End-to-End Non-Small-Cell Lung Cancer Prognostication Using Deep Learning Applied to Pretreatment Computed Tomography.

PURPOSE: Clinical TNM staging is a key prognostic factor for patients with lung cancer and is used to inform treatment and monitoring. Computed tomography (CT) plays a central role in defining the stage of disease. Deep learning applied to pretreatment CTs may offer additional, individualized prognostic information to facilitate more precise mortality risk prediction and stratification. METHODS: We developed a fully automated imaging-based prognostication technique (IPRO) using deep learning to predict 1-year, 2-year, and 5-year mortality from pretreatment CTs of patients with stage I-IV lung cancer. Using six publicly available data sets from The Cancer Imaging Archive, we performed a retrospective five-fold cross-validation using pretreatment CTs of 1,689 patients, of whom 1,110 were diagnosed with non-small-cell lung cancer and had available TNM staging information. We compared the association of IPRO and TNM staging with patients' survival status and assessed an Ensemble risk score that combines IPRO and TNM staging. Finally, we evaluated IPRO's ability to stratify patients within TNM stages using hazard ratios (HRs) and Kaplan-Meier curves. RESULTS: IPRO showed similar prognostic power (concordance index [C-index] 1-year: 0.72, 2-year: 0.70, 5-year: 0.68) compared with that of TNM staging (C-index 1-year: 0.71, 2-year: 0.71, 5-year: 0.70) in predicting 1-year, 2-year, and 5-year mortality. The Ensemble risk score yielded superior performance across all time points (C-index 1-year: 0.77, 2-year: 0.77, 5-year: 0.76). IPRO stratified patients within TNM stages, discriminating between highest- and lowest-risk quintiles in stages I (HR: 8.60), II (HR: 5.03), III (HR: 3.18), and IV (HR: 1.91). CONCLUSION: Deep learning applied to pretreatment CT combined with TNM staging enhances prognostication and risk stratification in patients with lung cancer.

SPIE-AAPM-NCI BreastPathQ challenge: an image analysis challenge for quantitative tumor cellularity assessment in breast cancer histology images following neoadjuvant treatment.

Purpose: The breast pathology quantitative biomarkers (BreastPathQ) challenge was a grand challenge organized jointly by the International Society for Optics and Photonics (SPIE), the American Association of Physicists in Medicine (AAPM), the U.S. National Cancer Institute (NCI), and the U.S. Food and Drug Administration (FDA). The task of the BreastPathQ challenge was computerized estimation of tumor cellularity (TC) in breast cancer histology images following neoadjuvant treatment. Approach: A total of 39 teams developed, validated, and tested their TC estimation algorithms during the challenge. The training, validation, and testing sets consisted of 2394, 185, and 1119 image patches originating from 63, 6, and 27 scanned pathology slides from 33, 4, and 18 patients, respectively. The summary performance metric used for comparing and ranking algorithms was the average prediction probability concordance (PK) using scores from two pathologists as the TC reference standard. Results: Test PK performance ranged from 0.497 to 0.941 across the 100 submitted algorithms. The submitted algorithms generally performed well in estimating TC, with high-performing algorithms obtaining comparable results to the average interrater PK of 0.927 from the two pathologists providing the reference TC scores. Conclusions: The SPIE-AAPM-NCI BreastPathQ challenge was a success, indicating that artificial intelligence/machine learning algorithms may be able to approach human performance for cellularity assessment and may have some utility in clinical practice for improving efficiency and reducing reader variability. The BreastPathQ challenge can be accessed on the Grand Challenge website.

Automated and Manual Quantification of Tumour Cellularity in Digital Slides for Tumour Burden Assessment.

The residual cancer burden index is an important quantitative measure used for assessing treatment response following neoadjuvant therapy for breast cancer. It has shown to be predictive of overall survival and is composed of two key metrics: qualitative assessment of lymph nodes and the percentage of invasive or in situ tumour cellularity (TC) in the tumour bed (TB). Currently, TC is assessed through eye-balling of routine histopathology slides estimating the proportion of tumour cells within the TB. With the advances in production of digitized slides and increasing availability of slide scanners in pathology laboratories, there is potential to measure TC using automated algorithms with greater precision and accuracy. We describe two methods for automated TC scoring: 1) a traditional approach to image analysis development whereby we mimic the pathologists' workflow, and 2) a recent development in artificial intelligence in which features are learned automatically in deep neural networks using image data alone. We show strong agreements between automated and manual analysis of digital slides. Agreements between our trained deep neural networks and experts in this study (0.82) approach the inter-rater agreements between pathologists (0.89). We also reveal properties that are captured when we apply deep neural network to whole slide images, and discuss the potential of using such visualisations to improve upon TC assessment in the future.

Comparing computer-generated and pathologist-generated tumour segmentations for immunohistochemical scoring of breast tissue microarrays.

BACKGROUND: Tissue microarrays (TMAs) have become a valuable resource for biomarker expression in translational research. Immunohistochemical (IHC) assessment of TMAs is the principal method for analysing large numbers of patient samples, but manual IHC assessment of TMAs remains a challenging and laborious task. With advances in image analysis, computer-generated analyses of TMAs have the potential to lessen the burden of expert pathologist review. METHODS: In current commercial software computerised oestrogen receptor (ER) scoring relies on tumour localisation in the form of hand-drawn annotations. In this study, tumour localisation for ER scoring was evaluated comparing computer-generated segmentation masks with those of two specialist breast pathologists. Automatically and manually obtained segmentation masks were used to obtain IHC scores for thirty-two ER-stained invasive breast cancer TMA samples using FDA-approved IHC scoring software. RESULTS: Although pixel-level comparisons showed lower agreement between automated and manual segmentation masks (κ=0.81) than between pathologists' masks (κ=0.91), this had little impact on computed IHC scores (Allred; =0.91, Quickscore; =0.92). CONCLUSIONS: The proposed automated system provides consistent measurements thus ensuring standardisation, and shows promise for increasing IHC analysis of nuclear staining in TMAs from large clinical trials.

Immunohistochemical analysis of breast tissue microarray images using contextual classifiers.

BACKGROUND: Tissue microarrays (TMAs) are an important tool in translational research for examining multiple cancers for molecular and protein markers. Automatic immunohistochemical (IHC) scoring of breast TMA images remains a challenging problem. METHODS: A two-stage approach that involves localization of regions of invasive and in-situ carcinoma followed by ordinal IHC scoring of nuclei in these regions is proposed. The localization stage classifies locations on a grid as tumor or non-tumor based on local image features. These classifications are then refined using an auto-context algorithm called spin-context. Spin-context uses a series of classifiers to integrate image feature information with spatial context information in the form of estimated class probabilities. This is achieved in a rotationally-invariant manner. The second stage estimates ordinal IHC scores in terms of the strength of staining and the proportion of nuclei stained. These estimates take the form of posterior probabilities, enabling images with uncertain scores to be referred for pathologist review. RESULTS: The method was validated against manual pathologist scoring on two nuclear markers, progesterone receptor (PR) and estrogen receptor (ER). Errors for PR data were consistently lower than those achieved with ER data. Scoring was in terms of estimated proportion of cells that were positively stained (scored on an ordinal scale of 0-6) and perceived strength of staining (scored on an ordinal scale of 0-3). Average absolute differences between predicted scores and pathologist-assigned scores were 0.74 for proportion of cells and 0.35 for strength of staining (PR). CONCLUSIONS: The use of context information via spin-context improved the precision and recall of tumor localization. The combination of the spin-context localization method with the automated scoring method resulted in reduced IHC scoring errors.