This thesis is made up of three parts: i) the development of a comprehensive computational model of the pulmonary (patho)physiology of healthy and diseased lungs, ii) the application of a novel optimisation-based approach to validate this computational model, and iii) the use of this model to optimise mechanical ventilator settings for patients with diseased lungs. The model described in this thesis is an extended implementation of the Nottingham Physiological Simulator (NPS) in MATLAB. An iterative multi-compartmental modelling approach is adopted, and modifications (based on physiological mechanisms) are proposed to characterise healthy as well as diseased states. In the second part of the thesis, an optimisation-based approach is employed to validate the robustness of this model. The model is subjected to simultaneous variations in the values of multiple physiologically relevant uncertain parameters with respect to a set of specified performance criteria, based on expected levels of variation in arterial blood gas values found in the patient population. Performance criteria are evaluated using computer simulations. Local and global optimisation algorithms are employed to search for the worst-case parameter combination that could cause the model outputs to deviate from their expected range of operation, i.e. violate the specified model performance criteria. The optimisation-based analysis is proposed as a useful complement to current statistical model validation techniques, which are reliant on matching data from in vitro and in vivo studies. The last section of the thesis considers the problem of optimising settings of mechanical ventilation in an Intensive Therapy Unit (ITU) for patients with diseased lungs. This is a challenging task for physicians who have to select appropriate mechanical ventilator settings to satisfy multiple, sometimes conflicting, objectives including i) maintaining adequate oxygenation, ii) maintaining adequate carbon dioxide clearance and iii) minimising the risks of ventilator associated lung injury (VALI). Currently, physicians are reliant on guidelines based on previous experience and recommendations from a very limited number of in vivo studies which, by their very nature, cannot form the basis of personalised, disease-specific treatment protocols. This thesis formulates the choice of ventilator settings as a constrained multi-objective optimisation problem, which is solved using a hybrid optimisation algorithm and a validated physiological simulation model, to optimise the settings of mechanical ventilation for a healthy lung and for several pulmonary disease cases. The optimal settings are shown to satisfy the conflicting clinical objectives, to improve the ventilation perfusion matching within the lung, and, crucially, to be disease-specific.

It is well known that multi-input multi-output (MIMO) wireless communication systemsthat employ precoding techniques are capable of meeting the high expectations of modernand future wireless communication standards. In order to fully utilize these techniques, thecommunication system typically requires information of the channel, commonly referred toas channel state information (CSI). In practice, the CSI at the transmitter (CSIT) is oftennot perfect which addresses the need for robust precoding designs, that can mitigate theeffects of precoding with imperfect CSIT. By modeling the imperfect CSIT as deterministic,it can be assumed that the estimated channel, as represented by the CSIT, belongs to aconvex uncertainty set. With this approach, the problem of finding a robust precoding designcan be formulated as a convex maximin problem, where the solution optimizes the systemperformance for the worst channel that belongs to the uncertainty set. How the uncertaintyset is modeled impacts the performance of the communication system, which calls for theevaluation of several robust precoding designs. While different characteristics of the convexuncertainty sets has been evaluated for MIMO flat-fading channels represented by i.i.d. zero-mean, unit variance Gaussian elements, it is of interest to apply the theory of worst-caserobust precoding designs on real-world measured MIMO channels.More concisely, this project investigates MIMO precoding designs with deterministic im-perfect CSIT for real-world measured channels that utilizes orthogonal frequency divisionmultiplexing (OFDM) schemes. The worst-case received signal-to-noise ratio (SNR) will bepresented as a result of using MIMO precoding designs on real-world channels, and the effectof the choice of model parameters and characteristics of the chosen uncertainty set will bevisualized and discussed. Furthermore, orthogonal space-time block code (OSTBC) transmis-sion designs will be employed to measure the worst case symbol error rate (SER) as a tool toevaluate the system performance in different scenarios. The results will be compared to thatwhen the channel is composed of i.i.d. zero-mean, unit variance Gaussian elements and forthe case when the channel is based on the Kronecker model.The results indicate that a further analysis of how the Kronecker model behaves in termsof capacity is required in order to draw accurate conclusions regarding the implementation ofrobust precoding strategies when each pair of antennas are correlated. Also, it is essential todevelop a framework that offers methods on how to accurately model the uncertainty set sothat it can represent errors that originates from both quantization errors, estimation errorsand outdated estimates. / Det är välkänt att trådlösa multi-input, multi-output (MIMO) system som använder förkodar-tekniker har kapabilitet att möta de höga förväntningar som är fastställt av moderna ochframtida kommunikationsstandarder. För att utnyttja dessa förkodartekniker till fullo be-hövs information om kanalen (CSI). I praktiska kommunikationssystem är kanalinformatio-nen hos sändaren (CSIT) ofta inte perfekt vilket adresserar betydelsen av att använda robustaförkodare som kan mildra den negativa effekten som uppstår av att förkoda med CSIT som in-nehåller fel. Genom att använda en deterministisk modell för CSIT med fel kan man anta attden skattade kanalen som är representerad av CSIT tillhör en konvex osäkerhetsregion. Meddetta tillvägagångssätt kan man formulera problemet att hitta en robust förkodardesign somett konvext maximin-problem, där lösningen optimerar systemets prestanda för den värstakanalskattningen i osäkerhetsregionen. Olika modeller av osäkerhetsregioner ger upphov tillolika systemprestanda vilket betyder att olika modeller med tillhörande robusta förkodare be-höver utvärderas. Medan tidigare forskningsrapporter behandlat MIMO flat fädnings-kanalerför i.i.d. Gaussisk fördelning av elementen finns det ett intresse att applicera teorin omvärsta-fallet robust förkodning på riktiga uppmätta MIMO-kanaler.Mer koncist undersöker detta projekt designs på förkodare för riktiga uppmätta MIMO-kanaler utifrån en deterministisk modell på felaktigt CSIT, där MIMO-kanalerna utnyttjarorthogonal frequency divsion multiplexing (OFDM) scheman. Värsta-fallet signal-to-noiseratio (SNR) kommer presenteras för olika förkodar-designs och MIMO-kanaler. Hur olika valav modellparametrar och karakteristik hos osäkerhetsregionerna påverkar systemprestandankommer att diskuteras. Vidare kommer även orthogonal space-time block codes (OSTBC)användas som transmissionsscheman för att mäta symbol error rate (SER). Resultaten kom-mer att jämföras med när MIMO-kanalen består av i.i.d. Gaussisk fördelning av elementenoch för fallet när kanalen är baserad på en Kronecker-modell.Resultaten indikerar att en fortsatt analys av hur Kronecker-modellen beter sig medavseende på kapacitet är nödvändig för att dra tillförlitliga slutsatser om systemprestan-dan för förkodar-designs när antennparen är korrelerade. Det är även väsentligt att utvecklaen teori som behandlar metoder för hur man kan på ett tillförlitligt sätt modellera osäker-hetsregionen så CSIT så att kvantiseringsfel, skattningsfel och utdaterade skattningar kanrepresenteras i den.