Prediction of spontaneous ureteral calculous passage by an artificial neural network.

PURPOSE: Although a consensus exists that small stones presenting in the distal ureter have a good probability of spontaneous passage, it is difficult to predict in individuals whether a particular ureteral stone would pass or require intervention. If an accurate judgment were made at presentation on the likelihood of stone passage, patients would receive immediate intervention for the stone or be notified of a more appropriate time at which to expect passage. We used an artificial neural network to evaluate data in patients with ureteral calculi to predict whether a stone would pass spontaneously or require intervention. MATERIALS AND METHODS: Data were collected from the records of 181 patients presenting with colic due to a ureteral calculus. Patient input factors included age, sex, race, marital status, insurance, stone side, level and size, hydronephrosis and obstruction grades, duration of symptoms before presentation, serum creatinine, history of stone passage or intervention and nausea, vomiting or fever. Outcomes evaluated were stone passage or intervention. Data were entered into a neural network created using commercially available computer software. RESULTS: A set of 125 patients from the database was used for training the network. The network correctly predicted outcome in 38 of the remaining 55 patients (76%) used for testing. In the 25 cases in which stones passed spontaneously sensitivity was 100%. Duration of symptoms before presentation was the most influential factor in network ability to predict accurately stone passage, followed by hydronephrosis grade. CONCLUSIONS: An artificial neural network may be used to predict accurately the probability of spontaneous ureteral stone passage. Using such a model at presentation may help to determine whether a patient should receive early intervention for a stone or expect a lengthy interval before stone passage.

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.

Prediction of trauma mortality using a neural network.

A neural network is a computerized construct consisting of input neurons (which process input data) connected to hidden neurons (to mathematically manipulate values they receive from all the input neurons) connected to output neurons (to output a prediction). Neural networks are created and trained via multiple iterations over data with known results. In 1993, 897 trauma patients were either declared dead in the emergency room (ER; 76 cases), admitted to the intensive care unit (427 cases, 36 deaths), or taken directly to the operating room (394 cases, 29 deaths). Using only data available from the ER, a neural network was created, and 628 cases were randomly selected for training. After 268 iterations, the network was trained to correctly predict death or survival in all 628 cases. This trained network was then tested on the other 269 cases without our providing the death or survival result. Its overall accuracy was 91 per cent (244 of 269 cases). It was able to predict correctly 60 per cent (12 of 20 cases) of the postoperative or post-intensive care unit admission deaths and 90 per cent (26 of 29 cases) of the deaths in the ER. Computerized neural networks can accurately predict a trauma patient's fate based on inital ER presentation. The theory and use of neural networks in predicting clinical outcome will be presented.

Simulated biologic intelligence used to predict length of stay and survival of burns.

From July 13, 1988, to May 14, 1995, 1585 patients with burns and no other injuries besides inhalation were treated; 4.5% did not survive. Artificial neural networks were trained on patient presentation data with known outcomes on 90% of the randomized cases. The remaining cases were then used to predict survival and length of stay in cases not trained on. Survival was predicted with more than 98% accuracy and length of stay to within a week with 72% accuracy in these cases. For anatomic area involved by burn, burns involving the feet, scalp, or both had the largest negative effect on the survival prediction. In survivors burns involving the buttocks, transport to this burn center by the military or by helicopter, electrical burns, hot tar burns, and inhalation were associated with increasing the length of stay prediction. Neural networks can be used to accurately predict the clinical outcome of a burn. What factors affect that prediction can be investigated.

Tumor necrosis factor-induced mortality is reversed with cyclooxygenase inhibition.

OBJECTIVE: The authors hypothesized that TNF would induce eicosanoid synthesis, and a cyclooxygenase inhibitor would attenuate both eicosanoid synthesis and improve survival in an LD90 TNF-induced (150 ng/kg/i.v./5 min) mortality model. SUMMARY BACKGROUND DATA: Tumor necrosis factor is a cardinal mediator in sepsis; however, little is known about its effects on arachidonate metabolism. METHODS: Conscious male rats with carotid arterial and jugular venous catheters were randomized for mortality: group I, TNF alone (150 kg/i.v./15 min, n = 30); group II, ibuprofen (30 mg/kg/i.v. at t = -20 and +240 min), plus TNF, (n = 28); and for hemodynamics, eicosanoid synthesis, blood gases: group III, TNF alone, (n = 8); group IV, ibuprofen + TNF (n = 8); group V, monoclonal antibody to TNF plus TNF (n = 8). Mortality was determined at 4-72 hr. Other parameters determined over 4 hours (0, 5, 60, 120, 240 min). RESULTS: TNF stimulated synthesis of (a) TXB2 (71 +/- 30 pg/ml, mean +/- SE at base vs. 117 +/- 18 at 4 hr, p < 0.02); (b) PGE2 (70 +/- 6 pg/ml at base vs. 231 +/- 68 at 4 hr, p < 0.02); (c) 6PGF (52 +/- 6 pg/ml at base vs. 250 +/- 80 at 4 hr, p < 0.02). Ibuprofen significantly (p < 0.05) inhibited eicosanoid synthesis from TNF. TNF-induced mortality (87%, 26/30) was dramatically decreased with ibuprofen (11%, 3/28), at 4, 24, and 72 hr (p < 0.01). Monoclonal antibody to TNF prevented all abnormalities and had 100% survival. Hemodynamic events were similar in both groups, but metabolic acidosis was attenuated with ibuprofen. CONCLUSIONS: TNF stimulates arachidonic acid metabolism in vivo. A cyclooxygenase inhibitor attenuates eicosanoid synthesis and dramatically improves survival. TNF appears to have different effect on tissues that synthesize certain eicosanoids. Hypotension from TNF is not mediated via the eicosanoids. TNF-induced mortality, like endotoxemia/sepsis may be mediated, in part, via arachidonic acid metabolites. These new findings support the notion that cyclooxygenase inhibitors may be used as adjunctive therapy in clinical sepsis.

Decreased contractility and compliance of the left ventricle as complications of thermal trauma.

To test the hypothesis that systemic complications of dermal burns encompass dysfunction of myocardial contractile mechanisms, we studied contraction-relaxation properties of isovolumic left ventricular (LV) preparations isolated from guinea pigs 24 hours after full-thickness burn to approximately 47% total body surface area. Compared to control hearts, hearts from burned subjects consistently generated significantly lower values for LV systolic pressure (94 +/- 2 vs 66 +/- 2 mm Hg; p less than 0.001) and maximal rates of LV pressure rise (+ dP/dtmax; 1296 +/-71 vs 1091 +/- 46 mm Hg X sec-1; p less than 0.05) and fall (-dP/dtmax; 1214 +/- 45 vs 856 +/- 34 mm Hg X sec-1; p less than 0.001). The LV contractile deficit of burn hearts was not correlated with changes in tissue water content, and it was not surmountable by excess glucose, insulin, increased coronary flow, or maximal preload elevation. In addition, end-diastolic pressure-volume relationships in burn hearts were shifted upward and to the left of controls in the direction of decreased compliance (p less than 0.05 to p less than 0.01). Thus, LV sequelae of thermal trauma manifest in isolated hearts as decreased contractility, slowed isovolumic relaxation, and decreased diastolic compliance; in the intact animal this combination would reduce ejection and impede filling of the ventricle, with diastolic pressures reflecting changes in compliance as well as in contractile function.