21 |
Generation of Synthetic Clinical Trial Subject Data Using Generative Adversarial NetworksLindell, Linus January 2024 (has links)
The development of new solutions incorporating artificial intelligence (AI) within the medical field is an area of great interest. However, access to comprehensive and diverse datasets is restricted due to the sensitive nature of the data. A potential solution to this is to generatesynthetic datasets based on real medical data. Synthetic data could protect the integrity of the subjects while preserving the inherent information necessary for training AI models and be generated in greater quantity than otherwise available. This thesis project aims to generate reliable clinical trial subject data using a generative adversarial network (GAN). The main data set used is a mock clinical trial dataset consisting of multiple subject visits, however an additional data set containing authentic medical data is also used for better insights into the model’s ability to learn underlying relationships. The thesis also investigates training strategies for simulating the temporal dimension and the missing values in the data. The GAN model used is an altered version of the Conditional Tabular GAN (CTGAN)made to be compatible with the preprocessed clinical trial mock data, and multiple model architectures and number of training epochs are examined. The results show great potential for GAN models on clinical trial datasets, especially for real-life data. One model, trained on the authentic dataset, generates near-perfect synthetic data with respect to column distributions and correlation between columns. The results also show that classification models trained on synthetic data and tested on real data have the potential to match the performance of classification models trained on real data. While the synthetic data replicates the missing values, no definitive conclusion can be drawn regarding the temporal characteristics due to the sparsity of the mock dataset and lack of real correlations in it. Although the results are promising, further experiments on authentic datasets with less sparsity are required.
|
22 |
Managing a digital transformation : A case study of digitizing functional operations in a sociotechnical system / Hantering av en digital transformation : En fallstudie av digitalisering av funktionella verksamheter i ett sociotekniskt systemFlodmark, Erik, Sävendahl, Carl January 2021 (has links)
Sweden has the ambition to be the world leading country leveraging the opportunities of digitalization in the healthcare sector. In parallel, the Swedish Research Council highlights that conducting more clinical studies is essential to improve the healthcare. Henceforth, considering the need for increased operational efficiency as an enabler for increased clinical activity, a digital transformation of the industry was identified as a potential catalyst. The study thus utilizes a cognitive work analysis framework to investigate the potential benefits and risks of digitizing the functional operations at a contract management department for clinical studies at a Swedish university hospital. The aim is thereafter to determine the appropriate properties necessary to consider managing a digital transformation. The analysis identified three key benefits from a digitization. 1) transparent data sharing, 2) standardized contract management and 3) efficient operations. These three aspects are currently insufficient at the department hindering the objective of increasing the clinical activity. The study found that a digital transformation would be suitable in order to mitigate these insufficiencies and consequently facilitating the achievement of the objectives. Thereafter, the study found the key properties to consider managing a digital transformation to be interoperability, quality, adaptability and usability. In addition, safety was found critical to be considered in the transformation as the contract management department acts under rigid laws and regulations on ethics and patient security with which digitized processes must comply. The results contribute to the field of cognitive systems engineering. However, the study has limitations regarding the reliability and generalizability of the results. The findings are based on a single case study, which may not be representative for the industry in general nor for university hospitals in particular. In addition, since no actual digitalization effort was performed at the organization during the study, appropriate properties key to consider in the digital transformation are speculative by design. Consequently, it is necessary to study an actual implementation process in future research and whether the proposed considerations are sufficient in order to realize the suggested benefits of such a digitalization. / Sverige har ett övergripande mål att vara det ledande landet när det gäller att dra nytta av digitaliseringens möjligheter inom sjukvården. Dessutom understryker Vetenskapsrådet att det är centralt för förbättrad sjukvård att öka antalet kliniska studier i landet. Följaktligen, med tanke på behovet av en ökad operativ effektivitet, identifierade författarna det av intresse att studera digitalisering av branschen. Studien tillämpar således ett kognitivt ramverk för arbetsanalys i syfte att undersöka de potentiella fördelarna eller riskerna med att digitalisera den funktionella verksamheten hos en kontrakthanteringsavdelning för kliniska studier vid ett stort svenskt universitetssjukhus. Målsättningen är därefter att ta fram lämpliga egenskaper som är nödvändiga att beakta vid hanteringen av den digitala transformationen. Kontrakthanteringsavdelningen fanns att inneha brister i sina arbetsprocesser gällande transparens, effektivitet och standardisering vilket hindrar målet avseende ökad klinisk aktivitet. Studien visade att en digital transformation skulle vara nödvändig för att motverka dessa brister, samt för att möjliggöra en uppskalning av organisationen. Ett annat specifikt förbättringsområde som skulle underlättas av en digital transformation visade sig vara förbättrad synkronisering mellan arbetsprocesser. Vidare fann studien att de mest kritiska egenskaperna nödvändiga att beakta, vid hantering av en digital transformation, skulle vara interoperabilitet, kvalitet, anpassningsförmåga och användbarhet. Dessutom är säkerhet en egenskap som visat sig vara kritisk att beakta vid digitalisering då kontrakthanteringsavdelningen lyder under stränga lagar och föreskrifter beträffande etik och patientsäkerhet. Resultaten bidrar till forskningsområdet cognitive systems engineering. Studien har dock vissa begränsningar gällande tillförlitlighet och generaliserbarhet. Resultaten är baserade på en enfallstudie, som eventuellt inte är representativ för branschen i allmänhet eller för universitetssjukhus i synnerhet. Dessutom, då ingen digitaliseringsinsats utfördes under studien är de viktiga egenskaperna att beakta i den digitala transformationen enbart spekulativa. Således är det i framtida forskning viktigt att studera en faktisk implementation och då studera om föreslagna beaktanden är tillräckliga för att utnyttja digitaliseringens möjligheter.
|
Page generated in 0.0764 seconds