“Machine learning is the process of discovering and interpreting meaningful information, such as new correlations, patterns and trends by sifting through large amounts of data stored in repositories, using pattern recognition technologies as well as statistical and mathematical techniques” (Larose, 2005). From my understanding, machine learning is a process of using different analysis techniques to observe previously unknown, potentially meaningful information, and discover strong patterns and relationships from a large dataset. Professor Kasabov (2007b) classified computational models into three categories (e.g. global, local, and personalised) which have been widespread and used in the areas of data analysis and decision support in general, and in the areas of medicine and bioinformatics in particular. Most recently, the concept of personalised modelling has been widely applied to various disciplines such as personalised medicine, personalised drug design for known diseases (e.g. cancer, diabetes, brain disease, etc.) as well as for other modelling problems in ecology, business, finance, crime prevention, and so on. The philosophy behind the personalised modelling approach is that every person is different from others, thus he/she will benefit from having a personalised model and treatment. However, personalised modelling is not without issues, such as defining the correct number of neighbours or defining an appropriate number of features. As a result, the principal goal of this research is to study and address these issues and to create a novel framework and system for personalised modelling. The framework would allow users to select and optimise the most important features and nearest neighbours for a new input sample in relation to a certain problem based on a weighted variable distance measure in order to obtain more precise prognostic accuracy and personalised knowledge, when compared with global modelling and local modelling approaches.

“Machine learning is the process of discovering and interpreting meaningful information, such as new correlations, patterns and trends by sifting through large amounts of data stored in repositories, using pattern recognition technologies as well as statistical and mathematical techniques” (Larose, 2005). From my understanding, machine learning is a process of using different analysis techniques to observe previously unknown, potentially meaningful information, and discover strong patterns and relationships from a large dataset. Professor Kasabov (2007b) classified computational models into three categories (e.g. global, local, and personalised) which have been widespread and used in the areas of data analysis and decision support in general, and in the areas of medicine and bioinformatics in particular. Most recently, the concept of personalised modelling has been widely applied to various disciplines such as personalised medicine, personalised drug design for known diseases (e.g. cancer, diabetes, brain disease, etc.) as well as for other modelling problems in ecology, business, finance, crime prevention, and so on. The philosophy behind the personalised modelling approach is that every person is different from others, thus he/she will benefit from having a personalised model and treatment. However, personalised modelling is not without issues, such as defining the correct number of neighbours or defining an appropriate number of features. As a result, the principal goal of this research is to study and address these issues and to create a novel framework and system for personalised modelling. The framework would allow users to select and optimise the most important features and nearest neighbours for a new input sample in relation to a certain problem based on a weighted variable distance measure in order to obtain more precise prognostic accuracy and personalised knowledge, when compared with global modelling and local modelling approaches.

“Machine learning is the process of discovering and interpreting meaningful information, such as new correlations, patterns and trends by sifting through large amounts of data stored in repositories, using pattern recognition technologies as well as statistical and mathematical techniques” (Larose, 2005). From my understanding, machine learning is a process of using different analysis techniques to observe previously unknown, potentially meaningful information, and discover strong patterns and relationships from a large dataset. Professor Kasabov (2007b) classified computational models into three categories (e.g. global, local, and personalised) which have been widespread and used in the areas of data analysis and decision support in general, and in the areas of medicine and bioinformatics in particular. Most recently, the concept of personalised modelling has been widely applied to various disciplines such as personalised medicine, personalised drug design for known diseases (e.g. cancer, diabetes, brain disease, etc.) as well as for other modelling problems in ecology, business, finance, crime prevention, and so on. The philosophy behind the personalised modelling approach is that every person is different from others, thus he/she will benefit from having a personalised model and treatment. However, personalised modelling is not without issues, such as defining the correct number of neighbours or defining an appropriate number of features. As a result, the principal goal of this research is to study and address these issues and to create a novel framework and system for personalised modelling. The framework would allow users to select and optimise the most important features and nearest neighbours for a new input sample in relation to a certain problem based on a weighted variable distance measure in order to obtain more precise prognostic accuracy and personalised knowledge, when compared with global modelling and local modelling approaches.

Analyse intégrative de données génomiques et pharmacologiques pour améliorer la prédiction de la réponse aux thérapies cibléesL'utilisation de thérapies ciblées dans le contexte de la médecine personnalisée du cancer a permis d’améliorer le traitement des patients dans différents types de cancer. Cependant, alors que la décision thérapeutique est basée sur une unique altération moléculaire (par exemple une mutation ou un changement du nombre de copies d’un gène), les tumeurs montrent différents degrés de réponse. Dans cette thèse, nous démontrons que la décision thérapeutique basée sur une unique altération n’est pas optimale et nous proposons un modèle mathématique intégrant des données génomiques et pharmacologiques pour identifier de nouveaux biomarqueurs prédictifs de la réponse thérapeutique. Le modèle a été construit à partir de deux bases de données de lignées cellulaires (the Genomics of Drug Sensitivity in Cancer, GDSC and the Cancer Cell Line Encyclopedia, CCLE) et validé avec des données de lignées et des données cliniques. De plus, nous avons également développé une nouvelle méthode pour améliorer la détection des mutations somatiques à partir de données de séquençage d'exomes complets et proposons un nouvel outil, cmDetect, disponible gratuitement pour la communauté scientifique. / Integrated analysis of genomic and pharmacological data to better predict the response to targeted therapiesThe use of targeted therapies in the context of cancer personalized medicine has shown great improvement of patients’ treatment in different cancer types. However, while the therapeutic decision is based on a single molecular alteration (for example a mutation or a gene copy number change), tumors will show different degrees of response. In this thesis, we demonstrate that a therapeutic decision based on a unique alteration is not optimal and we propose a mathematical model integrating genomic and pharmacological data to identify new single predictive biomarkers as well as combinations of biomarkers of therapy response. The model was trained using two public large-scale cell line data sets (the Genomics of Drug Sensitivity in Cancer, GDSC and the Cancer Cell Line Encyclopedia, CCLE) and validated with cell line and clinical data. Additionally, we also developed a new method for improving the detection of somatic mutations using whole exome sequencing data and propose a new tool, cmDetect, freely available to the scientific community.

Radioembolization (also called selective internal radiation therapy, SIRT) with yttrium-90 (90Y)-loaded microspheres has been broadly adopted as a locoregional therapy for primary and metastatic liver cancers. Although radioembolization is a well-established therapy, efforts to personalise and refine the planning and administration of therapy are ongoing. The ability to accurately predict, plan and deliver optimal doses to tumour and non-tumour tissues, including final validation of dose distribution, is essential for successful radiotherapy. Determining the true dose absorbed by tissue compartments is the primary way to safely individualise therapy for maximal response while respecting normal tissue tolerances. The overarching objective of this work was to expand our knowledge of dosimetry in 90Y-resin-microsphere radioembolization, with the ultimate goal of improving the clinical outcomes in our patients. Initially we sought to identify the patient- and treatment-related variables that predict radioembolization outcome in patients with intrahepatic cholangiocarcinoma (Chapter 2). Then, as a step toward personalised radioembolization in liver metastases from colorectal cancer patients, we evaluated the relationship between radioembolization real absorbed dose, as determined by 90Y positron emission tomography, and outcome (lesion-based and patient-based) (Chapter 3). In the work described in Chapter 4, we compared predictive (simulated) and post-treatment (real) dosimetry in liver metastases from colorectal cancer patients to pursue radioembolization personalisation. Finally, based on experience accumulated in previous studies and advances reported in the literature, we generated state-of-the-art recommendations to assist practitioners in performing personalised radioembolization with 90Y-resin microspheres in patients with primary and metastatic liver tumours (Chapter 5). / Doctorat en Sciences biomédicales et pharmaceutiques (Médecine) / info:eu-repo/semantics/nonPublished

Background: Tech-enabled personalised nutrition is an emerging area that has promise to improve health outcomes, widen access to nutrition expertise and reduce healthcare expenditure, yet uptake by registered dietitians remains low. This research programme aimed to identify levers and barriers that contribute to adoption of personalised nutrition in order to guide practice and policy for registered dietitians, educators and consumers. Methods: A mixed methods study with a sequential exploratory design was adopted to determine what the barriers to adoption of technologies are, and secondly, what needs to be in place to make tech-enabled personalised nutrition a reality. The research programme was conducted online using qualitative (focus groups and interviews) and quantitative measures (survey and secondary analysis). Thematic analysis, statistical and secondary analyses of data were performed respectively. Results: Using diffusion of innovation and entrepreneurial theories, findings indicate that barriers to integration of personalised nutrition technologies include intrinsic and extrinsic factors which relate to a low self-efficacy, high perception of risk, low perceived importance and usefulness of technologies to dietetic practice as well as a lack of an entrepreneurial mindset and regulatory environment. Conclusion: Uptake of tech-enabled personalised nutrition by registered dietitians will require a multi-stakeholder approach. Educational, professional, regulatory and health policies will need to be in place and strategies that open discussion between Registered Dietitians (RD’s) at all levels are needed.

No / Traditional Chinese medicine (TCM) investigates the clinical diagnosis and treatment regularities in a typical schema of personalized medicine, which means that individualized patients with same diseases would obtain distinct diagnosis and optimal treatment from different TCM physicians. This principle has been recognized and adhered by TCM clinical practitioners for thousands of years. However, the underlying mechanisms of TCM personalized medicine are not fully investigated so far and remained unknown. This paper discusses framework of TCM personalized medicine in classic literatures and in real-world clinical settings, and investigates the underlying mechanisms of TCM personalized medicine from the perspectives of network medicine. Based on 246 well-designed outpatient records on insomnia, by evaluating the personal biases of manifestation observation and preferences of herb prescriptions, we noted significant similarities between each herb prescriptions and symptom similarities between each encounters. To investigate the underlying mechanisms of TCM personalized medicine, we constructed a clinical phenotype network (CPN), in which the clinical phenotype entities like symptoms and diagnoses are presented as nodes and the correlation between these entities as links. This CPN is used to investigate the promiscuous boundary of syndromes and the co-occurrence of symptoms. The small-world topological characteristics are noted in the CPN with high clustering structures, which provide insight on the rationality of TCM personalized diagnosis and treatment. The investigation on this network would help us to gain understanding on the underlying mechanism of TCM personalized medicine and would propose a new perspective for the refinement of the TCM individualized clinical skills.

Yes / Personalised nutrition (PN) has the potential to reduce disease risk and optimise health and performance. Although previous research has shown good acceptance of the concept of PN in the UK, preferences regarding the delivery of a PN service (e.g. online v. face-to-face) are not fully understood. It is anticipated that the presence of a free at point of delivery healthcare system, the National Health Service (NHS), in the UK may have an impact on end-user preferences for deliverances. To determine this, supplementary analysis of qualitative data obtained from focus group discussions on PN service delivery, collected as part of the Food4Me project in the UK and Ireland, was undertaken. Irish data provided comparative analysis of a healthcare system that is not provided free of charge at the point of delivery to the entire population. Analyses were conducted using the ‘framework approach’ described by Rabiee (Focus-group interview and data analysis. Proc Nutr Soc 63, 655-660). There was a preference for services to be led by the government and delivered face-to-face, which was perceived to increase trust and transparency, and add value. Both countries associated paying for nutritional advice with increased commitment and motivation to follow guidelines. Contrary to Ireland, however, and despite the perceived benefit of paying, UK discussants still expected PN services to be delivered free of charge by the NHS. Consideration of this unique challenge of free healthcare that is embedded in the NHS culture will be crucial when introducing PN to the UK.

Yes / In order to determine the efficacy of behavior change techniques (BCT) applied in dietary and physical activity intervention studies, it is first necessary to record and describe techniques which have been used during such interventions. Published frameworks used in dietary and smoking cessation interventions undergo continuous development and most are not adapted for online delivery. The Food4Me study (N=1607) provided the opportunity to use existing frameworks to describe standardized online techniques employed in a large-scale internet-based intervention to change dietary behaviour and physical activity.

Yes / Social Cognitive Theory has been used to explain findings derived from focus group discussions (N = 4) held in the United Kingdom with the aim of informing best practice in personalised nutrition. Positive expectancies included weight loss and negative expectancies surrounded on-line security. Monitoring and feedback were crucial to goal setting and progress. Coaching by the service provider, family and friends was deemed important for self-efficacy. Paying for personalised nutrition symbolised commitment to behaviour change. The social context of eating, however, was perceived a problem and should be considered when designing personalised diets. Social Cognitive Theory could provide an effective framework through which to deliver personalised nutrition. / This work was supported by the European Commission under the Food, Agriculture, Fisheries and Biotechnology Theme of the 7th Framework Programme for Research and Technological Development (265494).