Please use this identifier to cite or link to this item: http://hdl.handle.net/10609/138466
Title: Entrenament d'una xarxa neuronal per al diagnòstic de lesions de la pell amb el dataset HAM10000
Author: Baranguer Codina, Albert
Tutor: Yu, Longlong
Others: Ventura, Carles  
Abstract: There's a variety of techniques availables in the skin cancer diagnosis. Moreover minor surgery techniques, as biopsy, it's possible to apply diagnosis through image. Image Diagnostics is, in essence, a classification image problem. So, it's a problem that can be solved using Machine Learning and Deep Learning techniques. The subject of this work is the image classification applied to diagnosis of skin lesions, as skin cancer is. But more important that achieving optimal results in classification, the main objective has been to acquire a general understanding of basic concepts and techniques in Machine Learning / Deep Learning applied to Medical Images Classification. And the knowledge and practical use o Machine Learning / Deep Learning basic techniques for improving the performance of the classifier networks. Having in consideration the limitations in time and hardware resources availables, it has been used a little neural netwok of the Resnet18 type. The Neural network has been trained using the HAM10000 dataset involving Supervised Machine Learning techniques, with the use of scripts and classes developped in Python and the PyTorch framework over differents development environments. The methodology followed has been iterative and incremental. This work compiles the results, and brief introductions to the different theroretical amb practical aspects treated.
Keywords: cancer
classification
diagnosis
Document type: info:eu-repo/semantics/masterThesis
Issue Date: Jan-2022
Publication license: http://creativecommons.org/licenses/by-nc-nd/3.0/es/  
Appears in Collections:Bachelor thesis, research projects, etc.

Files in This Item:
File Description SizeFormat 
abaranguerTFM0122memòria.pdfMemòria del TFM5,61 MBAdobe PDFThumbnail
View/Open
abaranguerTFM0122presentació.pdfPresentació del TFM1,56 MBAdobe PDFThumbnail
View/Open