Field-oriented assessment of agricultural crops through temporal segmentation of modis VI data.

Monitoring agricultural crops constitutes a vital task for the general understanding of land use spatio-temporal dynamics. This paper presents an approach for the enhancement of current crop monitoring capabilities on a regional scale, in order to allow for the analysis of environmental and socio-economic drivers and impacts of agricultural land use. This work discusses the advantages and current limitations of using 250m VI data from the Moderate Resolution Imaging Spectroradiometer (MODIS) for this purpose, with emphasis in the difficulty of correctly analyzing pixels whose temporal responses are disturbed due to certain sources of interference such as mixed or heterogeneous land cover. It is shown that the influence of noisy or disturbed pixels can be minimized, and a much more consistent and useful result can be attained, if individual agricultural fields are identified and each field's pixels are analyzed in a collective manner. As such, a method is proposed that makes use of image segmentation techniques based on MODIS temporal information in order to identify portions of the study area that agree with actual agricultural field borders. The pixels of each portion or segment are then analyzed individually in order to estimate the reliability of the temporal signal observed and the consequent relevance of any estimation of land use from that data. The proposed method was applied in the state of Mato Grosso, in mid-western Brazil, where extensive ground truth data was available. Experiments were carried out using several supervised classification algorithms as well as different subsets of land cover classes, in order to test the methodology in a comprehensive way. Results show that the proposed method is capable of consistently improving classification results not only in terms of overall accuracy but also qualitatively by allowing a better understanding of the land use patterns detected. It thus provides a practical and straightforward procedure for enhancing crop-mapping capabilities using temporal series of moderate resolution remote sensing data.

Comparazione performance di un'applicazione enterprise basata su service mesh

Ultimamente si stanno sviluppando tecnologie per rendere più efficiente la virtualizzazione a livello di sistema operativo, tra cui si cita la suite Docker, che permette di gestire processi come se fossero macchine virtuali. Inoltre i meccanismi di clustering, come Kubernetes, permettono di collegare macchine multiple, farle comunicare tra loro e renderle assimilabili ad un server monolitico per l'utente esterno. Il connubio tra virtualizzazione a livello di sistema operativo e clustering permette di costruire server potenti quanto quelli monolitici ma più economici e possono adattarsi meglio alle richieste esterne. Data l'enorme mole di dati e di potenza di calcolo necessaria per gestire le comunicazioni e le interazioni tra utenti e servizi web, molte imprese non possono permettersi investimenti su un server proprietario e la sua manutenzione, perciò affittano le risorse necessarie che costituiscono il cosiddetto "cloud", cioè l'insieme di server che le aziende mettono a disposizione dei propri clienti. Il trasferimento dei servizi da macchina fisica a cloud ha modificato la visione che si ha dei servizi stessi, infatti non sono più visti come software monolitici ma come microservizi che interagiscono tra di loro. L'infrastruttura di comunicazione che permette ai microservizi di comunicare è chiamata service mesh e la sua suddivisione richiama la tecnologia SDN. È stato studiato il comportamento del software di service mesh Istio installato in un cluster Kubernetes. Sono state raccolte metriche su memoria occupata, CPU utilizzata, pacchetti trasmessi ed eventuali errori e infine latenza per confrontarle a quelle ottenute da un cluster su cui non è stato installato Istio. Lo studio dimostra che, in un cluster rivolto all'uso in produzione, la service mesh offerta da Istio fornisce molti strumenti per il controllo della rete a scapito di una richiesta leggermente più alta di risorse hardware.

Machine Learning methods for hepatocellular malignancies segmentation and MVI prediction

The aim of this thesis project is to automatically localize HCC tumors in the human liver and subsequently predict if the tumor will undergo microvascular infiltration (MVI), the initial stage of metastasis development. The input data for the work have been partially supplied by Sant'Orsola Hospital and partially downloaded from online medical databases. Two Unet models have been implemented for the automatic segmentation of the livers and the HCC malignancies within it. The segmentation models have been evaluated with the Intersection-over-Union and the Dice Coefficient metrics. The outcomes obtained for the liver automatic segmentation are quite good (IOU = 0.82; DC = 0.35); the outcomes obtained for the tumor automatic segmentation (IOU = 0.35; DC = 0.46) are, instead, affected by some limitations: it can be state that the algorithm is almost always able to detect the location of the tumor, but it tends to underestimate its dimensions. The purpose is to achieve the CT images of the HCC tumors, necessary for features extraction. The 14 Haralick features calculated from the 3D-GLCM, the 120 Radiomic features and the patients' clinical information are collected to build a dataset of 153 features. Now, the goal is to build a model able to discriminate, based on the features given, the tumors that will undergo MVI and those that will not. This task can be seen as a classification problem: each tumor needs to be classified either as “MVI positive” or “MVI negative”. Techniques for features selection are implemented to identify the most descriptive features for the problem at hand and then, a set of classification models are trained and compared. Among all, the models with the best performances (around 80-84% ± 8-15%) result to be the XGBoost Classifier, the SDG Classifier and the Logist Regression models (without penalization and with Lasso, Ridge or Elastic Net penalization).

Hybrid ground-aerial mesh networks for IoT monitoring applications: network design and software platform development

The Internet of Things (IoT) has grown rapidly in recent years, leading to an increased need for efficient and secure communication between connected devices. Wireless Sensor Networks (WSNs) are composed of small, low-power devices that are capable of sensing and exchanging data, and are often used in IoT applications. In addition, Mesh WSNs involve intermediate nodes forwarding data to ensure more robust communication. The integration of Unmanned Aerial Vehicles (UAVs) in Mesh WSNs has emerged as a promising solution for increasing the effectiveness of data collection, as UAVs can act as mobile relays, providing extended communication range and reducing energy consumption. However, the integration of UAVs and Mesh WSNs still poses new challenges, such as the design of efficient control and communication strategies. This thesis explores the networking capabilities of WSNs and investigates how the integration of UAVs can enhance their performance. The research focuses on three main objectives: (1) Ground Wireless Mesh Sensor Networks, (2) Aerial Wireless Mesh Sensor Networks, and (3) Ground/Aerial WMSN integration. For the first objective, we investigate the use of the Bluetooth Mesh standard for IoT monitoring in different environments. The second objective focuses on deploying aerial nodes to maximize data collection effectiveness and QoS of UAV-to-UAV links while maintaining the aerial mesh connectivity. The third objective investigates hybrid WMSN scenarios with air-to-ground communication links. One of the main contribution of the thesis consists in the design and implementation of a software framework called "Uhura", which enables the creation of Hybrid Wireless Mesh Sensor Networks and abstracts and handles multiple M2M communication stacks on both ground and aerial links. The operations of Uhura have been validated through simulations and small-scale testbeds involving ground and aerial devices.

Supervised and weakly supervised counting-by-segmentation: the fluorescent microscopy use case

This thesis focuses on automating the time-consuming task of manually counting activated neurons in fluorescent microscopy images, which is used to study the mechanisms underlying torpor. The traditional method of manual annotation can introduce bias and delay the outcome of experiments, so the author investigates a deep-learning-based procedure to automatize this task. The author explores two of the main convolutional-neural-network (CNNs) state-of-the-art architectures: UNet and ResUnet family model, and uses a counting-by-segmentation strategy to provide a justification of the objects considered during the counting process. The author also explores a weakly-supervised learning strategy that exploits only dot annotations. The author quantifies the advantages in terms of data reduction and counting performance boost obtainable with a transfer-learning approach and, specifically, a fine-tuning procedure. The author released the dataset used for the supervised use case and all the pre-training models, and designed a web application to share both the counting process pipeline developed in this work and the models pre-trained on the dataset analyzed in this work.

Geometric deep learning for polygonal mesh denoising

Il seguente lavoro si propone come analisi degli operatori convoluzionali che caratterizzano le graph neural networks. ln particolare, la trattazione si divide in due parti, una teorica e una sperimentale. Nella parte teorica vengono innanzitutto introdotte le nozioni preliminari di mesh e convoluzione su mesh. In seguito vengono riportati i concetti base del geometric deep learning, quali le definizioni degli operatori convoluzionali e di pooling e unpooling. Un'attenzione particolare è stata data all'architettura Graph U-Net. La parte sperimentare riguarda l'applicazione delle reti neurali e l'analisi degli operatori convoluzionali applicati al denoising di superfici perturbate a causa di misurazioni imperfette effettuate da scanner 3D.

Geometric Deep Learning per il denoising di mesh 3D

La crescente disponibilità di scanner 3D ha reso più semplice l’acquisizione di modelli 3D dall’ambiente. A causa delle inevitabili imperfezioni ed errori che possono avvenire durante la fase di scansione, i modelli acquisiti possono risultare a volte inutilizzabili ed affetti da rumore. Le tecniche di denoising hanno come obiettivo quello di rimuovere dalla superficie della mesh 3D scannerizzata i disturbi provocati dal rumore, ristabilendo le caratteristiche originali della superficie senza introdurre false informazioni. Per risolvere questo problema, un approccio innovativo è quello di utilizzare il Geometric Deep Learning per addestrare una Rete Neurale in maniera da renderla in grado di eseguire efficacemente il denoising di mesh. L’obiettivo di questa tesi è descrivere il Geometric Deep Learning nell’ambito del problema sotto esame.

Virtual commissioning of automatic machines for mesh panels manufacturing

The purpose of this thesis is to present the concept of simulation for automatic machines and how it might be used to test and debug software implemented for an automatic machine. The simulation is used to detect errors and allow corrections of the code before the machine has been built. Simulation permits testing different solutions and improving the software to get an optimized one. Additionally, simulation can be used to keep track of a machine after the installation in order to improve the production process during the machine’s life cycle. The central argument of this project is discussing the advantage of using virtual commissioning to test the implemented software in a virtual environment. Such an environment is getting benefit in avoiding potential damages as well as reduction of time to have the machine ready to work. Also, the use of virtual commissioning allows testing different solutions without high losses of time and money. Subsequently, an optimized solution could be found after testing different proposed solutions. The software implemented is based on the Object-Oriented Programming paradigm which implies different features such as encapsulation, modularity, and reusability of the code. Therefore, this way of programming helps to get simplified code that is easier to be understood and debugged as well as its high efficiency. Finally, different communication protocols are implemented in order to allow communication between the real plant and the simulation model. By the outcome that this communication provides, we might be able to gather all the necessary data for the simulation and the analysis, in real-time, of the production process in a way to improve it during the machine life cycle.

LoRa Mesh: un caso di studio per la comunicazione tra imbarcazioni da diporto

Il progetto è stato sviluppato con l’idea di creare una rete attraverso la quale imbarcazioni da diporto relativamente vicine (10km), si possano scambiare informazioni sullo stato del mare e della navigazione, anche in assenza di una connessione a internet. In tal modo i dati dell’imbarcazione, come temperatura esterna, temperatura dell’acqua, vento, coordinate gps, AIS ecc... verrebbero condivisi attraverso la rete. In questo progetto è stata sviluppata un'infrastruttura in grado di far comunicare imbarcazioni da diporto su bande non licenziate, utilizzando solo materiale OpenSource, in particolare un protocollo chiamato LoRaMesh. Tale infrastruttura, non basandosi su uno standard definito, ha la possibilità di adattarsi a qualsiasi tipo di dato. Tutto il progetto si basa su schede PyCom, ed è stato sviluppato del codice in grado di fornire uno scambio di dati costante e un’interfacci BLE per comunicare con più dispositivi possibili. Per fornire un’esempio di come ci si può connettere con il BLE è stata scritta un’app per IOS che fornisce varie funzionalità, tra cui la possibilità di inviare dati GPS, molto utile per l’esecuzione dei vari test. Sono state svolte varie prove, in diversi luoghi e condizioni, utili a capire la portata massima dei dispositivi, e come la rete mesh si adatta e ripara.

A performance analysis of mesh models for cloud-based workflows

Nell'ambito della loro trasformazione digitale, molte organizzazioni stanno adottando nuove tecnologie per supportare lo sviluppo, l'implementazione e la gestione delle proprie architetture basate su microservizi negli ambienti cloud e tra i fornitori di cloud. In questo scenario, le service ed event mesh stanno emergendo come livelli infrastrutturali dinamici e configurabili che facilitano interazioni complesse e la gestione di applicazioni basate su microservizi e servizi cloud. L’obiettivo di questo lavoro è quello di analizzare soluzioni mesh open-source (istio, Linkerd, Apache EventMesh) dal punto di vista delle prestazioni, quando usate per gestire la comunicazione tra applicazioni a workflow basate su microservizi all’interno dell’ambiente cloud. A questo scopo è stato realizzato un sistema per eseguire il dislocamento di ognuno dei componenti all’interno di un cluster singolo e in un ambiente multi-cluster. La raccolta delle metriche e la loro sintesi è stata realizzata con un sistema personalizzato, compatibile con il formato dei dati di Prometheus. I test ci hanno permesso di valutare le prestazioni di ogni componente insieme alla sua efficacia. In generale, mentre si è potuta accertare la maturità delle implementazioni di service mesh testate, la soluzione di event mesh da noi usata è apparsa come una tecnologia ancora non matura, a causa di numerosi problemi di funzionamento.

Application of active and transfer learning to wound image segmentation

Wound management is a fundamental task in standard clinical practice. Automated solutions already exist for humans, but there is a lack of applications on wound management for pets. The importance of a precise and efficient wound assessment is helpful to improve diagnosis and to increase the effectiveness of treatment plans for the chronic wounds. The goal of the research was to propose an automated pipeline capable of segmenting natural light-reflected wound images of animals. Two datasets composed by light-reflected images were used in this work: Deepskin dataset, 1564 human wound images obtained during routine dermatological exams, with 145 manual annotated images; Petwound dataset, a set of 290 wound photos of dogs and cats with 0 annotated images. Two implementations of U-Net Convolutioal Neural Network model were proposed for the automated segmentation. Active Semi-Supervised Learning techniques were applied for human-wound images to perform segmentation from 10% of annotated images. Then the same models were trained, via Transfer Learning, adopting an Active Semi- upervised Learning to unlabelled animal-wound images. The combination of the two training strategies proved their effectiveness in generating large amounts of annotated samples (94% of Deepskin, 80% of PetWound) with the minimal human intervention. The correctness of automated segmentation were evaluated by clinical experts at each round of training thus we can assert that the results obtained in this thesis stands as a reliable solution to perform a correct wound image segmentation. The use of Transfer Learning and Active Semi-Supervied Learning allows to minimize labelling effort from clinicians, even requiring no starting manual annotation at all. Moreover the performances of the model with limited number of parameters suggest the implementation of smartphone-based application to this topic, helping the future standardization of light-reflected images as acknowledge medical images.

Multiple Desmoid Tumors In A Patient With Gardner's Syndrome - Report Of A Case.

Desmoid tumor (DT) is a common manifestation of Gardner's Syndrome (GS), although it is a rare condition in the general population. DT in patients with GS is usually located in the abdominal wall and/or intra-abdominal cavity. We report a case of a 32 years-old female patient with familial adenomatous polyposis (FAP), who was already submitted to total colectomy and developed multiple DT, located in the abdominal wall and in the left breast. The patient underwent several surgical procedures, with a multidisciplinary team of surgeons. Wide surgical resections of the left breast and the abdominal wall tumors were performed in separate steps. Polypropylene mesh reconstruction and muscle flaps were needed to cover the defects of the thoracic and abdominal walls. After partial necrosis of the adipose-cutaneous flap in the abdomen that required a new skin graft, she had a satisfactory outcome with complete healing of the surgical incisions. DT is frequent in GS, however, breast localization is very rare, with few cases reported in the literature. Recurrence of DT is not negligible, even after a wide surgical resection. GS patients must be followed up closely, and clinical examination, associated with imaging studies, should be performed to detect any signs of tumor. DT represents one of the most significant causes of the morbidity and mortality that affects FAP patients following colectomy. In general, the surgical procedures to excise DT are highly complex, requiring a multidisciplinary team.

Graded Meshes In Bio-thermal Problems With Transmission-line Modeling Method.

In this study, the transmission-line modeling (TLM) applied to bio-thermal problems was improved by incorporating several novel computational techniques, which include application of graded meshes which resulted in 9 times faster in computational time and uses only a fraction (16%) of the computational resources used by regular meshes in analyzing heat flow through heterogeneous media. Graded meshes, unlike regular meshes, allow heat sources to be modeled in all segments of the mesh. A new boundary condition that considers thermal properties and thus resulting in a more realistic modeling of complex problems is introduced. Also, a new way of calculating an error parameter is introduced. The calculated temperatures between nodes were compared against the results obtained from the literature and agreed within less than 1% difference. It is reasonable, therefore, to conclude that the improved TLM model described herein has great potential in heat transfer of biological systems.

Preference Mapping Of Dulce De Leche Commercialized In Brazilian Markets.

Dulce de leche samples available in the Brazilian market were submitted to sensory profiling by quantitative descriptive analysis and acceptance test, as well sensory evaluation using the just-about-right scale and purchase intent. External preference mapping and the ideal sensory characteristics of dulce de leche were determined. The results were also evaluated by principal component analysis, hierarchical cluster analysis, partial least squares regression, artificial neural networks, and logistic regression. Overall, significant product acceptance was related to intermediate scores of the sensory attributes in the descriptive test, and this trend was observed even after consumer segmentation. The results obtained by sensometric techniques showed that optimizing an ideal dulce de leche from the sensory standpoint is a multidimensional process, with necessary adjustments on the appearance, aroma, taste, and texture attributes of the product for better consumer acceptance and purchase. The optimum dulce de leche was characterized by high scores for the attributes sweet taste, caramel taste, brightness, color, and caramel aroma in accordance with the preference mapping findings. In industrial terms, this means changing the parameters used in the thermal treatment and quantitative changes in the ingredients used in formulations.

Lateral Pterygoid Muscle Volume And Migraine In Patients With Temporomandibular Disorders.

Lateral pterygoid muscle (LPM) plays an important role in jaw movement and has been implicated in Temporomandibular disorders (TMDs). Migraine has been described as a common symptom in patients with TMDs and may be related to muscle hyperactivity. This study aimed to compare LPM volume in individuals with and without migraine, using segmentation of the LPM in magnetic resonance (MR) imaging of the TMJ. Twenty patients with migraine and 20 volunteers without migraine underwent a clinical examination of the TMJ, according to the Research Diagnostic Criteria for TMDs. MR imaging was performed and the LPM was segmented using the ITK-SNAP 1.4.1 software, which calculates the volume of each segmented structure in voxels per cubic millimeter. The chi-squared test and the Fisher's exact test were used to relate the TMD variables obtained from the MR images and clinical examinations to the presence of migraine. Logistic binary regression was used to determine the importance of each factor for predicting the presence of a migraine headache. Patients with TMDs and migraine tended to have hypertrophy of the LPM (58.7%). In addition, abnormal mandibular movements (61.2%) and disc displacement (70.0%) were found to be the most common signs in patients with TMDs and migraine. In patients with TMDs and simultaneous migraine, the LPM tends to be hypertrophic. LPM segmentation on MR imaging may be an alternative method to study this muscle in such patients because the hypertrophic LPM is not always palpable.