PubMed Health⌕ Search

Biomedical subjects

A M Marchevsky

Publications and source records attributed to A M Marchevsky.

At least 37 records · Page 2Linked to original sources
PubMed1997

The application of image analysis and neural network technology to the study of large-cell liver-cell dysplasia and hepatocellular carcinoma.

Liver cell dysplasia (LCD) is considered a preneoplastic lesion, whose characterization and differentiation from hepatocellular carcinoma (HCC) and from the reactive changes seen in cirrhosis has been controversial. We studied 12 cases of LCD (large cell type) with image analysis techniques (IA) and compared the findings with those of HCC (n = 40), and a spectrum of non-neoplastic hepatic lesions including normal liver and cirrhosis (n = 49). A minimum of 200 Feulgen-stained nuclei were measured from each lesion with the CAS 200 image analysis system. The data were collected with the aid of CellSheet software. Thirty-four variables were measured, including geometric, textural, and photometric nuclear features and DNA ploidy. The data were analyzed with multivariate statistics and a backpropagation neural network (NN). Stepwise statistical analysis selected 22 variables that were statistically significant in the three groups with P values <.05. Various NN architectures were developed using these variables. The best NN architecture included a sigmoidal transfer function, 14 input, 16 hidden, and 3 output neurons. It trained to completion after 8,887 runs using 90% of the lesions. This NN yielded a 100% cross-validation rate for unknown cases. These data support the concept of LCD (large cell type) as a lesion that can be objectively distinguished from HCC and non-neoplastic liver. Our study also demonstrates the potential usefulness of IA for the evaluation of difficult histopathological problems.

Carcinoma, Hepatocellular↗
PubMed1997

Neural networks as a prognostic tool for patients with non-small cell carcinoma of the lung.

Patients with non-small cell carcinoma of the lung (NSCLC) have a poor prognosis (64 and 41% survival rates in Stages I and II). It is currently not possible to predict which patients with Stage I or II NSCLC will survive the disease. Sixty-seven patients with NSCLC, including 49 patients with Stage I NSCLC and 18 with Stage II disease (11 with squamous cell carcinomas, 35 with adenocarcinomas, and 21 with large cell carcinomas) were treated with lobectomy and followed for a minimum of 5 years. The tumors were studied with DNA flow cytometry and quantitative immunocytochemical studies for proliferation cell nuclear antigen, p53 protein, and MIB-1. The data were analyzed with backpropagation neural networks, univariate analysis of variance, the Kaplan-Meier survival method, and Cox proportional hazards model. The dependent variables were "free of disease" and "recurrence or dead from disease." Twenty neural network models were trained, using all cases but one, after 1883 to 2000 training cycles. At 5 years, 30 patients were free of disease and 37 were dead or had recurrence. Proliferating cell nuclear antigen was the only statistically significant prognostic factor by univariate analysis of variance and Cox proportional hazards analysis. The S phase was statistically significant by univariate analysis of variance (P <.05). All of the 20 models classified the test cases correctly. Study with backpropagation neural networks using multiple prognostic features from patients with NSCLC suggests that this technology might be useful for prediction of survival. This preliminary study must be validated with data from a larger group of patients with NSCLC before its clinical adequacy is established.

Adenocarcinoma↗
PubMed1996

Neuroendocrine tumors of the lung.

The lung can develop a spectrum of neuroendocrine neoplasms, whose histopathologic classification is difficult, controversial, and potentially confusing for the surgical pathologist. This chapter describes the various pulmonary endocrine tumors of the lung and provides guidelines for their differential diagnosis.

Carcinoma, Non-Small-Cell Lung↗
PubMed1995

Lung tumors derived from ectopic tissues.

Ectopic tissues rarely occur in the lung. They include endometriosis, neurological tissue, thyroid, pancreas, and adrenal. Choristomas are very rare developmental anomalies derived from these ectopic tissues. They develop as a mass composed of histologically normal tissues that are heterotopic in the lung. Various intrapulmonary neoplasms also rarely develop from ectopic tissues. They include malignant melanoma, thymoma, meningioma, glomus tumor, germ cell neoplasms (choriocarcinoma and teratoma), and ependymoma. The clinicopathological features of these unusual pulmonary neoplasms are discussed.

Choriocarcinoma↗
PubMed1994

Lymphoid-rich effusions. Diagnosis by morphometry using the CAS 200 System.

Lymphoid-rich effusions frequently present diagnostic problems in clinical cytology. In the authors' previous studies, most lymphoid-rich effusions had been correctly classified as benign lymphocytosis or malignant lymphoma by an experimental computerized interactive morphometry system, in which randomly selected lymphoid nuclear profile images were measured in Papanicolaou fixed and stained cytospin smears. The present study used the CAS 200 System and criteria from the previously described rule-based expert system to classify similar preparations of 134 lymphoid rich pleural, peritoneal, and pericardial effusions (90 benign lymphocytoses, 36 malignant lymphomas, and 8 chronic lymphocytic leukemias). A total of 98.9% of the benign lymphocytoses and 88.9% of the malignant lymphomas were correctly classified (predictive values of correct diagnoses 95.7% and 97.3%, respectively). Chronic lymphocytic leukemias could not be distinguished from benign lymphocytoses by nuclear profile areas. Optical density histograms of benign, lymphomatous, and chronic lymphocytic leukemias effusions are described. Advantages and limitations of image analysis and immunocytochemistry are discussed.

Ascitic Fluid↗
PubMed1994

Image analysis and diagnostic classification of hepatocellular carcinoma using neural networks and multivariate discriminant functions.

BACKGROUND: Hepatocellular carcinoma (HCC) is often difficult to diagnose in cytologic material and small tissue biopsies since histomorphologic information is minimal or absent. The potential for misdiagnosis is greatest in attempting to discriminate well-differentiated HCC from dysplastic hepatocytes in cirrhosis. We investigated the feasibility of developing artificial intelligence classification methods based on nuclear image analysis data for use as adjuncts to the morphologic diagnosis of HCC. EXPERIMENTAL DESIGN: Ninety hematoxylin-eosin stained histologic slides including 56 with well- to poorly differentiated HCC and 34 showing a morphologic continuum from normal to markedly dysplastic benign hepatocytes were assembled from four laboratories. A relatively inexpensive PC-based image analysis system was used to measure 35 nuclear morphometric and densitometric parameters of 100 nuclei in each specimen. The data were randomized into classification training and testing sets containing equal numbers of benign and HCC samples. Objective diagnostic classification criteria for HCC based on neural networks and multivariate discriminant functions (DFs) were developed for the most discriminatory subsets of morphometric, densitometric, and combined morphometric/densitometric variables as selected by stepwise discriminant analysis of training data. RESULTS: Morphometric parameters provided the best results with the following testing data positive and negative predictive values (PV+ and PV-) for HCC classification: 86.2% PV+ and 81.3% PV- for a linear DF, 85.7% PV+ and 76.5% PV- for a quadratic DF and 100% PV+ and 85.0% PV- for a neural network. CONCLUSIONS: Our results demonstrate that nuclear image analysis-based objective classification criteria for HCC can be developed using artificial intelligence methods and that histologic material prepared at different institutions can be reliably classified. Neural networks for HCC classification were superior to linear and quadratic DFs. Morphometric data yielded the best results compared with densitometric or combined morphometric/densitometric data.

Carcinoma, Hepatocellular↗
PubMed1993

Immunocytologic analysis of cells obtained from bronchoalveolar lavage in a model of rat lung allograft rejection.

Left lung transplantation (n = 30) between BN rat (RT1n) and LEW rat (RT1(1)) was performed to examine serial changes in the inflammatory cell profile and T-cell subsets occurring in bronchoalveolar lavage (BAL) obtained after transplantation. Transplanted animals were sacrificed on Days 1 to 7 post-transplantation. Previous studies show that lung allografts between these rat strains were strongly rejected within 7 days. The serial change in the cell profile of BAL showed a marked initial predominance of polymorphonuclear leukocytes, a decrease in macrophages, and a temporary increase in number of eosinophils on Day 2 post-transplantation. A gradual increase in lymphocytes coincident with progression of rejection was also noted. Immunocytologic studies using monoclonal antibodies specific for rat T-cell subsets and interleukin-2 receptor (IL-2R) showed significant increase in pan-T-cells on Days 3, 4, and 5 and T-suppressor/cytotoxic (CD8 positive) fraction on Days 4 and 5, whereas the T-helper (CD4 positive) fraction peaked on Day 2. The frequency of T-cells expressing IL-2R (55 kDa), indicating activated T-cells, significantly increased as early as Day 2 and maintained its high value thereafter. mRNA levels for IL-2R were detectable in the allografts on Day 2 and peaked on Day 5 post-transplantation. The value of CD4/CD8 T-cell ratios rose initially and then dropped below 1.0 on Days 4 and 5. These values differed markedly from those of syngeneic transplants (Lew-->Lew) examined on Day 5 post-transplantation. First, no significant changes in BAL cytology were seen when syngeneic transplants were compared with normal (Lew) lung.(ABSTRACT TRUNCATED AT 250 WORDS)

Animals↗
PubMed1993

A study of hepatocellular carcinoma using morphometric and densitometric image analysis.

Hepatocellular carcinoma is often difficult to diagnose in cytologic material and biopsy specimens. To demonstrate the utility of image analysis in discriminating benign and malignant hepatocytes, 42 malignant cell groups were compared with 26 benign cell groups with a wide range of nuclear morphology in hematoxylin and eosin-stained histologic sections from 42 patients with hepatocellular carcinoma. Nuclear measurements were performed with a relatively inexpensive microcomputer-based image analysis system using a highly flexible imaging software package. Twenty-two nuclear morphometric and densitometric parameters were evaluated. The best single discriminator of benign and malignant cells was the nuclear major axis. Classification of the test samples using optimized linear discriminant functions achieved the following positive predictive values (PV+) and negative predictive values (PV-) for hepatocellular carcinoma: 95.0% PV+ and 85.7% PV- for the major axis; 90.5% PV+ and 84.6% PV- for five densitometric parameters; 100% PV+ and 86.7% PV- for three morphometric parameters; and 95.5% PV+ and 100% PV- for nine combined morphometric/densitometric parameters. These results demonstrate that multivariate linear discriminant functions of nuclear features measured by image analysis can be used to classify benign and malignant hepatocytes accurately.

Adolescent↗
PubMed1993

Computerized interactive morphometry and the diagnosis of lymphoid-rich effusions.

An experimental computerized interactive morphometry system for the classification of lymphoid-rich effusions is presented. The relatively inexpensive microcomputer-based system was assembled in our laboratory with commercially available hardware and software. One hundred twenty-two effusions (86 benign lymphocytoses, 26 lymphomas, and 10 chronic lymphocytic leukemias) were studied. Lymphoid cells were selected randomly by the system from real-time images of Papanicolaou-fixed and stained cytospin smears of pleural, peritoneal, and pericardial effusions. Parameters of nuclear shape, area, and optical density were measured automatically. Multiparameter statistical procedures of discriminant classificatory analysis analyzed the distribution of lymphoid nuclear profile integrated optical density to yield three groups of effusions, each with a predictive value of diagnosis of 100%. Neither these statistical procedures nor a simple rule-based expert system accurately classified chronic lymphocytic leukemia effusions based on the distribution of lymphoid nuclear profile area. When chronic lymphocytic leukemia effusions were excluded, however, the predictive values of a diagnosis of lymphoma by statistical analysis and by rule-based expert system were 92.3% and 88.9%, respectively, whereas the predictive values of a diagnosis of benign lymphocytosis were 97.7% and 91.3%, respectively. Potential applications and limitations of this technology for the diagnosis of lymphoid-rich effusions are discussed.

Cell Nucleus↗
PubMed1993

An image analysis workstation for the pathology laboratory.

Computer-based image analysis (IA) is a technology gaining importance in diagnostic pathology. Applications of IA in pathology include DNA ploidy analysis, quantitative immunohistochemistry, three-dimensional reconstruction of tissue sections, motility studies, and chromosomal analysis. Morphometry, the quantitative measurement of size, shape, and textural features of cells and tissues, is another rapidly developing area of IA in pathology. Morphometric IA allows the objective evaluation of subtle histologic and cytologic features to yield useful diagnostic and prognostic information. Research is currently underway to develop diagnostically useful applications of morphometric IA. Several image analysis workstations designed for the pathology laboratory are currently available. However, the high cost and software inflexibility of these instruments are prohibitive to many potential users limiting the practicality of IA and hindering research. We present a relatively inexpensive pathology IA workstation assembled from commercially available hardware and software components. System features and basic image processing methods are described. A variety of practical applications for the surgical pathology laboratory are illustrated, including spatial measurements of tumors, nerve and muscle biopsy evaluation, and nuclear morphometry for classification of lymphoid effusions and hepatocellular carcinoma.

Biopsy↗
PubMed1993

Expert systems as an aid for the pathologist's role of clinical consultant: CANCER-STAGE.

The traditional role of the pathologist as a clinical consultant for various clinical specialists has become more difficult with the explosion of information that has taken place in the past decades. Rule-based expert systems offer the ability to store and quickly and efficiently retrieve information that is relevant to specific clinical situations. They are ideally suited for simple repetitive tasks such as the staging of patients with cancer and the retrieval of therapeutic and prognostic information. A microcomputer-based expert system, CANCER-STAGE, is described that allows a pathologist to perform these tasks with minimal computer training. The system was constructed by utilizing an inexpensive expert system development package (EXSYS). The knowledge base from the literature is stored in 370 IF-THEN rules that have notes with therapeutic information and references. The computer user interacting with the system answers questions and is quickly provided with the TNM stage of a particular patient and a short summary with therapeutic options. The conditions can be changed and the results of various WHAT-IF simulations compared for educational purposes.

Consultants↗
PubMed1991

Pulmonary cytology in lung transplant recipients: recent trends in laboratory utilization.

The value of bronchoscopy for the diagnosis of rejection and opportunistic infection in lung transplant recipients is controversial. We review our experience with pulmonary cytology obtained from 10 lung transplant recipients during the first 15 mos of the transplantation program at Cedars-Sinai Medical Center and compare the efficacy of pulmonary cytology for the diagnosis of opportunistic infectious agents with that of histology and microbiology. Our study encompasses 1,465 post-transplant days during which 70 bronchoscopies were performed yielding 94 cytologic specimens (44 bronchial washes, 25 bronchial brushings, and 25 bronchoalveolar lavages) and 55 transbronchial biopsies. The major advantages of cytology in this setting are short turn around time and high specificity for nonbacterial agents. All of the patients experienced episodes of bacterial pneumonia as well as fungal and viral infections. None developed Pneumocystis carinii infection during the study period. Simultaneous and concurrent infections were diagnosed. The initial diagnosis of bacterial pneumonia and herpes simplex virus preceded the diagnosis of cytomegalovirus; the former infections tended to persist and/or recur. Cytology was more effective than histology in establishing the diagnosis of Candida sp. and herpes simplex virus, while histology was more effective in establishing the diagnosis of cytomegalovirus. Increased numbers of polymorphonuclear cells did not constitute a consistent finding in cytologic or histologic samples during episodes of bacterial infection; cultures were most sensitive for detection of bacterial infection. Histochemical and immunohistochemical stains as well as in situ hybridization studies confirmed diagnoses rendered on routine Papanicolaou and hematoxylin and eosin stained material but did not provide additional diagnoses.(ABSTRACT TRUNCATED AT 250 WORDS)

Adult↗
PubMed1991

Vascular rejection and graft eosinophilia in rat lung allografts.

Allogeneic lung transplantation was performed using a rat model in order to assess the pathologic changes that developed during the process of rejection. The left lungs of 38 BN rats (RT-1n) were orthotopically transplanted into LEW rats (RT-1). The allografts developed the well-known changes of perivascular, peribronchial, and interstitial lymphocytic infiltrates resulting in necrosis of the pulmonary parenchyma at 7-8 days after transplantation. In addition, we document two findings that have not been reported previously in lung transplantation: vasculitis and eosinophilic infiltrates. Vasculitis with swelling and vacuolization of the endothelial cells was observed in transplants as early as 3 days following transplantation. Vasculitis with fibrinoid necrosis of the vessel wall was prominent at 7-8 days after grafting. The allografts also exhibited eosinophilia at 2 to 4 days following transplantation. The density of eosinophils in the inflammatory infiltrate reached a peak of 20% on Day 3 post-transplantation. These findings suggest an important role of humoral immunity and a possible involvement of eosinophils in lung allograft rejection.

Animals↗
PubMed1991

Expert systems for efficient handling of medical information. I. Lung cancer.

Rule-based expert systems offer the ability to quickly retrieve considerable amounts of information relevant to specific situations; the technology can be potentially very useful to physicians in an era characterized by an explosion of medical data. A simple rule-based expert system that allows a microcomputer user to obtain staging, prognostic and therapeutic information relevant to patients with lung cancer is presented. The program was constructed by utilizing an inexpensive expert system development package (EXSYS). Using information from the literature, 194 rules were formulated; these rules can be expanded or updated at any time. The computer user interacting with the system is asked sequential questions regarding the characteristics of the tumor (T) of a particular patient, the nodal status (N), the presence or absence of metastasis (M), how the staging information was derived (clinically or at surgery), the tumor cell type and the therapeutic options being considered (different surgical procedures, radiotherapy, chemotherapy and others). The system selects the appropriate answers and displays the stage of the tumor and relevant prognostic information. The microcomputer user can also examine the rules that were selected by the system. These rules have relevant comments that apply to the specific condition, as well as appropriate references. The microcomputer user can change all or some of the conditions (i.e., therapeutic options) and compare the results of the various "WHAT-IF" simulations.

Expert Systems↗
PubMed1989

Computerized interactive morphometry. An expert system for the diagnosis of lymphoid-rich effusions.

The authors present experimental techniques for the diagnosis of malignant lymphoma and benign lymphocytosis in lymphoid-rich effusions with the use of an inexpensive microcomputer-based video system for computerized interactive morphometry (CIM). Lymphoid cells were randomly selected by a trained observer from real-time images of Papanicolaou-fixed and -stained cytospin smears prepared from pleural, peritoneal, or pericardial effusions. The lymphoid cells were classified by the instrument, based on the size and shape of their nuclear profiles. The morphometric data collected by the instrument were interpreted by a simple rule-based expert system that classified the smears as benign or malignant. One hundred four cases, including 28 malignant lymphomas, 63 benign lymphocytoses, 8 chronic lymphocytic leukemias, and 5 cases with incomplete immunopathologic data, were studied retrospectively. Sixty-three of these effusions had been stained to detect light chain monoclonality. Ninety-one effusions were correctly classified by the expert system. There were four potential false negative diagnoses and one potential false positive diagnosis by the CIM system. Eight effusions from patients with chronic lymphocytic leukemia (CLL) were consistently classified as benign. Although the author's series of patients with a history of CLL is small, their results suggest that CIM is unsuitable for the diagnosis of malignancy in these effusions. If only those effusions from patients with a history of CLL are excluded, the predictive value of a diagnosis of malignant lymphoma was 96.5%, whereas the predictive value of a diagnosis of benign lymphocytosis was 94.0%.

Algorithms↗
PubMed1988

Septal thickness in human lungs assessed by computerized interactive morphometry.

We have studied the feasibility of using interactive computerized morphometric methods for the quantitation of septal thickness in the lung with the goal of introducing objective criteria for the diagnosis and grading of pulmonary fibrosis and edema. After reviewing available methods, we have decided in favor of linear intercept measurements of septal profiles which are limited to the tissue and therefore not affected by the level of lung inflation. Our method consisted in generating a graphic overlay of parallel lines over a real time video display of lung tissue. We had the options of measuring either uniformly oriented (random) or orthogonal intercepts. Randomly oriented intercepts (i.e., those with incidence angles ranging between 0 and 90 degrees) resulted in poor statistics because of excessive scattering of primary data; we resolved this difficulty by limiting the incident angle to the range of +/- 45 degrees from the normal to the septum at the point of sampling. This results in marked increase in consistency and reproducibility of the results with fewer measurements. Alternatively, we used intercepts measured over a line generated perpendicularly to the septal axis which approaches orthogonal thickness measurements. We found that both approaches yield acceptable results after as little as 50 measurements.

Computer Simulation↗