Martin Oberholzer

Islet Pathology and the Pathogenesis of Type 1 and Type 2 Diabetes mellitus Revisited

The present review draws attention to the diversity of islet lesions seen in human type 1 and type 2 diabetes. This heterogeneity of islet changes is best demonstrated by immunocytochemistry. In type 1 diabetes the endocrine pancreas is characterized by selective loss of B cells, which most likely results from a slowly acting autoimmune process depending on the presence of both genetic and environmental factors. The process starts years before overt diabetes develops and manifests when the B-cell volume is reduced by about 80%. In type 2 diabetes B cells are always present, regardless of the duration and severity of the disease, but lack any signs of functional activity. This reflects a secretory defect of the B cells which obviously becomes evident under the conditions of obesity, hyperinsulinism and insulin resistance. Obese but non-diabetic subjects show, in parallel to their hyperinsulinism, an increased B cell volume, suggesting that under prediabetic conditions the B cells have still the capacity to respond to increased functional demands by enhanced proliferation. In manifest diabetes the B cells have lost their proliferative potential. Whether this is due to an inherent defect or the consequence of a functional disturbance, is not clear. The development of islet amyloidosis most likely represents an associated functional abnormality of the B cell.

Methods in quantitative image analysis

The main steps of image analysis are image capturing, image storage (compression), correcting imaging defects (e.g. non-uniform illumination, electronic noise, glare effect), image enhancement, segmentation of objects in the image and image measurements. Digitisation is made by a camera. The most modern types include a frame-grabber, converting the analog-to-digital signal into digital (numerical) information. The numerical information consists of the grey values describing the brightness of every point within the image, named a pixel. The information is stored in bits. Eight bits are summarised in one byte. Therefore, grey values can have a value between 0 and 256 (28). The human eye seems to be quite content with a display of 5-bit images (corresponding to 64 different grey values). In a digitised image, the pixel grey values can vary within regions that are uniform in the original scene: the image is noisy. The noise is mainly manifested in the background of the image. For an optimal discrimination between different objects or features in an image, uniformity of illumination in the whole image is required. These defects can be minimised by shading correction [subtraction of a background (white) image from the original image, pixel per pixel, or division of the original image by the background image]. The brightness of an image represented by its grey values can be analysed for every single pixel or for a group of pixels. The most frequently used pixelbased image descriptors are optical density, integrated optical density, the histogram of the grey values, mean grey value and entropy. The distribution of the grey values existing within an image is one of the most important characteristics of the image. However, the histogram gives no information about the texture of the image. The simplest way to improve the contrast of an image is to expand the brightness scale by spreading the histogram out to the full available range. Rules for transforming the grey value histogram of an existing image (input image) into a new grey value histogram (output image) are most quickly handled by a look-up table (LUT). The histogram of an image can be influenced by gain, offset and gamma of the camera. Gain defines the voltage range, offset defines the reference voltage and gamma the slope of the regression line between the light intensity and the voltage of the camera. A very important descriptor of neighbourhood relations in an image is the co-occurrence matrix. The distance between the pixels (original pixel and its neighbouring pixel) can influence the various parameters calculated from the co-occurrence matrix. The main goals of image enhancement are climination of surface roughness in an image (smoothing), correction of defects (e.g. noise), extraction of edges, identification of points, strengthening texture elements and improving contrast. In enhancement, two types of operations can be distinguished: pixel-based (point operations) and neighbourhood-based (matrix operations). The most important pixel-based operations are linear stretching of grey values, application of pre-stored LUTs and histogram equalisation. The neighbourhood-based operations work with so-called filters. These are organising elements with an original or initial point in their centre. Filters can be used to accentuate or to suppress specific structures within the image. Filters can work either in the spatial or in the frequency domain. The method used for analysing alterations of grey value intensities in the frequency domain is the Hartley transform. Filter operations in the spatial domain can be based on averaging or ranking the grey values occurring in the organising element. The most important filters, which are usually applied, are the Gaussian filter and the Laplace filter (both averaging filters), and the median filter, the top hat filter and the range operator (all ranking filters). The principal advantage of ranking filters over averaging operators is that they do not reduce the brightness difference across steps. The edges remain in place and well-defined. Very important prerequisites for extracting quantitative information from digitised images are clearly identifiable segmented objects and knowledge about instrumental and technical influences on the results (glare effect and thickness of histological slides). Segmentation of objects is traditionally based on threshold grey values. The grey value histogram of the original or enhanced image is an important tool for setting threshold levels. For determining the threshold grey value, bidimensional histograms can be applied. Quantitative information can be extracted from images by mathematical operations on binary images or on grey scale images. Boolean operations on binary images are applied when one desires to combine the information contained in several binary images. The morphological operations on binary images mainly include erosion and dilation, and modifications of these operations. There are many methods allowing direct quantitation of segmented objects within grey scale images. They use different sets of parameters: planimetric, histogram-derived, densitometric, co-occurrence matrix-derived parameters, invariant moments, run lengths, parameters of quantitative immunohistochemistry and immunocytochemistry, parameters of silver-stained nucleolar organiser regions (AgNORs) and parameters of cellular sociology. Digital image analysis requires a distinction between two phases for the evaluation procedure: generation of fundamental data (x- andy-coordinates and grey values of the pixels, immediately after object segmentation) and calculation of parameters from these data. The data generated during segementation must remain always available and these data must be conceptually separated from the parameters deduced from them. With such a data organisation it is no longer necessary to repeat the object segmentation if new algorithms should be applied on objects which earlier were segmented. In dealing with methods or instruments for digital image analysis, it is always essential to know precisely the characteristics of both of them.

Morphometric evidence for a striking B-cell reduction at the clinical onset of type 1 diabetes

The distribution and volume of the pancreatic endocrine cells were studied in a case of type 1 diabetes with a duration of approximately 7 days. Immunocytochemical techniques combined with morphometry were used. The PP-cell rich lobe, making up about 10% of the total pancreatic volume, was not included in this study. The volume density and the absolute volume of the B-cells was found to be reduced to about one third to one seventh of the values determined in four controls of similar age and/or pancreatic volume. The A-cell volume was also diminished whereas the D- and PP-cell volume remained constant. As B-cell necroses could not be detected and insulitis was in the initial stages of development it is concluded that the destruction of B-cells proceeds slowly in type 1 diabetes. In the majority of cases it probably starts years before the clinical onset of the disease.

Diagnostic tools for differentiating between pleural mesothelioma and lung adenocarcinoma in paraffin embedded tissue. Part I: immunohistochemical findings

Specimens of 27 histologically definite mesotheliomas and 34 proven adenocarcinomas were examined with a panel of 14 antibodies: pan-epithelial antibody Lu-5, anti-keratin-18, anti-keratin-7, Ber-EP4, anti-Leu-M1, HEA-125, anti-carcino-embryonic antigen (CEA), anti-blood group-related antigens (anti-BGR A, B, H), B 72.3, anti-placental alkaline phosphatase (PLAP), anti-vimentin and BMA-120 used to determine their value in the differentiation between pleural mesothelioma and lung adenocarcinoma. Lu-5, anti-cytokeratin-7 and -18, B 72.3 and PLAP reacted in a high percentage of cases with both mesothelioma and adenocarcinoma. Anti-CEA and anti-Leu-Ml did not react with any of the 27 mesotheliomas tested but showed a reaction in 75% (anti-CEA) and 66% (anti-Leu-M1) of the lung adenocarcinomas. Seventeen percent of the adenocarcinomas and 96% of the mesotheliomas showed a positive reaction with anti-vimentin. Ber-EP4 was demonstrated in all lung adenocarcinomas, but only in 2 mesotheliomas in a focal manner (7%). HEA-125 and anti-BGR A, B, H reacted with 83% (HEA-125) and 75% (anti-BGR A, B, H) of the lung adenocarcinomas. The statistical parameters, sensitivity and efficiency were estimated and a normogram for judging the diagnostic power of a single antibody for the differential diagnosis of mesothelioma versus adenocarcinoma was developed. According to this, Ber-EP4, HEA-125, anti-BGR A, B, H and anti-CEA were, in descending order, the most powerful discriminatory antibodies.

Stereology of liver biopsies from healthy volunteers.

The stereologioal model and the base-line data of normal human liver needle biopsy-specimens are presented. Four reference systems were introduced: 1 cm3 of liver tissue, 1 cm3 of hepatoeyte, 1 cm3 of hepatocytic cytoplasm and the volume of an average “mononuclear” hepatocyte. The sampling was done at three levels of magnification (1,000 ×, 5,000 × and 10,000 ×). A lobular differentiation was not considered. The baseline data show strikingly small variations (s.e. less than 10%) within the individual biopsy specimen and within the group of four biopsies. There is no principal difference between human beings presented here, rats, mice and dogs. Only the mean individual volume of human hepatocytes is clearly larger than in rodents. The problems and limitations of stereological work on liver biopsy specimens are discussed.

Telepathology on the Solomon Islands — two years' experience with a hybrid Web- and email-based telepathology system

The National Referral Hospital in Honiara, Solomon Islands, has used an Internet-based system in Switzerland for telepathology consultations since September 2001. Due to the limited bandwidth of Internet connections on the Solomon Islands, an email interface was developed that allows users in Honiara to submit cases and receive reports by email. At the other end, consultants can use a more sophisticated Web-based interface that allows discussion of cases among an expert panel. The result is a hybrid email- and Web-based telepathology system. Over two years, 333 consultations were performed, in which 94% of cases could be diagnosed by a remote pathologist. A computer-assisted ‘virtual institute’ of pathologists was established. This form of organization helped to reduce the median time from submission of the request to a report from 28 h to 8.5 h for a preliminary diagnosis and 13h for a final report. A final report was possible in 77% of all submitted cases.

Diagnostic telepathology: long-term experience of a single institution

ObjectivesThe paper reviews the development of the application of telepathology in a department of surgical pathology between 1991 and 2003. The goal of the efforts during this time was to give up the concept of programming a single application, available only between two fixed workstations with sophisticated devices and special software, and to find the virtual “largest common denominator” for implementing as many different applications as possible with the same basic system.MethodsA new telepathology system was designed as a client–server system with a relational database at its centre. The clients interact together by transferring the questions (texts and images) to a record (case) in the database on the server and by transferring the answers to the same record on the database.ResultsThe new “open” telepathology system iPath (http://telepath.patho.unibas.ch) has been very well accepted by many groups around the world. The main application fields are: consultations between pathologists and medical institutions without a pathologist (e.g. for frozen section diagnoses or for surgical diagnoses in hospitals in South Asia or Africa), tumour boards, field studies and distance education (http://teleteach.patho.unibas.ch).ConclusionsHaving observed that with iPath we have succeeded in satisfying all our telepathology needs, we are inclined to put the emphasis on the nature of the tasks being performed, as opposed to the methods or technical means for performing a given task. The three organisation models proposed by Weinstein et al. (2001) [24] can be reduced to only two models: the model of discussion groups and the model of expert groups (virtual institutes).

Digital Imagery/Telecytology

ISSUES Optical digital imaging and its related technologies have applications in cytopathology that encompass training and education, image analysis, diagnosis, report documentation and archiving, and telecommunications. Telecytology involves the use of telecommunications to transmit cytology images for the purposes of diagnosis, consultation or education. This working paper provides a mainly informational overview of optical digital imaging and summarizes current technologic resources and applications and some of the ethical and legal implications of the use of these new technologies in cytopathology. CONSENSUS POSITION Computer hardware standards for optical digital imagery will continue to be driven mainly by commercial interests and nonmedical imperatives, but professional organizations can play a valuable role in developing recommendations or standards for digital image sampling, documentation, archiving, authenticity safeguards and teleconsultation protocols; in addressing patient confidentiality and ethical, legal and informed consent issues; and in providing support for quality assurance and standardization of digital image-based testing. There is some evidence that high levels of accuracy for telepathology diagnosis can be achieved using existing dynamic systems, which may also be applicable to telecytology consultation. Static systems for both telepathology and telecytology, which have the advantage of considerably lower cost, appear to have lower levels of accuracy. Laboratories that maintain digital image databases should adopt practices and protocols that ensure patient confidentiality. Individuals participating in telecommunication of digital images for diagnosis should be properly qualified, meet licensing requirements and use procedures that protect patient confidentiality. Such individuals should be cognizant of the limitations of the technology and employ quality assurance practices that ensure the validity and accuracy of each consultation. Even in an informal teleconsultation setting one should define the extent of participation and be mindful of potential malpractice liability. ONGOING ISSUES Digital imagery applications will continue to present new opportunities and challenges. Position papers such as this are directed toward assisting the profession to stay informed and in control of these applications in the laboratory. Telecytology is an area in particular need of studies of good quality to provide data on factors affecting accuracy. New technologic approaches to addressing the issue of selective sampling in static image consultation are needed. The use of artificial intelligence software as an adjunct to enhance the accuracy and reproducibility of cytologic diagnosis of digital images in routine and consultation settings deserves to be pursued. Other telecytology-related issues that require clarification and the adoption of workable guidelines include interstate licensure and protocols to define malpractice liability.

The iPath telemedicine platform.

An early, point-to-point telepathology system at the University of Basel developed into an open-source, Internet-based platform for telemedicine in 2001. The Internet Pathology Suite (iPath) is a Web-based telepathology platform that permits the online presentation and discussion of cases within user groups. It also allows realtime telemicroscopy across firewalls. After four years, the telepathology network has over 700 active users. More than 6300 cases with a total of about 39,000 images have been diagnosed. The diagnostic workload of all these cases is not exclusively handled by the Department of Pathology in Basel, but by a growing number of independent groups who also use the server simply as a case repository. What started as a small project for hospitals in Switzerland has become a global network.

Karyomegalic interstitial nephritis: Further support for a distinct entity and evidence for a genetic defect

Karyomegalic interstitial nephritis was first described in 1979 by Mihatsch, who was reporting three such cases. We report here four additional cases as well as two family investigations. Our findings support the association of karyomegaly and interstitial nephritis as a distinct entity. Typical clinical features are asymptomatic progressive renal failure in the third decade of life and recurrent infections, mostly of the upper respiratory tract. Histologic alterations consist of markedly enlarged and hyperchromic nuclei in many tubular epithelial cells throughout the nephron accompanied by interstitial fibrosis in the surrounding atrophic tubules. Karyomegaly is not limited to the kidneys. In one case, autopsy revealed karyomegaly in epithelial and mesenchymal cells of many other organs. However, no association of karyomegaly with further histologic damage is evident except in the kidneys. Because of the familial clustering, karyomegalic interstitial nephritis seems to be an inherited disease. Examination of the nuclear proliferation-associated structures proliferating cell nuclear antigen/cyclin, Ki 67, and p53 suggests an inhibition of mitosis in karyomegalic cells. The finding of the same HLA haplotype, A9/B35, in four of six HLA-typed cases suggests the possibility of a genetic defect on chromosome 6, which is inherited and linked to the HLA locus.