shape and stromal chronic inflammatory infiltration in squamous cell carcinomas of the oral tongue. , 2010, Journal of oral pathology & medicine : official publication of the International Association of Oral Pathologists and the American Academy of Oral Pathology.

[18]  Andrew H. Beck,et al.  Systematic Analysis of Breast Cancer Morphology Uncovers Stromal Features Associated with Survival , 2011, Science Translational Medicine.

[19]  Trevor Hastie,et al.  Regularization Paths for Cox's Proportional Hazards Model via Coordinate Descent. , 2011, Journal of statistical software.

[20]  Pu Chen,et al.  Effect of tumor shape and size on drug delivery to solid tumors , 2012, Journal of biological engineering.

[21]  F. Markowetz,et al.  Quantitative Image Analysis of Cellular Heterogeneity in Breast Tumors Complements Genomic Profiling , 2012, Science Translational Medicine.

[22]  Geoffrey E. Hinton,et al.  ImageNet classification with deep convolutional neural networks , 2012, Commun. ACM.

[23]  Thomas J Vogl,et al.  Factors influencing local tumor control in patients with neoplastic pulmonary nodules treated with microwave ablation: a risk-factor analysis. , 2013, AJR. American journal of roentgenology.

[24]  R Core Team,et al.  R: A language and environment for statistical computing. , 2014 .

[25]  K. Mimori,et al.  A single nucleotide polymorphism in fibronectin 1 determines tumor shape in colorectal cancer. , 2014, Oncology Report.

[26]  Hai Su,et al.  Novel image markers for non-small cell lung cancer classification and survival prediction , 2014, BMC Bioinformatics.

[27]  Kaamran Raahemifar,et al.  Effect of tumor shape, size, and tissue transport properties on drug delivery to solid tumors , 2014, Journal of biological engineering.

[28]  Robert J. Gillies,et al.  Test–Retest Reproducibility Analysis of Lung CT Image Features , 2014, Journal of Digital Imaging.

[29]  R. Machii,et al.  Morphological distribution of lung cancer from Cancer Incidence in Five Continents Vol. X. , 2015, Japanese journal of clinical oncology.

[30]  Jürgen Schmidhuber,et al.  Deep learning in neural networks: An overview , 2014, Neural Networks.

[31]  David R. Jones,et al.  Tumor Spread through Air Spaces is an Important Pattern of Invasion and Impacts the Frequency and Location of Recurrences after Limited Resection for Small Stage I Lung Adenocarcinomas , 2015, Journal of thoracic oncology : official publication of the International Association for the Study of Lung Cancer.

[32]  Robert J. Gillies,et al.  Quantitative Computed Tomographic Descriptors Associate Tumor Shape Complexity and Intratumor Heterogeneity with Prognosis in Lung Adenocarcinoma , 2015, PloS one.

[33]  E. Wang,et al.  Predictive genomics: a cancer hallmark network framework for predicting tumor clinical phenotypes using genome sequencing data. , 2014, Seminars in cancer biology.

[34]  Geoffrey E. Hinton,et al.  Deep Learning , 2015, Nature.

[35]  Navid Farahani,et al.  whole slide imaging in pathology: advantages, limitations, and emerging perspectives , 2015 .

[36]  J. Crowley,et al.  The IASLC Lung Cancer Staging Project: Proposals for Revision of the TNM Stage Groupings in the Forthcoming (Eighth) Edition of the TNM Classification for Lung Cancer , 2016, Journal of thoracic oncology : official publication of the International Association for the Study of Lung Cancer.

[37]  Dayong Wang,et al.  Deep Learning for Identifying Metastatic Breast Cancer , 2016, ArXiv.

[38]  Sergey Ioffe,et al.  Rethinking the Inception Architecture for Computer Vision , 2015, 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[39]  Jinfeng Zou,et al.  Identification and Construction of Combinatory Cancer Hallmark-Based Gene Signature Sets to Predict Recurrence and Chemotherapy Benefit in Stage II Colorectal Cancer. , 2016, JAMA oncology.

[40]  Ce Zhang,et al.  Predicting non-small cell lung cancer prognosis by fully automated microscopic pathology image features , 2016, Nature Communications.

[41]  Michel E. Vandenberghe,et al.  Relevance of deep learning to facilitate the diagnosis of HER2 status in breast cancer , 2017, Scientific Reports.

[42]  Junzhou Huang,et al.  Comprehensive Computational Pathological Image Analysis Predicts Lung Cancer Prognosis , 2017, Journal of thoracic oncology : official publication of the International Association for the Study of Lung Cancer.

[43]  Edwin Wang,et al.  Network Analysis Reveals A Signaling Regulatory Loop in the PIK3CA-mutated Breast Cancer Predicting Survival Outcome , 2017, Genom. Proteom. Bioinform..

[44]  Aleksey Boyko,et al.  Detecting Cancer Metastases on Gigapixel Pathology Images , 2017, ArXiv.

[45]  Hui Li,et al.  Deep learning in breast cancer risk assessment: evaluation of convolutional neural networks on a clinical dataset of full-field digital mammograms , 2017, Journal of medical imaging.

[46]  Anant Madabhushi,et al.  Accurate and reproducible invasive breast cancer detection in whole-slide images: A Deep Learning approach for quantifying tumor extent , 2017, Scientific Reports.

[47]  Sebastian Thrun,et al.  Dermatologist-level classification of skin cancer with deep neural networks , 2017, Nature.