Search PubMed⌕ Search

SEARCH · Search PubMed

Results for “CLASSIFICATION”

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
PubMed1997

The "Tic-Tac-Toe" classification system for mutilating injuries of the hand.

Several classifications of mutilating hand injuries exist in the literature. Unfortunately, each of these provides a categorization that is arbitrarily grouped according to the part of the hand predominantly involved. It is imperative that a comprehensive classification system incorporate the degree and precise location of soft-tissue and/or bony destruction and the vascular integrity in addition to the predominantly involved part of the hand. We therefore devised a new classification system for mutilating injuries of the hand which categorizes them into seven types: (I) dorsal mutilation, (II) palmer mutilation, (III) ulnar mutilation, (IV) radial mutilation, (V) transverse amputations, (VI) degloving injuries, and (VII) combination injuries. These types are subcategorized into three subtypes: (A) soft-tissue loss, (B) bony loss, and (C) combined tissue loss. Vascular integrity is recorded with subscript notation: (0) vascularization intact or (1) devascularization. The hand is then systematically divided into nine numerical zones in "tic-tac-toe" fashion with radial, central, and ulnar columns and proximal, central, and distal rows. The "Tic-Tac-Toe" classification system allows the examining surgeon to describe precisely any mutilating injury of the hand. This system permits accurate assessment of each hand injury by assignment of the appropriate classification type, subtype, vascular status, and zone involvement. Clinical examples illustrate the user-friendliness and practicality of this new classification system.

Adolescent↗
PubMed1997

Classification of low back pain from dynamic motion characteristics using an artificial neural network.

STUDY DESIGN: Data were collected from 183 subjects who were randomly assigned to the training and test groups. During testing of the classification system, knowledge of the low back pain condition or motion characteristics of the patients in the test group was not made available to the system. OBJECTIVES: To determine specific characteristics of trunk motion associated with different categories of spinal disorders and to determine whether a neural network analysis system can be effective in distinguishing patterns. SUMMARY OF BACKGROUND DATA: Numerous studies have established the difficulty of evaluating lower back pain. Imaging techniques are expensive and ineffective in many cases. A technique for evaluation of lower back pain was developed on the basis of analysis of such dynamic motion features as shape, velocity, and symmetry of movements, using a neural network classification system. METHODS: Dynamic motion data were collected from 183 subjects using a triaxial goniometer. Features of the movement were extracted and provided as input to a two-stage neural network classifier governed by a radial basis function architecture. After training, the output of the classifier was compared with Québec Task Force pain classifications obtained for the patients. Linear and nonlinear classification techniques were compared. RESULTS: The system could determine low back pain classification from motion characteristics. The neural network classifier produced the best results with up to 85% accuracy on novel "validation" data. CONCLUSIONS: A neural network based on kinematic data is an excellent predictive model for classification of lower back pain. Such a system could markedly improve the management of lower back pain in the individual patient.

Adult↗
PubMed2001

Interobserver and intraobserver agreement of radiograph interpretation with and without pedicle screw implants: the need for a detailed classification system in posterolateral spinal fusion.

STUDY DESIGN: A prospective randomized clinical study in which four observers evaluated radiographs of posterolateral fusion masses. OBJECTIVES: To evaluate the accuracy of radiograph interpretation of the posterolateral spinal fusion mass when using a detailed classification system and to analyze the influence of metallic internal fixation devices on radiologic inaccuracy. SUMMARY OF BACKGROUND DATA: In general, the literature describing the classification criteria used for radiograph interpretation of spinal posterolateral fusion has serious deficiencies. There is a need for a detailed classification system. METHODS: Seventy patients were randomly allocated to receive no instrumentation (n = 36) or Cotrel-Dubousset instrumentation (n = 34) in posterolateral lumbar fusion. All four observers participated in a prestudy discussion and evaluated the radiographs (anteroposterior, lateral) taken at the 1-year follow-up evaluation. The observers scored the radiographs twice (30 days apart). Each level on each side was judged separately. A continuous intertransverse bony bridge involving at minimum one of the two sides indicated a fusion at that level. "Fusion" indicated this quality of fusion at all intended levels. If the fusion was doubtful on both sides of the interspace, the individual case could not be classified as "fused." RESULTS: The mean interobserver agreement was 86% (Kappa 0.53), and the mean intraobserver agreement was 93% (Kappa 0.78). No difference in interobserver and intraobserver agreement was found between patients with and without supplementary pedicle screw fixation. All mean Kappa values were classified as fair or good. The four observers identified a mean fusion rate of 81%. CONCLUSION: It is extremely difficult to interpret radiographic lumbar posterolateral fusion success. Such an assessment needs to be performed by use of a detailed radiographic classification system. The classification system presented here revealed good interobserver and intraobserver agreement, both with and without instrumentation. The classification showed acceptable reliability and may be one way to improve interstudy and intrastudy correlation of radiologic outcomes after posterolateral spinal fusion. Instrumentation did not influence reproducibility but may result in slightly underestimated fusion rates.

Adult↗
PubMed2002

Multisurgeon assessment of coronal pattern classification systems for adolescent idiopathic scoliosis: reliability and error analysis.

STUDY DESIGN: Three scoliosis surgeons and one orthopedic fellow were presented the anteroposterior radiographs of 70 patients with adolescent idiopathic scoliosis. All the reviewers assigned a type to each curve according to the classification systems of H. A. King and R. W. Coonrad. OBJECTIVES: To compare multisurgeon reliability in applying the classification systems of H. A. King and R. W. Coonrad, and to analyze controversially classified curve patterns. SUMMARY OF BACKGROUND DATA: The system most commonly used to classify adolescent idiopathic scoliosis is King's classification. However, because of poor interobserver reliability, the validity of this system is questioned. In contrast, high interobserver reliability is reported for Coonrad's classification system, which is used less frequently in clinical practice. METHODS: Interobserver agreement and intraobserver reproducibility were tested. Kappa coefficients were used to test reliability. Between the observers, the divergent assignments to curve patterns were analyzed in both quantitative and qualitative terms. An error analysis was performed. RESULTS: Paired comparisons showed a mean interobserver kappa coefficient of 0.45 for King's and 0.38 for Coonrad's classification systems. According to Svanholm et al, these values indicate poor reliability in terms of interobserver agreement. Error analyses for both classification systems showed that the reason for poor reproducibility is disagreement among the observers about structural upper thoracic and structural lumbar curves. CONCLUSIONS: Neither the King nor the Coonrad method appears to have sufficient interobserver reliability. To improve reliability, the authors recommend that the structural stigmas of the upper thoracic and lumbar curves be unequivocally described.

Adolescent↗
PubMed2002

Discriminative and predictive validity assessment of the quebec task force classification.

STUDY DESIGN: A prospective cohort study of workers with low back pain who had been absent from work for more than 4 weeks was conducted. OBJECTIVE: To assess the discriminative and predictive validity of the Quebec Task Force Classification for workers during the subacute phase of disability from back pain. SUMMARY OF BACKGROUND DATA: The Quebec Task Force Classification was designed for clinical decision making, prognosis establishment, quality of care evaluation, and scientific research in low back pain. METHODS: For this study, 104 workers absent from work because of back pain were classified according to the first four categories of the Quebec Task Force Classification 4 weeks after their first day of work absence. They then were randomized into four treatment groups: standard care (control), clinical-rehabilitation intervention, occupational intervention, and the Sherbrooke model (a combination of the clinical-rehabilitation and occupational interventions). Functional status, pain level, and work status were assessed at baseline and after 1 year. Duration of full compensation and back-related costs were calculated over a mean follow-up period of 6.5 years. The discriminative validity of the Quebec Task Force Classification was evaluated using Kendall tau correlation coefficients. Predictive validity was evaluated using logistic regression analyses. Age, gender, comorbidities, body mass index, and treatment group were considered as potential confounders. RESULTS: Significant but low correlation coefficients were found between Quebec Task Force Classification categories and functional status scores at baseline. Subjects classified as having distal radiating pain (categories 3 and 4) at baseline were more likely to have a lower functional status, higher pain level, and no return to regular work at the 1-year follow-up evaluation. They also were more likely to accumulate more days of full compensation and to cost more after a mean follow-up period of 6.5 years. CONCLUSION: The Quebec Task Force Classification demonstrated good predictive ability by discriminating between subjects with and those without distal radiating pain.

Adult↗
PubMed2006

Three-dimensional classification of spinal deformities using fuzzy clustering.

STUDY DESIGN: A prospective study of a large set of three-dimensional (3D) reconstructions of spinal deformities in adolescent idiopathic scoliosis (AIS). OBJECTIVES: To determine the value of fuzzy clustering techniques to automatically detect clinically relevant 3D curve patterns within this set of 3D spine models. SUMMARY OF BACKGROUND DATA: Classification is important for the assessment of AIS and has been mainly used to guide surgical treatment. Current classification systems are based on visual curve pattern identification using two-dimensional radiologic measurements but remain controversial because of their low interobserver and intraobserver reliability. A clinically useful 3D classification remains to be found. METHODS: An unsupervised learning algorithm, fuzzy k-means clustering, was applied on 409 3D spine models. Analysis of data distribution using clinical parameters was performed by studying similar curve patterns, near each cluster center identified. RESULTS: The algorithm determined that the entire sample of models could be segmented in five easily differentiated curve patterns similar to those of the Lenke and King classifications. Furthermore, a system with 12 classes made possible the identification of subpatterns of spinal deformity with true 3D components. CONCLUSIONS: Automatic and clinically relevant 3D classification of AIS is possible using an unsupervised learning algorithm. This approach can now be used to build a relevant 3D classification of AIS using appropriate key features of 3D models selected by a panel of expert spinal deformity surgeons.

Adolescent↗
PubMed2006

Retinal nerve fiber layer analysis in the diagnosis of glaucoma.

PURPOSE OF REVIEW: The detection of optic disc and retinal nerve fiber layer damage and change is the cornerstone of glaucoma management. Assessment of the retinal nerve fiber layer for localized and diffuse damage has been traditionally based on clinical examination, with documentation of change primarily qualitative. With the latest improvements in optical imaging instruments, objective and quantitative measurements of the retinal nerve fiber layer are now possible. This review summarizes the results from recent cross-sectional studies evaluating the discriminating ability of automated retinal nerve fiber layer measurements to detect glaucoma, and from longitudinal studies assessing the ability to predict and monitor glaucomatous changes. RECENT FINDINGS: Numerous cross-sectional studies have documented good diagnostic accuracy of a scanning laser polarimeter (GDx VCC), the optical coherence tomograph (Stratus), and the Heidelberg Retina Tomograph retinal nerve fiber layer measurements for differentiating between healthy and glaucoma eyes. There are only limited data available on the ability of these retinal nerve fiber layer measurements to document change over time. SUMMARY: It is essential that the clinician understand the specific strengths and weaknesses of each technique so that only good quality retinal nerve fiber layer information will be used in conjunction with careful clinical examination and visual function testing for glaucoma management decisions. Longitudinal studies are needed to evaluate the ability of these instruments to document retinal nerve fiber layer change over time.

Diagnostic Imaging↗
PubMed2004

Multi-institutional validation of a symptom based classification for renal cell carcinoma.

PURPOSE: We validate the prognostic value of a symptom based classification (S classification) in a multi-institutional study. MATERIALS AND METHODS: A total of 2,242 patients from 5 European centers were included in this study. Based on symptoms at diagnosis, patients were stratified into 3 groups of S1-asymptomatic tumors, S2-tumors with local symptoms and S3-tumors with systemic symptoms. Variables such as age, gender, tumor size, TNM stage, Fuhrman grade, Eastern Cooperative Oncology Group (ECOG) performance status, perinephric fat, renal vein and adrenal invasion were also considered for prognostic value. The end point of the study was cancer specific survival. Survival assessment was made with univariate and multivariate analyses using the Kaplan-Meier method and Cox regression analysis. RESULTS: Of the patients 1,018 (45.4%) were classified as S1, 865 (38.6%) S2 and 339 (16.0%) S3. The S classification correlated to tumor stage, grade and ECOG (p <0.001). On univariate analysis ECOG performance status, S classification, tumor size, TNM stage, Fuhrman grade, and adrenal, perinephric fat or vein invasion were significant prognostic factors (p <0.001). The S classification provided a significant prognostic stratification in the aggregate as well at each of the 5 centers. On multivariate analysis the S classification, TNM stage, Fuhrman grade, and perinephric fat and renal vein invasion remained independent prognostic factors (p <0.001). CONCLUSIONS: This study confirms that it is possible to graduate symptoms for a prognostic purpose. The proposed symptom score should be evaluated for its integration in prognostic algorithms.

Adolescent↗
PubMed2006

Describing computed tomography findings in acute necrotizing pancreatitis with the Atlanta classification: an interobserver agreement study.

OBJECTIVES: The 1992 Atlanta classification is a clinically based classification system that defines the severity and complications of acute pancreatitis. A study was undertaken to assess the interobserver agreement of categorizing peripancreatic collections on computed tomography (CT) using the Atlanta classification. METHODS: Preoperative contrast-enhanced CTs from 70 consecutive patients (49 men; median age, 59 years; range, 29-79 years) operated for acute necrotizing pancreatitis (2000-2003) in 11 hospitals were reviewed. Five abdominal radiologists independently categorized the peripancreatic collections according to the Atlanta classification. Radiologists were aware of the timing of the CT and the clinical condition of the patient. Interobserver agreement was determined. RESULTS: Interobserver agreement among the radiologists was poor (kappa, 0.144; SD, 0.095). In 3 (4%) of 70 cases, the same Atlanta definition was chosen. In 13 (19%) of 70 cases, 4 radiologists agreed, and in 42 (60%) of 70 cases, 3 radiologists agreed on the definition. In 21 cases (30%), 1 or more of the radiologists classified a collection as "pancreatic abscess," whereas 1 or more radiologist used another Atlanta definition. CONCLUSION: The interobserver agreement of the Atlanta classification for categorizing peripancreatic collections in acute pancreatitis on CT is poor. The Atlanta classification should not be used to describe complications of acute pancreatitis on CT.

Adult↗
PubMed2006

Diagnosis and classification of congenital craniofacial cleft deformities.

Classification and diagnosis of congenital craniofacial cleft deformities are helpful in discerning the severity of the deformity and providing guidance for surgical repair. Eighty-one cases of congenital craniofacial cleft deformity were analyzed using the Tessier classification. Depending on the location, status of the deformity, and results of examinations such as computed tomography, according to the range affected, the location and status of the deformity were designated by the STO classification, with S for skin, T for soft tissue, and O for os (craniofacial bone). The severity of the deformity is delineated by Arabic numerals. The analysis of 81 cases by the STO classification method showed that suborbital deformities mainly were Tessier 3 and 4 clefts (24.70%) and supraorbital deformities mainly were Tessier 9 and 10 clefts (38.27%). There was no definite regular pattern for the affected extent of tissues. STO classification can be a supplement to Tessier classification and can provide references for the surgical repair of craniofacial cleft deformity.

Adolescent↗
PubMed2006

Orbital fractures: a new classification and staging of 190 patients.

The orbit is located in the middle third of the face, composed of several bones and surrounded by complex anatomic structures so that orbital fractures (OF) often involve other parts of the face. A staging system for classifying OF is of paramount importance in order to exchange information between trauma centers. Several classifications have been proposed for describing OF but they have not a single method applicable to the whole orbit. Here, a classification for OF that can be summarized with four abbreviations is proposed. Four letters define the localization (F = frontal, N = nasal, M = maxillary and Z = zygomatic bone fracture), two acronyms describe fragment shift (in = blow-in or out = blow-out), four numbers define ocular movement impairment (1 = superior, 2 = internal, 3 = inferior, and 4 = external extrinsic muscular deficit) and two acronyms describe eye position (EX = exophthalmos and ENO = enophthalmos). To evaluate the suitability of the proposed classification a retrospective study on a series of 190 OFs is performed. Age, gender, new stage, clinical diagnosis at admission, type of surgery, and need for graft for orbital reconstruction are considered. A good correlation between the proposed classification and the studied variables is detected. In conclusion, the proposed classification is a simply and precise method to stage OF. It can summarize OF and be used in the daily practice. However, it is our belief that a multi-center study should be performed before the effectiveness of the proposed classification can be clearly stated.

Adolescent↗
PubMed2006

Disparities in the classification of esophageal and cardia adenocarcinomas and their influence on reported incidence rates.

OBJECTIVE: To evaluate the diagnostic accuracy of esophageal and cardia adenocarcinoma in the Swedish Cancer Register. SUMMARY BACKGROUND DATA: Based on cancer registers, a rising incidence of esophageal and cardia adenocarcinoma has been reported in several populations, but possible influence of differences in tumor classification has not been evaluated. METHODS: In a nationwide study in 1995 through 1997, all Swedish patients, born in Sweden and younger than 80 years with esophageal or cardia adenocarcinoma and half of all patients with esophageal squamous cell carcinoma, were prospectively, uniformly, and thoroughly classified. This study classification was compared with the tumor classification in the Swedish Cancer Register, which is based on routine clinical practice. RESULTS: The overall completeness of the Cancer Register was high (98.3%), whereas the site-specific completeness of the Register was 63% for esophageal adenocarcinoma, 74% for cardia adenocarcinoma, and 91% for esophageal squamous cell carcinoma. The incidence of esophageal adenocarcinomas was 16% higher in the study classification compared with that of the Register during the study period, whereas the incidence of cardia adenocarcinoma was 2% lower in the study classification. CONCLUSIONS: There is a diagnostic mismatch between esophageal and cardia adenocarcinoma in the clinical setting and, therefore, also in Cancer Registers. In etiologic and therapeutic research, this problem needs consideration, since these tumors have distinct risk factor profiles and could be subjected to different treatment strategies. The increasing incidence rate of esophageal adenocarcinoma in Sweden is unlikely to be explained by such differences in tumor classification, however.

Adenocarcinoma↗
PubMedDate not supplied

Physeal fractures: Part 3. Classification.

Over the past 100 years, several attempts to classify physeal fractures have been made. Each new classification has made changes to the previously existing classifications. After review of these classifications and of data collected from a population-based study (see Physeal Fractures: Part 1. Epidemiology in Olmsted County, Minnesota, 1979-1988, pp. 423-30), a new classification was constructed. This classification includes two new fractures (see Physeal Fractures: Part 2. Two Previously Unclassified Types, pp. 431-38). This classification has sound anatomic, epidemiologic, and prognostic bases.

Child↗
PubMed1990

Comparison of pattern recognition methods for computer-assisted classification of spectra of heart sounds in patients with a porcine bioprosthetic valve implanted in the mitral position.

The diagnostic performance of two pattern recognition methods (or classifiers) to detect valvular degeneration was evaluated in 48 patients with a porcine bioprosthetic heart valve inserted in the mitral position. Twenty patients had a normal porcine bioprosthetic valve and 28 patients had a degenerated bioprosthetic valve. One method was based on the Gaussian-Bayes model and the second on the "nearest neighbor" algorithm using three distance measurements. Eighteen diagnostic features were extracted from the sound spectrum of each patient and, for each method, a two-class supervised learning approach was used to determine the most discriminant diagnostic patterns composed of 6 features or less. The probability of error of the classifiers was estimated with the leave-one-out approach. The performance of each method to discriminate between normal and degenerated bioprosthetic valves was verified by clinical evaluation of the valves. The best performance in evaluation of the sound spectrum (98% correct classifications) was obtained with the Bayes classifier and two patterns of six features each. The percentage of false positive classifications of valve degeneration was 0% and the percentage of false negative classifications was 4%. Sensitivity for the detection of valve degeneration was 96%, specificity was 100%, positive predictive value was 100%, and negative predictive value was 95%. The best performance of the nearest neighbor method (94% correct classifications) was obtained by using the Mahalanobis distance and five patterns composed of three, four, five, or six diagnostic features. Using a pattern composed of only three features, the percentage of false positive classifications for degeneration was 10% and the percentage of false negative classifications was 4%.(ABSTRACT TRUNCATED AT 250 WORDS)

Bayes Theorem↗
PubMed2006

Texture characterization for joint compression and classification based on human perception in the wavelet domain.

Today's multimedia applications demand sophisticated compression and classification techniques in order to store, transmit, and retrieve audio-visual information efficiently. Over the last decade, perceptually based image compression methods have been gaining importance. These methods take into account the abilities (and the limitations) of human visual perception (HVP) when performing compression. The upcoming MPEG 7 standard also addresses the need for succinct classification and indexing of visual content for efficient retrieval. However, there has been no research that has attempted to exploit the characteristics of the human visual system to perform both compression and classification jointly. One area of HVP that has unexplored potential for joint compression and classification is spatial frequency perception. Spatial frequency content that is perceived by humans can be characterized in terms of three parameters, which are: 1) magnitude; 2) phase; and 3) orientation. While the magnitude of spatial frequency content has been exploited in several existing image compression techniques, the novel contribution of this paper is its focus on the use of phase coherence for joint compression and classification in the wavelet domain. Specifically, this paper describes a human visual system-based method for measuring the degree to which an image contains coherent (perceptible) phase information, and then exploits that information to provide joint compression and classification. Simulation results that demonstrate the efficiency of this method are presented.

Algorithms↗
PubMed2006

Principal components null space analysis for image and video classification.

We present a new classification algorithm, principal component null space analysis (PCNSA), which is designed for classification problems like object recognition where different classes have unequal and nonwhite noise covariance matrices. PCNSA first obtains a principal components subspace (PCA space) for the entire data. In this PCA space, it finds for each class "i," an Mi-dimensional subspace along which the class' intraclass variance is the smallest. We call this subspace an approximate null space (ANS) since the lowest variance is usually "much smaller" than the highest. A query is classified into class "i" if its distance from the class' mean in the class' ANS is a minimum. We derive upper bounds on classification error probability of PCNSA and use these expressions to compare classification performance of PCNSA with that of subspace linear discriminant analysis (SLDA). We propose a practical modification of PCNSA called progressive-PCNSA that also detects "new" (untrained classes). Finally, we provide an experimental comparison of PCNSA and progressive PCNSA with SLDA and PCA and also with other classification algorithms-linear SVMs, kernel PCA, kernel discriminant analysis, and kernel SLDA, for object recognition and face recognition under large pose/expression variation. We also show applications of PCNSA to two classification problems in video--an action retrieval problem and abnormal activity detection.

Algorithms↗
PubMed2006

Activity classification using realistic data from wearable sensors.

Automatic classification of everyday activities can be used for promotion of health-enhancing physical activities and a healthier lifestyle. In this paper, methods used for classification of everyday activities like walking, running, and cycling are described. The aim of the study was to find out how to recognize activities, which sensors are useful and what kind of signal processing and classification is required. A large and realistic data library of sensor data was collected. Sixteen test persons took part in the data collection, resulting in approximately 31 h of annotated, 35-channel data recorded in an everyday environment. The test persons carried a set of wearable sensors while performing several activities during the 2-h measurement session. Classification results of three classifiers are shown: custom decision tree, automatically generated decision tree, and artificial neural network. The classification accuracies using leave-one-subject-out cross validation range from 58 to 97% for custom decision tree classifier, from 56 to 97% for automatically generated decision tree, and from 22 to 96% for artificial neural network. Total classification accuracy is 82 % for custom decision tree classifier, 86% for automatically generated decision tree, and 82% for artificial neural network.

Activities of Daily Living↗
PubMed2005

The new World Health Organization-European Organization for Research and Treatment of Cancer classification for cutaneous lymphomas: a practical marriage of two giants.

Following consensus meetings of the two parent organizations, a new World Health Organization-European Organization for Research and Treatment of Cancer (WHO-EORTC) classification for primary cutaneous lymphomas has recently been published. This important development will now end the ongoing debate as to which of these was the preferred classification. The new classification will facilitate more uniformity in diagnosis, management and treatment of cutaneous lymphomas. In particular, it provides a useful distinction between indolent and more aggressive types of primary cutaneous lymphoma and provides practical advice on preferred management and treatment regimens. This will thereby prevent patients receiving high-grade treatment for low-grade biological disease. This review focuses on those diseases which have found new consensus agreement compared with the original WHO and EORTC classifications. In cutaneous T-cell lymphomas, these include folliculotropic mycosis fungoides, defining features of Sézary syndrome, primary cutaneous CD30+ lymphoproliferative disorders (primary cutaneous anaplastic large cell lymphoma, lymphomatoid papulosis and borderline lesions) and subcutaneous panniculitis-like T-cell lymphoma. Primary cutaneous CD4+ small/medium-sized pleomorphic T-cell lymphoma, primary cutaneous aggressive epidermotropic CD8+ T-cell lymphoma and cutaneous gamma/delta T-cell lymphoma are allocated provisional entry status and thereby afford better definitions for some cases of currently unspecified primary cutaneous peripheral T-cell lymphoma. In cutaneous B-cell lymphomas, diseases which have found new consensus agreement include primary cutaneous marginal zone B-cell lymphoma, primary cutaneous follicular centre lymphoma, primary cutaneous diffuse large B-cell lymphoma, leg type and primary cutaneous diffuse large B-cell lymphoma, other. CD4+/CD56+ haematodermic neoplasm (early plasmacytoid dendritic cell leukaemia/lymphoma) now appears as a precursor haematological neoplasm and replaces the previous terminology of blastic NK-cell lymphoma. Other haematopoietic and lymphoid tumours involving the skin, as part of systemic disease, will appear in the forthcoming WHO publication Tumours of the Skin. The new classification raises interesting new problems and questions about primary cutaneous lymphoma and some of these are discussed in this article. It is, however, a splendid signpost indicating the direction in which research in cutaneous lymphoma needs to go. In the interim, we have an international consensus classification which is clinically meaningful.

Humans↗