Options
Deep Learning-Based Prostate Cancer Risk Prediction on Electronic Health Records and Histopathology Images
Citation Link: https://doi.org/10.15480/882.14817
Publikationstyp
Doctoral Thesis
Date Issued
2025
Sprache
English
Author(s)
Fuhlert, Patrick
Advisor
Referee
Title Granting Institution
Medizintechnische und Intelligente Systeme E-1
Place of Title Granting Institution
Hamburg
Examination Date
2025-02-05
TORE-DOI
Citation
Technische Universität Hamburg (2025)
For the selection of optimal patient treatment, survival prediction methods that estimate the expected time to an event of interest can be utilized. These models aim to provide accurate disease predictions such as for cancer patients based on characteristics of an individual patient or group. This thesis analyzes how deep learning-based survival prediction models can be utilized in the context of prostate cancer diagnostics. A survival prediction model called Discrete Calibrated Survival was developed using medical datasets and applied on a high quality dataset of radical prostatectomy patients. The results showed that the developed approach outperforms the commonly used Cox model regarding relapse-free survival prediction. Moreover, the thesis investigates one of the most important factors in predicting relapse-free survival of prostate cancer patients. To assess prostate cancer severity, pathologists traditionally use Gleason grading that involves evaluating individual tissue samples. However, this grading system suffers from interobserver variability. This thesis demonstrates that a newly developed deep learning-based cancer risk prediction model, called Prostate Cancer Aggressiveness Index, can surpass the Gleason grading system. Instead of relying on subjective annotations by human experts, this model utilizes the objective endpoint of relapse-free survival to assess the risk of individual tissue images. Additionally, this approach can be applied to biopsies when appropriate robustness measures are applied.
Subjects
survival analysis
prostate cancer
deep learning
digital histopathology
electronic health records
DDC Class
610: Medicine, Health
616: Deseases
006: Special computer methods
Publication version
publishedVersion
Loading...
Name
patrick_fuhlert_deep_learning-based_prostate_cancer_risk_prediction_on_electronic_health_records_and_histopathology_images.pdf
Size
28.44 MB
Format
Adobe PDF