The diabetic foot represents one of the most severe and debilitating consequences of diabetes, which can drastically impact the quality of life for patients and result in significant annual healthcare expenses worldwide. The formation of ulcers is often an outcome associated with the improper use of footwear, which, in combination with bone and joint deformities, can lead to their development. The International Working Group of the Diabetic Foot (IWGDF) outlines guidelines for selecting the ideal footwear for diabetic patients, advocating the use of therapeutic footwear and orthotic insoles as a preventive measure against ulcer formation. The objective of this thesis is to develop an artificial intelligence-based algorithm capable of matching the shape of the foot with the ideal shoe for its treatment, starting from the automated acquisition of characteristic foot measurements, such as various circumferences, length, and width. Traditionally, these measurements required manual intervention by a skilled operator. To achieve this goal, a database was created, comprising measurement values obtained through custom insoles or 3D scanners. These measurements were then processed using machine learning techniques to create a model capable of parameterizing and classifying foot shapes with various associated pathologies. This research introduces an innovative automation model for a process that has historically been predominantly manual, with the aim of significantly reducing production times, containing costs, and eliminating human error.
Il piede diabetico rappresenta una delle più gravi e invalidanti conseguenze derivanti dal diabete, che può influenzare drasticamente la qualità di vita dei pazienti, oltre che comportare un'ingente spesa annuale per il sistema sanitario mondiale. La formazione di ulcere è spesso un risultato associato all'utilizzo inappropriato di calzature, il quale, in combinazione con deformità ossee e articolari, può condurre alla loro comparsa. L'International Working Group of the Diabetic Foot (IWGDF) infatti delinea le linee guide per la scelta della calzatura ideale per il piede diabetico e indica l'utilizzo di calzature terapeutiche e ortesi plantari come metodo per la prevenzione nella formazione di ulcere. L'obiettivo di questa tesi è sviluppare un algoritmo basato sull'intelligenza artificiale in grado di trovare un match tra la forma del piede e la scarpa ideale per il suo trattamento, partendo da un'acquisizione automatizzata delle misure caratteristiche del piede, come le varie circonferenze, lunghezza e larghezza, che fino ad ora richiedevano l'intervento manuale di un esperto operatore. A tale scopo si è creato un Database composto dai valori delle misurazioni acquisite tramite schede su misura o scanner 3D. Queste misure sono poi state processate con tecniche di Machine Learning al fine di creare un modello capace di parametrizzare e classificare le forme di piede con le varie patologie associate. Questa ricerca introduce un modello di automazione innovativo per un processo tutt'ora gestito principalmente manualmente, con l'obiettivo di abbreviare notevolmente i tempi di produzione, contenere i costi ed eliminare l'errore umano.
Sviluppo di un modello parametrico 3D del piede e della calzatura attraverso tecniche di artificial intellingence per applicazione nella prevenzione piede diabetico.
MELE, FRANCESCA
2022/2023
Abstract
The diabetic foot represents one of the most severe and debilitating consequences of diabetes, which can drastically impact the quality of life for patients and result in significant annual healthcare expenses worldwide. The formation of ulcers is often an outcome associated with the improper use of footwear, which, in combination with bone and joint deformities, can lead to their development. The International Working Group of the Diabetic Foot (IWGDF) outlines guidelines for selecting the ideal footwear for diabetic patients, advocating the use of therapeutic footwear and orthotic insoles as a preventive measure against ulcer formation. The objective of this thesis is to develop an artificial intelligence-based algorithm capable of matching the shape of the foot with the ideal shoe for its treatment, starting from the automated acquisition of characteristic foot measurements, such as various circumferences, length, and width. Traditionally, these measurements required manual intervention by a skilled operator. To achieve this goal, a database was created, comprising measurement values obtained through custom insoles or 3D scanners. These measurements were then processed using machine learning techniques to create a model capable of parameterizing and classifying foot shapes with various associated pathologies. This research introduces an innovative automation model for a process that has historically been predominantly manual, with the aim of significantly reducing production times, containing costs, and eliminating human error.File | Dimensione | Formato | |
---|---|---|---|
Mele_Francesca.pdf
accesso riservato
Dimensione
2.72 MB
Formato
Adobe PDF
|
2.72 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/20.500.12608/54927