Behzad Eftekhar

Comparison of artificial neural network and logistic regression models for prediction of mortality in head trauma based on initial clinical data

BackgroundIn recent years, outcome prediction models using artificial neural network and multivariable logistic regression analysis have been developed in many areas of health care research. Both these methods have advantages and disadvantages. In this study we have compared the performance of artificial neural network and multivariable logistic regression models, in prediction of outcomes in head trauma and studied the reproducibility of the findings.Methods1000 Logistic regression and ANN models based on initial clinical data related to the GCS, tracheal intubation status, age, systolic blood pressure, respiratory rate, pulse rate, injury severity score and the outcome of 1271 mainly head injured patients were compared in this study. For each of one thousand pairs of ANN and logistic models, the area under the receiver operating characteristic (ROC) curves, Hosmer-Lemeshow (HL) statistics and accuracy rate were calculated and compared using paired T-tests.ResultsANN significantly outperformed logistic models in both fields of discrimination and calibration but under performed in accuracy. In 77.8% of cases the area under the ROC curves and in 56.4% of cases the HL statistics for the neural network model were superior to that for the logistic model. In 68% of cases the accuracy of the logistic model was superior to the neural network model.ConclusionsANN significantly outperformed the logistic models in both fields of discrimination and calibration but lagged behind in accuracy. This study clearly showed that any single comparison between these two models might not reliably represent the true end results. External validation of the designed models, using larger databases with different rates of outcomes is necessary to get an accurate measure of performance outside the development population.

Are the distributions of variations of circle of Willis different in different populations? – Results of an anatomical study and review of literature

BackgroundPrevious studies have proposed correlation between variants of the cerebral arterial circle (also known as circle of Willis) and some cerebrovascular diseases. Differences in the incidence of these diseases in different populations have also been investigated. The study of variations in the anatomy of the cerebral arterial circle may partially explain differences in the incidence of some of the cerebrovascular diseases in different ethnic or racial groups.While many studies have investigated the variations in the anatomy of each segment of the cerebral arterial circle, few have addressed the variants of the cerebral arterial circle as a whole. Similarly, the frequency of occurrence of such variants in different ethnic or racial groups has not been compared.Methods102 brains of recently deceased Iranian males were dissected, in order to observe variations in the anatomy of the cerebral arterial circle. The dissection process was recorded on film and digitized. One resized picture from each dissection, showing complete circle has been made available online. The variations of the circle as whole and segmental variations were compared with previous studies.ResultsOn the whole, the frequencies of the different variants of the entire cerebral arterial circle and segmental variations were comparable with previous studies.More specifically variants with uni- and bilateral hypoplasia of posterior communicating arteries were the most common in our study, similar to the previous works. No hypoplasia of the precommunicating part of the left anterior cerebral artery (A1), aplasia of A1 or the precommunicating part of the posterior cerebral artery (P1) was seen. In 3% both right and left posterior communcating arteries were absent.ConclusionThe anatomical variations found in the cerebral arterial circle of the Iranian males in the current study were not significantly different to those of more diverse populations reported in the literature. While taking into account potential confounding factors, the authors conclude that based on available studies, there is no evidence suggesting that the distributions of the variations of cerebral arterial circle differ in different populations.

Surgical simulation software for insertion of pedicle screws

OBJECTIVE As the first step toward finding noninvasive alternatives to the traditional methods of surgical training, we have developed a small, stand-alone computer program that simulates insertion of pedicle screws in different spinal vertebrae (T10–L5). METHODS We used Delphi 5.0 and DirectX 7.0 extension for Microsoft Windows. This is a stand-alone and portable program. RESULTS The program can run on most personal computers. It provides the trainee with visual feedback during practice of the technique. At present, it uses predefined three-dimensional images of the vertebrae, but we are attempting to adapt the program to three-dimensional objects based on real computed tomographic scans of the patients. The program can be downloaded at no cost from the web site: www.tums.ac.ir/downloads CONCLUSION As a preliminary work, it requires further development, particularly toward better visual, auditory, and even proprioceptive feedback and use of the individual patient’s data.

Spinal subdural hematoma revealing hemophilia A in a child: A case report

BackgroundIntraspinal bleeding especially in the form of subdural hematoma is rare in hemophiliacs. In the present case, we report a neglected hemophilic A child with such a problem and discuss its management options.Case PresentationA 9-year old hemophilic A boy presented with quadriparesis, confusion and meningismus after a fall 4 days previously. There was no sign of direct trauma to his back. His CT Scan and MRI showed spinal extramedullary hematoma extended from C5 to L2. We corrected the factor VIII level, but two days later, the patients lower limbs weakened to 1/5 proximally as well as distally. We performed a laminectomy from T11 to L2, according to the level of the maximal neurological deficit and recent deterioration course. The subdural hematoma was evacuated. The hematoma in other spinal levels was managed conservatively. In the week following the operation, the patients neurological status approached normal.ConclusionThis case calls attention to the clinical manifestation, radiological features and management options of the rarely reported intraspinal hematoma in hemophilic children. Although this case has been managed operatively for its hematoma in the thoracolumbar region, at the same time it can be considered a successful case of conservative management of intraspinal hematoma in the cervicothoracic region. Both conservative and surgical management could be an option in managing these patients considering their neurological course.

Prophylactic antibiotic for prevention of posttraumatic meningitis after traumatic pneumocephalus: design and rationale of a placebo-controlled randomized multicenter trial [ISRCTN71132784]

BackgroundThe purpose of this study is to compare the efficacy of prophylactic antibiotic for prevention of meningitis in acute traumatic pneumocephalus patients.MethodsIn this prospective, randomized controlled clinical trial, 200 selected head injury patients with traumatic pneumocephalus are randomly assigned to receive intravenous antibiotics (2 grams Ceftriaxone twice a day), oral antibiotics (Azithromycin) or placebo for at least 7 days after trauma. The patients will be followed for one month posttrauma.ConclusionThe authors hope that this study helps clarifying the effectiveness and indications of antibiotics in prevention of meningitis in traumatic pneumocephalus after head injury and in specific subgroup of these patients.

Play dough as an educational tool for visualization of complicated cerebral aneurysm anatomy

BackgroundImagination of the three-dimensional (3D) structure of cerebral vascular lesions using two-dimensional (2D) angiograms is one of the skills that neurosurgical residents should achieve during their training. Although ongoing progress in computer software and digital imaging systems has facilitated viewing and interpretation of cerebral angiograms enormously, these facilities are not always available.MethodsWe have presented the use of play dough as an adjunct to the teaching armamentarium for training in visualization of cerebral aneurysms in some cases.ResultsThe advantages of play dough are low cost, availability and simplicity of use, being more efficient and realistic in training the less experienced resident in comparison with the simple drawings and even angiographic views from different angles without the need for computers and similar equipment. The disadvantages include the psychological resistance of residents to the use of something in surgical training that usually is considered to be a toy, and not being as clean as drawings or computerized images.ConclusionAlthough technology and computerized software using the patients own imaging data seems likely to become more advanced in the future, use of play dough in some complicated cerebral aneurysm cases may be helpful in 3D reconstruction of the real situation.

Vestibular schwannoma with contralateral facial pain – case report

BackgroundVestibular schwannoma (acoustic neuroma) most commonly presents with ipsilateral disturbances of acoustic, vestibular, trigeminal and facial nerves. Presentation of vestibular schwannoma with contralateral facial pain is quite uncommon.Case presentationAmong 156 cases of operated vestibular schwannoma, we found one case with unusual presentation of contralateral hemifacial pain.ConclusionThe presentation of contralateral facial pain in the vestibular schwannoma is rare. It seems that displacement and distortion of the brainstem and compression of the contralateral trigeminal nerve in Meckels cave by the large mass lesion may lead to this atypical presentation. The best practice in these patients is removal of the tumour, although persistent contralateral pain after operation has been reported.

Bilateral asynchronous acute epidural hematoma : a case report.

BackgroundBilateral extradural hematomas have only rarely been reported in the literature. Even rarer are cases where the hematomas develop sequentially, one after removal of the other. Among 187 cases of operated epidural hematomas during past 4 years in our hospital, we found one case of sequentially developed bilateral epidural hematoma.Case PresentationAn 18-year-old conscious male worker was admitted to our hospital after a fall. After deterioration of his consciousness, an emergency brain CT scan showed a right temporoparietal epidural hematoma. The hematoma was evacuated, but the patient did not improve afterwards. Another CT scan showed contralateral epidural hematoma and the patient was reoperated. Postoperatively, the patient recovered completely.ConclusionsThis case underlines the need for monitoring after an operation for an epidural hematoma and the need for repeat brain CT scans if the patient does not recover quickly after removal of the hematoma, especially if the first CT scan has been done less than 6 hours after the trauma. Intraoperative brain swelling can be considered as a clue for the development of contralateral hematoma.