PubMed Health⌕ Search

Biomedical subjects

P Willshaw

Publications and source records attributed to P Willshaw.

At least 19 recordsLinked to original sources
PubMed2005

Comparison of different methods for hemodialysis evaluation by means of ROC curves: from artificial intelligence to current methods.

BACKGROUND: The National Kidney Foundation Guidelines (DOQI) and the European Renal Association (ERA) have set standards for adequacy of hemodialysis treatment. They recommended minimum single pool doses of 1.2 (Kt/Vsp DOQI), and 1.4 (Kt/Vsp ERA) and a "standard" urea removal ratio (URR) of 65%. Here, we compare an Artificial Intelligence Method (AIM) based on an Artificial Neural Network (ANN) and the usual methods for hemodialysis treatment follow-up such as Smye, Daugirdas, standard urea reduction ratio (URR using post-dialysis urea concentration) and modified URR [Cheng et al. 2001] against equilibrated Kt/V and URR calculated using a 60 min post-dialysis urea concentration. METHODS: We used ROC analysis to evaluate and compare these methodologies. We also propose a method to find a minimum target dose that maximizes the sensitivity, specificity and positive predictive values of the diagnostic tool. RESULTS: From a URR point of view, the ANN, stdURR and mURR perform almost equally well with an area under the curve (AUC) of 0.90, 0.93 and 0.92, respectively, but the ANN achieved the lowest false positive rate (FPR = 7.94%) and error rate (ER = 12.7%). When Kt/V is used as a dose index, the logarithmic single-and double-pool equations perform almost equally (AUC 0.957 and 0.962), and the ANN method achieves an AUC of 0.934. The lowest FPR was for ANN and Kt/Vsp (4.76%), which also achieved the lowest ER of 6.39%. CONCLUSIONS: For both cases (URR and Kt/V), the minimum doses required to achieve the lowest FPR and ER for the standard methods (stdURR and Kt/Vsp) were higher than those reported by the DOQI guidelines, being 70% for stdURR and 1.35 for Kt/Vsp, whereas for those methods using the double-pool Kt/V or equilibrated URR, the dose targets were close to those recommended by DOQI and ERA. Our proposed method for target dose selection is easy to understand, and it takes into account both accuracy and confidence of the adequacy tool. We found the ANN method to be superior to the Smye method for estimation of equilibrated urea, and the results presented here suggest that ANN methods could be useful tools in the analysis of nephrology data.

Artificial Intelligence↗
PubMed2003

Dialysate-side urea kinetics. Neural network predicts dialysis dose during dialysis.

Determination of the adequacy of dialysis is a routine but crucial procedure in patient evaluation. The total dialysis dose, expressed as Kt/V, has been widely recognised to be a major determinant of morbidity and mortality in haemodialysed patients. Many different factors influence the correct determination of Kt/V, such as urea sequestration in different body compartments, access and cardiopulmonary recirculation. These factors are responsible for urea rebound after the end of the haemodialysis session, causing poor Kt/V estimation. There are many techniques that try to overcome this problem. Some of them use analysis of blood-side urea samples, and, in recent years, on-line urea monitors have become available to calculate haemodialysis dose from dialysate-side urea kinetics. All these methods require waiting until the end of the session to calculate the Kt/V dose. In this work, a neural network (NN) method is presented for early prediction of the Kt/V dose. Two different portions of the dialysate urea concentration-time profile (provided by an on-line urea monitor) were analysed: the entire curve A and the first half B, using an NN to predict the Kt/V and compare this with that provided by the monitor. The NN was able to predict Kt/V is the middle of the 4h session (B data) without a significant increase in the percentage error (B data: 6.69% +/- 2.46%; A data: 5.58% +/- 8.77%, mean +/- SD) compared with the monitor Kt/V.

Adult↗
PubMed2001

Detection of abnormality in the electrocardiogram without prior knowledge by using the quantisation error of a self-organising map, tested on the European ischaemia database.

Most systems for the automatic detection of abnormalities in the ECG require prior knowledge of normal and abnormal ECG morphology from pre-existing databases. An automated system for abnormality detection has been developed based on learning normal ECG morphology directly from the patient. The quantisation error from a self-organising map 'learns' the form of the patient's ECG and detects any change in its morphology. The system does not require prior knowledge of normal and abnormal morphologies. It was tested on 76 records from the European Society of Cardiology database and detected 90.5% of those first abnormalities declared by the database to be ischaemic. The system also responded to abnormalities arising from ECG axis changes and slow baseline drifts and revealed that ischaemic episodes are often followed by long-term changes in ECG morphology.

Electrocardiography↗
PubMed2001

Using artificial intelligence to predict the equilibrated postdialysis blood urea concentration.

Total dialysis dose (Kt/V) is considered to be a major determinant of morbidity and mortality in hemodialyzed patients. The continuous growth of the blood urea concentration over the 30- to 60-min period following dialysis, a phenomenon known as urea rebound, is a critical factor in determining the true dose of hemodialysis. The misestimation of the equilibrated (true) postdialysis blood urea or equilibrated Kt/V results in an inadequate hemodialysis prescription, with predictably poor clinical outcomes for the patients. The estimation of the equilibrated postdialysis blood urea (eqU) is therefore crucial in order to estimate the equilibrated (true) Kt/V. In this work we propose a supervised neural network to predict the eqU at 60 min after the end of hemodialysis. The use of this model is new in this field and is shown to be better than the currently accepted methods (Smye for eqU and Daugirdas for eqKt/V). With this approach we achieve a mean difference error of 0.22 +/- 7.71 mg/ml (mean % error: 1.88 +/- 13.46) on the eqU prediction and a mean difference error for eqKt/V of -0.01 +/- 0.15 (mean % error: -0.95 +/- 14.73). The equilibrated Kt/V estimated with the eqU calculated using the Smye formula is not appropriate because it showed a great dispersion. The Daugirdas double-pool Kt/V estimation formula appeared to be accurate and in agreement with the results of the HEMO study.

Adult↗
PubMed1994

Left ventricular end-systolic elastance is incorrectly estimated by the use of stepwise afterload variations in conscious, unsedated, autonomically intact dogs.

BACKGROUND: End-systolic elastance (Ees), the slope parameter of the end-systolic pressure (ESP)-volume (ESV) relation (ESPVR), is usually estimated in patients by producing stepwise, steady-state pharmacological afterload variations and collecting one ESP-ESV point from each step. The ESPVR is then constructed by fitting a linear equation to these points. In sedated, autonomically blocked dogs, it has been shown that when one point from control, one point from a state of increased afterload, and one point from a state of decreased afterload are used, the resulting Ees incorrectly estimates true Ees, defined as the slope of the ESPVR obtained by transient vena caval occlusion. We investigated if this was also the case in unsedated, autonomically intact dogs when the points used belonged to steady states of progressively decreasing or progressively increasing afterload pressure. METHODS AND RESULTS: In 10 conscious dogs instrumented with left ventricular (LV) endocardial sonomicrometers to measure LV volume, a LV pressure transducer, and an inferior vena caval (IVC) occluder, two protocols were carried out on separate days. In each protocol, an ESPVR was generated by IVC occlusion in the control state and in two steady-state levels of afterload change produced by stepwise infusion of nitroprusside (protocol 1, afterload decrease) and angiotensin II (protocol 2, afterload increase). In each protocol, steady-state ESP-ESV data points were averaged from the control state and from each level of afterload variation. Linear equations were fitted to the three steady-state points from each protocol, and the estimated Ees values obtained (EesEST) were compared with the Ees values of the control ESPVRs obtained by IVC occlusion (EesTRUE). In protocol 1, EesEST underestimated EesTRUE by about 16% (EesEST, 6.49 +/- 1.55 mm Hg/mL; EesTRUE, 7.48 +/- 1.29 mm Hg/mL; P < .02). In protocol 2, EesEST overestimated EesTRUE by about 37% (EesEST, 9.99 +/- 3.97 mm Hg/mL; EesTRUE, 6.43 +/- 3.88 mm Hg/mL; P < .007). CONCLUSIONS: In conscious, autonomically intact dogs, the use of stepwise, steady-state afterload variations to obtain ESP-ESV data points to construct the ESPVR incorrectly estimates Ees. In the case of afterload reduction, EesTRUE is underestimated an average of 16.3%, and in the case of afterload increase, EesTRUE is overestimated an average of 37.1%. These errors should be taken into account when interpreting clinical studies using this methodology.

Animals↗
PubMed1992

Differential effects of left anterior descending coronary occlusion on left and right ventricular anterior wall thickening in the conscious pig.

OBJECTIVE: In humans, the left anterior descending coronary artery supplies the left ventricular wall, anterior septum and the paraseptal part of the right ventricular anterior wall. Our aim was to study the effects of acute left anterior descending coronary occlusion on wall thickening in the regions of the left and right ventricular anterior walls supplied by the artery, and in remote, non-ischaemic regions of both ventricles. METHODS: Systolic wall thickening (defined as percent thickening with respect to end diastolic wall thickness) was studied in eight conscious pigs every 15 s during 1 min of acute left anterior descending coronary occlusion by a cuff occluder, and every 30 s during 4 min of reperfusion. Pigs were instrumented with ultrasonic microcrystals measuring wall thickness in the anterior walls (left anterior descending artery territory) and lateral walls (left circumflex or right coronary artery territory) of both ventricles, and a left ventricular pressure microtransducer. RESULTS: During control and reperfusion, both anterior walls displayed similar systolic thickening. During coronary occlusion, the left ventricular anterior wall showed paradoxical systolic thinning (dyskinesia) whereas the right ventricular anterior wall showed only hypokinesia. CONCLUSIONS: In the presence of equal blood flow deprivation, the right ventricular anterior wall supplied by the left anterior descending coronary artery displays a significantly lesser degree of functional impairment than the left ventricular anterior wall supplied by the same artery. This differential effect may be due to mechanical unloading of the right ventricular anterior wall resulting from left ventricular anterior wall ischaemia. This afterload reduction due to decreased mechanical interaction between the two walls would allow the right ventricular anterior wall to express its contractile reserve in the form of systolic thickening.

Animals↗
PubMed1991

Interventricular coronary steal induced by stenosis of left anterior descending coronary artery in exercising pigs.

BACKGROUND: In pigs and humans, the left anterior descending coronary artery (LAD) supplies the left ventricular anterior wall (LVAW), anterior septum, and paraseptal band of the right ventricular anterior wall (RVAW). The purposes of our study were 1) to study the LAD flow distribution in these walls during preexercise, exercise, and exercise with LAD stenosis and 2) to analyze regional wall motion under these conditions. METHODS AND RESULTS: Nine pigs were instrumented with sonomicrometers for measuring percent wall thickening (%WTh) in LVAW, RVAW, and lateral (control) walls of both ventricles, a hydraulic occluder at the LAD origin, an LV pressure transducer, and catheters for radioactive microsphere injection (left atrium) and blood withdrawal (aorta). One month later, regional %WTh and flows were measured during preexercise, exercise, and continuing exercise with LAD stenosis resulting in more than 50% reduction in systolic LVAW %WTh with regard to exercise. LAD stenosis caused a dramatic decrease in total mean +/- SD LVAW subendocardial flow with regard to exercise (28.7 +/- 8 to 9.1 +/- 3.2 ml.min-1, p less than 0.0001) but not significant changes in either LVAW subepicardial flow or RVAW flow. The transmural distribution of flows within the LAD bed (as percentages of the total LAD flow in each experimental condition) showed that LAD stenosis redistributed flows with regard to exercise such that the LVAW subendocardial flow decreased from 26.4 +/- 4.2% of the total LAD flow to 11.8 +/- 4.3% (p less than 0.0001), whereas LVAW subepicardial flow increased from 32.9 +/- 2.3% of the total LAD flow to 45.5 +/- 7.9% (p less than 0.0001) and RVAW increased from 12 +/- 4.9% of the total LAD flow to 18.7 +/- 7.2% (p less than 0.0005). With exercise plus LAD stenosis, LVAW %WTh decreased from 43.2 +/- 8.4% to 17.2 +/- 9.7% (p less than 0.0001), but RVAW %WTh did not change. CONCLUSIONS: In the LAD bed of exercising pigs, LAD stenosis induces, in addition to transmural steal, an interventricular steal favoring the RVAW at the expense of the LVAW subendocardium. This steal results in preserved RVAW thickening despite severe LVAW hypokinesia.

Animals↗
PubMed1988

Detection of left ventricular regional myocardial ischaemia in dogs by intraventricular conductance catheter.

Nine adult mongrel dogs were instrumented with ultrasonic microcrystals to measure left ventricular basal anteroposterior diameter and midwall myocardial segment length near the cardiac apex. Pneumatic cuff occluders were positioned around the left circumflex coronary artery near its origin and around the left anterior descending coronary artery two thirds of the way along its length. A pressure microtransducer was implanted into the left ventricle. Ten days after instrumentation the animals were anaesthetised with morphine chlorhydrate and pentobarbital sodium. An eight electrode catheter was advanced into the left ventricle to measure ventricular apical and basal regional and total electrical conductance. Minor ischaemia caused by occlusion of the left anterior descending artery was detected only by the electrode pair located near the apex, as decreased local ejection fraction. Major ischaemia caused by left circumflex artery occlusion was detected by both apical and basal electrode pairs and by total conductance, the three conductance signals indicating reduced ejection fractions compared with control values. The basal diameter signal indicated that basal regional motility changed only during major ischaemia, thus confirming the specificity of the changes in the basal conductance signals. The apical segment length signal confirmed the altered motility indicated by the apical conductance signal. These results suggest that regional wall motion abnormalities may be detected by the use of a multielectrode conductance catheter.

Animals↗
PubMed1988

End-systolic pressure-volume relationships in dogs during ventilation with PEEP.

Whether left ventricular (LV) contractility changes during ventilation with positive end-expiratory pressure (PEEP) remains controversial. To assess LV inotropic state during PEEP using a load-independent index, we generated end-systolic pressure-volume relationships (ESPVRs) in eight closed-chest, chronically instrumented, anesthetized dogs undergoing 0 [zero end-expiratory pressure for the 1st time (ZEEP1)], 5 (PEEP-5), 10 (PEEP-10), and again 0 (ZEEP2) cmH2O PEEP. LV volume was calculated from three orthogonal internal diameters (sonomicrometry), and LV pressure was measured using an implanted transducer. ESPVRs at each level of PEEP were generated by transient inflation of a vena caval occluder. Despite significant decreases in cardiac output with PEEP-5 (1.81 +/- 0.38 l/min, means +/- SE; P less than 0.05) and PEEP-10 (1.70 +/- 0.46; P less than 0.01) with respect to ZEEP1 (2.12 +/- 0.41), no change was found in the slope (ZEEP1: 6.99 +/- 1.03 mmHg/ml; PEEP-5: 7.48 +/- 1.20; PEEP-10: 7.17 +/- 1.02; ZEEP2: 7.38 +/- 1.02), the volume intercept (ZEEP1: 7.4 +/- 3.4 ml; PEEP-5: 6.6 +/- 3.0; PEEP-10: 7.2 +/- 4.0; ZEEP2: 6.6 +/- 3.6), or the new index area beneath the ESPVR (ZEEP1: 304 +/- 98; PEEP-5: 329 +/- 104; PEEP-10: 310 +/- 98; ZEEP2: 343 +/- 114). We conclude that these levels of PEEP do not affect LV contractility as assessed by the ESPVR.

Animals↗
PubMed1987

Inconsistency of the slope and the volume intercept of the end-systolic pressure-volume relationship as individual indexes of inotropic state in conscious dogs: presentation of an index combining both variables.

We tested the ability of the slope (Emax) and the volume intercept (Vo) of the end-systolic pressure-volume relationship (ESPVR) to indicate contractility changes in conscious dogs instrumented with sonomicrometers measuring left ventricular diameter in three orthogonal axes and a left ventricular pressure microtransducer. ESPVRs were generated by inferior vena caval occlusion under control conditions (C1 and C2) and during enhanced (I+) and depressed (I-) inotropic states achieved by infusion of dobutamine and injection of propranolol, respectively. No significant difference between the first control (C1) and I+ or between the second control (C2) and I- were found for either Emax (C1, 5.31 +/- 1.68 mm Hg/ml, mean +/- SD; I+, 5.37 +/- 1.44; C2, 5.20 +/- 1.62; I-, 4.18 +/- 1.32) or Vo (C1, 10.3 +/- 9.6 ml; I+, 7.3 +/- 9.1; C2, 9.9 +/- 9.0; I-, 12.7 +/- 12.5), despite significant changes in other indexes of contractility. Comparison of changes in Emax in individual animals in response to I+ and I- revealed that 63% were nonsignificant, 28% were significant and expected, and 9% were significant and paradoxical. Within defined volume limits and irrespective of individual changes in Emax and Vo, in all animals I+ shifted the ESPVR above and to the left of C1 and I- shifted the ESPVR below and to the right of C2. We thus integrated the changes in Emax and Vo by measuring the area beneath each ESPVR between defined limits of end-systolic volume. The values for area were: C1, 612 +/- 150 mm Hg.ml; I+, 745 +/- 191 (p less than .001); C2, 520 +/- 198; I-, 420 +/- 139 (p less than .001). We conclude that (1) neither Emax nor Vo are individually reliable indexes of changed contractility, and (2) the area beneath the ESPVR between defined end-systolic volume limits is a consistent indicator of variations in inotropic state.

Animals↗
PubMed1987

Animal model of acute pulmonary thromboembolism treated by local recirculation of streptokinase through the lung.

Massive acute pulmonary thromboembolism has a high mortality within the first few hours. Surgical intervention can remove only larger thrombi. Systemic fibrinolytic administration requires many hours for adequate treatment. We describe an anesthetized dog model of acute, massive, disseminated pulmonary thromboembolism achieved by injection of 1.7 ml/kg of 1-hour-old thrombi directly into the pulmonary artery. The emboli were lysed with 50,000 IU streptokinase recirculated for 20 minutes through the isolated pulmonary vascular bed by use of a roller pump while the systemic bed was supported by conventional cardiopulmonary bypass. On reestablishing natural circulation all hemodynamic parameters returned to preembolism values. Success of lysis was histologically confirmed. Perfusion of the pulmonary vascular bed without inclusion of streptokinase in the perfusate worsened the hemodynamic state of the animals compared with an untreated nonperfused control group.

Acute Disease↗
PubMed1986

A comparative in vitro study of the closing characteristics of Björk-Shiley and Bicer-Val tilting disc mitral valve prostheses.

The replacement of mitral valves has become a relatively common procedure in cardiac surgery. Information concerning losses due to backflow through the valve may be used to assess and compare the physical characteristics of commercially available prostheses. A simple apparatus based on an artificial ventricle was used to measure closing volume, leakage volume, total volume, leakage rate and closing time on a beat by beat basis for Björk-Shiley and Bicer Val prosthetic mitral valves. The Björk-Shiley valve opened to 60 degrees and displayed a smaller closing volume than the Bicer Val, which opened to 75 degrees; and the Björk-Shiley displayed higher leakage than the Bicer Val, thus reducing its advantage in terms of closing volume. Overall, the Bicer-Val showed about 1 ml greater reflux per beat than the Björk-Shiley. Study of a Bicer-Val, modified to have an opening angle of 80 degrees, confirmed that valve closing volume was a function of opening angle.

Cardiac Output↗
PubMed1985

Experimental acute right ventricular failure and right ventricular assist in the dog.

We describe a technique for the production of acute progressive right ventricular failure in experimental animals that mimics the hemodynamic characteristics of right ventricular failure found in some patients being weaned from extracorporeal circulation after surgical repair of left ventricular abnormalities. The technique combines three alterations of right ventricular state: excision of the tricuspid valve, ventriculotomy, and ligation of the right coronary artery. Seven control dogs died within 3 hours after this intervention. Death was due to low cardiac output as a result of low left atrial and pulmonary arterial pressures. Right atrial pressure was high. Use of a right ventricular assist device in an additional seven dogs to pump blood from the right atrium to the pulmonary artery confirmed good preservation of left ventricular function by reestablishing adequate left ventricular filling pressure. All seven dogs survived for more than 3 hours. The validity of the technique in restricting failure principally to the right ventricle was thus demonstrated.

Animals↗
PubMed1984

Loss of peripheral chemoreflexes to hypoxia after carotid body removal in the rat.

Young Sprague-Dawley female rats (16-18 days old) were subjected to bilateral carotid glomectomy or the sham operation under halothane anaesthesia. After recovery, the rats were placed with non-operated peers. One to six months later glomectomised, sham-operated and control animals were anaesthetized with sodium pentobarbital. Respiratory minute volume and arterial pressure were recorded. Respiratory responses to few-breath administration of oxygen or nitrogen, and arterial pressure responses to carotid occlusion or tugging were tested. Oxygen produced transient hypoventilation and nitrogen transient hyperventilation in control, sham and 3 of 31 glomectomised rats. Bilateral vagotomy did not abolish these responses. In only those 3 glomectomised rats was carotid glomus tissue histologically identifiable. Carotid occlusion raised and tugging lowered arterial pressure in all animals. Glomectomy did not affect serum levels of GH, FSH, LH or PRL hormones but produced right ventricular hypertrophy. We conclude that peripheral chemoreception requires the presence of glomus tissue.

Animals↗
PubMed1984

Tri-axial graphics as an aid to understanding left ventricular function in conscious dogs.

Ultrasonic distance gauges were implanted in the left ventricle of dogs to measure continuously ventricular wall thickness and small axis diameter. A miniature transducer was implanted to measure intraventricular pressure. The circumflex artery could be occluded mechanically to mimic coronary spasm. Recordings were made on tape after allowing eight days post-operative recuperation. The taped signals were digitised and processed to show graphical relationships between selected variables. We found that triaxial (3D) graphics greatly enhanced our ability to interpret moment by moment changes in these relationships, and recommend this procedure to others.

Animals↗
PubMed1984

A cardiac output monitor and diagnostic unit for pneumatically driven artificial hearts.

The pneumatic artificial heart uses a flexible diaphragm to separate blood from air. The air space behind the diaphragm is connected to the atmosphere during diastole. Blood entering the heart during diastole displaces an equal volume of air into the atmosphere. A pneumotachograph is used to sense this flow of air. Integration of the airflow signal with respect to time yields the volume of air displaced during diastole, which is equal to the volume of blood entering the heart over the same period. This measured filling volume is equal to the cardiac output when averaged over a number of cardiac cycles. Allowance is made for regurgitant losses owing to the prosthetic heart valves. The output of the pneumotachograph is independent of absolute pressure, and so the device may also be used during the systolic (air compression) phase of the cardiac cycle. Analysis of the display of the airflow curves during diastole and systole has great diagnostic value. This device is accurate, easy to calibrate, stable, noninvasive, and does not require modification of any existing air-driven artificial hearts.

Air↗