The AO comprehensive classification of pediatric long-bone fractures: a web-based multicenter agreement study.

The first AO comprehensive pediatric long-bone fracture classification system has been proposed following a structured path of development and validation with experienced pediatric surgeons. A Web-based multicenter agreement study involving 70 surgeons in 15 clinics and 5 countries was conducted to assess the reliability and accuracy of this classification when used by a wide range of surgeons with various levels of experience. Training was provided at each clinic before the session. Using the Internet, participants could log in at any time and classify 275 supracondylar, radius, and tibia fractures at their own pace. The fracture diagnosis was made following the hierarchy of the classification system using both clinical terminology and codes. kappa coefficients for the single-surgeon diagnosis of epiphyseal, metaphyseal, or diaphyseal fracture type were 0.66, 0.80, and 0.91, respectively. Median accuracy estimates for each bone and type were all greater than 80%. Depending on their experience and specialization, surgeons greatly varied in their ability to classify fractures. Pediatric training and at least 2 years of experience were associated with significant improvement in reliability and accuracy. Kappa coefficients for diagnosis of specific child patterns were 0.51, 0.63, and 0.48 for epiphyseal, metaphyseal, and diaphyseal fractures, respectively. Identified reasons for coding discrepancies were related to different understandings of terminology and definitions, as well as poor quality radiographic images. Results supported some minor adjustments in the coding of fracture type and child patterns. This classification system received wide acceptance and support among the surgeons involved. As long as appropriate training could be performed, the system classification was reliable, especially among surgeons with a minimum of 2 years of clinical experience. We encourage broad-based consultation between surgeons' international societies and the use of this classification system in the context of clinical practice as well as prospectively for clinical studies.

Comparison between standard radiography and spiral CT with 3D reconstruction in the evaluation, classification and management of tibial plateau fractures.

The aim of this study was to compare the diagnostic efficiency of plain film and spiral CT examinations with 3D reconstructions of 42 tibial plateau fractures and to assess the accuracy of these two techniques in the pre-operative surgical plan in 22 cases. Forty-two tibial plateau fractures were examined with plain film (anteroposterior, lateral, two obliques) and spiral CT with surface-shaded-display 3D reconstructions. The Swiss AO-ASIF classification system of bone fracture from Muller was used. In 22 cases the surgical plans and the sequence of reconstruction of the fragments were prospectively determined with both techniques, successively, and then correlated with the surgical reports and post-operative plain film. The fractures were underestimated with plain film in 18 of 42 cases (43%). Due to the spiral CT 3D reconstructions, and precise pre-operative information, the surgical plans based on plain film were modified and adjusted in 13 cases among 22 (59%). Spiral CT 3D reconstructions give a better and more accurate demonstration of the tibial plateau fracture and allows a more precise pre-operative surgical plan.

AWSum-Combining Classification with Knowledge Aquisition

Many classifiers achieve high levels of accuracy but have limited applicability in real world situations because they do not lead to a greater understanding or insight into the^way features influence the classification. In areas such as health informatics a classifier that clearly identifies the influences on classification can be used to direct research and formulate interventions. This research investigates the practical applications of Automated Weighted Sum, (AWSum), a classifier that provides accuracy comparable to other techniques whilst providing insight into the data. This is achieved by calculating a weight for each feature value that represents its influence on the class value. The merits of this approach in classification and insight are evaluated on a Cystic Fibrosis and Diabetes datasets with positive results.

Active learning methods for remote sensing image classification

In this paper, we propose two active learning algorithms for semiautomatic definition of training samples in remote sensing image classification. Based on predefined heuristics, the classifier ranks the unlabeled pixels and automatically chooses those that are considered the most valuable for its improvement. Once the pixels have been selected, the analyst labels them manually and the process is iterated. Starting with a small and nonoptimal training set, the model itself builds the optimal set of samples which minimizes the classification error. We have applied the proposed algorithms to a variety of remote sensing data, including very high resolution and hyperspectral images, using support vector machines. Experimental results confirm the consistency of the methods. The required number of training samples can be reduced to 10% using the methods proposed, reaching the same level of accuracy as larger data sets. A comparison with a state-of-the-art active learning method, margin sampling, is provided, highlighting advantages of the methods proposed. The effect of spatial resolution and separability of the classes on the quality of the selection of pixels is also discussed.

Analysis, modelling and classification of geospatial data using machine learning

The research considers the problem of spatial data classification using machine learning algorithms: probabilistic neural networks (PNN) and support vector machines (SVM). As a benchmark model simple k-nearest neighbor algorithm is considered. PNN is a neural network reformulation of well known nonparametric principles of probability density modeling using kernel density estimator and Bayesian optimal or maximum a posteriori decision rules. PNN is well suited to problems where not only predictions but also quantification of accuracy and integration of prior information are necessary. An important property of PNN is that they can be easily used in decision support systems dealing with problems of automatic classification. Support vector machine is an implementation of the principles of statistical learning theory for the classification tasks. Recently they were successfully applied for different environmental topics: classification of soil types and hydro-geological units, optimization of monitoring networks, susceptibility mapping of natural hazards. In the present paper both simulated and real data case studies (low and high dimensional) are considered. The main attention is paid to the detection and learning of spatial patterns by the algorithms applied.

Monitoring network optimisation for spatial data classification using support vector machines

The paper presents a novel method for monitoring network optimisation, based on a recent machine learning technique known as support vector machine. It is problem-oriented in the sense that it directly answers the question of whether the advised spatial location is important for the classification model. The method can be used to increase the accuracy of classification models by taking a small number of additional measurements. Traditionally, network optimisation is performed by means of the analysis of the kriging variances. The comparison of the method with the traditional approach is presented on a real case study with climate data.

Comparison and validation of tissue modelization and statistical classification methods in T1-weighted MR brain images.

This paper presents a validation study on statistical nonsupervised brain tissue classification techniques in magnetic resonance (MR) images. Several image models assuming different hypotheses regarding the intensity distribution model, the spatial model and the number of classes are assessed. The methods are tested on simulated data for which the classification ground truth is known. Different noise and intensity nonuniformities are added to simulate real imaging conditions. No enhancement of the image quality is considered either before or during the classification process. This way, the accuracy of the methods and their robustness against image artifacts are tested. Classification is also performed on real data where a quantitative validation compares the methods' results with an estimated ground truth from manual segmentations by experts. Validity of the various classification methods in the labeling of the image as well as in the tissue volume is estimated with different local and global measures. Results demonstrate that methods relying on both intensity and spatial information are more robust to noise and field inhomogeneities. We also demonstrate that partial volume is not perfectly modeled, even though methods that account for mixture classes outperform methods that only consider pure Gaussian classes. Finally, we show that simulated data results can also be extended to real data.

Octopus Standard Automated Perimetry, Pulsar Perimetry, Moorfields Motion Displacement Test and Heidelberg Retinal Tomography in Glaucoma Detection - A Comparison of Diagnostic Accuracy

Purpose: To investigate the accuracy of 4 clinical instruments in the detection of glaucomatous damage. Methods: 102 eyes of 55 test subjects (Age mean = 66.5yrs, range = [39; 89]) underwent Heidelberg Retinal Tomography (HRTIII), (disc area<2.43); and standard automated perimetry (SAP) using Octopus (Dynamic); Pulsar (TOP); and Moorfields Motion Displacement Test (MDT) (ESTA strategy). Eyes were separated into three groups 1) Healthy (H): IOP<21mmHg and healthy discs (clinical examination), 39 subjects, 78 eyes; 2) Glaucoma suspect (GS): Suspicious discs (clinical examination), 12 subjects, 15 eyes; 3) Glaucoma (G): progressive structural or functional loss, 14 subjects, 20 eyes. Clinical diagnostic precision was examined using the cut-off associated with the p<5% normative limit of MD (Octopus/Pulsar), PTD (MDT) and MRA (HRT) analysis. The sensitivity, specificity and accuracy were calculated for each instrument. Results: See table Conclusions: Despite the advantage of defining glaucoma suspects using clinical optic disc examination, the HRT did not yield significantly higher accuracy than functional measures. HRT, MDT and Octopus SAP yielded higher accuracy than Pulsar perimetry, although results did not reach statistical significance. Further studies are required to investigate the structure-function correlations between these instruments.

Epidemiology of multiple congenital anomalies in Europe: a EUROCAT population-based registry study.

BACKGROUND: This study describes the prevalence, associated anomalies, and demographic characteristics of cases of multiple congenital anomalies (MCA) in 19 population-based European registries (EUROCAT) covering 959,446 births in 2004 and 2010. METHODS: EUROCAT implemented a computer algorithm for classification of congenital anomaly cases followed by manual review of potential MCA cases by geneticists. MCA cases are defined as cases with two or more major anomalies of different organ systems, excluding sequences, chromosomal and monogenic syndromes. RESULTS: The combination of an epidemiological and clinical approach for classification of cases has improved the quality and accuracy of the MCA data. Total prevalence of MCA cases was 15.8 per 10,000 births. Fetal deaths and termination of pregnancy were significantly more frequent in MCA cases compared with isolated cases (p < 0.001) and MCA cases were more frequently prenatally diagnosed (p < 0.001). Live born infants with MCA were more often born preterm (p < 0.01) and with birth weight < 2500 grams (p < 0.01). Respiratory and ear, face, and neck anomalies were the most likely to occur with other anomalies (34% and 32%) and congenital heart defects and limb anomalies were the least likely to occur with other anomalies (13%) (p < 0.01). However, due to their high prevalence, congenital heart defects were present in half of all MCA cases. Among males with MCA, the frequency of genital anomalies was significantly greater than the frequency of genital anomalies among females with MCA (p < 0.001). CONCLUSION: Although rare, MCA cases are an important public health issue, because of their severity. The EUROCAT database of MCA cases will allow future investigation on the epidemiology of these conditions and related clinical and diagnostic problems.

Urban image classification with semisupervised multiscale cluster kernels

This paper presents a semisupervised support vector machine (SVM) that integrates the information of both labeled and unlabeled pixels efficiently. Method's performance is illustrated in the relevant problem of very high resolution image classification of urban areas. The SVM is trained with the linear combination of two kernels: a base kernel working only with labeled examples is deformed by a likelihood kernel encoding similarities between labeled and unlabeled examples. Results obtained on very high resolution (VHR) multispectral and hyperspectral images show the relevance of the method in the context of urban image classification. Also, its simplicity and the few parameters involved make the method versatile and workable by unexperienced users.

Accuracy of Brain Multimodal Monitoring to Detect Cerebral Hypoperfusion After Traumatic Brain Injury.

OBJECTIVE:: To examine the accuracy of brain multimodal monitoring-consisting of intracranial pressure, brain tissue PO2, and cerebral microdialysis-in detecting cerebral hypoperfusion in patients with severe traumatic brain injury. DESIGN:: Prospective single-center study. PATIENTS:: Patients with severe traumatic brain injury. SETTING:: Medico-surgical ICU, university hospital. INTERVENTION:: Intracranial pressure, brain tissue PO2, and cerebral microdialysis monitoring (right frontal lobe, apparently normal tissue) combined with cerebral blood flow measurements using perfusion CT. MEASUREMENTS AND MAIN RESULTS:: Cerebral blood flow was measured using perfusion CT in tissue area around intracranial monitoring (regional cerebral blood flow) and in bilateral supra-ventricular brain areas (global cerebral blood flow) and was matched to cerebral physiologic variables. The accuracy of intracranial monitoring to predict cerebral hypoperfusion (defined as an oligemic regional cerebral blood flow < 35 mL/100 g/min) was examined using area under the receiver-operating characteristic curves. Thirty perfusion CT scans (median, 27 hr [interquartile range, 20-45] after traumatic brain injury) were performed on 27 patients (age, 39 yr [24-54 yr]; Glasgow Coma Scale, 7 [6-8]; 24/27 [89%] with diffuse injury). Regional cerebral blood flow correlated significantly with global cerebral blood flow (Pearson r = 0.70, p < 0.01). Compared with normal regional cerebral blood flow (n = 16), low regional cerebral blood flow (n = 14) measurements had a higher proportion of samples with intracranial pressure more than 20 mm Hg (13% vs 30%), brain tissue PO2 less than 20 mm Hg (9% vs 20%), cerebral microdialysis glucose less than 1 mmol/L (22% vs 57%), and lactate/pyruvate ratio more than 40 (4% vs 14%; all p < 0.05). Compared with intracranial pressure monitoring alone (area under the receiver-operating characteristic curve, 0.74 [95% CI, 0.61-0.87]), monitoring intracranial pressure + brain tissue PO2 (area under the receiver-operating characteristic curve, 0.84 [0.74-0.93]) or intracranial pressure + brain tissue PO2+ cerebral microdialysis (area under the receiver-operating characteristic curve, 0.88 [0.79-0.96]) was significantly more accurate in predicting low regional cerebral blood flow (both p < 0.05). CONCLUSION:: Brain multimodal monitoring-including intracranial pressure, brain tissue PO2, and cerebral microdialysis-is more accurate than intracranial pressure monitoring alone in detecting cerebral hypoperfusion at the bedside in patients with severe traumatic brain injury and predominantly diffuse injury.

Accuracy of a new transmittance-reflectance pulse oximetry sensor in critically ill neonates.

OBJECTIVE: To test the accuracy of a new pulse oximeter sensor based on transmittance and reflectance. This sensor makes transillumination of tissue unnecessary and allows measurements on the hand, forearm, foot, and lower limb. DESIGN: Prospective, open, nonrandomized criterion standard study. SETTING: Neonatal intensive care unit, tertiary care center. PATIENTS: Sequential sample of 54 critically ill neonates (gestational age 27 to 42 wks; postnatal age 1 to 28 days) with arterial catheters in place. MEASUREMENTS AND MAIN RESULTS: A total of 99 comparisons between pulse oximetry and arterial saturation were obtained. Comparison of femoral or umbilical arterial blood with transcutaneous measurements on the lower limb (n = 66) demonstrated an excellent correlation (r2 = .96). The mean difference was +1.44% +/- 3.51 (SD) % (range -11% to +8%). Comparison of the transcutaneous values with the radial artery saturation from the corresponding upper limb (n = 33) revealed a correlation coefficient of 0.94 with a mean error of +0.66% +/- 3.34% (range -6% to +7%). The mean difference between noninvasive and invasive measurements was least with the test sensor on the hand, intermediate on the calf and arm, and greatest on the foot. The mean error and its standard deviation were slightly larger for arterial saturation values < 90% than for values > or = 90%. CONCLUSION: Accurate pulse oximetry saturation can be acquired from the hand, forearm, foot, and calf of critically ill newborns using this new sensor.

Computer-assisted placement technique in hip resurfacing arthroplasty: improvement in accuracy?

Freehand positioning of the femoral drill guide is difficult during hip resurfacing and the surgeon is often unsure of the implant position achieved peroperatively. The purpose of this study was to find out whether, by using a navigation system, acetabular and femoral component positioning could be made easier and more precise. Eighteen patients operated on by the same surgeon were matched by sex, age, BMI, diagnosis and ASA score (nine patients with computer assistance, nine with the regular ancillary). Pre-operative planning was done on standard AP and axial radiographs with CT scan views for the computer-assisted operations. The final position of implants was evaluated by the same radiographs for all patients. The follow-up was at least 1 year. No difference between both groups in terms of femoral component position was observed (p > 0.05). There was also no difference in femoral notching. A trend for a better cup position was observed for the navigated hips, especially for cup anteversion. There was no additional operating time for the navigated hips. Hip navigation for resurfacing surgery may allow improved visualisation and hip implant positioning, but its advantage probably will be more obvious with mini-incisions than with regular incision surgery.