| KEYWORD |
Design and development of machine learning models to analyze clinical images for early diagnosis diseases
Thesis in external company
keywords ARTIFICIAL INTELLIGENCE, DECISION SUPPORT, MACHINE LEARNING, MEDICAL IMAGING, NEURAL NETWORKS
Reference persons SANTA DI CATALDO, EDOARDO PATTI
External reference persons Daniele Conti (daniele.conti@syndiag.ai)
Research Groups DAUIN - GR-06 - ELECTRONIC DESIGN AUTOMATION - EDA, ELECTRONIC DESIGN AUTOMATION - EDA, GR-06 - ELECTRONIC DESIGN AUTOMATION - EDA, ICT4SS - ICT FOR SMART SOCIETIES
Thesis type EXPERIMENTAL, IN COMPANY
Description SynDiag in an Italian startup born to enable early diagnosis of ovarian cancer with artificial intelligence applied to medical imaging.
Ovarian cancer is a pathology presenting high mortality due to late diagnosis: 75% of clinical cases are detected when already developed and with survival rate at 30%. Performing an early diagnosis would increase the survival probability as high as 90%. For such a reason SynDiag wants to equip al gynecologists with a medical device based on AI that speeds up the diagnostic process.
SynDiag is a young team composed of researchers, medical doctors and entrepreneurs, hosted at I3P – Incubator of Politecnico di Torino. We collaborate with hospitals in Italy and Israel.
The thesis here proposed, in the field of machine learning, is focused on the development of neural network models, such as multi layer perceptrons, deep convolutional neural networks, region-based convolutional neural networks.
The thesis will focus on:
- Analysis of neural networks algorithms and available learning protocols
- Training of neural networks with medical images data
- Testing and evaluation of clinical accuracy of trained NN models
Deadline 15/02/2022
PROPONI LA TUA CANDIDATURA