University of Twente Student Theses
Enhancing Cancer Treatment Planning : A Combined Approach of Process Mining and Machine Learning with a Focus on Colorectal Cancer (CRC)
Lin, C.Y. (2024) Enhancing Cancer Treatment Planning : A Combined Approach of Process Mining and Machine Learning with a Focus on Colorectal Cancer (CRC).
|
PDF
4MB |
| Abstract: | Colorectal cancer (CRC) is a major health issue globally, highlighting the need for better treatment planning methods. Traditional decision-making processes in CRC treatment face obstacles due to the complex nature of individual patient cases, making it difficult to tailor treatment effectively. The combination of process mining and machine learning offers a fresh perspective for enhancing the accuracy of treatment decisions. This approach analyzes patient data to forecast the outcomes of different treatment options with greater precision. The study uses clinical records from the MIMIC-III database, alongside sophisticated process mining tools (ProM), to find crucial information and patient features that impact CRC treatment results. It then trains machine learning models, particularly LSTM networks improved with transfer learning, to predict the survival rates of patients following various treatments. This method stands out for its use of process mining to map out treatment paths and machine learning to estimate the success of different treatments. The result based on the specified evaluation method offers a confidence range of ±0.14 (Avg\_EM), indicating that machine learning helps in analyzing and selecting the optimal treatment plans. This approach gives doctors a reliable range to customize patient care. Moreover, it helps patients to make informed choices about their treatment, reducing dependence on guesswork and uncertainty of the future. In conclusion, this research not only demonstrates the practicality of merging process mining and machine learning to better CRC treatment planning but also paves the way for future studies, such as addressing data sparsity issues or further expanding this research direction. Therefore, this integrated approach could influence the future of personalized medicine and achieve the goal of establishing a data-driven healthcare system. |
| Item Type: | Essay (Master) |
| Faculty: | EEMCS: Electrical Engineering, Mathematics and Computer Science |
| Subject: | 58 process technology |
| Programme: | Business Information Technology MSc (60025) |
| Link to this item: | https://purl.utwente.nl/essays/98595 |
| Export this item as: | BibTeX EndNote HTML Citation Reference Manager |
Show download statistics for this publication
Show download statistics for this publicationRepository Staff Only: item control page