Diabetic foot is one of the most common complications of diabetes. The condition causes anatomical and functional alterations of the foot and ankle, manifesting in the onset of injuries and ulcers that, in severe cases, can lead to amputation of the limb. In addition, the person may also incur impaired sensory abilities and fail to notice the presence of wound, burns, frostbite, as the ability to perceive stimuli reaching the level of the foot and respond appropriately to pain in loss. The following thesis illustrates the process by which a database of 3D scanners of shapes was created, finalized for the production of custom footwear. It is necessary, in fact, to provide the patient with a shoe adapted to his or her health condition in order to prevent and limit the formation of further ulcers, which cause pain and motor difficulty. The shoe must adapt to the biomechanics and deformity of the patient’s foot. In this study, scans of the limb and 3D shapes created during processing are made with the aim of expanding the database in favor of learning the configurator, which starting from the shape and measurements of the foot, must provide as output the most suitable shoe for the subject. Patients with diabetes, rheumatic patients and patients who have undergone surgical amputations were examined. This study was part of a collaborative project between the start-up BBSoF and the companies T&B and Podartis, in which the internship experience took place. These companies came together with the aim of creating software that would go a long way toward limiting operators’ errors in taking measurements and speeding up the process of choosing custom footwear. In the following, the methodology and tools possessed, such as possible machine learning algorithms and scans obtained during the manufacturing process, are studied and analyzed. The result is the preparation of data acquired from patients and devices owned by companies to generate software with the best possible learning and functionality.
Il piede diabetico è una delle complicanze più diffuse del diabete. La patologia causa alterazioni anatomiche e funzionali di piede e caviglia, manifestandosi con l’insorgenza di lesioni e ulcere che, nei casi più gravi, possono portare all’amputazione dell’arto. Inoltre la persona può incorrere anche in un’alterazione delle capacità sensitive e non accorgersi della presenza di ferite, ustioni, congelamento, poiché viene meno la capacità di percepire gli stimoli che giungono al livello del piede e di rispondere in modo adeguato al dolore. Nella seguente tesi viene illustrato il procedimento con il quale si è andato a creare un database di scanner di forme 3D, finalizzate alla produzione di calzature su misura. È necessario, infatti, fornire al paziente una calzatura adatta alla sua condizione di salute al fine di prevenire e limitare le formazioni di ulteriori ulcere, che possono degenerare in amputazioni. La scarpa si deve perciò adattare alla biomeccanica e alla deformità del piede del paziente. In questo studio, vengono realizzate scansioni dell’arto e delle forme 3D create durante la lavorazione con l’obiettivo di ampliare il database a favore dell’apprendimento del configuratore, il quale partendo dalla forma e dalle misure del piede, deve fornire come output la calzatura più idonea al soggetto. Sono stati presi in analisi pazienti affetti da diabete, pazienti reumatici e pazienti che hanno subito amputazioni chirurgiche. Questo studio si inserisce in un progetto di collaborazione tra la start-up BBSoF e le aziende T&B e Podartis, nelle quali è stata svolta l’esperienza di tirocinio. Queste realtà si sono unite con lo scopo di creare un software che andasse a limitare gli errori relativi alla presa di misure da parte degli operatori e di velocizzare il processo di scelta della calzatura custom. Nel seguito, vengono studiati e analizzati la metodologia e gli strumenti posseduti, come i possibili algoritmi di machine learning e le scansioni ottenute durante il processo di lavorazione. Il risultato finale è la preparazione dei dati acquisiti dai pazienti e dei dispositivi posseduti dalle aziende per generare un software con il miglior apprendimento e con la massima funzionalità possibile.
Creazione di un database di scanner 3D di forme per la produzione di calzature per piedi diabetici
MUSSONI, IRENE
2023/2024
Abstract
Diabetic foot is one of the most common complications of diabetes. The condition causes anatomical and functional alterations of the foot and ankle, manifesting in the onset of injuries and ulcers that, in severe cases, can lead to amputation of the limb. In addition, the person may also incur impaired sensory abilities and fail to notice the presence of wound, burns, frostbite, as the ability to perceive stimuli reaching the level of the foot and respond appropriately to pain in loss. The following thesis illustrates the process by which a database of 3D scanners of shapes was created, finalized for the production of custom footwear. It is necessary, in fact, to provide the patient with a shoe adapted to his or her health condition in order to prevent and limit the formation of further ulcers, which cause pain and motor difficulty. The shoe must adapt to the biomechanics and deformity of the patient’s foot. In this study, scans of the limb and 3D shapes created during processing are made with the aim of expanding the database in favor of learning the configurator, which starting from the shape and measurements of the foot, must provide as output the most suitable shoe for the subject. Patients with diabetes, rheumatic patients and patients who have undergone surgical amputations were examined. This study was part of a collaborative project between the start-up BBSoF and the companies T&B and Podartis, in which the internship experience took place. These companies came together with the aim of creating software that would go a long way toward limiting operators’ errors in taking measurements and speeding up the process of choosing custom footwear. In the following, the methodology and tools possessed, such as possible machine learning algorithms and scans obtained during the manufacturing process, are studied and analyzed. The result is the preparation of data acquired from patients and devices owned by companies to generate software with the best possible learning and functionality.File | Dimensione | Formato | |
---|---|---|---|
TESI1 - depositata.pdf
accesso riservato
Dimensione
2.95 MB
Formato
Adobe PDF
|
2.95 MB | Adobe PDF |
The text of this website © Università degli studi di Padova. Full Text are published under a non-exclusive license. Metadata are under a CC0 License
https://hdl.handle.net/20.500.12608/67371