An artificial neural network approach to diagnosing epilepsy using lateralized bursts of theta EEGs.

Determining the cause of seizures is a significant medical problem, as misdiagnosis can result in increased morbidity and even mortality of patients. The reported research evaluates the efficacy of using an artificial neural network (ANN) for determining epileptic seizure occurrences for patients with lateralized bursts of theta (LBT) EEGs. Training and test cases are acquired from examining records of 1,500 consecutive adult seizure patients. The small resulting pool of 92 patients with LBT EEGs requires using a jack-knife procedure for developing the ANN categorization models. The ANNs are evaluated for accuracy, specificity, and sensitivity on classification of each patient into the correct two-group categorization: epileptic seizure or non-epileptic seizure. The original ANN model using eight variables produces a categorization accuracy of 62%. Following a modified factor analysis, an ANN model utilizing just four of the original variables achieves a categorization accuracy of 68%.

Redesigning the medical office for improved efficiency: an object-oriented event-driven messaging system.

Medical practitioners are under ever increasing pressure to maximize patient care, while minimizing costs. One productivity area that has not previously undergone thorough investigation is the efficient utilization of time for intra-office communication. Medical office personnel typically need to communicate patient information and resource requests, as well as personal messages. An intra-office communication system is designed that reduces time-waste typically incurred in medical office environments. Redesigning medical offices with intra-office communication systems provides time savings of several man hours per day. The subsequent increase in time efficiency enables higher quality of patient care and larger patient loads to be managed by existing medical staff.

Predicting pediatric length of stay and acuity of care in the first ten minutes with artificial neural networks.

OBJECTIVE: To evaluate the efficacy of artificial neural networks in categorizing pediatric trauma patients into four distinct acuity of care groups and in determining the length of stay (LOS) within specific areas of the hospital. DESIGN: Using historical information from >8,000 pediatric trauma patient records, train and evaluate artificial neural networks to predict the injury severity and LOS for each patient in pediatric intensive care units (PICUs), step-down units, and floor units. Each artificial neural network is evaluated for categorization accuracy and mean absolute error difference on the predicted LOS. SUBJECTS: A total of 10,353 patient records from the National Pediatric Trauma registry, representing all pediatric trauma patients treated at affiliated hospitals from April 1994 through December 1996. Records with incomplete information were eliminated from the study, leaving 8,081 usable patient records. MEASUREMENTS: A total of 14 variables are selected from the 81 values present in the National Pediatric Trauma Registry as independent variables for the artificial neural networks. Each neural network produces nine output values: five for categorizing the patient's injury severity, three for the LOS in the PICU, step-down unit, and floor units, and one for the patient's total LOS. RESULTS: A fuzzy ARTMAP neural network accurately categorizes 88% of mortality patients and 58.3% of critical PICU patients. A backpropagation neural network succeeded in predicting the total LOS to within 1 day for 51.4% and the ICU LOS to within 1 day for 70.4% of all evaluated patients. CONCLUSION: Information available in the first 10 mins of a patient's presentation at the emergency room can be used by an artificial neural network to predict injury severity and LOS. Artificial neural networks enable more effective resource planning and patient management.

A comparison of posttraumatic stress disorder in veterans with and without spinal cord injury.

The authors assessed effects of paraplegic and quadriplegic spinal cord injuries (SCIs) on posttraumatic stress disorder (PTSD) by comparing severity and prevalence of PTSD in these groups to a sample of controls who experienced traumatic injuries other than SCI. The authors found that veterans with quadriplegia reported significantly less severe current PTSD symptoms than controls who were not significantly different from veterans with paraplegia. These results suggest that sustaining a quadriplegic SCI decreases risk of current PTSD, whereas sustaining a paraplegic SCI is associated with greater risk of PTSD, although the risk is no greater than that incurred from experiencing the trauma itself.

Factors influencing stay in the postanesthesia care unit: a prospective analysis.

STUDY OBJECTIVE: To identify indicators of prolonged length of stay (LOS) in the postanesthesia care unit (PACU) and to test the following hypotheses: (1) that patient age, pain medication administration at the time of PACU admission, length of surgery, and cardiovascular, pulmonary, and pain responses postoperatively predict prolonged PACU LOS and (2) that cardiovascular and pulmonary symptoms preoperatively predict cardiovascular and pulmonary symptoms postoperatively. DESIGN: Prospective, observational analysis. SETTING: PACU of a university teaching hospital. PATIENTS: 1,067 patients scheduled for surgery with general anesthesia between February and September 1996, 18 years of age or older. MEASUREMENT AND MAIN RESULTS: 11.2% of the variation in prolonged PACU LOS can be predicted by age, pain medication at the time of PACU admission, and postoperative cardiovascular, pulmonary, and pain symptoms. A significant number of patients who did not report a prior history experienced postoperative cardiovascular and pulmonary symptoms. CONCLUSION: Patient history and postoperative symptoms predict only a small percentage of prolonged PACU stays. Organizational factors may be a more important predictor of prolonged PACU stay. Additionally, assessment of cardiovascular and pulmonary history needs refinement to improve prediction of patient responses postoperatively.

Use of an artificial neural network to predict length of stay in acute pancreatitis.

Length of stay (LOS) predictions in acute pancreatitis could be used to stratify patients with severe acute pancreatitis, make treatment and resource allocation decisions, and for quality assurance. Artificial neural networks have been used to predict LOS in other conditions but not acute pancreatitis. The hypothesis of this study was that a neural network could predict LOS in patients with acute pancreatitis. The medical records of 195 patients admitted with acute pancreatitis were reviewed. A backpropagation neural network was developed to predict LOS >7 days. The network was trained on 156 randomly selected cases and tested on the remaining 39 cases. The neural network had the highest sensitivity (75%) for predicting LOS >7 days. Ranson criteria had the highest specificity (94%) for making this prediction. All methods incorrectly predicted LOS in two patients with severe acute pancreatitis who died early in their hospital course. An artificial neural network can predict LOS >7 days. The network and traditional prognostic indices were least accurate for predicting LOS in patients with severe acute pancreatitis who died early in their hospital course. The neural network has the advantage of making this prediction using admission data.

Posttraumatic stress disorder in veterans with spinal cord injury: trauma-related risk factors.

Trauma-related risk factors for posttraumatic stress disorder (PTSD) were examined in a sample of 125 veterans with spinal cord injury. Category of injury was found to be the most consistent predictor of PTSD diagnosis and symptom severity with paraplegia predicting more PTSD symptoms than quadriplegia. The occurrence of a head injury at the time of the trauma was found to predict PTSD symptom severity measures, but not PTSD diagnosis. Trauma recency consistently predicted Impact of Event score (IES) and was found to be related to current PTSD severity and lifetime PTSD diagnosis in multiple but not simple regression models. Trauma severity was found to be significantly related to self-reported PTSD symptoms and lifetime PTSD diagnosis in simple but not in multiple regression analyses. Type of trauma, alcohol or other drug (AOD) use during the trauma and loss of consciousness (LOC) during the trauma were not consistently associated with PTSD symptom severity or diagnosis.

Neural network models for a resource allocation problem.

University admissions and business personnel offices use a limited number of resources to process an ever-increasing quantity of student and employment applications. Application systems are further constrained to identify and acquire, in a limited time period, those candidates who are most likely to accept an offer of enrolment or employment. Neural networks are a new methodology to this particular domain. Various neural network architectures and learning algorithms are analyzed comparatively to determine the applicability of supervised learning neural networks to the domain problem of personnel resource allocation and to identify optimal learning strategies in this domain. This paper focuses on multilayer perceptron backpropagation, radial basis function, counterpropagation, general regression, fuzzy ARTMAP, and linear vector quantization neural networks. Each neural network predicts the probability of enrolment and nonenrolment for individual student applicants. Backpropagation networks produced the best overall performance. Network performance results are measured by the reduction in counsellors student case load and corresponding increases in student enrolment. The backpropagation neural networks achieve a 56% reduction in counsellor case load.

Longitudinal follow-up of lung function from childhood to adolescence in prematurely born patients with neonatal chronic lung disease.

We investigated whether early lung function abnormalities in prematurely born children with a history of chronic lung disease improve in late childhood and adolescence. We performed a prospective, longitudinal evaluations of pulmonary function over an 8 year period. In seventeen patients from the age (mean +/- SD) of 8.2 +/- 1.2 years to the age of 15.1 +/- 1.6 years. They had been born at 29.1 +/- 1.9 weeks of gestation, with a birthweight of 1120 +/- 190 g, and they had received supplemental oxygen, with or without mechanical ventilation, for 40.4 +/- 23.8 days during the neonatal period. They all had radiographic evidence of chronic lung disease at 4 weeks of age. Annual measurements of lung volumes using the helium dilution technique, and of airway function with spirometry and maximal expiratory flow-volume curves over a 5 to 8 year period, were obtained. The results indicated that total lung capacity (TLC) and vital capacity (VC) were within the predicted normal range in all patients and increased over time. In contrast, the initially abnormal residual volume (RV) and RV/TLC ratio decreased over time, suggesting gradual resolution of air-trapping. The peak expiratory flow rate (PEFR), forced expiratory volume in 1 second (FEV1), and the ratio FEV1/FVC remained at or above the predicted normal range in all patients. FEF25-75, FEF50, and FEF75 were within normal limits in eight patients and abnormally low (more than 2 SD below the predicted normal value) in the remaining nine patients, indicating small airway obstruction. Eight of the nine patients with lower airway obstruction showed significant response to inhaled bronchodilator, and four responded to a histamine challenge. None of the eight patients with normal airway function responded to histamine, but four responded to bronchodilators. The perinatal history, family history of asthma, and exposure to smoking were similar in patients with and without airway obstruction. The height and weight were and remained within the normal range. We conclude that gradual normalization of air-trapping continues well into adolescence in virtually all patients with a history of prematurity and chronic lung disease. in contrast, airflow obstruction may persist but does not get worse later in life. Although chronic airflow obstruction probably is the consequence of injury to the small airways during the neonatal period, it is present in only some of the children, and it does not appear to be directly related to the perinatal history. Finally, there is evidence that airway hyperresponsiveness may be a contributing factor to the development and/or persistence of airflow obstruction in chronic lung disease of prematurity.

[Assessment of threatening behaviors in patients admitted to mental hospital]. / Ocena zachowan bezposrednio zagrazajacych u pacjentów przyjmowanych do szpitala psychiatrycznego.

Imminent threat presented by patients admitted to mental hospital was studied. The research data came from a questionnaire filled out by psychiatrists on duty at the time of admission. The study was carried out during 3 months in seven mental hospitals. It concerns only 1001 patients assessed by psychiatrists as imminently threatening. These assessments were compared with Lessard's criteria of dangerousness. Two types of assessments were made; type A--consistent with Lessard and type B not-consistent. Results obtained indicate that every third assessment belonged to type B i.e. could be seen as imminently threatening. Our findings show that psychiatrists taking part in the study were inclined to unduly broad or unduly discretional assessment of imminent threat.

Gastrin and epidermal growth factor induction of ornithine decarboxylase in rat colonic explants.

An organ culture system was utilized to examine the effect of gastrin (G-17-I) and epidermal growth factor (EGF) on colonic mucosal ornithine decarboxylase (ODC) activity, and the expression of the ODC gene. Exposure of colonic mucosal explants to either gastrin or EGF (50-500 ng/ml) for only 4 h resulted in a profound stimulation (150-600%) in ODC activity over the basal level. These increases were essentially abolished by difluoromethylornithine (DFMO; 2 nmol/ml) or CaCl2 (2 umol/ml). Gastrin also activated the ODC gene in the colonic mucosa as evidenced by increased steady-state ODC mRNA levels in the colonic mucosal explants after 4 h exposure to the hormone, when compared with the controls. It is concluded that colonic mucosal ODC is responsive to both gastrin and EGF.

Attenuation of azoxymethane-induced colonic mucosal ornithine decarboxylase and tyrosine kinase activity by calcium in rats.

Two in vivo and one in vitro studies were performed to evaluate the chemoprotective role of calcium during the early period of azoxymethane (AOM) induction. In the first set of experiments, groups of male Fischer 344 rats were s.c. injected with either AOM (20 mg/kg) or water (controls) and sacrificed immediately (0 time), and 1, 3, 5, and 7 days postinjection. In the second set of experiments, animals were injected with the same dose of AOM and subsequently pair-fed with rat chow containing either calcium carbonate or diet devoid of added calcium. The amount of calcium consumed was calculated to be 250 mg/kg b.w. In both experiments, colonic mucosa was assayed for ornithine decarboxylase (ODC). In addition, tyrosine kinase (Tyr-k) activity as well as tyrosine specific phosphorylation of membrane proteins were determined. Results revealed that maximal stimulation by AOM of ODC and Tyr-k activity occurred 5 days postinjection. This stimulation was significantly suppressed by calcium. AOM also produced an increase in the rate of tyrosine specific phosphorylation of two distinct colonic mucosal membrane proteins with Mr of 57,000 and 59,000. Again, dietary calcium suppressed the stimulation. In the third set of experiments, organ culture was utilized. Methylazoxymethanol, the active metabolite of AOM, was used instead of AOM in this part of the study. Four hour exposure of mucosal explants to methylazoxymethanol (1 microgram/ml) resulted in a significant (20-30%) increase in ODC and Tyr-k activity when compared to controls. Addition of either CaCl2 (2 mumol/ml) or difluoromethylornithine (2 nmol/ml) the irreversible inhibitor of ODC, significantly suppressed the methylazoxymethanol-induced activity of both ODC and Tyr-k. We conclude that calcium may have a chemoprotective role and tyrosine kinases may have a regulatory role in the early stages of AOM induction of colon cancer.

Acetylmethadol metabolites influence opiate receptors and adenylate cyclase in amygdala.

The activity of acetylmethadol and two major metabolites of this drug, noracetylmethadol and dinoracetyl-methadol were studied in monkey brain amygdaloid tissue. This tissue contains opiate receptors which may be assessed by direct binding studies; also these receptors are coupled to and capable of modulating a dopamine-stimulated adenylate cyclase system. Etrophine and D-ALa2-Met-enkephalin exhibited similar potencies when assessed for inhibition of dopamine-stimulated adenylate cyclase or competition for [3H] D-Ala2-Met-enkephalin binding sites. While acetylmethadol displaced [3H] D-Ala2-Met-enkephalin binding sites with a Ki of 7.4 X 10(-7) M, it had no detectable activity in the opiate receptor coupled adenylate cyclase system. Noracetylmethadol and dinoracetylmethodol, however, were capable of both binding (Ki values, 5.6 and 1140 nM respectively) as well as inhibiting the dopamine stimulated adenylate cyclase system (IC50 values, 1.2 and 800 nM respectively). Thus, it appears that metabolism of acetylmethadol to its mono-demethylated form results in a compound more active in both assay systems. The further demethylation of noracetylmethadol results in a second, but less potent, active metabolite. This process of biological activation can be assumed to account for the slow onset and long duration of action of acetylmethadol.

Interaction of morphine, etorphine and enkephalins with dopamine-stimulated adenylate cyclase of monkey amygdala.

Adenylate cyclase activity (AC) of homogenates of monkey amygdaloid nucleus was approximately doubled in the presence of dopamine (10 micrometer). Morphine, etorphine, and several enkephalin analogs (met-enkephalin, D-ala2-met-enkephalin, and D-met2, pro5-enkephalinamide) were capable of inhibiting the stimulation of AC produced by dopamine (90-100% with etorphine or D-ala2-met-enkephalin). Unlike morphine and etorphine, the peptides exhibited bell-shaped dose-response curves for this inhibition with maximal effects at approximately 1 X 10(-7) M, but negligible effects at 1 X 10(-5) M. Under the conditions studied, only etorphine inhibited basal AC. Naloxone antagonized the inhibitory effects of each of the opioids tested, and dextrorphan, an inactive L-(+)-opiate, failed to inhibit the dopamine response. Together these data indicate that the effects were mediated via the classically described stereospecific opiate receptor. The relative order of potency (etorphine greater than enkephalins greater than morphine) was similar to that previously reported for the binding affinities of these drugs in rat brain homogenates. The influence of narcotic agents on dopamine stimulated AC was eliminated by either freezing the amygdaloid tissue or preincubating the homogenate at 4 degrees C; the dopamine responses, however, could still be elicited. The narcotic receptor interaction with the adenylate cyclase thus appears to be distinct from and more labile than that of the dopamine receptor. Gpp(NH)p-stimulated AC was not inhibited by morphine. It is postulated that the inhibition involves interaction of opiate receptors with catalytic units of dopamine-stimulated AC, but not with other cyclase species which may provide the major component of Gpp(NH)p-stimulated activity in amygdala.