polito.it
Politecnico di Torino (logo)

Deep Learning approaches for prediction of Hepatocellular Carcinoma Recurrence post Liver Transplantation

Carmine De Stefano

Deep Learning approaches for prediction of Hepatocellular Carcinoma Recurrence post Liver Transplantation.

Rel. Alfredo Benso, Quirino Lai, Gianfranco Michele Maria Politano. Politecnico di Torino, Corso di laurea magistrale in Data Science and Engineering, 2022

[img]
Preview
PDF (Tesi_di_laurea) - Tesi
Licenza: Creative Commons Attribution Non-commercial No Derivatives.

Download (2MB) | Preview
Abstract:

In recent years, several criteria have been identified for the selection of hepatocellular cancer (HCC) patients waiting for liver transplantation (LT). These criteria, like the Milan Criteria, are also the foundation of models for predicting the risk of relapse after transplantation. However, these models are severely limited in considering many variables and their non-linear interactions. The starting point is TRAIN-AI, based on the DeepSurv neural network: a Cox proportional hazards model. This model was developed starting from an International Cohort which will also be the main dataset of this study. Furthermore new deep learning models based on hazard rate parametrization and probability mass function (PMF) parametrization are proposed and compared, performing an appropriate hyperparameter tuning for each network.

Relators: Alfredo Benso, Quirino Lai, Gianfranco Michele Maria Politano
Academic year: 2022/23
Publication type: Electronic
Number of Pages: 89
Subjects:
Corso di laurea: Corso di laurea magistrale in Data Science and Engineering
Classe di laurea: New organization > Master science > LM-32 - COMPUTER SYSTEMS ENGINEERING
Aziende collaboratrici: Universita' degli Studi di Roma La Sapienza
URI: http://webthesis.biblio.polito.it/id/eprint/24746
Modify record (reserved for operators) Modify record (reserved for operators)