Ammonia, a colorless gas with a characteristic pungent odor, is produced by various industrial and agricultural activities. Emissions of ammonia into the atmosphere not only cause a nuisance in the vicinity of the sources, but also have various environmental effects, such as eutrophication and acidification of terrestrial and aquatic ecosystems, and visibility problems resulting from the formation of aerosols. The traditional treatment of ammonia emissions is based on physical and/or chemical processes, both of which are expensive and produce secondary pollutants. Biological methods are effective and economical for biodegradable odorants and VOC contaminants. This study used fixed-film bioreactors, a biofilter and a biotrickling filter, to remove and oxidize gas-born ammonia. Firstly, a pilot-scale biofilter consisted of two columns (40 cmW ¡Ñ 40 cmL ¡Ñ 70 cmH acrylic column) arranged in series. A medium consisting solely of fern chips, on which biofilms were cultivated, was used as a packing material. The biofilter was tested continuously for 110 days, measuring the removal efficiency, empty bed residence time (EBRT), removal capacity, pressure drop, moisture content and pH. Most of ammonia was eliminated in the first biofiltration column and the removal efficiency increased with the increase in EBRT. Complete removal of the influent ammonia (20-120 ppm) was obtained with an ammonia loading as high as 5.4 g N kg-1 dry media d-1 during the experiment. The Michaelis-Menten equation was tested to be adequate for modeling the ammonia elimination kinetics in the biofilter and the maximum removal rate (Vm) and the half-saturation constant (Ks) were estimated to be 28.2 g N kg-1 dry media d-1 and 129 ppm, respectively. Secondly, a pilot-scale reactor, consisting of a set of two-stage-in-series biotrickling filters, an influent gas supply system and a liquid recirculation system, was utilized to treat ammonia in an air stream. Each stage of the biotrickling filter was constructed from a 20 cm ¡Ñ 200 cm (inner diameter ¡Ñ height) acrylic column packed with cokes (average diameter = 3.0 cm, specific area = 150 m2/m3) of 125 cm height. Experimental results indicate that a time of 30 days is required for development of biofilms for nitrification of the absorbed ammonia from the gas. Long-term (187 days) experimental results show that, in the conditions of EBRT (empty bed gas retention time) = 7.25 s, ¡§circulation liquid/gas¡¨ flow rate ratio = 7.7 L m-3, and liquid pH = 6.65, the level of ammonia in the influent gas was reduced from 230 to 4.0 ppm. With the volumetric ammonia loading of less than 7.37 g NH3-N m-3 hr-1, the system could achieve ammonia removal and nitrification efficiencies of 98 and 94%, respectively, without supplementary glucose as a carbon source. However, with a loading of 13.1 g NH3-N m3 h-1, both decreased gradually due to a lake of carbon source and an accumulation of ammonium and nitrite ions in the recirculation liquid.

This study aimed to develop a biofilter packed only with fern chips for the removal of air-borne low concentration VOCs (volatile organic compounds) emitted from various industries such as semiconductor manufacturing and electronic ones. The fern chip biofilters could avoid the shortcomings of traditional media, such as compaction, drying, and breakdown, which lead to the performance failure of the biofilters. The study contains two topics. The first is a performance test on the elimination of mixed VOCs used in semiconductor manufacturing industries in an air stream. The second is the one on the elimination of a single VOC (methyl ethyl ketone) in a waste gas drawn from a CCL (copper clad laminate) plant. Two pilot-scale biofilters consisted of two columns (0.40 mW¡Ñ0.40 mL¡Ñ0.70 mH acrylic column) arranged in series were used for the performance tests. Each of the two columns was packed with fern chips to a packing volume of around 56 L (0.40 mW¡Ñ0.40 mL¡Ñ0.35 mH). A sprinkler was set over the packed fern chips for providing them with water and nutrition solutions. Liquid leached from both layers of chips were collected in the bottom container of the column. In the first topic, tests were performed for biofiltration removal of VOCs in simulated semiconductor manufacturing emitted gases which consisted of IPA (isopropyl alcohol), acetone, HMDS (hexamethylene disilazane), PGME (propylene glycol monomethyl ether), and PGMEA (propylene glycol monomethyl ether acetate). From the results, it could be proposed that for achieving over 94% of the VOC removal, appropriate operation conditions are media moisture content = 52-68%, media pH = 7-8, influent VOC concentration = 150-450 mg/Am3, empty bed residence time (EBRT) = 0.75 min, and volumetric organic loading L to the whole media = 11.4-34.1 g/m3.h. In the second topic, performances of biofiltration for the removal of methyl ethyl ketone (MEK) in a gas stream from a copper clad laminate (CCL) manufacturing process were tested. Experimental results indicate that with L of <115 g /m3.h., EBRT = 0.5-1.28 min , media pH = 5.3-6.8, influent MEK concentration = 215-1,670 mg/Am3, MEK removal efficiencies of over 91% were obtained. Instant milk powder was essential to the good and stable performance of the biofilter for MEK removal.

This thesis addresses three problems related to image registration, prediction and tracking, applied to Angiography and Oncology. For image analysis, various probabilistic models have been employed to characterize the image deformations, target motions and state estimations. (i) In Digital Subtraction Angiography (DSA), having a high quality visualization of the blood motion in the vessels is essential both in diagnostic and interventional applications. In order to reduce the inherent movement artifacts in DSA, non-rigid image registration is used before subtracting the mask from the contrast image. DSA image registration is a challenging problem, as it requires non-rigid matching across spatially non-uniform control points, at high speed. We model the problem of sub-pixel matching, as a labeling problem on a non-uniform Markov Random Field (MRF). We use quad-trees in a novel way to generate the non uniform grid structure and optimize the registration cost using graph-cuts technique. The MRF formulation produces a smooth displacement field which results in better artifact reduction than with the conventional approach of independently registering the control points. The above approach is further improved using two models. First, we introduce the concept of pivotal and non-pivotal control points. `Pivotal control points' are nodes in the Markov network that are close to the edges in the mask image, while 'non-pivotal control points' are identified in soft tissue regions. This model leads to a novel MRF framework and energy formulation. Next, we propose a Gaussian MRF model and solve the energy minimization problem for sub-pixel DSA registration using Random Walker (RW). An incremental registration approach is developed using quad-tree based MRF structure and RW, wherein the density of control points is hierarchically increased at each level M depending of the features to be used and the required accuracy. A novel numbering scheme of the control points allows us to reuse the computations done at level M in M + 1. Both the models result in an accelerated performance without compromising on the artifact reduction. We have also provided a CUDA based design of the algorithm, and shown performance acceleration on a GPU. We have tested the approach using 25 clinical data sets, and have presented the results of quantitative analysis and clinical assessment. (ii) In External Beam Radiation Therapy (EBRT), in order to monitor the intra fraction motion of thoracic and abdominal tumors, the lung diaphragm apex can be used as an internal marker. However, tracking the position of the apex from image based observations is a challenging problem, as it undergoes both position and shape variation. We propose a novel approach for tracking the ipsilateral hemidiaphragm apex (IHDA) position on CBCT projection images. We model the diaphragm state as a spatiotemporal MRF, and obtain the trace of the apex by solving an energy minimization problem through graph-cuts. We have tested the approach using 15 clinical data sets and found that this approach outperforms the conventional full search method in terms of accuracy. We have provided a GPU based heterogeneous implementation of the algorithm using CUDA to increase the viability of the approach for clinical use. (iii) In an adaptive radiotherapy system, irrespective of the methods used for target observations there is an inherent latency in the beam control as they involve mechanical movement and processing delays. Hence predicting the target position during `beam on target' is essential to increase the control precision. We propose a novel prediction model (called o set sine model) for the breathing pattern. We use IHDA positions (from CBCT images) as measurements and an Unscented Kalman Filter (UKF) for state estimation. The results based on 15 clinical datasets show that, o set sine model outperforms the state of the art LCM model in terms of prediction accuracy.