Neural network-based segmentation and classification system for automated grading of histologic sections of bladder carcinoma

OBJECTIVE: To develop an image analysis system for automated nuclear segmentation and classification of histologic bladder sections employing quantitative nuclear features. STUDY DESIGN: Ninety-two cases were classified into three classes by experienced pathologists according to the WHO grading system: 18 cases as grade 1, 45 as grade 2, and 29 as grade 3. Nuclear segmentation was performed by means of an artificial neural network (ANN)-based pixel classification algorithm, and each case was represented by 36 nuclei features. Automated grading of bladder tumor histologic sections was performed by an ANN classifier implemented in a two-stage hierarchic tree. RESULTS: On average, 95% of the nuclei were correctly detected. At the first stage of the hierarchic tree, classifier performance in discriminating between cases of grade 1 and 2 and cases of grade 3 was 89%. At the second stage, 79% of grade 1 cases were correctly distinguished from grade 2 cases. CONCLUSION: The proposed image analysis system provides the means to reduce subjectivity in grading bladder tumors and may contribute to more accurate diagnosis and prognosis since it relies on nuclear features, the value of which has been confirmed.     

A comparative study in morphometric grading of transitional cell carcinoma of the urinary bladder

To overcome the considerable observer inconsistency in the histologic grading of transitional cell carcinomas, the value of four different morphometric grading methods was investigated in 61 tumors of the bladder. Only two methods showed satisfactory reproducibility. Both methods, one based on random nuclear sampling and the other on selective nuclear sampling, showed an increase in the mean and standard deviation of the nuclear area with higher tumor grades (P less than .00001). Morphometric classification of the learning set (44 cases) was in agreement with the unequivocally assessed histologic grade in 35 cases (79.5%) using random sampling and in 38 cases (86.4%) using selective sampling. By reducing the grading classes to "low" (grades 1 and 2) and "high" (grade 3) and by introducing a classification probability threshold (0.80), an accurate morphometric classification was achieved in 38 cases (86.4%) using random sampling and in 41 cases (93.2%) using selective sampling. Of the 17 cases with histologic grading discrepancies, all 10 low-grade tumors (with discrepancies of grade 1 versus grade 2) were correctly classified as low-grade carcinomas by both of the morphometric methods; in the remaining 7 cases, with low-versus-high discrepancies (grade 2 versus grade 3), the selective method yielded better correlation with the tumor stage and clinical follow-up. It is concluded that morphometric classification is an acceptable alternative for histologic grading by pathologists, provided that the reproducibility of the method is confirmed. Although both random and selective sampling yielded satisfactory classifications, the selective method gave more reliable results as confirmed by the clinical behavior.     

Further testing of morphometric grading in T(A,1) urothelial carcinomas of the urinary bladder and the additional value of p53 immunoquantitation

OBJECTIVE: To analyze the value for grading of a previously developed quantitative morphometric/cytometric multivariate grading model (consisting of the mean nuclear area of the 10 largest nuclei (MNA-10, mitotic activity index = MAI and Ki-67 area% = Ki-67) in two new independent test sets of urothelial carcinomas (UCs) of the urinary bladder and to evaluate the additional value of p53 area% (p53) in this model. STUDY DESIGN: Ki-67 immunoquantitation, morphometric MAI and MNA-10 assessments using a previously described, strict protocol and matching of the resulting morphometric grade with subjective grade in two test sets of 154 T(A,1) UCs of the bladder (consensus grade between two independent observers). Further testing of this morphometric grading model was performed in 57 cases that lacked initial interobserver agreement on grade. Single and multivariate analysis of all features (including p53) was performed. RESULTS: With the previously developed morphometric/cytometric grading model, 93% (grade 1 vs. 2) and 91% (grade 2 vs. 3) of the consensus cases were correctly classified. These percentages were very similar to previous results, suggesting that the model is robust. Of the 57 cases that lacked initial interobserver agreement on grade, 53/57 (93%) were classified unambiguously as grade 1, 2 or 3 with the quantitative morphometric/ cytometric grading model. In the exploratory analysis, p53 was significant but with more overlap than the other features had. In multivariate analysis p53 did not improve the overall classification result of the original morphometric/cytometric model. CONCLUSION: The value of MNA-10, MAI and Ki-67 for grading in T(A,1) urothelial carcinomas of the urinary bladder was confirmed. p53 Did not improve overall grading classification of this combination.     

Grading of ductal breast carcinoma by cytomorphology and image morphometry with histologic correlation

OBJECTIVE: To investigate the relevance of image analysis for grading breast carcinoma. STUDY DESIGN: Twenty-five ductal breast carcinoma cases were chosen randomly from routine fine needle aspiration clinics. The results of cytomorphologic grading and image morphometry were correlated with those of histologic grading. The five image morphometric parameters studied were nuclear diameter, nuclear area, nuclear roundness, nuclear perimeter and grey level to compare with chromatin texture. RESULTS: Cytologic grading alone had a high correlation with histologic grading. The lowest correlation was found in grade 2 tumors. When cytologic grading was supplemented with image morphometric parameters, the correlation was higher than that of cytologic grading alone. CONCLUSION: Cytologic grading has a high correlation with histologic grading. The correlation improves further on supplementation with image morphometric parameters.     

A modern reproducible method for the histologic grading of astrocytomas with statistical classification tools

OBJECTIVE: To investigate whether statistical classification tools can infer the correct World Health Organization (WHO) grade from standardized histologic features in astrocytomas and how these tools compare with GRADO-IGL, an earlier computer-assisted method. STUDY DESIGN: A total of 794 human brain astrocytomas were studied between January 1976 and June 2005. The presence of 50 histologic features was rated in 4 categories from 0 (not present) to 3 (abundant) by visual inspection of the sections under a microscope. All tumors were also classified with the corresponding WHO grade between I and IV. We tested the prediction performance of several statistical classification tools (learning vector quantization [LVQ], supervised relevance neural gas [SRNG], support vector machines [SVM], and generalized regression neural network [GRNN]) for this data set. RESULTS: The WHO grade was predicted correctly from histologic features in close to 80% of the cases by 2 modern classifiers (SRNG and SVM), and GRADO-IGL was predicted correctly in > 84% of the cases by a GRNN. CONCLUSION: A standardized report, based the 50 histologic features, can be used in conjunction with modern classification tools as an objective and reproducible method for histologic grading of astrocytomas.     

Computer-assisted grading of adenocarcinoma in prostatic aspirates

Conventional cytologic grading of fine needle aspirates of prostatic adenocarcinoma has been shown neither to be reproducible nor to correlate well with histologic grading. This study developed a tumor grade classification based on computerized cytomorphometric features and compared the results to conventional grading of companion tissue sections. The image analysis system evaluated architectural features of the aspirates (mainly cell cluster features and interrelationships) as well as nuclear features. Thirty-five prostatic adenocarcinomas (8 well, 19 moderately and 8 poorly differentiated) were evaluated. Discriminant functions based on data collected at medium and high resolution distinguished between aspirates from low-grade (well-differentiated) and high-grade (poorly differentiated) adenocarcinomas with 81% accuracy. Moderately differentiated cancers could not be classified as a distinct group. This study suggests that accurate grading of prostatic adenocarcinoma in fine needle aspirate smears requires the evaluation of medium-resolution features related to specimen cellularity and uniformity or crowding of cell clusters as well as of high-resolution features of nuclear area, perimeter and coarseness of chromatin texture. These findings are compared to those of other schemes for the cytologic grading of prostatic aspirates.     

Tumour ploidy, morphometry, histological grading and clinical features in ovarian carcinoma: mutual relations

The relationship between tumour ploidy and qualitative and quantitative histopathology was assessed in a series of 95 ovarian carcinomas. 67% of the tumours were non-diploid (DNA aneuploid). 56% of the early stage (I-II) tumours were non-diploid and 81% of the tumours in advanced (III-IV) stages were aneuploid. Histological grading failed to show a clear relationship between increasing malignancy grade and ploidy. There was a close association between DNA ploidy and nuclear perimeter, area and shortest and longest nuclear diameter: the nuclei of non-diploid tumours were generally larger. Also the number of mitotic figures per square millimeter of epithelium in the microscope image (volume-corrected mitotic index, M/V-index) differed significantly between near-diploid and non-diploid tumours. Discriminant analysis showed that 74% of the learning-set tumours (67% of the test set tumours) could be correctly classified in low-ploidy and high-ploidy categories with morphometric features (nuclear perimeter, M/V-index and volume percentage of epithelium). Characteristic features of non-diploid ovarian tumours--rapid proliferation and large nuclear size--could be assessed with morphometric methods which allowed a relatively large aneuploid tumour group to be distinguished.     

A morphometric filter improves the diagnostic value of morphometric analyses of frozen histopathological sections from mammary tumours.

Frozen sections of 202 consecutive breast tumour cases were analyzed by morphometric quantitation of nuclear features. Nuclei were selected at random. Conventional light microscope examination of the paraffin-embedded specimens revealed 144 cases of cancer and 56 benign tumours. Using multivariate discriminant analysis of morphometric features, all but two of the benign cases and 79% of the malignant tumours were correctly classified. When a morphometrically based dynamic filter set to exclude 'non-diagnostic' nuclei was used, the correctly classified malignant cases rose to 86% Morphometry is a fast, reproducible and efficient method that can be used in conjunction with the histomorphological diagnosis of mammary frozen sections. The combination of systematic sampling and an objective dynamic filter may be a powerful approach to quantitative analyses of tumours from other sites. However, it is also likely that efficiency can be improved by combining nuclear morphometric features with structural, histochemical and molecular biological data. The combination of traditional histomorphological examination with quantitative information may well increase the diagnostic accuracy in individual patients.     

Distinguishing cortical adrenal gland adenomas from carcinomas by their quantitative nuclear features

OBJECTIVE: To explore data from a set of cases of adrenal cortical adenomas with different endocrine syndromes and carcinomas to determine whether quantitative image analysis of nuclear features might be used to separate the groups. STUDY DESIGN: Fifteen adrenal cortical tumors in which clinical information and optimally preserved, paraffin-embedded tissue were available were used. There were 10 adenomas and 5 carcinomas. Among the adenomas, five were associated with primary hyperaldosteronism (Conn's syndrome) and five with Cushing's syndrome. Five-micrometer-thick sections were stained with hematoxylin and eosin. In each case 50 nuclei were measured, and a number of morphometric and densitometric features were extracted. The data were subjected to multivariate analysis. RESULTS: Quantitative analysis showed that nuclei from adrenal carcinomas are larger than those from adenomas. Total optical density (OD) had a near-diploid distribution in the adenomas, while it was clearly aneuploid in the carcinomas. The pixel OD histograms were almost identical for all groups. Differences in nuclear chromatin texture were found between adenomas and carcinomas and also between the two adenoma categories. Multivariate analysis showed good discrimination between carcinomas and adenomas (Wilks lambda = .628, P < .0001) and between adenomas. However, based on Bayesian decision boundaries, 20-25% of carcinoma nuclei could be expected to be in the range of adenoma, and about 12% of Cushing's adenoma nuclei and 15% of Conn's adenoma nuclei would be classified as carcinoma. CONCLUSION: Computer-assisted analysis of nuclear characteristics proved useful in identifying and describing differences between groups of tumors arising in the adrenal cortex.     

Morphometry of bladder carcinoma: morphometry and grading complement each other

Subjective grading of bladder carcinoma is a good predictor of the clinical outcome in those patients whose tumours are grade 1 or grade 3. However, in grade 2 tumours, which account for 45% of cases, grading has little predictive value in an individual patient. We have complemented the use of subjective grading with measurement of nuclear area and used a calculation of the distribution of nuclear sizes as a predictor of the clinical course. When subjective grading was complemented by morphometry the outcome was correctly predicted in 55 of 58 cases and all cases with poor clinical outcome were identified.     

Morphometry of nucleoli as an indicator for grade of malignancy of bladder tumors

To establish a new indicator for the classification of human urinary bladder cancers, the nucleoli of normal epithelial and neoplastic cells were analyzed, using morphometric techniques. By electron microscopy, the nucleolar profiles of cells from grade 2 and 3 transitional cell carcinomas were often small and irregular. Morphometry showed that the nucleolar volumes, nucleolar/nuclear volume ratios, volume densities of various nucleolar components, and the numbers of fibrillar centers (FCs) altered significantly with an increase in tumor grade. In particular, an increase in FC numbers in the nuclei of higher grade tumors was associated with a decrease in individual volume. The number of FCs in intact urothelial cells obtained from patients with bladder tumors is significantly larger than in the normal urothelial cells. This may be related to the multicentric origin of bladder cancers. These results suggest that morphometric analysis of nucleoli is useful in evaluating the degree of differentiation and invasive capacity of human bladder tumor cells. In particular, the number and individual volume of FCs may be an indicator of tumor malignancy.     

Cytological grading of breast cancer in Giemsa-stained fine needle aspiration smears

A potential cytological nuclear grading based on a semi-quantitative evaluation of three basic nuclear features, size of cell nuclei, anisonucleosis and the proportion of nucleoli-containing-nuclei, was tested on 74 Giemsa-stained fine needle aspiration of breast smears for its reliability in establishing the malignant potential of breast cancer. The prognostic impact of DNA-ploidy and S-phase fraction was also assessed. A good correlation between the three basic nuclear features, DNA-ploidy, S-phase fraction, cytological nuclear grade and histological grade, was shown. Using the cytological nuclear grade proposed, correct classification of cases between low histological grade (HG I) and high histological grade (HG II + HG III) was achieved in 79.73%. A statistically significant difference in 5-year survival rate was also observed between low malignancy grade and high malignancy grade breast cancer patients, regardless of the grading method used. DNA-ploidy and S-phase fraction were not statistically significant in establishing the malignant potential of breast cancer.     

DNA assessment in thin sections of lymphomas by densitometric and stereological methods: comparison with imprint and flow cytometric results

The nuclear DNA content was estimated in 2 microns sections of 18 lymphoma cases by two methods: (1) Feulgen densitometry using QTM 900 with correction by Bins' procedure which allows size-independent DNA distributions; (2) stereological unfolding as proposed by Cruz-Orive giving sphere-size distributions. A general correlation was found between results and DNA measurements obtained by imprint and flow cytometric techniques in the same specimens. When histologic DNA profiles were compared to cytologic histograms, a high correlation was found between the distribution of ploidy classes by correspondence analysis. However two highly proliferating lymphomas were erroneously classified as aneuploid. Conversely, sphere-size distributions allowed the identification of the majority of aneuploid lymphomas but failed to recognize proliferating ones. It appears that when cytologic specimens are not available, densitometric studies on sections may provide valuable information on DNA content, with complementary data obtained from stereological procedures.     

DNA ploidy and nuclear morphometry for the classification of dysplastic nevi

OBJECTIVE: To examine the diagnostic value of DNA ploidy and nuclear morphometric features in sporadic dysplastic nevi as compared to those in compound nevi and melanoma. STUDY DESIGN: DNA ploidy profiles plus seven direct and three derived nuclear features were obtained in a series of 120 melanocytic skin neoplasms (30 dysplastic nevi [DN], 30 melanomas [MM], 60 compound nevi [CN]) and the results compared. RESULTS: DNA ploidy separated melanomas from benign melanocytic skin neoplasms with 96.5% accuracy in classifying the grouped cases. The derived nuclear shape factor Form PE and nuclear axis ratio were the most successful discriminants separating DN from MM but allowed only 73.3% correct classification of cases. Separation of DN from CN was best achieved using Form PE and mean nuclear area (74.4% correctly classified). Results from compound nevi in subjects < 25 years of age fell between those for DN and MM. CONCLUSION: Quantitative nuclear cytologic characteristics in sporadic dysplastic nevi span a range seen in common nevi through to those in thin melanomas. Cytologic changes in sporadic dysplastic nevi overlap those seen in other melanocytic skin neoplasms. Therefore, other reproducible morphometric features need to be assessed in order to further refine the histopathologic diagnosis of this entity.     

Application of artificial neural network for classification of thyroid follicular tumors

OBJECTIVE: To analyze smears of 197 thyroid follicular tumors (adenoma and carcinoma). STUDY DESIGN: Several types of artificial neural networks (ANN) of various designs were used for diagnosis of thyroid follicular tumors. The typical complex of cytologic features, some nuclear morphometric parameters (area, perimeter, shape factor) and density features of chromatin texture (mean value and SD of gray levels) were defined for each tumor. RESULTS: The ANN was trained by means of cytologic features characteristic for a thyroid follicular adenoma and a follicular carcinoma. At subsequent testing, the correct cytologic diagnosis was established in 93% (25 of 27) of cases. The morphometry increased the accuracy of diagnosis for follicular tumors in up to 97% (75 of 78) of cases. ANN correctly distinguished an adenoma or a carcinoma in 87% (73 of 84) of cases when using color microscopic images of tumors. CONCLUSION: The usage of ANN has raised sensitivity of cytologic diagnosis of follicular tumors to 90%, compared with a usual cytologic method (sensitivity of 56%). The automatic classification of thyroid follicular tumors by means of ANN is prospective.     

The significance of DNA measurements in a histologically defined subset of infiltrating ductal carcinomas of the breast with long-term follow-up

The nuclear DNA content and other karyometric parameters were evaluated in a histologically homogeneous group of invasive ductal carcinomas of the breast from 13 patients who survived 25 years after radical mastectomy and from 13 controls matched for histologic tumor grade, lymph node status, tumor size and patient age. The nuclear DNA content and other morphometric features were evaluated by image analysis (using a modified TICAS system) on 12-microns-thick, Feulgen-stained sections. The DNA content of the tumors of both the long-term survivors and the controls varied from the diploid range to highly aneuploid (with a large proportion of the cells having a DNA content above 5N). Overall, the tumors of the controls exhibited a higher ploidy, a greater deviation from the diploid range and a greater variation of nuclear size than did the tumors of the long-term survivors. These results suggest that these measurements may be helpful in yielding prognostic information among sets of histologically identical breast tumors of similar pathologic stage.     

High-resolution and contextual analysis for the diagnosis of fine needle aspirates of breast.

A study was undertaken to confirm earlier work on a smaller number of patients that had suggested that medium-resolution contextual analysis complements high-resolution individual cell analysis for cytomorphometric classification of fine needle aspirate smears of breast. The objectives of this study were to improve and verify the method. Sixty-one biopsy-confirmed hematoxylin and eosin-stained aspirate smears of breast were restained using the Feulgen technique. Individual nuclei were digitized at a resolution of 0.25 micron. Features describing size, shape, density and texture were extracted from the images. Individual cell analysis correctly classified 84% of cases, contextual analysis correctly classified 70% of cases, and the combined use of both techniques resulted in 87% classification accuracy. However, if fibroadenoma cases are excluded, the combined correct classification rate is 93%. Geometric and densitometric features contributed most to correct classification in individual cell analysis, while the most important contextual feature was the number of clusters per scene. We conclude that the addition of quantitative measures of smear patterns, termed "contextual analysis," improves automated classification schemes.     

Prediction of radiosensitivity in human bladder cell lines using nuclear chromatin phenotype.

BACKGROUND: Nuclear texture analysis measures phenotypic changes in chromatin distribution within a cell nucleus, while the alkaline Comet assay is a sensitive method for measuring the extent of DNA breakage in individual cells. The authors aim to use both methods to provide information about the sensitivity of cells to ionizing radiation. METHODS: The alkaline Comet assay was performed on six human bladder carcinoma cell lines and one human urothelial cell line exposed to gamma-radiation doses from 0 to 10 Gy. Nuclear chromatin texture analysis of 40 features was then performed in the same cell lines exposed to 0, 2, and 6 Gy to explore if nuclear phenotype was related to radiation sensitivity. RESULTS: Comet assay results demonstrated that the cell lines exhibited different levels of radiosensitivity and could be divided into a radiosensitive and a radioresistant group at >6 Gy. Using stepwise discriminant analysis, a subset of important nuclear texture features that best discriminated between sensitive and resistant cell lines were identified A classification function, defined using these features, correctly classified 81.75% of all cells into their radiosensitive or radioresistant groups based on their pretreatment chromatin phenotype. Posttreatment chromatin changes also varied between cell lines, with sensitive cell lines showing a relaxed chromatin conformation following radiation, whereas resistant cell lines exhibited chromatin condensation. CONCLUSIONS: The authors conclude that the alkaline Comet assay and nuclear texture methodologies may prove to be valuable aids in predicting the response of tumor cells to radiotherapy.     

Cytologic grading and DNA image cytometry of breast carcinoma on fine needle aspiration cytology smears

OBJECTIVE: To correlate the cytologic grade of breast carcinoma with DNA image cytometry (ICM) and nuclear area on fine needle aspiration cytology (FNAC) smears. STUDY DESIGN: In this prospective study, FNAC material from 28 breast carcinomas were studied for cytologic grade and DNA ICM. Breast carcinomas were classified as grade 1-3 (low to high). DNA histograms were classified by the modified Auer method. Degree of hyperploidy (DH), ploidy balance (PB) and nuclear area (NA) were measured on Feulgen-stained smears by a CAS 200 image cytometer. Cytologic grade was correlated with DNA ICM findings and NA. RESULTS: There were 3 cytologic grade 1, 13 grade 2 and 12 grade 3 breast carcinomas. Seven of eight cases of hypertetraploid aneuploidy were grade 3 tumors. All cytologic grade 1 tumors were diploid. There were significant differences in DH, PB and NA in different grades of breast carcinoma (one-way ANOVA). CONCLUSION: DNA image cytometry in combination with cytologic grading might offer additional information for the characterization of breast carcinomas diagnosed by FNAC. These observations are of particular interest with the introduction of preoperative chemotherapy.     

Classifying breast carcinomas with DNA measurements and morphometry

The relation between the DNA distribution pattern and the variables DNA content and nuclear area was investigated in 73 breast carcinomas. According to the DNA patterns, the tumors were subclassified into groups I, II, III, and IV (Auer's classification). The Auer's classes correlate with prognosis. In this investigation it was shown that Auer's classes strongly correlate with the standard deviation of DNA content and mean nuclear area. The carcinomas were subclassified using data of DNA content and nuclear area. For the establishment of this subclassification, multivariate analysis was applied. It was evident that Auer's group III was a hybrid group. Therefore a second classification rule was calculated without the data of Auer's group III. This second classification rule was even more successful in determining the Auer's group of each individual case. Our investigation shows that it is possible to subclassify breast carcinomas objectively using DNA content and nuclear area of the tumor cells.