Effect of Quantitative Nuclear Image Features on Recurrence of Ductal Carcinoma In Situ (DCIS) of the Breast

Abstract

BackgroundNuclear grade has been associated with breast DCIS recurrence and progression to invasive carcinoma; however, our previous study of a cohort of patients with breast DCIS did not find such an association with outcome. Fifty percent of patients had heterogeneous DCIS with more than one nuclear grade. The aim of the current study was to investigate the effect of quantitative nuclear features assessed with digital image analysis on ipsilateral DCIS recurrence.MethodsHematoxylin and eosin stained slides for a cohort of 80 patients with primary breast DCIS were reviewed and two fields with representative grade (or grades) were identified by a Pathologist and simultaneously used for acquisition of digital images for each field. Van Nuys worst nuclear grade was assigned, as was predominant grade, and heterogeneous grading when present. Patients were grouped by heterogeneity of their nuclear grade: Group A: nuclear grade 1 only, nuclear grades 1 and 2, or nuclear grade 2 only (32 patients), Group B: nuclear grades 1, 2 and 3, or nuclear grades 2 and 3 (31 patients), Group 3: nuclear grade 3 only (17 patients). Nuclear fine structure was assessed by software which captured thirty-nine nuclear feature values describing nuclear morphometry, densitometry, and texture. Step-wise forward Cox regressions were performed with previous clinical and pathologic factors, and the new image analysis features.ResultsDuplicate measurements were similar for 89.7% to 97.4% of assessed image features. The rate of correct classification of nuclear grading with digital image analysis features was similar in the two fields, and pooled assessment across both fields. In the pooled assessment, a discriminant function with one nuclear morphometric and one texture feature was significantly (p = 0.001) associated with nuclear grading, and provided correct jackknifed classification of a patient’s nuclear grade for Group A (78.1%), Group B (48.4%), and Group C (70.6%). The factors significantly associated with DCIS recurrence were those previously found, type of initial presentation (p = 0.03) and amount of parenchymal involvement (p = 0.05), along with the morphometry image feature of ellipticity (p = 0.04).ConclusionAnalysis of nuclear features measured by image cytometry may contribute to the classification and prognosis of breast DCIS patients with more than one nuclear grade.

Automated Colorectal Cancer Diagnosis for Whole-Slice Histopathology

“…A gland, which has lumina at the center and is surrounded by the stroma and nuclei, loses its regular structure with progressing malignancy [3,4]. In addition to glands, the number of nuclei and also the arrangement of the nuclei in the tissue can have diagnostic significance for some kind of malignancy in histopathology [5,6]. Numerous methods have been used to detect the nuclei in histopathology images and various automated detection results were compared against manual segmentation [7,8].…”

“…Besides, recent studies in histopathology image analysis were mostly focused on prostate and breast cancer [6,5] detection but very limited study was encountered on computational diagnosis of colorectal cancer [13,14]. In general, the computational histopathological image analysis is performed on patch level rather than slice-level due to their relatively large size [6,13,3,8]. Serter et.al [11] proposed a multi-scale analysis system to classify the nervous tissue images into stroma-rich and stroma-poor regions for neuroblastoma cancer detection.…”

Automatic Segmentation of Molecular Pathology Images Using a Robust Mixture Model with Markov Random Fields

“…Drastically changes in the density and structure of glands, nuclei, lumina, and stroma in tissue have diagnostic significance for some kind of malignancy in histopathology [32]. The usefulness of this automatic segmentation method can be further justified by comparing segmentation results with existing techniques over a wide variety of pathology images.…”

“…In applications where the attempt is to identify the nuclear structure from histopathology images for the diagnosis of cancer [15], [32], the proposed segmentation method can be readily applied to cluster digital histological images that are acquired using the Feulgen staining technique to specifically isolate nuclei for computational analysis [15]. Alternatively, our method can be adopted to firstly cluster the pixels into different tissue components and then process a second-step clustering of pixel groups that have the estimated parameters close to that of nuclei obtained from normal or tumour tissues.…”

Learning invariant features of tumor signatures

“…In some cases, tumor grading has been associated with recurrence, progression, and invasion carcinoma (e.g., breast DCIS), but such an association is highly dependent on tumor heterogeneity and mixed grading (e.g., presence of more than one grade), which offers significant challenges to the pathologists as mixed grading appears to be present in 50 percent of patients [8]. A recent study indicates that detailed segmentation and multivariate representation of nuclear features from H&E stained sections can predict DCIS recurrence [9] in patients with more than one nuclear grade. In this study, nuclei in the H&E stained samples were manually segmented and a multidimensional representation was computed for differential analysis between the cohorts.…”