Computational methods for the detection of facial palsy

We are developing a telemedicine application which offers automated diagnosis of facial (Bell's) palsy through a Web service. We used a test data set of 43 images of facial palsy patients and 44 normal people to develop the automatic recognition algorithm. Three different image pre-processing methods were used. Machine learning techniques (support vector machine, SVM) were used to examine the difference between the two halves of the face. If there was a sufficient difference, then the SVM recognized facial palsy. Otherwise, if the halves were roughly symmetrical, the SVM classified the image as normal. It was found that the facial palsy images had a greater Hamming Distance than the normal images, indicating greater asymmetry. The median distance in the normal group was 331 (interquartile range 277-435) and the median distance in the facial palsy group was 509 (interquartile range 334-703). This difference was significant (P

Prediction of height from knee height in children with cerebral palsy and non-disabled children

Measurement of height or length is essential in the assessment of nutritional status. In some conditions, for example cerebral palsy (CP), such measurements may be difficult or impossible. Proxy measurements such as knee height have been used to predict height in such cases. We have evaluated two equations in the literature that predict stature from knee height in a group of 17 children with CP and 20 non-disabled children. The two equations performed well on average in the non-disabled children, with the mean predicted height being within 1% of the mean measured height. Nevertheless, the limits of agreement were relatively large. This was also the case for the children with CP. Thus the equations may be accurate at the group level; however they may lead to unacceptable error at the individual level. © 2006 Informa UK Ltd.

Maternal antecedents for cerebral palsy in extremely preterm babies: A case-controlled study

The study aimed to identify significant antenatal risk factors for cerebral palsy (CP) among extremely preterm infants with a matched case-control design. Infants born between 1989 and 1996 at 24 to 27 weeks' gestation who survived to hospital discharge were evaluated: 30 with a proven diagnosis of CP at 2 years corrected for prematurity and 120 control children matched for gestational age without CP. Information on maternal obstetric risk factors and medication was obtained. Matched analyses were performed and odds ratios (OR) and 95% confidence intervals (CI) were calculated. An antenatal diagnosis of intrauterine growth restriction was associated with an increased risk of CP (OR 6.6; 95% CI 1.8 to 25.2), while maternal administration of corticosteroids was associated with a reduced risk of CP (OR 0.4; 95% CI 0.1 to 0.98). A high rate of placental histopathology was achieved but no relation between clinical or histological chorioamnionitis or funisitis and CP was demonstrated. Maternal preeclampsia was not associated with a statistically significant reduction in the risk of CP. It is concluded that a reduced risk of CP in extremely preterm infants is associated with the antenatal use of corticosteroids.

Violation of multiparticle Bell inequalities for low- and high-flux parametric amplification using both vacuum and entangled input states

We show how polarization measurements on the output fields generated by parametric down conversion will reveal a violation of multiparticle Bell inequalities, in the regime of both low- and high-output intensity. In this case, each spatially separated system, upon which a measurement is performed, is comprised of more than one particle. In view of the formal analogy with spin systems, the proposal provides an opportunity to test the predictions of quantum mechanics for spatially separated higher spin states. Here the quantum behavior possible even where measurements are performed on systems of large quantum (particle) number may be demonstrated. Our proposal applies to both vacuum-state signal and idler inputs, and also to the quantum-injected parametric amplifier as studied by De Martini The effect of detector inefficiencies is included, and weaker Bell-Clauser-Horne inequalities are derived to enable realistic tests of local hidden variables with auxiliary assumptions for the multiparticle situation.

Leonard Bell Cox (1894-1976) - pioneer of Australian clinical neurology

The name of Leonard Bell Cox (1894-1976) will long be associated with a number of very significant areas in the intellectual and cultural life of the Australian State of Victoria. A quarter of a century after his death, his cultural achievements, and the enduring products of these achievements, continue to be celebrated in his native city, Melbourne. However his enormous contributions in these cultural fields were matched by his perhaps less widely known achievements in medicine, in particular in the neurosciences. In his time he not only pioneered the foundation and progressive development of the speciality of clinical neurology in Australia, but at the same time became a recognised world expert on the pathology of brain tumours.