381023 research outputs found
Sort by
Novel teachers, lovers, and readers : praeceptores amoris in Achilles Tatius and Longus amidst Greek and Latin literary traditions
No full textMy dissertation examines the erōtodidactic scenes in two Greek love novels, Longus’ Pastorals and Achilles Tatius’ Leucippe and Clitophon. It makes two hypotheses: first, that these scenes are indebted to scenes of erotic tuition in Latin Augustan poetry; second, that they self-reflexively comment on the strategies of the text to control and manipulate the reader’s emotional state, thereby actualizing strategies from Latin erōtotidactic tradition. Since until recently the direction of exchange Latin>Greek is controversial, I dedicate my first two chapters to methodological discussions. In chapter 1, I explain my choice of a reader-based methodology and in chapter 2, I isolate one reader of Greek novels with knowledge of Latin poetry, Apuleius. In chapter 3, I discuss Clinias’ erōtodidaxis in connection with Ovid’s Ars Amatoria. I argue that just as the speaker in the Ars Amatoria, Clinias teaches Clitophon both how to love and how to write about love. Furthermore, Clitophon uses the rhetorical guidelines taught by Clinias in his speech to Leucippe and in his narration to the primary narrator/reader, thereby suggesting that he attempts to seduce the reader into the story. Again, this actualises a rhetorical strategy of Ovid’s Ars Amatoria. My chapter 4 consists in a systematic study of the reception of the Georgics in Longus’ novel. I argue that the beginning of the second book of the novel, and in particular the scene in which Philetas, the most important praeceptor amoris in the novel, recounts his pursuit of Eros, could be read as alluding to the Aristaeus epyllion in the fourth book of the Georgics. I then explore how this possible allusion to the Georgics is rewarding for our interpretation of the themes of education and myth, didactic authority and the possibility of speech to control powerful emotions such as love in this passage of Longus. Longus appears to play with his reader by apparently depriving language of pragmatic effect, while all the same celebrating his novel’s ability to change his readers’ emotional state. This is perhaps the sign of a polemical discussion about the relative efficiency of poetry and sophistic rhetoric
Recovery and substitute addictions in the Western Cape, South Africa : a multi-perspective approach
No full text
Physiological basis for revascularization of left main coronary artery lesions and chronically occluded coronary arteries
No full textThis thesis sought evidence from coronary physiology that selected patients with a left main coronary artery (LM) lesion or a chronic total occlusion (CTO) of a coronary artery may benefit from myocardial revascularization (MR) and that physiological assessments may assist in guiding and/or optimizing treatment.
In the first part, we investigated the subset of patients with a LM lesion.
We first showed by analyzing the nationwide Quality-oriented Electronic Registration of Medical Implant Devices (QERMID) registry data, that patients undergoing LM percutaneous coronary intervention (PCI) had a significant risk of short- and long-term mortality. ST-elevation myocardial infarction (STEMI) patients presenting in cardiogenic shock and undergoing LM PCI also had a higher mortality compared to that in the whole group of cardiogenic shock patients.
Second, we reviewed the current evidence on physiology-based revascularization of LM lesions. We concluded that the recommended physiological assessment techniques, fractional flow reserve (FFR) and instantaneous wave-free ratio (iFR), have only been studied relatively limitedly in the setting of LM disease and are probably insufficient on their own to correctly predict the prognosis of the individual patient.
Third, we presented the first results of the Prospective Left Main Physiology (PHYNAL) registry that was established to collect more physiological, imaging and outcome data on this type of patients. Physiological measurements yielded contradictory results in more than a quarter of patients and even more so if concomitant downstream disease is present.
In the second part, we investigated the subset of patients with a CTO.
First, we looked at the evidence underlying CTO PCI. Most proven today is its potential to better reduce ischemic symptoms compared to optimal medical treatment alone.
Next, we showed our analysis of the Belgian Working Group on CTO (BWGCTO) registry, illustrating that today CTO PCI can be performed with a high success rate and few complications.
Finally, we presented the DISTAL CTO study results. We were able to demonstrate with physiological assessments that CTO patients do indeed have ischemic myocardium and that PCI can alleviate this, which may explain symptomatic improvement. We also showed that residual ischemia can persist in up to 30% of patients. To improve the clinical outcomes of CTO PCI, one should strive to optimize PCI results as best as possible, and our findings suggest that physiological assessment can help with that
Interpersonal dynamics and therapeutic relationship in patients with functional somatic syndromes
No full textAgainst all odds : explaining the construction of the Nord Stream 2 gas pipeline project
No full text
Deep generative models for navigation using active inference
No full textDe wereld interpreteren om te kunnen beslissen wat te doen, is vanzelfsprekend voor mensen en is ook iets dat we al vanaf geboorte leren. Maar, voor robots blijft het een grote uitdaging om zowel te leren observeren als om te leren welke acties te nemen in de wereld. Recent groeide de populariteit van een procestheorie van de hersenen genaamd Active Inference, waarin acties en observaties beschouwd worden als twee zijden van dezelfde munt. In dit raamwerk vormen de hersenen een bepaalde geloofstoestand van de wereld om te kunnen afleiden welke acties te nemen met als doel hun verrassing en onzekerheid over de wereld te minimaliseren. In dit proefschrift wordt een nieuwe manier gepresenteerd om active inference modellen te leren uit data afkomstig uit de echte wereld door middel van recente ontwikkelingen in deep learning en generatieve modelering. Verder wordt er aangetoond dat zo een geleerd model toepasbaar is om robotica problemen op te lossen zoals navigatie en anomalie detectie
