Stereological characteristics of the equine accessory nerve.

Stereological techniques have been increasingly employed for assessment and characterization of neuromuscular diseases in humans and animals. As an adjunct to histopathology, morphometrical algorithms provide quantitative evidence of the peripheral nerve composition, thereby shedding light on its fibre characteristics and basic electrophysiological properties. In the horse, stereological investigations already have focussed on the recurrent laryngeal, deep peroneal and lateral palmar nerves (LPN). Of these, only the latter is suitable for taking biopsies in clinical settings, however, it does not contain any motor fibres and Ia-afferents. On account of its virtually mixed fibre qualities, most researchers today recommend the cervical branch of the equine accessory nerve (AN) for harvesting diagnostic samples. Thus, the present study was carried out to gain morphometrical proof of the AN composition and to obtain stereological base values in healthy individuals using state-of-the-art technology. All parameters were compared to the common peroneal nerve (CPN), known to harbour all myelinated fibre classes. As this second biopsy site is located farther distally to the neuro-axis, attention was paid to possible length-dependent features. Taken together, digital image analysis could be accurately applied on all AN samples. Stereology supported the histological and clinical evidence that the AN contains all myelinated fibre types. The huge range and scatter of fibre counts and density (3351-17,812/mm(2)) per fascicle were comparable to that measured in the equine common peroneal, deep peroneal, lateral palmar and recurrent laryngeal nerves. Similar to those, fibre diameter distribution was bimodal with slow Abeta- and Agamma-mechanoceptor afferents outnumbering large myelinated Aalpha-fibres by a factor of about 1.5. With a g-ratio at 0.55 +/- 0.001, the overall degree of myelination in the AN is highly consistent and insignificantly ranges between that of the equine common peroneal and LPNs. Apart from this subtle deviation, a statistically relevant difference between the more proximal AN and the distal CPN could not be documented. By obtaining morphometrical standard parameters and even more sophisticated distribution indices, stereology is a valuable tool for detection of subtle changes that are likely to escape from the investigators' eyes. The AN serves as a reliable source for advanced peripheral nerve research and should be accompanied by farther distal nerve probes for assessment of neuropathies that present with a proximodistal gradient.

Histological and ultrastructural evidence that recurrent laryngeal neuropathy is a bilateral mononeuropathy limited to recurrent laryngeal nerves.

REASONS FOR PERFORMING STUDY: Recurrent laryngeal neuropathy (RLN) is a common and debilitating peripheral nerve disease of horses, but it remains unclear if this disease is a mono- or polyneuropathy. An understanding of the distribution of the neuropathological lesions in RLN affected horses is fundamental to studying the aetiology of this very significant disease of tall horses. OBJECTIVE: To determine whether RLN should be classified as a mono- or polyneuropathy. METHODS: Multiple long peripheral nerves and their innervated muscles were examined systematically in 3 clinically affected RLN horses RESULTS: Severe lesions were evident in the left as well as right recurrent laryngeal nerves in all horses, both distally and, in one case, also proximally. No primary axonal lesions were evident in other nerves nor were changes found in their innervated muscles. CONCLUSIONS: RLN is not a polyneuropathy but should be classified as a bilateral mononeuropathy. POTENTIAL RELEVANCE: Genetic and local factors specifically affecting the recurrent laryngeal nerves in RLN-affected horses should now be investigated further.

Diagnosis and prognosis of neuroendocrine tumours of the lung by means of high resolution image analysis.

Neuroendocrine tumours (NET) of the lung are divided in subtypes with different malignant potential. The first is the benign or low-grade malignant tumours, well-differentiated, called typical carcinoids (TC) and the second is the high-grade malignant tumours, poorly differentiated of small (SCLC) or large cell type (LCLC). Between these tumour types lies the well-differentiated carcinoma with a lower grade of malignancy (WDNEC). In clinical routine it is very important with regard to prognosis to distinguish patients with low malignant potential from those with higher ones. In this study 32 cases of SCLC, 13 of WDNEC and 14 of TC with a follow-up time up to 7 years were collected. Sections 4 microm thick from paraffin embedded tissue were Feulgen stained. By means of high resolution image analysis 100 nuclei per case were randomly gathered to extract morphometric, densitometric and textural quantitative features. To investigate the ploidy status of the tumour the corrected DNA distribution was calculated. Stepwise linear discriminant analysis to differentiate the classes and Cox regression analysis for the survival time analysis were applied. Using chromatin textural and morphometric features in two two-class discriminations, 11 of the 14 TC cases and 8 of the 13 WDNEC cases were correctly classified and 11/13 WDNEC cases and 28/32 SCLC cases, respectively. The WDNEC cases are more similar in chromatin structure to TC than to SCLC. For the survival analysis, only chromatin features were selected to differentiate patients with better and worse prognosis independent of staging and tumour type.

Discrimination of esophageal dysplasia with progression and nonprogression. High-resolution image analysis for surrogate end point biomarkers.

OBJECTIVE: To detect textural nuclear features correlated with nonprogression and progression in esophageal dysplasia. STUDY DESIGN: Asymptomatic adults from Heshun Commune, Linxian County, China were examined with a balloon sampler in 1983 fifty cases of moderate esophageal dysplasia and 68 cases of mild were selected for study. By means of an Axiomat microscope equipped with a TV camera, 100 visually normal intermediate squamous cell nuclei per specimen were randomly measured from routinely Papanicolaou-stained slides. RESULTS: Of 50 esophageal moderate dysplasia cases, 24 and 7 progressed to carcinoma within three and nine years, respectively. The other 19 cases remained stable or regressed to normal and were used as the control group. By means of chromatin features, correct specimen classification rates of 79.2% (19/24), 73.7% (14/19), 85.5% (6/7) and 84.2% (16/19) were achieved, respectively (P < .001). Of 68 cases classified as mild dysplasia, 16, 13 and 12 progressed to carcinoma within three, five and nine years, respectively. The other 27 cases remained stable or regressed to normal and were used as the control group. The correct specimen classification rates were 93.8% (15/16), 88.9% (24/27), 69.2% (9/13), 74.1% (20/27), 83.3% (10/12) and 77.8% (21/27), respectively, using chromatin features of the nuclei (P < .001). CONCLUSION: In this study, nuclear chromatin features measured by high-resolution image analysis could sufficiently well forecast the outcome of precancerous lesions and discriminate precancerous lesions with progression and nonprogression. It also can be employed as surrogate end point biomarkers in clinical chemoprevention trials. Stoichiometric staining and standard preparations should increase the correct classification rates in further studies.

Automatic cell segmentation in cyto- and histometry using dominant contour feature points.

Automatic cell segmentation has various application potentials in cytometry and histometry. In this paper, an automatic cluster (touching) cell segmentation approach using the dominant contour feature points has been presented. Dominant feature points are the locations of indentation on the contour of the cluster. First, dominant feature points on the contour of the cluster are detected by distance profile. Next, using shape features of the cells, these feature points are selected for segmentation. We compared the results of the proposed method with manual segmentation and observed that the method has an overall accuracy about to 82%.

Modeling, definition and applications of histogram features based on DNA values weighed by sine functions.

OBJECTIVE: To model new DNA histogram features that weigh DNA values with values of curves of a sine function and to show the definition and applications of such features. STUDY DESIGN: A simple example of a sine feature can be modeled to yield the value zero if all cells are diploid or polyploid, with values of 2c, 4c or 8c, and to yield the value 100 if all cells are aneuploid, with DNA values of 3c, 6c or 12c-e.g., cells that are probably from a malignant lesion or indicate proliferation. All other values are multiplied by the corresponding sine value. We folded the logarithmic DNA histogram with a sine curve with positive values only. RESULTS: Correlation with ploidy balance was -0.94, demonstrating the similarity of both features. The sine features, however, avoid cutpoints between diploid and aneuploid values and are therefore less influenced by minor mistakes in standardization of DNA histograms. We introduced deviation factors as variants; that led to higher sine values for higher c values. For breast carcinoma (N = 306) the sine values were spread from very low to very high values, whereas esophageal carcinomas (N = 125) were centered at a sine value of 50. In breast carcinoma the sine features also correlated with prognostic factors, including hormone receptor status. CONCLUSION: Description of DNA histogram features by graphic demonstration of their weight functions improves understanding of features. Since functions respect the cyclic events in proliferation and are not influenced by polyploidization.

Canine neuroendocrine tumors of the pancreas: a study using image analysis techniques for the discrimination of metastatic versus nonmetastatic tumors.

Canine pancreatic neuroendocrine tumors were studied using different image analysis techniques (nuclear image histometry, analysis of argyrophilic proteins of nucleolar organizer regions, determination of the mouse anti-Ki 67 antigen proliferation index, and DNA densitometry) to correlate their biological behavior with objective phenotypic markers. The methods were compared to determine the best method for distinguishing between metastatic and nonmetastatic tumors. Discrimination between the two types of tumor was possible using nuclear image histometry in combination with morphometric analysis of argyrophilic proteins of nucleolar organizer regions. In contrast, the mouse anti-Ki 67 antigen proliferation index, DNA measurement, and immunohistochemical parameters revealed no significant difference between the two types of tumors.

Comparative FISH analysis of numerical chromosome 7 abnormalities in 5-micron and 15-micron paraffin-embedded tissue sections from prostatic carcinoma.

Interphase fluorescence in situ hybridization (FISH) was performed on 15-micron-thick paraffin sections from prostatic carcinomas using a chromosome 7-specific alpha-satellite DNA probe. A confocal laser scanning microscope (CLSM) was used for optical sectioning of the thick sections and reconstruction of 3D images. The number of FISH signals was determined by a gallery of optical sections evaluating only complete nuclei. To investiate the influence of section thickness and truncation and nuclei on scoring results, we compared the FISH data from 15-micron sections with signal counts obtained from 5-micron sections. The latter were evaluated by conventional fluorescence microscopy in the same tumor regions previously defined and marked on the slides. After statistical analysis of spot frequencies in tumor and non-tumorous cells (chi 2 test), we transferred the signal frequencies into a cytogenetic classification (-7, +7, polysomy 7). Based on this classification, most cases showed more than one chromosome 7 aberration type. Trisomy 7 (+7) became apparent in 15-micron thick sections in all 19 tumors, polysomy 7 (> 3 spots) in 18/19 cases, and monosomy 7 (-7) in 13/19 cases. In 5-micron sections, however, trisomy 7 and polysomy 7 were found in only 7/19 and 13/19 cases, respectively, and monosomy 7 in 7/19 cases. When comparing the classification results of tumor cells of the same tumor regions originating either from 5-micron or 15-micron sections, the following discrepancies were noted: in 15-micron sections exclusively, in 12/19 tumors, trisomy 7 was found; in 6/19 cases, polysomy 7; in 8/19 cases, monosomy 7. The high proportion of cases with tumor nuclei expressing only one hybridization signal of chromosome 7 in 15-micron sections could be confirmed as monosomy 7 in five selected cases by double-hybridization using centromere-specific probes for chromosomes 7 and 12. These results demonstrate that numerical chromosome 7 aberrations are more frequently observed in thick (15-micron) paraffin-embedded tissue sections by evaluating only complete nuclei. The use of routine sections (5-micron) for interphase cytogenetic analyses is compromised by a remarkable underestimation of the real chromosome copy numbers.

Groping for quantitative digital 3-D image analysis: an approach to quantitative fluorescence in situ hybridization in thick tissue sections of prostate carcinoma.

In molecular pathology numerical chromosome aberrations have been found to be decisive for the prognosis of malignancy in tumours. The existence of such aberrations can be detected by interphase fluorescence in situ hybridization (FISH). The gain or loss of certain base sequences in the desoxyribonucleic acid (DNA) can be estimated by counting the number of FISH signals per cell nucleus. The quantitative evaluation of such events is a necessary condition for a prospective use in diagnostic pathology. To avoid occlusions of signals, the cell nucleus has to be analyzed in three dimensions. Confocal laser scanning microscopy is the means to obtain series of optical thin sections from fluorescence stained or marked material to fulfill the conditions mentioned above. A graphical user interface (GUI) to a software package for display, inspection, count and (semi-)automatic analysis of 3-D images for pathologists is outlined including the underlying methods of 3-D image interaction and segmentation developed. The preparative methods are briefly described. Main emphasis is given to the methodical questions of computer-aided analysis of large 3-D image data sets for pathologists. Several automated analysis steps can be performed for segmentation and succeeding quantification. However tumour material is in contrast to isolated or cultured cells even for visual inspection, a difficult material. For the present a fully automated digital image analysis of 3-D data is not in sight. A semi-automatic segmentation method is thus presented here.

Relevance of chromatin features in the progression of esophageal epithelial severe dysplasia.

Since 1983, a long-term clinical trial of esophageal carcinoma chemoprevention has been conducted in a high-risk area in China. From this study, 25 esophageal severe dysplasia patients without therapy were selected for analysis. After 5-year follow-ups, 14 cases progressed to esophageal carcinoma, while the other 11 cases remained stable. Three Papanicolaou's smears were used for each case, including one from the esophageal cytological examination at the beginning, two from the re-examinations three and five years later respectively. About 100 visually normal intermediate cells were randomly collected per slide by high resolution image analysis. More than 100 features (morphologic, densitometric, textural) were extracted. The classifications were made by means of stepwise linear discriminate analysis at the single cell level as on the specimen level using up to ten features. In all three comparisons of patients with progression and with regression at time of diagnosis, three years after diagnosis and five years later, the correct cell classification rates were about 70%. The subsequent specimen classifications by means of the a posteriori probability (APOP) distribution of the cells in each case led to 80% correct classification. All selected features reflected the chromatin structure of nuclei. The result demonstrated that the chromatin structures of esophageal epithelial cells in severely dysplasic patients are different between cases with and without progression. These results suggest the possibility of the application of image analysis in the clinical trials to find the dysplasia patients with higher risk of progression, in order to reduce the number of patients for therapy.

Identification of a low-risk group of stage I breast cancer patients by cytometrically assessed DNA and nuclear texture parameters.

Image cytometrical measurements were performed on Feulgen-stained cells from 329 stage I breast cancers (pT1pN0,M0,R0). For each patient, several DNA (ploidy, S-phase fraction, exceeding rates, 2c deviation index, ploidy balance, entropy, and histogram typing), morphometric (area and radius of nuclei), and textural parameters (mainly co-occurrence and run-length) were calculated. The prognostic value of these parameters was investigated by multivariate Cox regression analysis, considering a distant recurrence-free survival of 8 years as the prognostic criterion. In the multivariate analysis, one DNA parameter (histogram type) and two textural parameters (co-occurrence and variation of the average heterochromatin area) were proven to have independent prognostic value. Using a linear combination of these variables, a prognostic factor was calculated for each individual patient. Patients were stratified using this factor into several groups according to their risk for distant recurrence. Thus, a low-risk group of stage I patients was identified, remaining distant recurrence-free for 8 years. In addition, a group of patients with a worse prognosis and an 8-year recurrence rate of about 26 per cent was identified, compared with the average distant recurrence rate of all stage I patients of 13 per cent. A combination of DNA and textural parameters can provide powerful prognostic information in stage I breast carcinomas and may allow a better selection of patients for different therapy protocols.

Invariance of textural features in image cytometry under variation of size and pixel magnitude.

Textural features of the granular structure of stained cell nuclei and nuclear sections derived from co-occurrence and run length matrices are often used for correlation with external (clinical) parameters. The representation of cell nuclei and nuclear sections vary considerably under changes of preparation and fixation conditions. Most obvious are changes in size e.g. by fixation as well as changes in the amount of bound stain. Computer-simulated variations in size and pixel magnitude of a set of images of cell nuclei stained with Feulgen were featured and compared. Resulting feature dependencies are used for the setting of parameters of feature extraction procedures as well as for the selection of invariant features for texture quantification of material with variations examined.

Improved prognostication in small (pT1) breast cancers by image cytometry.

Feulgen-stained samples from 460 small (pT1) primary breast cancers were investigated by means of an image analysis system. Several DNA, morphometrical and textural parameters were evaluated for each patient, and the prognostic meaning of these parameters was then investigated by the Cox regression analysis. As prognostic criterion a distant recurrence-free survival of five years was considered. All investigated DNA- and morphometrical parameters as well as several textural parameters showed a significant univariate correlation with the clinical course. In a multivariate approach the axillary nodal status was the most important prognostic parameter, followed by a morphometric parameter (anisokaryosis) and two textural parameters (runlength and co-occurrence). None of the DNA histogram derived parameters could add prognostic information in this multivariate approach. By the linear combination of the four selected variables, an individual prognostic factor was calculated. Using this factor the patients could be split into several groups according to their risk for distant metastases. Thus a low risk group of pT1 patients could be identified with a distant recurrence rate of only 2% after 5 years, and also a group of patients with a considerably worse prognosis and a 5-year distant recurrence rate of 53%. In contrast, using the nodal status as single parameter allows the identification of a low risk group of patients (pN0pT1) with a distant recurrence rate of 10.6%. Therefore, morphometrical and textural parameters can provide powerful prognostic information in small breast carcinomas and may allow a better selection of patients for adjuvant therapy.

Malignancy associated changes in squamous epithelium of the head and neck region.

Subtle cellular changes are known to exist in normal host tissue adjacent to tumours. These are called malignancy associated changes (MAC). To get more insight into the degree of expression and local spread of such changes we performed high resolution image cytometry on visually normal intermediate cell nuclei in smears from patients with laryngeal or pharyngeal squamous cancer. The smears were taken from the tumour surface, from a border region of the tumour and from a distant unsuspicious buccal site. In addition buccal smears from healthy control persons were examined. In a pilot study smears from 12 cancer patients and 11 control persons and in a succeeding validation study 63 controls, 18 non-tumour patients and 25 cancer patients were investigated. In both studies the occurrence of MACs was demonstrated quantitatively. In cancer patients normal appearing intermediate cells from the three different sampling sites could be discriminated with 65% in the pilot study and with 53% correct classification in the validation study. In addition the influences of smoking behaviour and sex were investigated in the control group. Only in the latter case there was a significant difference between female and male with a 63% correct cell and 71% correct specimen classification.

A knowledge acquisition tool in analytical pathology based on multi-media relational database.

The need for a knowledge based expert system for efficient decision making in the field of pathology has been well accepted. To build the knowledge-base for such an expert system is a painstaking task. This work is an attempt to provide the pathologists a powerful and user-friendly tool that will help them in the process of building the knowledge-base for medical diagnosis by closely looking at the specimen images and their extracted feature values. The tool is based on multi-media relational database and software like VAX Rally (4GL), ILIAD and SAS to provide a wide-range comparative study between feature data and image data; and also their statistical analysis.

Problems concerning the quality of DNA measurements on Feulgen-stained imprints. A study of five fixation techniques.

Feulgen-stained rat liver imprints were investigated, and hepatocytes, lymphocytes and granulocytes were measured. Additionally, chicken erythrocytes placed on the slides were measured as an external DNA standard. The imprints were treated according to five different fixation protocols. The measured integrated optical density (IOD) was normalized according to the leukocytes and afterwards scaled according to the diploid hepatocytes. The mean IOD, coefficient of variation (CV), standard error of the mean (SEM) and IOD ratios of distinct cell groups were calculated. The CV of the IOD for hepatocytes was slightly better for air-dried preparations. It was larger for leukocytes than for hepatocytes and worst for chicken erythrocytes. The SEM of hepatocytes did not show remarkable differences between the fixation groups; in general it was near 1%. In all cases the IOD ratios of 2c, 4c and 8c hepatocytes reasonably well followed the expected ratio of 1:2:4 except for wet, formalin-fixed hepatocytes (3.8). The ratios of leukocytes to 2c hepatocytes were about 1.0 for air-dried preparations but considerably lower (0.85) for wet fixation in formalin. The IOD ratios of chicken erythrocytes to 2C hepatocytes varied from 0.33 to 0.35. The deviation of the ratios of single specimens from the estimated mean of a fixation group are described.

Nucleolus organizer regions (AgNORs) in ductal mammary carcinoma. Comparison with classifications and prognosis.

The relevance of silver-stained NORs for classifications and prognosis was investigated in breast tissue. Paraffin sections from 137 cases of invasive ductal breast carcinomas and 12 cases with non-tumorous ductus epithelium as controls were stained according to a modified technique and analysed. From the cancer cases follow-up data up to 10 years (45 to 165 months) and in addition clinical, histological and several DNA distribution parameters were available. The nuclei and the silver grains were measured by means of a semiautomatic image analysis system. Significant differences in AgNOR features were found between controls and diploid tumors (p < or = 0.001), diploid and aneuploid tumors (p < or = 0.001), Bloom-Richardson-gradings I, II, and III (p < or = 0.001), and between the tumor cells from patients developing metastases within 5 years and those without (p < or = 0.002). The prognostic significance of AgNORs was estimated using Cox regression analysis. Four AgNOR features were correlated significantly with survival time. In a multivariate approach offering all parameters available an AgNOR parameter (CV of relative area AgNORs) ranked at the third position beyond the SD of DNA distribution and pTNM-staging. Considering the metastases-free interval of patients the same AgNOR feature showed an independent prognostic validity.

[Problems concerning standardization and quality assurance in DNA cytometry]. / Fragen der Standardisierung und Qualitätssicherung in der DNA-Zytometrie.

Rat liver imprints were treated with five different fixation techniques. Chicken erythrocytes stored over an extended period in the refrigerator were dropped on each slide before Feulgen staining. By means of an image analysis system chicken erythrocytes, rat leukocytes and hepatocytes were measured and the integrated optical density (IOD) was calculated for each nucleus. The coefficient of variation (CV) of IOD was about 15% for chicken erythrocytes. Leukocytes showed a CV of up to 10%. The CV was lower for hepatocytes than for leukocytes, and lowest for air-dried hepatocytes. The standard error of the mean (SEM) did not show remarkable differences between the fixation groups. For hepatocytes it was in general less than 1% of the respective mean value. The hepatocytes showed a linear staining except the wet formalin-fixed cells. This holds also for the wet formalin fixed leukocytes which showed only 85% of the IOD of 2c hepatocytes. The ratios of erythrocytes to 2c hepatocytes varied between 0.33 and 0.35. The IOD ratios of standard cells to 2c hepatocytes from single specimens showed remarkable differences from the above mentioned ratios of the pooled slides of a fixation group. The use of external standard cells was proved to be problematic, especially because of their variation in staining intensity independent of the staining intensity of hepatocytes.

Cytometry in histological sections of colon carcinoma.

A short description of a project of cytometry in histological sections of colon carcinoma is given with emphasis on the methodical aspects. Possible strategies of cytometric measurement and problems related to it (focus, overlap, segmentation of objects) are described. The main effort concerns interactive selection of tumor cells and the segmentation in cases of densely distributed and overlapping nuclei. All other succeeding processing steps are performed fully automatically. The resulting quantitative features are stored together with the original images on an optical disk for further examinations and reexaminations, allowing the direct relation of feature values to visual image content. The evaluation of the features as well as their interpretation is only at the beginning. Especially the problem of relating section information with true 3-dimensional information is not described here and necessitates further research. In a first investigation only a few tumors without and with metastases were analyzed. The preliminary results correspond with findings of Kunze et al.