SEARCH · PubMed Health

Results for “Classification”

Explore indexed PubMed citations for clinical trials, systematic reviews and public health research. Read source abstracts and follow each citation to its original PubMed record.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 847 records · Page 47Linked to original sources

PubMed1987

Reliability and validity of the classification of death in general practice.

The general practitioner in the Netherlands is solely responsible for assigning the natural causes of a patient's death when it occurs at home and for entering it on the certificate of death B, which is then forwarded to the CSO (Central Statistics Office). The CSO however, records the cause of death in the appropriate category of the ICD-9 of the WHO. The value of these data, therefore, depends upon the agreement between the general practitioner's cause of death and the CSO categories of causes of death. To assess the degree to which the CSO classification reflects the general practitioner's cause of death, form B from eight general practitioners practicing in Hoorn were compared with the classification of the CSO during a period of five years, 1979-83. There was a 65% (237/365) agreement between the classifications. Of the remaining 128 cases 68% (87/128) of the classification differences were caused by a wrongly described cause of death on the B certificate. In 32% (41/128) the classification rules of WHO made it hardly possible for the general practitioner to classify uncertain causes of death. The conclusions of this survey are that the quality of the statistics on causes of death will improve if the general practitioners are more careful with filling in the certificate of death B; the classification rules of the ICD-9 makes it possible to classify unknown causes of death; and if the certificate of death B becomes more 'user friendly'.

Classification↗

PubMed2003

Clinicians' use of the International Classification of Sleep Disorders: results of a national survey.

OBJECTIVE: To determine clinicians' patterns of use of the International Classification of Sleep Disorders (ICSD) and their ratings of its utility and organization. METHODS: A telephone survey was conducted in one clinician practicing at each of 206 sleep centers accredited by the American Academy of Sleep Medicine (AASM). Descriptive analyses were performed, and ratings for various sleep classifications were compared with ANOVA and post- hoc tests. RESULTS: Clinicians at 230/251 (92%) of contacted centers agreed to participate, and data from 206/251 (82%) were actually collected. The ICSD was used by 91.7% of clinicians for clinical decision making but by lower percentages for billing or medical record keeping. Features of the ICSD pertaining to clinical descriptions and differential diagnosis were rated as most useful. The Association of Sleep Disorders Centers (ASDC) classification was rated more highly than ICSD in organization and ease of use. The ICSD and ASDC systems were rated more highly than other sleep disorders classifications in terms of organization, "fit" to patients, and ease of use. CONCLUSIONS: The ICSD is used widely by clinicians, primarily for diagnostic purposes, and is preferred over more general sleep disorders classification systems. However, more favorable ratings for the ASDC compared to ICSD indicate the value clinicians place on easy-to-use, symptom-oriented classifications.

Adolescent↗

PubMed1998

Facial fracture classification according to skeletal support mechanisms.

OBJECTIVE: To construct, propose, and evaluate the usefulness of a new clinical facial fracture classification scheme to accurately denote, communicate, and compare facial fractures. DESIGN: A retrospective, consecutive sample study with application of the proposed classification scheme to denote maxillary and zygomatic fractures with computed tomography. SETTING: Metropolitan tertiary care trauma center. PATIENTS: A total of 213 consecutive adult patients with facial fractures evaluated by means of 2-dimensional computed tomography. RESULTS: The classification scheme is defined according to fractures of vertical buttresses and horizontal beams. The scheme uses 3 primary descriptors of laterality and support sites to denote the clinical pattern of the fractures. This scheme was accurately applied and sufficient to describe 87 midfacial fracture patterns in this study. In addition, 118 (98%) of 120 mock fracture patterns were correctly transcribed and reproducibly communicated among 12 participating physicians. CONCLUSIONS: This newly proposed facial fracture classification scheme provides a convenient, specific, descriptive, and reproducible method of denoting fracture patterns. This scheme may be used to accurately communicate and compare, in greater detail than permitted using current independent classification schemes, the essential site and degree-of-severity characteristics of facial fractures critical to their surgical reduction and reconstruction. The usefulness of this classification scheme in determining optimal methods and subsequent outcomes in midfacial fracture reduction requires further investigation.

Adolescent↗

PubMed1993

Progress toward achieving a common language in psychiatry. Results from the field trial of the clinical guidelines accompanying the WHO classification of mental and behavioral disorders in ICD-10.

In preparing for the 10th revision of the International Classification of Diseases (ICD-10), the Division of Mental Health of the World Health Organization organized an international field trial to help evaluate draft clinical descriptions and diagnostic guidelines that were produced to facilitate use of the chapter dealing with mental and behavioral disorders. These clinical guidelines were prepared in equivalent versions in most of the world's widely spoken languages. The field trial aimed to obtain data that would help in assessing whether the classification fits the diagnoses made in different countries, whether it is easy to use, and whether psychiatrists after a short period of familiarization with the classification can reach agreement about their diagnoses and classification. The field trial was carried out at 112 clinical centers in 39 countries by 711 clinicians who conducted 15,302 individual assessments. The trial included joint clinical assessments of patients and case history exercises. The results of the joint assessment part of the trial are reported here. Most clinicians reported that the draft document was easy to use and that the classification provided a good fit for the vast majority of the clinical conditions encountered. While interrater reliability was satisfactory for most categories, some (for example, those dealing with personality disorders) were somewhat difficult to use, and reliability of assignment for those was lower. The trial demonstrated that the ICD-10 chapter dealing with mental and behavioral disorders is on the whole suitable for general use. It provided valuable indications about changes needed for subsequent versions and demonstrated the feasibility of large-scale international research on classification and diagnosis in psychiatry.

Adult↗

PubMed1996

The World Health Organization classification of adenohypophysial neoplasms. A proposed five-tier scheme.

BACKGROUND: Although numerous attempts have been made, the classification of pituitary neoplasms remains controversial. METHODS: The present proposal is based on correlative clinical and pathologic (histologic, immunocytochemical, and electron microscopic) studies of more than 8000 surgically removed human pituitary tumors. RESULTS: Anterior pituitary tumors have been variously classified on the basis of their clinical presentation, biochemical findings, histology (growth pattern, tinctorial characteristics), proliferation indices, immunocytochemical profile, and ultrastructural features. Herein we propose a five-tier classification, clinicopathologic in nature, which focuses on endocrine activity, imaging, operative findings, histology, immunocytochemistry, and ultrastructure. The integration of these five complementary approaches into what is fundamentally a pathologic classification is as practical as it is novel. Its importance lies in the fact that the collected data provide valuable information to the clinical endocrinologist, neurosurgeon, and oncologist involved in the assessment of a tumor's biologic behavior, growth potential, therapeutic responsiveness, and prognosis. Due to financial restraints, lack of facilities, and unavailability of well trained personnel, we realize that at present the five approaches cannot be fully implemented in all institutions. Nonetheless, clinical and biochemical data, imaging, and operative findings, as well as basic histologic parameters are generally readily available. Collectively they are indispensable in establishing the correct diagnosis and in directing patient management. Conversely, the cost-effectiveness of immunocytochemistry and electron microscopy is debatable in that their performance and interpretation requires time, financial resources, and expertise. Nonetheless, it should be clear that any correlation between endocrine activity, growth potential, morphologic features, immunocytochemical profile, and ultrastructural features provides greater insight into the pathobiology of adenohypophysial tumors than is gained by routine histology alone. Thus we feel justified in including immunocytochemical and ultrastructural investigation of pituitary tumors in a modern classification. At present, more sophisticated molecular biologic methods represent investigative tools that play no significant role in the classification. CONCLUSIONS: A five-tier classification of adenohypophysial neoplasms based on clinical and biochemical results, imaging, operative findings, histology, immunocytochemistry, and electron microscopy is proposed. It is recommended to the World Health Organization for acceptance.

Humans↗

PubMed1997

Automated classification of patients with chronic lymphocytic leukemia and immunocytoma from flow cytometric three-color immunophenotypes.

The goal of this study was the discrimination between chronic lymphocytic leukemia (B-CLL), clinically more aggressive lymphoplasmocytoid immunocytoma (LP-IC) and other low-grade non-Hodgkin's lymphomas (NHL) of the B-cell type by automated analysis of flow cytometric immunophenotypes CD45/14/20, CD4/8/3, kappa/CD19/5, lambda/CD19/5 and CD10/23/19 from peripheral blood and bone marrow aspirate leukocytes using the multiparameter classification program CLASSIF1. The immunophenotype list mode files were exhaustively evaluated by combined lymphocyte, monocyte, and granulocyte (LMG) analysis. The results were introduced into databases and automatically classified in a standardized way. The resulting triple matrix classifiers are laboratory and instrument independent, error tolerant, and robust in the classification of unknown test samples. Practically 100% correct individual patient classification was achievable, and most manually unclassifiable patients were unambiguously classified. It is of interest that the single lambda/CD19/5 antibody triplet provided practically the same information as the full set of the five antibody triplets. This demonstrates that standardized classification can be used to optimize immunophenotype panels. On-line classification of test samples is accessible on the Internet: http://www.biochem.mpg.de/valet/leukaem1.html Immunophenotype panels are usually devised for the detection of the frequency of abnormal cell populations. As shown by computer classification, most the highly discriminant information is, however, not contained in percentage frequency values of cell populations, but rather in total antibody binding, antibody binding ratios, and relative antibody surface density parameters of various lymphocyte, monocyte, and granulocyte cell populations.

Antibodies↗

PubMed1985

Comparison of the working formulation of non-Hodgkin's lymphoma with the Rappaport, Kiel, and Lukes & Collins classifications. Translational value and prognostic significance based on review of 658 patients treated at a single institution.

Six hundred fifty-eight cases of previously untreated non-Hodgkin's lymphoma seen between 1970 and 1979 at the Medical Department, the Finsen Institute, were the basis for a comparative study of the prognostic value of the Rappaport, Kiel, and Lukes & Collins classifications and the new translation system, the Working Formulation of Non-Hodgkin's Lymphoma. Each histopathologic system proved equally effective in separating patients into subgroups with a spectrum of prognoses ranging from a median survival of less than 1 year to greater than 7 years. The established classifications were compared with the Working Formulation in order to evaluate its translational value. The Working Formulation was more similar to the Rappaport and the Lukes & Collins systems than to the Kiel system, since 82%, 89%, and 75% of the cases, respectively, were translatable following the guidelines outlined in the National Cancer Institute (NCI)-sponsored study. Similarities among the four systems were demonstrated in lymphomas with follicular growth pattern, and in diffuse lymphomas composed of small mature appearing lymphocytes or small cleaved lymphocytes. Incongruity among the systems was more marked in lymphomas composed of large lymphoid cells or in lymphomas of mixed cellular composition. A comparison was performed for each classification against the Working Formulation. All such subdivided subsets were tested for prognostic heterogeneity and the following conclusions were reached: the diffuse poorly differentiated lymphocytic category of Rappaport was separated into two subgroups (malignant lymphoma [ML] small cleaved cell and ML lymphoblastic) with different prognoses (P = 0.01); the diffuse "histiocytic" lymphomas were prognostically homogeneous, since none of the newer systems were able to identify subpopulations with significantly different prognoses; the subtypes of the Kiel classification were prognostically homogeneous; the only weakness of the Lukes & Collins classification was the undefined cell subtype, encompassing two populations with different prognoses; and (5) the importance of follicular growth pattern was confirmed for small cleaved cell and mixed cell cytology, whereas large cell cytology implied a poor prognosis regardless of pattern. By the use of the Cox regression model it could be demonstrated that the Working Formulation can substitute any of the established classifications in terms of prognostic value.

Female↗

PubMed2001

International neuroblastoma pathology classification for prognostic evaluation of patients with peripheral neuroblastic tumors: a report from the Children's Cancer Group.

BACKGROUND: The International Neuroblastoma Pathology Classification was established in 1999 for the prognostic evaluation of patients with neuroblastic tumors (NTs). METHODS: Pathology slides from 746 NTs (the Children's Cancer Group [CCG]-3881 and CCG-3891 studies) were evaluated according to the International Classification. First, prognostic effects of the morphologic indicators (grade of neuroblastic differentiation: undifferentiated [U], poorly differentiated [PD] and differentiating [D]; and mitosis-karyorrhexis index [MKI]: low [L-MKI], intermediate [I-MKI], and high [H-MKI]) for tumors in the neuroblastoma (NB) category were tested. Then, prognostic significance of the International Classification for all NTs in four categories (neuroblastoma [NB]; ganglioneuroblastoma, intermixed [GNBi]; ganglioneuroma [GN]; and ganglioneuroblastoma, nodular [GNBn]) was analyzed. Finally, age distribution of the patients in the four categories as well as three subtypes (based on the grade of differentiation) in the NB category was compared. RESULTS: There were 630 NB tumors, 30 GNBi tumors, 10 GN tumors, and 76 GNBn tumors. In the NB category, prognostic effects of the indicators (three grades of differentiation and three mitosis-karyorrhexis index [MKI] classes: low [L], intermediate [I], and high [H]) were affected significantly by the age of the patients. The age-linked evaluation of the indicators according to the International Classification successfully distinguished two prognostic subgroups: the favorable histology (FH) subgroup (PD/D and L/I-MKI tumors in patients age < 1.5 years, D and L-MKI tumors in patients ages 1.5-5.0 years; 90.4% 5-year event free survival [EFS]) and the unfavorable histology (UH) subgroup (U and/or H-MKI tumors in patients of any age, PD and/or I-MKI tumors in patients ages 1.5-5.0 years, any grade of differentiation, and any MKI class in patients age > or = 5 years; 26.9% EFS) (P < 0.0001). The International Classification also distinguished the FH group (FH subgroup with NB, GNBi, and GN tumors) and the UH group (UH subgroup with NB and GNBn tumors) for all NTs (90.8% EFS and 31.2% EFS, respectively; P < 0.0001) and provided independent prognostic information on both patient age and disease stage (P < 0.0001). Among patients with FH tumors, the median ages of patients with the PD and D subtype tumors in the NB category were 0.43 years (range, 0-1.50 years) and 1.50 years (range, 0.02-4.65 years), respectively, and the median ages of patients with GNBi and GN tumors were 3.51 years (range, 0.96-14.85 years) and 4.80 years (range, 1.94-17.05 years), respectively. In contrast, patients with UH tumors generally were older when they were diagnosed, and with median ages of 2.99 years (range, 1.30-8.84 years) for patients with U subtype tumors, 2.59 years (range, 0.0-12.57 years) for patients with PD subtype tumors, 2.16 years (range, 0.35-9.90) for patients with D subtype tumors, and 3.26 years (range, 0.57-15.90 years) for patients with GNBn tumors. CONCLUSIONS: This study confirmed the prognostic significance of the International Classification, substantiated age-linked prognostic effects of the morphologic indicators for patients with the tumors in the NB category, and supported the concept of an age-appropriate framework of maturation for patients with the tumors in the FH group.

Adolescent↗

PubMed1978

Utilization of the Goldberg MMPI profile classification rules for the assessment of psychopathology in different clinical populations.

MMPI profiles of 545 psychiatric inpatients and 560 incarcerated offenders were separated sequentially into normal, sociopathic, neurotic and psychotic groups by means of Goldberg's profile classification rules. Patient-prisoner differences in both rates of classification and profile patterns of groups within diagnostic categories were assessed quantitatively, and profiles were interpreted by use of standard MMPI codebooks. For the hospitalized Ss comparisons also were made between clinical diagnoses and Goldberg-MMPI classifications. The resulting differences in classification rates, similarities of profiles within diagnostic categories, correspondence between obtained and codebook-expected profiles, and agreement between clinical and Goldberg-MMPI classifications were not such that this approach could be used with confidence as a basis for establishing diagnostic group membership. Although Goldberg's system appears to represent certain improvements over previous criterion-related methods of profile classification, it was concluded that its value nonetheless is limited by the assumption of an invariant relationship between test patterns and nontest variables.

Adult↗

PubMed1991

The American College of Rheumatology criteria for the classification and reporting of osteoarthritis of the hip.

Clinical criteria for the classification of patients with hip pain associated with osteoarthritis (OA) were developed through a multicenter study. Data from 201 patients who had experienced hip pain for most days of the prior month were analyzed. The comparison group of patients had other causes of hip pain, such as rheumatoid arthritis or spondylarthropathy. Variables from the medical history, physical examination, laboratory tests, and radiographs were used to develop different sets of criteria to serve different investigative purposes. Multivariate methods included the traditional "number of criteria present" format and "classification tree" techniques. Clinical criteria: A classification tree was developed, without radiographs, for clinical and laboratory criteria or for clinical criteria alone. A patient was classified as having hip OA if pain was present in combination with either 1) hip internal rotation greater than or equal to 15 degrees, pain present on internal rotation of the hip, morning stiffness of the hip for less than or equal to 60 minutes, and age greater than 50 years, or 2) hip internal rotation less than 15 degrees and an erythrocyte sedimentation rate (ESR) less than or equal to 45 mm/hour; if no ESR was obtained, hip flexion less than or equal to 115 degrees was substituted (sensitivity 86%; specificity 75%). Clinical plus radiographic criteria: The traditional format combined pain with at least 2 of the following 3 criteria: osteophytes (femoral or acetabular), joint space narrowing (superior, axial, and/or medial), and ESR less than 20 mm/hour (sensitivity 89%; specificity 91%). The radiographic presence of osteophytes best separated OA patients and controls by the classification tree method (sensitivity 89%; specificity 91%). The "number of criteria present" format yielded criteria and levels of sensitivity and specificity similar to those of the classification tree for the combined clinical and radiographic criteria set. For the clinical criteria set, the classification tree provided much greater specificity. The value of the radiographic presence of an osteophyte in separating patients with OA of the hip from those with hip pain of other causes is emphasized.

Hip Joint↗

PubMed2003

Guidelines for case classification for the National Birth Defects Prevention Study.

BACKGROUND: Previous studies have suggested that etiologic heterogeneity may complicate epidemiologic analyses designed to identify risk factors for birth defects. Case classification uses knowledge of embryologic and pathogenetic mechanisms to make case groups more homogeneous and is important to the success of birth defects studies. METHODS: The goal of the National Birth Defects Prevention Study (NBDPS), an ongoing multi-site case-control study, is to identify environmental and genetic risk factors for birth defects. Information on environmental risk factors is collected through an hour-long maternal interview, and DNA is collected from the infant and both parents for evaluation of genetic risk factors. Clinical data on infants are reviewed by clinical geneticists to ensure they meet the detailed case definitions developed specifically for the study. To standardize the methods of case classification for the study, an algorithm has been developed to guide NBDPS clinical geneticists in this process. RESULTS: Methods for case classification into isolated, multiple, and syndrome categories are described. Defects considered minor for the purposes of case classification are defined. Differences in the approach to case classification for studies of specific defects and of specific exposures are noted. CONCLUSIONS: The case classification schema developed for the NBDPS may be of value to other clinicians working on epidemiologic studies of birth defects etiology. Consideration of these guidelines will lead to more comparable case groups, an important element of careful studies aimed at identifying risk factors for birth defects.

Adult↗

PubMed2003

Mortality associated with birth defects: influence of successive disease classification revisions.

INTRODUCTION: The National Vital Statistics System is the fundamental source of mortality statistics in the United States. The cause-of-death classification which is used to assign medical terms to a standard taxonomy is revised periodically, and it is necessary to account for these changes when investigating trends. This paper presents birth defects mortality statistics and preliminary information on the most recent transition between revisions of the classification. METHODS: Descriptive statistics are presented using multiple cause and underlying cause counts, rankings of leading causes of death, infant mortality rates, and age-adjusted death rates. Comparability results are based upon records that have been coded using two separate classifications. RESULTS: Birth defects remain the leading cause of death for infants and among the leading causes for younger age groups. The trend for birth defect mortality shows that the risk of dying from birth defects decreased between 1950 and 2000. The effect of implementing successive revisions of the cause-of-death classification was relatively minor until the implementation of the most recent revision. Fewer records are assigned to birth defects in the latest revision than in the previous revision. DISCUSSION: Researchers investigating mortality trends related to birth defects need to be aware that the cause-of-death classification changes periodically. The effect of the changes between two successive classifications can be measured and explained.

Cause of Death↗

PubMed2004

A combination of HER-2 status and the St. Gallen classification provides useful information on prognosis in lymph node-negative breast carcinoma.

BACKGROUND: Adjuvant chemotherapy for patients with lymph node-negative breast carcinoma is being recommended currently based on the St. Gallen classification. The prognostic importance of HER-2 status in patients with lymph node-negative breast carcinoma has been investigated extensively, with contradictory results. The authors investigated the clinical relevance of HER-2 overexpression when combined with the St. Gallen classification in lymph node-negative breast carcinoma. METHODS: The medical records of patients with breast carcinoma negative for lymph node involvement who underwent surgery between January 1995 and December 2000 at the Seoul National University College of Medicine (Seoul, Korea) were reviewed retrospectively. Risk groups based on the St. Gallen classification were categorized as average or minimal risk. The prognostic values of HER-2 in combination with the St. Gallen classification were analyzed with respect to disease-free survival (DFS) rates. RESULTS: A total of 906 patients were eligible for analysis. The overall 7-year DFS rate was 87.5%. The 7-year DFS rates for patients with HER-2-positive and HER-2-negative tumors were, respectively, 77.9% and 91.2% (P = 0.002). The 7-year DFS rates for patients with average and minimal risk group were 85.0% and 97.9%, respectively. The authors found that HER-2 overexpression significantly predicted the risk of disease recurrence (odds ratio = 3.03 [95% confidence interval, 1.63-5.63]). Furthermore, when HER-2 status was combined with the St. Gallen classification, the DFS rate of the HER-2-positive average risk group was 73.3% compared with 88.4% for the HER-2-negative average risk group (P = 0.007). CONCLUSIONS: The combination of HER-2 overexpression and the St. Gallen classification was more useful than either alone to predict the risk of disease recurrence in patients with lymph node-negative breast carcinoma.

Adult↗

PubMed1990

Automatically inferred Markov network models for classification of chromosomal band pattern structures.

A structural pattern recognition approach to the analysis and classification of metaphase chromosome band patterns is presented. An operational method of representing band pattern profiles as sharp edged idealized profiles is outlined. These profiles are nonlinearly scaled to a few, but fixed number of "density" levels. Previous experience has shown that profiles of six levels are appropriate and that the differences between successive bands in these profiles are suitable for classification. String representations, which focuses on the sequences of transitions between local band pattern levels, are derived from such "difference profiles." A method of syntactic analysis of the band transition sequences by dynamic programming for optimal (maximal probability) string-to-network alignments is described. It develops automatic data-driven inference of band pattern models (Markov networks) per class, and uses these models for classification. The method does not use centromere information, but assumes the p-q-orientation of the band pattern profiles to be known a priori. It is experimentally established that the method can build Markov network models, which, when used for classification, show a recognition rate of about 92% on test data. The experiments used 200 samples (chromosome profiles) for each of the 22 autosome chromosome types and are designed to also investigate various classifier design problems. It is found that the use of a priori knowledge of Denver Group assignment only improved classification by 1 or 2%. A scheme for typewise normalization of the class relationship measures prove useful, partly through improvements on average results and partly through a more evenly distributed error pattern. The choice of reference of the p-q-orientation of the band patterns is found to be unimportant, and results of timing of the execution time of the analysis show that recent and efficient implementations can process one cell in less than 1 min on current standard hardware. A measure of divergence between data sets and Markov network models is shown to provide usable estimates of experimental classification performance.

Chromosome Banding↗

PubMed2004

Image analysis for automatic segmentation of cytoplasms and classification of Rac1 activation.

BACKGROUND: Rac1 is a GTP-binding molecule involved in a wide range of cellular processes. Using digital image analysis, agonist-induced translocation of green fluorescent protein (GFP) Rac1 to the cellular membrane can be estimated quantitatively for individual cells. METHODS: A fully automatic image analysis method for cell segmentation, feature extraction, and classification of cells according to their activation, i.e., GFP-Rac1 translocation and ruffle formation at stimuli, is described. Based on training data produced by visual annotation of four image series, a statistical classifier was created. RESULTS: The results of the automatic classification were compared with results from visual inspection of the same time sequences. The automatic classification differed from the visual classification at about the same level as visual classifications performed by two different skilled professionals differed from each other. Classification of a second image set, consisting of seven image series with different concentrations of agonist, showed that the classifier could detect an increased proportion of activated cells at increased agonist concentration. CONCLUSIONS: Intracellular activities, such as ruffle formation, can be quantified by fully automatic image analysis, with an accuracy comparable to that achieved by visual inspection. This analysis can be done at a speed of hundreds of cells per second and without the subjectivity introduced by manual judgments.

Animals↗

PubMed2002

Automated assignment of SCOP and CATH protein structure classifications from FSSP scores.

We present an automated procedure to assign CATH and SCOP classifications to proteins whose FSSP score is available. CATH classification is assigned down to the topology level, and SCOP classification is assigned to the fold level. Because the FSSP database is updated weekly, this method makes it possible to update also CATH and SCOP with the same frequency. Our predictions have a nearly perfect success rate when ambiguous cases are discarded. These ambiguous cases are intrinsic in any protein structure classification that relies on structural information alone. Hence, we introduce the "twilight zone for structure classification." We further suggest that to resolve these ambiguous cases, other criteria of classification, based also on information about sequence and function, must be used.

Algorithms↗

PubMed2006

A new classification rule for incomplete doubly multivariate data using mixed effects model with performance comparisons on the imputed data.

A mixed effects model, enhanced by a Kronecker product structure for the residual variance-covariance matrix, is used in conjunction with a discriminant analysis technique, to devise a new statistical classification method on incomplete doubly multivariate data. The proposed method is efficient in small scale clinical trials that use relatively few patients. The new classification method is also applied to multiply imputed data sets. The misclassification error rates (MERs) are compared in order to investigate the effectiveness of the new classification rule on an incomplete data set. The classification method is applied to a real data set. The error rates on the incomplete data set are found to be much less than the median error rate on the multiply imputed data sets. Non-parametric methods, such as kernel method and k-nearest neighbourhood method, are also applied to multiply imputed data sets. Results illustrating the advantages of the new classification method over classic non-parametric classification methods are presented.

Bone Density↗

PubMed1996

Carcinogen classification systems: similarities and differences.

An overview of regulatory classification systems on carcinogens in the Organization for Economic Cooperation and Development (OECD) countries is presented based on a questionnaire study. Most OECD countries have implemented legislation including classification systems and lists of carcinogens. Basically, there are two types of classifications systems. The major difference between the two is that in one system carcinogens are classified according to the weight of evidence for carcinogenic effects in humans, whereas in the other carcinogens are allocated to various groups according to potency. Even if the classification systems may differ, the substances classified as carcinogens are to a large extent the same. Classification of carcinogens will in many countries require hazard labeling. This labeling, i.e., the limit for labeling of substances and preparations, and risk phrases show considerable similarities, but differ in certain aspects. Several countries have restrictions on sale and/or use of carcinogens. There is a trend toward introducing more mechanistic considerations in the classification of carcinogens.

Carcinogens↗