Introduction: The Peripheral Arterial Disease (PAD) is a disease caused by the obstruction of arteries in peripheral sites of the human body. These diseases cause a reduction in the blood flow in the limbs that can cause claudication and in the most severe cases the loss of the limb. The early diagnosis of PAD is crucial to prevent the loss of the limb. Aim of the work: The goal of this thesis is to validate an imaging-Photo-plethysmographic system (iPPG) that can be used to assess the reduction of blood flow in the limb caused by PAD. Materials and methods: To validate the iPPG system, experiments on 25 healthy volunteers were carried out. During the experiments, measurements with ABI, Duplex Ultrasound, Laser Speckle Contrast Analysis (LASCA) and iPPG were acquired on the subjects. To simulate the reduction of flow, typical of PAD, external cuffs were inflated on the ankles and on the arms of the healthy subjects. The protocol was divided in 5 different stages of progressively inflated cuffs from 0% to 120% of systolic pressure. After the processing of the data, a database that includes information about the subjects and the features derived by the measurements was created. That database was statistically analysed to identify similar trends in the measurements from different instruments. Machine Learning classification models were trained on the recorded data to classify the different inflation stages of the cuffs. Results: The statistical analysis shows that inflating the cuffs allows to identify a variation of the Perfusion. In particular, the Kruskal-Wallis test on all the distributions of data showed that data from baseline stage and inflated stages belong to different distributions. Among the trained classification algorithms the iPPG has the best accuracy and the best sensitivity and specificity for all the classes. The LASCA resulted as the instrument with the best repeatability. Conclusions: The iPPG is a promising technology: assuming the experiments are representative for PAD symptoms, it has the potential to find an implementation in clinical practice in the future. In order to effectively prove its efficiency in PAD assessment, trails on patients are needed.
Introduzione Le Malattie Arteriose Periferiche (PAD) sono causate da ostruzioni circolatorie nei siti periferici del corpo umano. Queste malattie causano una riduzione del flusso sanguigno negli arti e possono causare claudicazione e nei casi più gravi la perdita dell'arto. Scopo della tesi Lo scopo della tesi è validare un sistema di imaging Foto-Pletismografico che possa caratterizzare la riduzione del flusso sanguigno dovuta alle PAD. Questo progetto di tesi è stato svolto al Philips Research Center di Eindhoven. Materiali e Metodi Per validare lo strumento sono stati condotti esperimenti su 25 soggetti sani. Durante gli esperimenti sono state acquisite misure con ABI, Duplex Ultrasound, Laser Speckle Contrast Analysis e iPPG. Per simulare la riduzione di flusso tipica delle PAD delle cuffie esterne sono state gonfiate progressivamente dallo 0% al 120% della pressione sistolica. Dopo l'elaborazione dei dati, è stato creato un database che includesse tutti i dati personali e le misure effettuate. Il database è stato analizzato statisticamente per identificare andamenti simili nelle misure di strumenti diversi. Algoritmi di classificazione sono stati allenati sui dati presenti nel database per classificare i diversi stati di gonfiaggio delle cuffie. Risultati L'analisi dei dati registrati mostra che gonfiando le cuffie possiamo rilevare una variazione nella Perfusione. Il test di Kruskal-Wallis ha dimostrato che i dati provenienti dalla situazione di Baseline e dalle situazioni Inflated provengono da distribuzioni diverse. Tra tutti gli algoritmi di classificazione, l'iPPG dimostra avere la migliore accuratezza, sensitività e specificità per tutte le classi. Il LASCA è risultato lo strumento con la migliore ripetibilità. Conclusioni L'iPPG è una tecnologia promettente che permette di rilevare variazioni nella perfusione nei volontari testati. Per validare la reale efficacia dell'iPPG nella caratterizzazione delle PAD test su soggetti patologici saranno necessari.
Validation of a non-contact photo-plethysmographic imaging (iPPG) system for peripheral arterial disease assessment
DICORATO, CLAUDIO SPIRIDIONE
2018/2019
Abstract
Introduction: The Peripheral Arterial Disease (PAD) is a disease caused by the obstruction of arteries in peripheral sites of the human body. These diseases cause a reduction in the blood flow in the limbs that can cause claudication and in the most severe cases the loss of the limb. The early diagnosis of PAD is crucial to prevent the loss of the limb. Aim of the work: The goal of this thesis is to validate an imaging-Photo-plethysmographic system (iPPG) that can be used to assess the reduction of blood flow in the limb caused by PAD. Materials and methods: To validate the iPPG system, experiments on 25 healthy volunteers were carried out. During the experiments, measurements with ABI, Duplex Ultrasound, Laser Speckle Contrast Analysis (LASCA) and iPPG were acquired on the subjects. To simulate the reduction of flow, typical of PAD, external cuffs were inflated on the ankles and on the arms of the healthy subjects. The protocol was divided in 5 different stages of progressively inflated cuffs from 0% to 120% of systolic pressure. After the processing of the data, a database that includes information about the subjects and the features derived by the measurements was created. That database was statistically analysed to identify similar trends in the measurements from different instruments. Machine Learning classification models were trained on the recorded data to classify the different inflation stages of the cuffs. Results: The statistical analysis shows that inflating the cuffs allows to identify a variation of the Perfusion. In particular, the Kruskal-Wallis test on all the distributions of data showed that data from baseline stage and inflated stages belong to different distributions. Among the trained classification algorithms the iPPG has the best accuracy and the best sensitivity and specificity for all the classes. The LASCA resulted as the instrument with the best repeatability. Conclusions: The iPPG is a promising technology: assuming the experiments are representative for PAD symptoms, it has the potential to find an implementation in clinical practice in the future. In order to effectively prove its efficiency in PAD assessment, trails on patients are needed.| File | Dimensione | Formato | |
|---|---|---|---|
|
2020_06_DICORATO.pdf
non accessibile
Descrizione: Testo della tesi
Dimensione
15.76 MB
Formato
Adobe PDF
|
15.76 MB | Adobe PDF | Visualizza/Apri |
I documenti in POLITesi sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
https://hdl.handle.net/10589/164868