Statistical methods for convergence detection of multi-objective evolutionary algorithms.

In this paper, two approaches for estimating the generation in which a multi-objective evolutionary algorithm (MOEA) shows statistically significant signs of convergence are introduced. A set-based perspective is taken where convergence is measured by performance indicators. The proposed techniques fulfill the requirements of proper statistical assessment on the one hand and efficient optimisation for real-world problems on the other hand. The first approach accounts for the stochastic nature of the MOEA by repeating the optimisation runs for increasing generation numbers and analysing the performance indicators using statistical tools. This technique results in a very robust offline procedure. Moreover, an online convergence detection method is introduced as well. This method automatically stops the MOEA when either the variance of the performance indicators falls below a specified threshold or a stagnation of their overall trend is detected. Both methods are analysed and compared for two MOEA and on different classes of benchmark functions. It is shown that the methods successfully operate on all stated problems needing less function evaluations while preserving good approximation quality at the same time.

[Atropine versus glycopyrrolate in eye surgery. A comparison of rhythm disorders and heart frequency using Holter-ECG]. / Atropin versus Glycopyrrolat bei ophthalmochirurgischen Eingriffen. Vergleich der Rhythmusstörungen und der Herzfrequenz mit Holter-EKG.

Anticholinergics are indicated in ophthalmic surgery to inhibit oculocardiac reflexes (OCR). Although all anticholinergics exert an arrhythmic effect, comparative data on occurrence and severity of arrhythmias are not available. The aim of this investigation was to compare atropine (A) and glycopyrrolate (G) in adults undergoing ophthalmic surgery. One hundred patients (ASA class I and II, age range 20-60 yrs.) were given equipotent doses of either A (10 micrograms.kg-1) or G (5 micrograms.kg-1) in a randomized, double-blind fashion, before induction of anaesthesia. Continuous Holter monitoring was performed. The procedure was divided into the following phases: O (5 min before drug), I (5 min after drug), II (up to induction), III (induction until intubation), IV (intubation until operation), V (operation). Supraventricular, junctional, and ventricular arrhythmias were analyzed. Severe arrhythmia was judged to be greater than 5 events per min, or ventricular beats Lown classes III-V. The beat to beat analog signal was digitalized. The heart rate (HR) and the occurrence rate of OCR (greater than 20% decrease in heart rate, arrhythmias) was calculated by a computerized program. Analysed were: 1. the frequency of OCR, 2. the mean HR for each phase, 3. the frequency of all the various arrhythmias during the whole period, 4. as well as for each phase. The number of patients with severe arrhythmias 5. either for all the time or 6. for the individual phase were registered separately. There were no differences 1. in the frequency of OCR and 2. in mean HR during all phases between A and G.(ABSTRACT TRUNCATED AT 250 WORDS)

O2 administration by a nasal probe improves respiration in cataract surgery after retrobulbar anesthesia.

We measured certain respiratory conditions (respiratory rate [RR], oxygen saturation [SO2], and end-expiratory carbon dioxide partial pressure [pCO2]) of 31 patients undergoing planned cataract surgery using local anesthesia in order to determine the effects of administering pure oxygen (3 L/min) by a nasal probe in 10 of them. In the patients who did not receive pure oxygen, at the end of surgery the mean RR was 15.8 +/- 4.4/min (maximum, 21; minimum, 6.4/min); the mean SO2 was 86.9% +/- 6.6% (maximum, 98%; minimum, 74%; in 11/25 patients, the SO2 was lower than 90%); and the mean pCO2 was 34.9 +/- 7.7 mm Hg (maximum, 46.5; minimum, 12.15; in 4/25 patients, the pCO2 was greater than 45 mm Hg). In the patients who received pure oxygen by a nasal probe, the mean SO2 increased intraoperatively from 80.6% +/- 5.8% to 96.9 +/- 2.9% (in no patients was SO2 lower than 90%). Therefore we recommend an intraoperative administration of pure oxygen by a nasal probe.

[Possibilities and limitations of intravenous anesthesia]. / Möglichkeiten und Grenzen der intravenösen Anästhesie.

The increasing importance of intravenous anaesthesia is based on two developments, namely the synthesis of substances capable of acting selectively and over the short term, and which are rapidly eliminated (good control), and a desire on the part of the anaesthetist to have an anaesthetic whose actions can be controlled in various ways. Efforts in this area are aimed at ensuring not only a minimum of stressing of organs by the anaesthetic, but also a minimisation of risks during the intra-operative phase, and a non-problematic maximally pleasant post-operative course. There are two problems that militate against the realization of these objectives. Although the substances presented here ideally permit the realization of some of these aims, at the same time they are associated with side effects that prevent their use from ever being completely non-problematical, and which should always prompt the exercise of particular care when employing these medications. In addition, the differentiable control made possible by the use of these drugs is (partly) offset by inadequate monitoring with respect to the qualities of the anaesthetic--a fact that modifies the potential advantages of intravenous anaesthesia.

[Semiautomatic preparation of anesthesia records using the DATATRAC]. / Semiautomatische Protokollierung mit dem DATATRAC.

In this paper we report on our experience with clinical use of the electronic clipboard DATATRAC, which is designed for semiautomatic recording during anesthesia, i.e., for automated recording of cardiocirculatory parameters during anesthesia. In a period of 3 years, about 50 and 800 DATATRAC anesthesia records have been produced during vascular surgery and ophthalmic surgery, respectively. We discuss the advantages and short-comings of this type of anesthesia recording and the objections raised to it, and we analyze the malfunctions that can occur and their significance.

[Risk factors for the vis a tergo of the "open eye" in cataract extraction with local anesthesia]. / Risikofaktoren für die Vis a tergo am "offenen Auge" bei Katarakt-Extraktionen in Lokalanästhesie.

In a prospective study, 244 patients who were scheduled to undergo extracapsular cataract extraction with implantation of the intraocular lens (ECCE/IOL) under local anesthesia were examined to evaluate the risk factors for vis a tergo (VAT) at the "open eye". The intraoperative blood pressure was found to be the most important factor for the occurence of VAT. We think that a reversible suprachoroidal effusion syndrome is responsible for this effect. In order to minimize intraoperative complications, we recommend routine monitoring and treatment if the intraoperative blood pressure is too high during cataract surgery with local anesthesia.

[Perspectives of an electronic data processing-controlled anesthesia protocol]. / Perspektiven eines EDV-gerechten Anaesthesie-Protokolls.

There are two ways to introduce electronic data processing in anesthesia recording, which should be combined in the future: (1) computer-aided data collection (during anesthesia) and (2) data analysis. Both procedures have their own advantages and disadvantages. The first step in data collection is a system whereby the on-line registered data are automatically plotted and the discrete data are noted by hand (semi-automatic recording). The second step is to keep the minutes on a display screen instead of on paper, thus producing a protocol in digital form (automatic recording). We discuss the problems of these computer-aided recording systems and future trends, in particular the problems caused by the "human-computer interface" and by uncertainty with respect to the validity of the stored data. For computer-aided data analysis of anesthesia records, one has to select appropriate data in order to build up data bases. This selection is necessary whether the protocol is in analogical or in digital form, and we attempt to develop some general rules, the concrete selection depends, of course, on the aim of the evaluation. As an example we discuss evaluations for administrative purposes. Evaluations for scientific questions are even more affected by the quality of data definitions, and the efforts involved in data management are considerably higher. At the end of this paper we sketch a hybrid information system for computer-aided anesthesia recording that combines data collection and data analysis.