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 469 records · Page 26Linked to original sources

PubMed2001

The new World Health Organization classification of lymphomas: the past, the present and the future.

The REAL Classification of lymphomas, proposed in 1994, represents a new paradigm in lymphoma classification, consisting of a list of biologic entities defined by clinicopathologic and immunogenetic features. The non-Hodgkin's lymphomas comprise precursor lymphoblastic and mature cell neoplasms of B, T or putative natural killer cell lineage. An individual entity can exhibit a range of morphologic appearances and a range of clinical behavior. The categories in Hodgkin's lymphomas are identical to the widely used Rye classification except for the additional of a new category termed 'lymphocyte-rich classical Hodgkin's lymphoma'. The REAL classification has been validated by a major multi-institutional study involving 1378 cases (The Non-Hodgkin's Lymphoma Classification Project), showing that it is both reproducible and clinically relevant. The new World Health Organization classification of hematopoietic and lymphoid tumors, to be published in 2001, is a joint project of the Society for Hematopathology and European Association of Hematopathologists, under the auspices of the World Health Organization. This classification includes not only lymphoid neoplasms, but also myeloid, histiocytic and mast cell neoplasms. The lymphoma component of the classification is merely an update of the REAL classification, with minor changes necessitated by new information that has become available since its proposal. A conceptual grouping of the non-Hodgkin's lymphomas into four categories (indolent, aggressive, highly aggressive, and localized indolent) is also presented in this review. The next major impetus influencing the approach to lymphoma classification will no doubt be molecular genetics, in particular DNA microarrays, which will yield an enormous amount of new data that will aid in the understanding of lymphomas.

Forecasting↗

PubMed1985

Basis of the present classification of diabetes.

The present classification of diabetes most widely used is that recommended by the National Diabetes Data Group and subsequently endorsed by the World Health Organization. This classification is primarily a clinical classification of diabetes because in most instances the etiology is unknown. The need for a standardized classification arose out of the recognition that diabetes was a syndrome rather than a single disease and the different terminologies which emerged. While certain types of diabetes can be classified according to specific etiology or associations with specific syndromes, the vast majority cannot. Insulin-dependent and noninsulin-dependent diabetes usually represent syndromes whose etiopathology is believed to differ and their clinical characteristics are usually distinctive. As evidence of etiological heterogeneity has increased there has been a tendency to adopt the terms Type I and Type II diabetes to indicate different etiologies, although the original usage of these terms was as a clinical classification to differentiate between insulin dependent and non-insulin-dependent disease. At present the use of the four terms to describe the common types of diabetes leads to confusion, which could readily be resolved by arriving at agreed definitions for each of these terms. While the NDDG-WHO classification has served to standardize terminology and stimulate research into the different causes of diabetes, some further refinement of the classification, together with some additional definition of terms, should be considered. The classification of diabetes most widely used at the present time is that suggested by the National Diabetes Data Group (NDDG) in the United States in 1979, which was subsequently recommended by the World Health Organization (WHO) Expert Committee on Diabetes Mellitus in 1980. It should be stressed that this classification was intended to be a uniform framework for clinical and epidemiological research, and that the classification would almost certainly have to be modified on the basis of new knowledge in the future.

Age Factors↗

PubMed2007

Classification and localization of acetabular labral tears.

OBJECTIVE: The purpose of this study was to compare the findings on hip MR arthrography (MRA) with the published MRA and arthroscopic classifications of hip labral tears and to evaluate a clock-face method for localizing hip labral tears. DESIGN/PATIENTS: We retrospectively reviewed 65 hip MRA studies with correlative hip arthroscopies. Each labrum was evaluated on MRA using the classification system of Czerny and an MRA modification of the Lage arthroscopic classification. In addition, each tear was localized on MRA by using a clock-face description where 6 o'clock was the transverse ligament and 3 o'clock was anterior. These MRA findings were then correlated with the arthroscopic findings using the clock-face method of localization and the Lage arthroscopic classification of labral tears. RESULTS: At MRA, there were 42 Czerny grade 2 and 23 grade 3 labral tears and 22 MRA Lage type 1, 11 type 2, 22 type 3 and 10 type 4 tears. At arthroscopy, there were 10 Lage type 1 flap tears, 20 Lage type 2 fibrillated tears, 18 Lage type 3 longitudinal peripheral tears and 17 Lage type 4 unstable tears. The Czerny MRA classification and the modified MRA Lage classification had borderline correlation with the arthroscopic Lage classification. Localization of the tears using a clock-face description was within 1 o'clock of the arthroscopic localization of the tears in 85% of the patients. CONCLUSIONS: The Lage classification, which is the only published arthroscopic classification system for hip labral tears, does not correlate well with the Czerny MRA or an MRA modification of the Lage classification. Using a clock-face description to localize tears provides a way to accurately localize a labral tear and define its extent.

Acetabulum↗

PubMed2006

Comparison of the grade evaluated by "Liver damage" of Liver Cancer Study Group of Japan and Child-Pugh classification in patients with hepatocellular carcinoma.

BACKGROUND: The "Liver damage" classification proposed by the Liver Cancer Study Group of Japan and Child-Pugh classification are both useful classifications for hepatic function. However, the factors responsible for the difference between the two classifications have not been fully investigated. METHODS: The medical records of 594 admissions of 220 patients with hepatocellular carcinoma (HCC) were retrospectively analyzed for encephalopathy, ascites, serum bilirubin and albumin, plasma retention rate (%) at 15min after injection of 0.5mg/kg of indocyanine green (ICG R15), and prothrombin time. RESULTS: Of 594 admissions, ICG R15 was tested in 337 (56.7%). The Child-Pugh classification was evaluated in all 594 admissions, but the "Liver damage" could be evaluated in 510 (85.9%) due to the lack of ICG R15 results. Of the 594 admissions, 372 (62.6%), 162 (27.3%), and 60 (10.1%) were Child-Pugh grade A, B, and C, respectively. Of the 510 admissions, 219 (42.9%), 202 (39.6%), and 89 (17.5%) were "Liver damage" grade A, B, and C, respectively. The grade of "Liver damage" was similar to that of Child-Pugh classification in 369 (72.4%), under-evaluated in 138 (27.1%), and over-evaluated in 3 (0.6%). The Child-Pugh classification was statistically a better classification for predicting outcome than "Liver damage", but the "Liver damage" had better stratification ability than Child-Pugh classification in patients with relatively good liver function. CONCLUSIONS: Although the "Liver damage" could not be evaluated in some cases due to the lack of ICG R15 results, this classification system is useful in the evaluation and prediction of outcome of patients with early-stage liver diseases.

Journal Article↗

PubMed2006

Classification and scoring systems in myelodysplastic syndromes: a retrospective analysis of 311 patients.

The main objective of this study was to evaluate the role of the recent World Health Organization (WHO) classification for assessing prognosis in patients with myelodysplastic syndromes (MDS). To this effect, we analyzed the prognostic impact of the WHO and French-American-British (FAB) morphologic classifications and of four different scoring systems in a series of 311 patients with primary MDS diagnosed between October 1990 and June 2001. Both the FAB and WHO classifications identified groups with different prognoses (p<0.0001), those presenting refractory anemia (RA) and refractory anemia with ringed sideroblasts (RARS) showing the best prognosis. The WHO classification subdivided RA into RA with only red cell dysplasia, and refractory cytopenia with multilineage dysplasia (RCMD), and RARS into RARS plus refractory cytopenia with multilineage dysplasia and ringed sideroblast (RCMD-RS). In our population, we have shown that the two subtypes characterized by dysplasia affecting exclusively the erythroid population (RA and RARS) have a better prognosis, with a median survival of 122.2 and 81.9 months, respectively, than those with multilineage dysplasia (RCMD and RCMD-RS) with a median survival of 32.3 and 43.2 months, respectively. There were no significant differences in median survival comparing RA with RAS (p<0.95), or comparing RCMD with RSCMD (p<0.97). Besides, the four scoring systems discriminated our MDS patients in terms of survival, and an increase in prognostic capacity was achieved on adding the score to the morphological classifications. Risk scoring had a greater prognostic impact than the FAB and WHO classifications. Prognostic scoring systems may be an important tool for risk stratification in hematological practice, and add significance to morphological classification. Combined application of the WHO classification and score system is useful for improving the identification of patients with a poorer prognosis. The WHO classification establishes more homogeneous subcategories than the FAB classification and is also able to identify groups with different prognoses.

Adult↗

PubMed2004

Immunohistological comparison of the World Health Organization (WHO) and Ljubljana classifications on the grading of preneoplastic lesions of the larynx.

There is still controversy about the classification of laryngeal preneoplastic lesions. In this study, we compared the World Health Organization (WHO) and Ljubljana histological classifications with regard to laryngeal epithelial hyperplastic-dysplastic lesions in a single series to determine their relation to statistically significant criteria when the mechanisms of neoplastic progression are considered. Emphasis was put on p53 expression, proliferative activity, and angiogenesis. Fifty-four laryngeal biopsies with preneoplastic changes were re-evaluated and classified according to both classifications. The streptavidin-biotin method was used for immunohistochemical staining for Ki-67, p53, and CD34 antibodies. A positive correlation was obtained between the histological categories and Ki-67, p53, and CD34 expressions using both classifications (Spearman's Correlation test). There was a significant difference between the histological categories of both the WHO and the Ljubljana classifications, when the expression of the three markers was compared (Kruskal Wallis test, p = 0.000 for each). Further evaluation revealed a statistically significant difference between all categories of both classifications, excluding the p53 overexpression scores and vessel counts in mild and moderate dysplasia categories (Mann-Whitney U-test, p = 0.209, and 0.091 respectively), and the p53 overexpression scores in severe dysplasia and carcinoma in situ categories (Mann-Whitney U-test, p = 0.249) of the WHO classification. Similarly, no significant differences were found between severe dysplasia and carcinoma in situ, as well as between atypical hyperplasia and carcinoma in situ categories using both classifications for the Ki-67 expression (Mann-Whitney U-test, p = 0.806, and 0.111, respectively). Our results suggest that regarding the mechanisms of neoplastic progression such as p53 expression and angiogenesis, the histological categories of the Ljubljana classification seem to depend on additional evidence. Therefore, we support the use of the Ljubljana classification.

Adenocarcinoma↗

PubMedDate not supplied

The Ishikawa classification of cavernous sinus lesions by clinico-anatomical findings.

PURPOSE: The Jefferson classification has been used to localize cavernous sinus lesions. However, this classification occasionally showed dissociation between identified localization and clinical findings. We investigated the clinical applicability of the newly proposed Ishikawa classification based on serial topographic sections of human cavernous sinus and the clinical findings. METHODS: In the Ishikawa classification, the cavernous sinus is divided into three portions, that is, anterior, middle, and posterior, demarcated by the location of the intracranial orifice of the optic canal and the entry of the maxillary nerve into the cavernous sinus. A total of 162 patients with cavernous sinus lesions were classified using both the Jefferson and the Ishikawa classifications and the clinical applicability of these two classifications was studied. Characteristics of the localization of lesions were also examined in each etiological type. RESULTS: By the Jefferson classification, 11% of the 162 patients had the anterior type of lesion, 12% the middle, 8% the posterior type, and 69% the unclassifiable type. However, by the Ishikawa classification, 35% had the anterior type, 10% the middle type, 22% the posterior type, 5% the whole type, and 28% the unclassifiable type of lesion. Furthermore, the Ishikawa classification revealed that the etiology of the anterior type was mainly inflammation, and that the etiology of the posterior and whole types was tumors. CONCLUSION: The Ishikawa classification is clinically useful to identify and classify the localization of cavernous sinus lesions.

Carotid Artery, Internal↗

PubMed2005

Primary tumour-vessel tumour-nodal tumour classification for patients with invasive ductal carcinoma of the breast.

There are many studies that show biological differences between invasive ductal carcinoma (IDC) with and without nodal metastasis, but no prognostic classification taking into consideration any biological differences between them is currently available. We previously investigated the histological characteristics that play an important role in tumour progression of IDCs according to their nodal status, and a new prognostic histological classification, the primary tumour-vessel tumour-nodal tumour (PVN) classification, was devised based on the histological characteristics of IDCs with and without nodal metastasis. Multivariate analyses using the Cox proportional hazard regression models were used to compare the ability of the PVN classification to predict tumour recurrence and death in 393 IDC patients based on the following histological classifications: (1) the pTNM classification, (2) the Nottingham Prognostic Index, (3) the modified Nottingham Prognostic Index, and (4) the histologic grade. In IDCs without nodal metastasis, only the PVN classification significantly increased the hazard rates (HRs) of tumour recurrence and death (P<0.05), independent of the hormone receptor status. Similarly, in IDCs with nodal metastases, only the PVN classification significantly increased the HRs of tumour recurrence and death (P<0.05), independent of the hormone receptor status. We conclude that the PVN prognostic histological classification is the best classification available for IDC of the breast.

Breast Neoplasms↗

PubMed1999

Clinical evaluation of lymph node metastasis in gastric cancer defined by the fifth edition of the TNM classification in comparison with the Japanese system.

BACKGROUND: This study compared the classification of lymph node metastasis according to the number of involved nodes based on the new tumour node metastasis (TNM) system (fifth edition) with the classification by the Japanese Research Society for Gastric Cancer from an anatomical perspective. METHODS: The two classifications were related to long-term results in 1489 patients with gastric cancer who underwent gastrectomy with systematic extended lymphadenectomy. RESULTS: Both classifications performed well as prognostic indicators (5-year survival rates: pathological (p) N0, 89 per cent; pN1, 66 per cent; pN2, 34 per cent; pN3, nil; and M1, 10 per cent by the TNM classification; n0, 89 per cent; n1, 63 per cent; n2, 46 per cent; n3, 20 per cent; and n4, 8 per cent by the Japanese classification). For regional lymph nodes, the TNM classification was a better index of the prognosis. Significant survival differences were observed among patients with M1 disease according to the number of involved lymph nodes (between one and six nodes, 48 per cent; seven to 15 nodes, 12 per cent; more than 15 nodes, 2 per cent), indicating that patients with distant metastatic lymph nodes (M1) should also be classified by the number of involved nodes. On the other hand, the Japanese classification has the added benefit of being a good indicator of the anatomical extent of lymphadenectomy. CONCLUSION: The new TNM classification provided a better index of the prognosis of patients who underwent systematic lymph node dissection. However, both classifications have specific benefits in the surgical treatment of gastric cancer.

Female↗

PubMed1999

Classifications in psychiatry: a conceptual history.

BACKGROUND: Historical accounts of psychiatric classifications have hitherto been written in terms of a 'received view'. This contains two assumptions, that: (i) the activity of classifying is inherent to the human mind; and (ii) psychiatric 'phenomena' are stable natural objects. OBJECTIVES: The aim of this article is to provide an outline of the evolution of psychiatric classifications from the perspective of conceptual history. This is defined as a theoretical and empirical inquiry into the principles, sortal techniques and contexts in which alienists carried out their task. It assumes that all psychiatric classifications are cultural products, and endeavours to answer the question of whether classificatory models imported from the natural sciences can be applied to man-made constructs (such as mental illness) definitionally based on 'personalised semantics'. METHODS: Exemplars of classificatory activity are first mapped and contextualised. Then, it is suggested that in each historical period crafting classifications has been like playing a game of chess with each move being governed by rules. This is illustrated by offering an analysis of the 1860-1861 French debate on classification. RESULTS AND CONCLUSIONS: (1) Medicine is not a contemplative but a modificatory activity and hence classifications are only valuable if they can release new information about the object classified. (2) It should not be inferred from the fact that psychiatric classifications are not working well (i.e. that they only behave as actuarial devices) that they must be given up. Conceptual work needs to continue to identify 'invariants' (i.e. stable elements that anchor classifications to 'nature'. (3) Because mental disorders are more than unstable behavioural epiphenomena wrapped around stable molecular changes, 'neurobiological' invariants may not do. Stability depends upon time frames. Furthermore, it is unlikely that gene-based classifications will ever be considered as classifications of mental disorders. For once, they would have low predictive power because of their lack of information about the defining codes of mental illness. 'Social' and 'psychological' invariants have problems of their own.

History, 18th Century↗

PubMed1990

[Considerations and studies of the classification of glottis cancers].

Since 1972, the principles of the T classification of glottic carcinomas have remained unchanged. Nevertheless, discrepancies between T-classification and post-therapeutic pT classification are reported repeatedly and doubts have been expressed about the reliability of this system. Our T classification is based on pathohistological investigations performed 20 years ago. Therefore, the origin of the above-mentioned discrepancies must be sought in the definition and the clinical diagnostics. First, problems of the T classification of glottic cancer are analyzed and discussed on the basis of conditions which must be fulfilled by each classification: (1) clearly defined boundaries of the anatomical regions; (2) clearly defined rules for classification; (3) practicability for the clinician as well for the pathologist and comparability for both, and (4) prognostic relevance. Second, a new definition of the glottic region is proposed. It should be confined to the extension of the vocal ligaments, their anterior commissure and the medial surface of the arytenoid cartilages, including Reinke's space and the epithelial layer covering all the described areas. This is because the glottic region is confined to anatomical areas with reduced lymphatic drainage and a connective tissue barrier. The proposal is based on an anatomical and morphometric investigation of 154 larynx specimens. Third, 403 glottic carcinomas, treated between 1970 and 1984 were classified by the current T classification and histologically by pT classification (UICC 1987). By comparing these two classifications it was established that the early carcinomas had been overestimated and the advanced carcinomas underestimated.(ABSTRACT TRUNCATED AT 250 WORDS)

Follow-Up Studies↗

PubMed2001

Prognostic value of the Quebec Classification of Whiplash-Associated Disorders.

STUDY DESIGN: Retrospective cohort. OBJECTIVES: 1) Evaluate the utility of the Québec Classification of Whiplash-Associated Disorders as an initial assessment tool; 2) assess its ability to predict persistence of symptoms at 6, 12, 18, and 24 months postcollision; 3) examine one potential modification to the Classification. SUMMARY OF BACKGROUND DATA: In 1995, a task force from Québec, Canada, developed the Québec Classification of Whiplash-Associated Disorders to assist health care workers in making therapeutic decisions. The Classification was applied to an inception cohort of patients presenting for emergency medical care following their involvement in a rear-end motor vehicle collision. METHODS: All patients (n = 446) presenting to the only two emergency departments serving Kingston, Ontario, between October 1, 1995 and March 31, 1998 were considered for inclusion in the study. Eligible patients (n = 380) were categorized according to the Classification based on signs and symptoms documented in their emergency medical chart. Attempts were made to interview all patients shortly following and again 6 months after their collision. Patients were contacted at 12, 18, and 24 months postinjury only if sufficient time had elapsed between recruitment into and cessation of the study. Data were gathered regarding symptoms, treatments received, effects on usual activities, crash circumstances, and personal factors. Associations between initial Classification grade and the frequency/intensity of follow-up symptoms were quantified via multivariable analyses. RESULTS: The Classification was prognostic in that risk for Whiplash-Associated Disorders at 6, 12, 18, and 24 months increased with increasing grade. Analyses supported modification of the Classification to distinguish between Grade II cases of Whiplash-Associated Disorders with normal or limited range of motion. The greatest risk for long-term symptoms was seen among the group of patients with both point tenderness and limited range of motion. CONCLUSION: The analyses of this study support the use of the Québec Classification of Whiplash-Associated Disorders as a prognostic tool for emergency department settings, and the authors propose a modification of the Classification using a subdivision of the Grade II category.

Accidents, Traffic↗

PubMed2002

Identifying sources of variability in scoliosis classification using a rule-based automated algorithm.

STUDY DESIGN: Use of a rule-based automated algorithm to determine sources of variability in radiographic classification. OBJECTIVES: To determine whether unambiguous rules encoded in a computer program would ensure reliable classification. SUMMARY OF BACKGROUND DATA: Reliability problems have been identified in classifications used in surgical planning for patients with thoracic idiopathic scoliosis, but the sources of unreliability are not understood. METHODS: Objective classification methodology was tested on the King et al (1983) scheme. There were two novel components: 1) positions of the corners of vertebrae in radiographs were digitized relative to a defined axis system and used in automated evaluation of spinal shape parameters required for classification; and 2) the assignment of a classification was done with a rule-based algorithm. The algorithm was implemented after some ambiguities and absence of precise definitions in the King et al classification scheme had been resolved. The algorithm was tested with radiographs of patients having adolescent idiopathic scoliosis. RESULTS: The automated procedure could encounter reliability problems in cases in which a lumbar curve was very close to crossing the midline, thoracic and lumbar curves were of approximately equal value, when the apex level in the thoracolumbar region was ambiguous, when a Cobb angle was close to 10 degrees, or when the flexibility index was close to unity. CONCLUSION: Objective measurements and rule-based algorithms can eliminate some sources of interobserver and intraobserver errors in classification of spinal deformity. When classification parameters fall close to the boundaries for classification, reliability problems will persist.

Algorithms↗

PubMed2005

Interobserver and intraobserver reliability in the load sharing classification of the assessment of thoracolumbar burst fractures.

STUDY DESIGN: The Load Sharing Classification of spinal fractures was evaluated by 5 observers on 2 occasions. OBJECTIVE: To evaluate the interobserver and intraobserver reliability of the Load Sharing Classification of spinal fractures in the assessment of thoracolumbar burst fractures. SUMMARY OF BACKGROUND DATA: The Load Sharing Classification of spinal fractures provides a basis for the choice of operative approaches, but the reliability of this classification system has not been established. METHODS: The radiographic and computed tomography scan images of 45 consecutive patients with thoracolumbar burst fractures were reviewed by 5 observers on 2 different occasions 3 months apart. Interobserver reliability was assessed by comparison of the fracture classifications determined by the 5 observers. Intraobserver reliability was evaluated by comparison of the classifications determined by each observer on the first and second sessions. Ten paired interobserver and 5 intraobserver comparisons were then analyzed with use of kappa statistics. RESULTS: All 5 observers agreed on the final classification for 58% and 73% of the fractures on the first and second assessments, respectively. The average kappa coefficient for the 10 paired comparisons among the 5 observers was 0.79 (range 0.73-0.89) for the first assessment and 0.84 (range 0.81-0.95) for the second assessment. Interobserver agreement improved when the 3 components of the classification system were analyzed separately, reaching an almost perfect interobserver reliability with the average kappa values of 0.90 (range 0.82-0.97) for the first assessment and 0.92 (range 0.83-1) for the second assessment. The kappa values for the 5 intraobserver comparisons ranged from 0.73 to 0.87 (average 0.78), expressing at least substantial agreement; 2 observers showed almost perfect intraobserver reliability. For the 3 components of the classification system, all observers reached almost perfect intraobserver agreement with the kappa values of 0.83 to 0.97 (average, 0.89). CONCLUSIONS: Kappa statistics showed high levels of agreement when the Load Sharing Classification was used to assess thoracolumbar burst fractures. This system can be applied with excellent reliability.

Humans↗

PubMed2005

Independent validation of the 2002 American Joint Committee on cancer primary tumor classification for renal cell carcinoma using a large, single institution cohort.

PURPOSE: The primary tumor classification for renal cell carcinoma (RCC) was updated by the American Joint Committee on Cancer in 2002. To date the new classification has not been validated using an independent group of patients and, therefore, its accuracy for predicting patient outcome is unknown. In the current study we evaluated the 2002 primary tumor classification and compared its predictive ability with that of the 1997 classification. MATERIALS AND METHODS: We studied 2,746 patients treated with radical nephrectomy or nephron sparing surgery for unilateral, sporadic RCC between 1970 and 2000. Cancer specific survival was estimated using the Kaplan-Meier method. The predictive abilities of the 1997 and 2002 classifications were compared using the concordance index. RESULTS: There were 812 deaths from RCC a mean of 3.3 years following nephrectomy. Median followup in patients still alive at last followup was 9 years. Estimated 5-year cancer specific survival rates by the 2002 tumor classification were 97%, 87%, 71%, 53%, 44%, 37% and 20% in patients with pT1a, pT1b, pT2, pT3a, pT3b, pT3c and pT4 RCC, respectively. The concordance index for the association between the 2002 classification and death from RCC was 0.752 compared with 0.737 for the 1997 classification, indicating that the 2002 version contained more predictive ability. CONCLUSIONS: Our data suggest that the 2002 primary tumor classification with pT1 cancers subclassified into pT1a and pT1b provides excellent stratification of patients according to cancer specific survival and it has a predictive ability that is superior to that of the 1997 classification.

Adolescent↗

PubMed2005

Automated decision tree classification of corneal shape.

PURPOSE: The volume and complexity of data produced during videokeratography examinations present a challenge of interpretation. As a consequence, results are often analyzed qualitatively by subjective pattern recognition or reduced to comparisons of summary indices. We describe the application of decision tree induction, an automated machine learning classification method, to discriminate between normal and keratoconic corneal shapes in an objective and quantitative way. We then compared this method with other known classification methods. METHODS: The corneal surface was modeled with a seventh-order Zernike polynomial for 132 normal eyes of 92 subjects and 112 eyes of 71 subjects diagnosed with keratoconus. A decision tree classifier was induced using the C4.5 algorithm, and its classification performance was compared with the modified Rabinowitz-McDonnell index, Schwiegerling's Z3 index (Z3), Keratoconus Prediction Index (KPI), KISA%, and Cone Location and Magnitude Index using recommended classification thresholds for each method. We also evaluated the area under the receiver operator characteristic (ROC) curve for each classification method. RESULTS: Our decision tree classifier performed equal to or better than the other classifiers tested: accuracy was 92% and the area under the ROC curve was 0.97. Our decision tree classifier reduced the information needed to distinguish between normal and keratoconus eyes using four of 36 Zernike polynomial coefficients. The four surface features selected as classification attributes by the decision tree method were inferior elevation, greater sagittal depth, oblique toricity, and trefoil. CONCLUSION: Automated decision tree classification of corneal shape through Zernike polynomials is an accurate quantitative method of classification that is interpretable and can be generated from any instrument platform capable of raw elevation data output. This method of pattern classification is extendable to other classification problems.

Cornea↗

PubMed1994

Influence of diagnostic classification on outcomes and charges in geriatric assessment and rehabilitation.

OBJECTIVE: To determine if diagnostic classification is associated with different outcomes from treatment on a geriatric assessment unit (GAU) compared with usual care for elderly patients with acute illnesses. STUDY DESIGN: Reanalysis after post-hoc diagnostic classification of a randomized clinical trial with one year follow up. SETTING: Community hospital with rehabilitation facility. SUBJECTS: A total of 155 subjects, mean age 78 years and 78% female, of whom 58 subjects had a rehabilitation classification of their diagnoses and 97 had a medical or surgical classification. OUTCOME MEASURES: Mortality, nursing home use, activities of daily living, and charges for subsequent medical services. RESULTS: Mortality was decreased in persons with a rehabilitation classification who had received the GAU intervention. The GAU increased point-prevalence residence in the community for persons with a rehabilitation classification. For persons with a medical-surgical classification, the GAU intervention was associated with decreased cumulative use of nursing homes. However, GAU intervention was also associated with trends for increased charges in both diagnostic classification groups, even when adjusted for differential survival. CONCLUSION: Greater benefit from the GAU intervention on mortality and maintenance of residence in the community was observed in a group with a rehabilitation classification when compared with a group with a medical-surgical classification. These improved outcomes in the rehabilitation group were not accompanied by decreased charges for later medical services that could compensate for initial rehabilitation charges.

Activities of Daily Living↗

PubMed2005

Assessment of two thoracolumbar fracture classification systems as used by multiple surgeons.

BACKGROUND: The reproducibility and repeatability of modern systems for classification of thoracolumbar injuries have not been sufficiently studied. We assessed the interobserver and intraobserver reproducibility of the AO (Arbeitsgemeinschaft für Osteosynthesefragen) classification and compared it with that of the Denis classification. Our purpose was to determine whether the newer, AO system had better reproducibility than the older, Denis classification. METHODS: Anteroposterior and lateral radiographs and computerized tomography scans (axial images and sagittal reconstructions) of thirty-one acute traumatic fractures of the thoracolumbar spine were presented to nineteen observers, all trained spine surgeons, who classified the fractures according to both the AO and the Denis classification systems. Three months later, the images of the thirty-one fractures were scrambled into a different order, and the observers repeated the classification. The Cohen kappa (kappa) test was used to determine interobserver and intraobserver agreement, which was measured with regard to the three basic classifications in the AO system (types A, B, and C) as well as the nine subtypes of that system. We also measured the agreement with regard to the four basic types in the Denis classification (compression, burst, seat-belt, and fracture-dislocation) and with regard to the sixteen subtypes of that system. RESULTS: The AO classification was fairly reproducible, with an average kappa of 0.475 (range, 0.389 to 0.598) for the agreement regarding the assignment of the three types and an average kappa of 0.537 for the agreement regarding the nine subtypes. The average kappa for the agreement regarding the assignment of the four Denis fracture types was 0.606 (range, 0.395 to 0.702), and it was 0.173 for agreement regarding the sixteen subtypes. The intraobserver agreement (repeatability) was 82% and 79% for the AO and Denis types, respectively, and 67% and 56%, for the AO and Denis subtypes, respectively. CONCLUSIONS: Both the Denis and the AO system for the classification of spine fractures had only moderate reliability and repeatability. The tendency for well-trained spine surgeons to classify the same fracture differently on repeat testing is a matter of some concern.

Adult↗