Population-Based Advanced Optimisation Algorithms for Electrical Impedance Tomography Image Reconstruction

Publication Type:
Thesis
Issue Date:
2020
Full metadata record
The necessity of preventing living tissues' direct exposure to ionising radiation has resulted in tremendous growth in the area of medical imaging and e-health, enhancing intensive care of perilous patients, and help to improve quality of life. Moreover, the practice of image-reconstruction instruments that utilise ionising radiation has a significant impact on the health of the patients. Long or frequent exposure to ionizing radiation is linked to several illnesses like Cancer. These factors urged to enhance the endeavours to advance non-invasive approaches, for instance, Electrical Impedance Tomography (EIT) which is a portable, non-invasive, low-cost, and safe imaging method. Nevertheless, EIT image reconstruction still demands more exploitation, as it is an inverse and ill-conditioned problem. Numerous numerical techniques are used to answer this problem without producing anatomically, unpredictable outcomes. Evolutionary Computational techniques can be used as substitutes to the conventional methods that usually create low-resolution blurry images. EIT reconstruction techniques work on the principle of optimising the relative error of reconstruction utilising population-based optimisation methods that have been presented in this work. Three advanced optimisation methods have been developed to facilitate the iterative procedure for avoiding anatomically erratic solutions. Three different optimising techniques namely, a) Advanced Particle Swarm Optimisation Algorithm (APSO), b) Advanced Gravitational Search Algorithm (AGSA), and c) Hybrid Gravitational Search Particle Swarm Optimization Algorithm (HGSPSO) are used. By utilizing the advantages of these proposed techniques, the performance in terms of convergence and solution stability is improved. […] EIT images were obtained from the EIDORS library database for two case studies. The image reconstruction was optimized using the three proposed algorithms. EIDORS library was used for generating and solving forward and reverse problems. Two case studies were undertaken, i.e. circular tank simulation and gastric emptying. The results thus obtained are analysed and presented as a real-world application of population-based optimization methods. Results obtained from the proposed methods are quantitatively assessed with ground truth images by using the relative mean squared error, confirming that a low error value is reached in the results. HGSPSO algorithm has superior performance as compared to the other proposed methods in terms of solution quality and stability.
Please use this identifier to cite or link to this item: