Search PubMed⌕ Search

SEARCH · Search PubMed

Results for “classifier”

Search indexed PubMed citations on genomics, clinical trials, systematic reviews and public health. Explore titles, authors and supplied subject terms, then open the PubMed record.

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

At least 973 records · Page 54Linked to original sources
PubMed1992

Chemicals classified by IARC: their potency in tests for carcinogenicity in rodents and their genotoxicity and acute toxicity.

Chemicals classified by the IARC to its groups 1, 2A, 2B and 3 were examined in an attempt to identify characteristics of their behaviour in experimental studies of carcinogenicity, genotoxicity and acute mammalian toxicity that correlate with those categories. Only those agents for which information on carcinogenic potency was available were studied. In both mice and rats, more chemicals were potent carcinogens if they had been categorized in Group 1 (human carcinogens) than if they had been put into one of the other categories. Not surprisingly, there was a weak association between carcinogenic potency and acute toxicity. Mice were especially sensitive to tumour induction by halides; the lower sensitivity of rats to any carcinogenic effect of halides could be due in part to their higher systemic toxicity in this species: a reduced differential of toxic and carcinogenic doses decreases the dose window in which carcinogenic effects may be demonstrated. It was notable that the human carcinogens were active in those genotoxicity tests with higher specificity for identifying rodent carcinogens. Predictive assays for carcinogenicity that were considered to be highly specific were tests for cytogenetic effects in vivo, unscheduled DNA synthesis in hepatocytes, mutation in any of the five commonly used strains of Salmonella typhimurium and mutation at the hprt locus in mammalian cells. None of the relationships was strong enough to form the basis of a simple categorization, but they could serve to alert investigators to chemicals of special toxicological interest and importance.

Animals↗
PubMed2004

Computer-based malignancy grading of astrocytomas employing a support vector machine classifier, the WHO grading system and the regular hematoxylin-eosin diagnostic staining procedure.

OBJECTIVE: To investigate and develop an automated technique for astrocytoma malignancy grading compatible with the clinical routine. STUDY DESIGN: One hundred forty biopsies of astrocytomas were collected from 2 hospitals. The degree of tumor malignancy was defined as low or high according to the World Health Organization grading system. From each biopsy, images were digitized and segmented to isolate nuclei from background tissue. Morphologic and textural nuclear features were quantified to encode tumor malignancy. Each case was represented by a 40-dimensional feature vector. An exhaustive search procedure in feature space was utilized to determine the best feature combination that resulted in the smallest classification error. Low and high grade tumors were discriminated using support vector machines (SVMs). To evaluate the system performance, all available data were split randomly into training and test sets. RESULTS: The best vector combination consisted of 3 textural and 2 morphologic features. Low and high grade cases were discriminated with an accuracy of 90.7% and 88.9%, respectively, using an SVM classifier with polynomial kernel of degree 2. CONCLUSION: The proposed methodology was based on standards that are common in daily clinical practice and might be used in parallel with conventional grading as a second-opinion tool to reduce subjectivity in the classification of astrocytomas.

Artificial Intelligence↗
PubMed2004

Medicare program; changes to the criteria for being classified as an inpatient rehabilitation facility. Final rule.

This final rule responds to public comments on the September 9, 2003 proposed rule, and revises the classification criterion, commonly known as the "75 percent rule," used to classify a hospital as an inpatient rehabilitation facility (IRF). This final rule also modifies and expands the medical conditions listed in the regulatory requirements as well as temporarily lowers the percentage of patients required to fall within one of the specified list of medical conditions.

Humans↗
PubMed2004

Non-Hodgkin lymphoma in Jordan. Types and patterns of 111 cases classified according to the WHO classification of hematological malignancies.

OBJECTIVE: Non-Hodgkin lymphoma (NHL) is one of the most frequent malignancies in Jordan. The aims of this study are: 1. To classify NHL cases in Jordan, using the new World Health Organization (WHO) classification system, 2. To identify the most common types of NHL in Jordan, and 3. To compare lymphoma types and patterns in Jordan with those in surrounding countries and the West. METHODS: We studied all NHL cases, diagnosed during 1996 through to 1999 inclusive, at 2 major medical centers in Jordan, in order to identify their main types and patterns. One hundred and eleven cases of confirmed NHLs were reexamined and immunophenotyped in the year 2000, at the Department of Pathology, Jordan University of Science and Technology, Irbid, Jordan, using an immunohistochemical panel of antibodies, including CD3, CD15, CD20, CD30, CD43, CD45 and CD45RO. Confirmed NHL cases were reclassified according to the recently proposed WHO system of Hematological Malignancies. RESULTS: The median age of NHL cases was 44-years (range 2-85). The vast majority of cases were of B-cell phenotype; only 14% of the cases were T-cell lymphomas. Most of the cases were of the aggressive intermediate to high-grade large cell type. Diffuse large B-cell lymphoma (DLBCL) comprised 53% of NHLs and 62% of B-cell NHLs. Indolent lymphomas were uncommon, comprising 14% of all NHL cases. Twenty-nine cases were seen among patients less than 20-years. Burkitt lymphoma represented the largest group (55%) of the childhood NHLs followed by diffuse large cell and lymphoblastic types. CONCLUSION: Indolent lymphomas are rare in Jordan and account for less than 15% of all NHLs. Aggressive lymphomas; on the other hand, account for the majority of NHLs in Jordan. Burkitt lymphoma affected children less than 10-years of age with a median of 4.5-years. These observations indicate that NHLs in Jordan have different type distribution and patterns from those seen in the West.

Adolescent↗
PubMed2004

Using symbolic knowledge in the UMLS to disambiguate words in small datasets with a naïve Bayes classifier.

Current approaches to word sense disambiguation use and combine various machine-learning techniques. Most refer to characteristics of the ambiguous word and surrounding words and are based on hundreds of examples. Unfortunately, developing large training sets is time-consuming. We investigate the use of symbolic knowledge to augment machine-learning techniques for small datasets. UMLS semantic types assigned to concepts found in the sentence and relationships between these semantic types form the knowledge base. A naïve Bayes classifier was trained for 15 words with 100 examples for each. The most frequent sense of a word served as the baseline. The effect of increasingly accurate symbolic knowledge was evaluated in eight experimental conditions. Performance was measured by accuracy based on 10-fold cross-validation. The best condition used only the semantic types of the words in the sentence. Accuracy was then on average 10% higher than the baseline; however, it varied from 8% deterioration to 29% improvement. In a follow-up evaluation, we noted a trend that the best disambiguation was found for words that were the least troublesome to the human evaluators.

Abstracting and Indexing↗
PubMed1992

How do we classify functional status?

The original question "How do we classify functional status?" is rephrased as "How do we order or arrange the limitations of function of primary care patients in classes?". After a review of the functions to be considered, the concept of functional status is presented using empirical data from the research project "Morbidity and Functional Status of the Elderly." The group studied consisted of 5,502 patients older than age 65 and 25 general practitioners in 12 practices. Functional status was assessed using five COOP charts: physical status, psychological status, daily activities, social status, and change. Morbidity was registered using the International Classification of Primary Care (ICPC). At the beginning and end of the study, 30% of the elderly patients assessed their physical functions as seriously limited, whereas 6% to 8% reported psychological problems and limitations in daily activities or social contacts. During the encounters, more serious limitations were recorded: 35% of encounters involved serious physical limitations; 18% involved serious limitations in activities of daily living, and 11% involved constant psychological problems or limitations in social contact. Women reported more physical limitations than men. Older patients reported more physical limitations than younger ones, and people living in nursing homes reported more limitations than patients living independently. From the disease-specific health profiles, we concluded that the greatest limitation of all aspects of function was scored during encounters for cerebrovascular disease, dementia, and cancer of the lung, stomach, intestine, and breast. Hypertension, "no disease," and "common cold" elicited the lowest functional limitations.(ABSTRACT TRUNCATED AT 250 WORDS)

Activities of Daily Living↗
PubMed2002

Performance evaluation of artificial intelligence classifiers for the medical domain.

The application of artificial intelligence systems is still not widespread in the medical field, however there is an increasing necessity for these to handle the surfeit of information available. One drawback to their implementation is the lack of criteria or guidelines for the evaluation of these systems. This is the primary issue in their acceptability to clinicians, who require them for decision support and therefore need evidence that these systems meet the special safety-critical requirements of the domain. This paper shows evidence that the most prevalent form of intelligent system, neural networks, is generally not being evaluated rigorously regarding classification precision. A taxonomy of the types of evaluation tests that can be carried out, to gauge inherent performance of the outputs of intelligent systems has been assembled, and the results of this presented in a clear and concise form, which should be applicable to all intelligent classifiers for medicine.

Artificial Intelligence↗
PubMed2004

[A study on the variation in classifying chest radiographs for pneumoconiosis].

OBJECTIVE: To study the interobserver variation and its sources in classifying chest radiographs of pneumoconiosis for improving diagnostic quality. METHODS: A set of 240 high-voltage chest radiographs including different stages of pneumoconiosis was developed and 5 sets of copies were made in the same condition. It was read by 7 experienced readers independently in eligible situation. Results of reading were recorded in unified reading sheets. The profusion of small opacities and the stage of pneumoconiosis were recorded as corresponding numerical value for analysis. RESULTS: The coefficients of variation of 7 readers in total profusion of small opacities, average profusion of small opacities and stage of pneumoconiosis were 15.17%, 15.91%, 12.06%, respectively. The judgement of the presence or absence of small opacities, large opacities and coalescence of small opacities were closer, and the rates of consistency were 91.09%, 91.93% and 76.41% respectively. Technical quality of the radiographs in this study could not make a considerable impact on the reading results. If the shape of small opacities were recorded consistently by most readers as small rounded opacities or small irregular opacities, the judgement of total profusion of small opacities and stage of pneumoconiosis were more consistent; if the shape of small opacities were recorded inconsistently, the difference of judgement was higher. CONCLUSION: Interobserver variation exists actually among different readers and the main sources of the variation come from uncertain judgement of the shape of small opacities.

Humans↗
PubMed2004

[Features of grassland resources and their classified management in Alatai region of Xinjiang].

Grassland degradation, an important problem in grassland management, has affected the sustainable development of animal husbandry in Alatai region of Xinjiang. Based on the analysis of the community features, ecological service value and multiple functions, and main function of grassland types in the regional development, the classified management for grassland was designed. The results showed that Alatai grassland could be divided into three sectors, i.e., ecological function region, economic function region and mixed function region. The ecological function region made up 164.66 x 10(4) hm2 or 16.73% of total Alatai grassland, and the grassland types included alpine meadow, alpine rangeland, swamp, mountain desert steppe, mountain steppe desert and part of plain desert. For the ecological function region, the main management strategy was to prohibit grazing and cropping. The economic function region of 116.33 x 10(4) hm2 accounted for 11.82% of the total Alatai grassland, the grassland types included mountain meadow, flat meadow and meadow steppe, and the main management strategy to enhance the productivity was agricultural measures, such as fertilization and irrigation. The mixed function region which included mountain steppe, plain desert steppe and most part of plain desert was 703.21 x 10(4) hm2 or 71.45% of total Alatai grassland, and the main management strategy was rotational grazing.

Animal Husbandry↗
PubMed2004

Classifying clinical pathways.

BACKGROUND: Clinical pathways are commonly developed for homogenous patient groups. We were wondering if the traditional patient classification systems could be used for classifying clinical pathways. METHODOLOGY: To examine the utility of patient classification systems for clinical pathways, a sample of 13 clinical pathways was analyzed, involving a total of 412 patients. Three classification systems were tested: International Classification of Diseases, Ninth Revision (ICD9-CM), Clinical Coding System (CCS) data and All-Patient Redefined Diagnosis Related Groups (APR-DRG). RESULTS: Categorization with ICD9-CM and CCS shows rather wide variation. However, when restricting for the principal codes, CCS classification shows an almost homogeneous relationship with clinical pathways. APR-DRG's are already corrected for secondary procedures and are difficult to assess. Categorization with the Risk Of Mortality (ROM) is more homogeneous than with the Severity Of Illness (SOI). CONCLUSION: Patient groups in clinical pathways are rather heterogeneous. When restricting for the principal procedures, the strongest relationship seems to exist between clinical pathways and CCS. Further research is needed to refine this relationship.

Belgium↗
PubMed2005

Histologic implications of Pap smears classified as atypical glandular cells.

OBJECTIVE: To evaluate the histologic outcome of Pap smears reported as atypical glandular cells (AGC). STUDY DESIGN: Cervical cytology smears reported as AGC were reviewed and correlated with histologic follow-up. RESULTS: Of a total of 41,500 Pap smears performed at Alexandra Hospital during 2000-2002, 113 (0.2%) were reported as AGC. Of these, 64 were classified as AGC not otherwise specified (NOS) (56.6%), 48 AGC favor neoplasia (42.5%) and 1 (0.9%) adenocarcinoma in situ. All these women underwent histological evaluation. The mean age of the patients was 52 years. Significant abnormality was found in 30 women (26.5%). Eleven percent of women with smears reported as AGC NOS and 45.8% of those with AGC favor neoplasia had significant abnormality. Of these patients, 14 (12.4%) had endometrial involvement, whereas cervical abnormalities were identified in the remaining 16 (14.1%). Older women were more likely to have involvement of the endometrium and younger women, involvement of the cervix. CONCLUSION: Cytologic criteria and patient age might effectively direct the clinician toward the likely origin of AGC.

Adenocarcinoma↗
PubMed2005

[The diagnoses and classifying of congenital craniofacial cleft].

OBJECTIVE: To diagnose and classify the congenital craniofacial cleft with a uniform scale is helpful to evaluate the abnormality and select the repairing methods. METHODS: We analyzed 81 cases of congenital craniofacial cleft basically using Tessier craniofacial cleft classification. Furthermore, according to the position of soft tissue or bone, the character and degree of clefts or dysplasia and the results of CT scanning, we subdivided the congenital deformities based on S (skin), T (tissue), and O (OS). Arabic numerals were used to express the degree of the abnormality. RESULTS: Of all the cases analyzed with the STO classification, No. III and IV clefts are often seen in the infraorbital region (24.70%). No. IX and X clefts are mostly seen in the supraorbital region (38.27%). The relationship between the cleft types and involved tissue has not been found. CONCLUSIONS: The STO classification reinforces Tessier classification. It offers the basis for craniofacial cleft repair.

Adolescent↗
PubMed2006

[Relationship between the polymorphism of alpha-adducin gene and the two phenotypes of constitutions in patients with essential hypertension classified by TCM].

OBJECTIVE: To investigate the relationship between the polymorphism of alpha-adducin (ADD1) gene and the two phenotypes of constitution in patients with essential hypertension, the Yang-hyperactive (YH) type and phlegm-dampness (PD) type, classified by traditional Chinese medicine (TCM) approach. METHODS: Two hundred and seven patients differentiated by TCM approach as YH type (113 cases) or PD type (94 cases) were observed, with the systolic blood pressure (SBP), diastolic blood pressure (DBP), body mass index (BMI), fasting blood glucose (FBG), serum creatinine (Cr), uric acid (UA), total cholesterol (TC) and triglycerides (TG) as the criteria of observation. Gly460Trp polymorphism of the ADD1 gene was detected by MALDI-TOF mass spectrometry. Results The levels of BMI, DBP, FBG and UA, etc. in the PD group were significantly higher than those in the YH group respectively. The rate of GG, GT and TT type of ADD1 gene was 29.2%, 41.6% and 29.2% in the YH group, 28.7%, 48.9% and 22.3% in the PD group, showing no significant difference in ADD1 genotype distribution between the two groups, while there was also no difference in the hypertension phenotype distribution among different genotypes (both P > 0.05). For the patients with TT genotype, there were significant differences between the YH group and the PD group in BMI (24.11 +/- 3.04 kg/m2 vs 26.20 +/- 2.30 kg/m2), DBP (96.79 +/- 4.05 mmHg vs 99.56 +/- 3.90 mmHg), FBG (5.01 +/- 0.53 mmol/L vs 5.51 +/- 1.07 mmol/L) and UA level (302.22 +/- 71.95 micromol/L vs 358.25 +/- 88.75 micromol/L, all P < 0.05). CONCLUSION: There was no relation between ADD1 gene polymorphism and the TCM genotype of constitution in patients with essential hypertension. However, it is likely that for hypertension patients with TT genotype, those of PD type are more susceptible to cardiovascular disease and have worse prognosis than those of YH type.

Adult↗
PubMedDate not supplied

Neonatal mortality clusters: a new tool for classifying neonatal outcomes.

BACKGROUND: A method for assessing general hospital neonatal care performance is needed that is simple, is easy to use, and requires minimal data. METHODS: All neonatal deaths in Washington State obstetric hospitals from 1980 to 1983 were assigned to 10 mutually exclusive neonatal mortality clusters, a new classification method derived from information available on the death certificate. RESULTS: More than one-third (35.3 percent) of all neonatal deaths fell within one of the seven clusters considered to represent potentially preventable causes of death. The rate of possibly preventable deaths was much higher in level III hospitals than in level II or level I hospitals, a finding similar to that observed in other states using different analytic approaches. CONCLUSIONS: Neonatal mortality clusters offer a less complex method of classifying neonatal deaths and assessing hospital performance than other currently used techniques.

Birth Weight↗
PubMed1991

[Should aphasias be classified?].

In the same sense as classical monuments are classified? The more so as the classification described here goes back to the end of the 19th century, for it is the one which was designed by Wernicke (Breslau, Germany) and Lichtheim (Bern, Switzerland) and was remarketed by the Boston school (Geschwind, Goodglass, Kaplan, etc.) which invaded even the Japanese market. In view of their diversity, disorders of language in patients with cortical and subcortical lesions must be reorganized into syndromes that do not express a nuclear linguistic deficit. Interindividual differences are still the rule, which is not surprising in view of the numerous variance factors.

Aphasia↗
PubMed1990

[Screening for venous thrombosis and pulmonary embolisms in patients with stroke using an adaptive classifier].

The use of adaptive linear neuron is referred to in recognizing veinous thromboses and lung embolism in a total of 88 acute palsies. As the input signs, the accessible clinical and laboratory data were used. Both training and adaptation of neuron have been performed in a part of total [35 patients], and 53 patients were submitted to the classification of patients with unknown distribution. In classified group with present complications, the screened diseases have been recognized in 16/20 patients. In 29 from 33 patients without thromboembolic states, the distribution by classificator appeared as correct one. The level of correct resolution was as high as 88% in all.

Cerebrovascular Disorders↗
PubMed1990

[An investigation of X-ray examinations classified according to cases of dental diseases at Fukuoka Dental College Hospital].

This paper reports the results of an investigation on the types and frequencies of X-ray examinations given to patients with various types of their dental diseases from their first visit to the hospital to the end of treatment. The subjects were 337 patients (not including orthodontic patients and in-patients) randomly sampled out of all those who visited the Fukuoka Dental College Hospital in 1986. Out of these 337, the number of those who were given X-ray examinations was 284. 1. The Cases of dental diseases for which X-ray examinations were given were classified into the following seven: diseases caused by dental caries, trauma, inflammation, diseases of temporomandibular joint, diseases of salivary gland and other diseases, and other examinations. Out of these X-ray examinations, those given to the patients of diseases caused by dental caries and marginal periodontitis amounted to the 60% of the total X-ray examinations. 2. On the average, 3.3 radiographs were taken per patient. 3. In the examination of luxation, a case of trauma, 7.8 radiographs were taken per patient, the largest number of all diseases on this report. In the examination of marginal periodontitis, a case of inflammation, 5.4 dental radiographs and 0.5 panoramic radiograph were taken per patient. It is necessary to make the most of panoramic examination effective for this disease to decrease radiation exposure. 4. The diseases that have more difficulty of diagnosis and more necessity of observing the prognosis had a greater tendency to depend on X-ray examinations. 5. The percentage of the patients who did not receive X-ray examination was 15.7%. Out of these patients, the patients in pedodontics accounted for 34.5%.

Adolescent↗
PubMed1990

[Process of the tooth brushing habit formation in children. 1. Period of the start and present circumstances classified by age].

The purpose of this study was to investigate the process by which the tooth brushing habit was formed in children. The study is based on questionnaires about the habit of tooth brushing from mothers of kindergarten pupils aged from one to six years living in Nagoya and the suburbs. The results of this study were summarised as follows: 1. Period of the start: 1) The number of children who started tooth brushing before the age of one year and six months was 75.7%, and before the age of two, 88.3%. 2) The item of "Eruption of the teeth" showed the highest ratio (49.2%) regarding the motive for the start of tooth brushing. 3) The children who started tooth brushing with the motive which construction about caries prevention with mothers as the main tended to start the earliest. 4) The ratio of persons who actually brush the children's teeth were as follows, "mother brushes (51.6%)", "mother after her child (37.1%)" and "child brushes (6.8%)". 5) The number of children who started using a tooth-brush before the age of two years was 78.8%, and before the age of two years and six months, 92.5%. 2. Present circumstances classified by age: 1) The number of children whose teeth are brushed by themselves or their mothers more than once a day before they were six years old was 61.8%. 2) In all ages, the group of once or twice a day showed a high ratio for frequency of tooth brushing. 3) "mother brushes (47.4%)" for one year olds, and "mother helps her child's brushing (85.7-53.6%)" for two to five year olds, and "child brushes (64.8%)" for six years olds were the highest ratios regarding the persons who actually brush the children's teeth. 4) Co-operative attitude of the children towards their mother's help is bad for one year olds, but good above two years old. 5) Attitude of the children towards tooth brushing is negative under three years old, and positive above four years old. 6) There is little hope that children can brush well by themselves under two years old. 7) children who have experienced guidance in tooth brushing from their mothers before they were three years old numbered 84.7%, and from a dentist or dental hygienist before they were six years old numbered 49.8%.

Age Factors↗