Migraine is the third most common medical condition in the world. Xiaogai Li have a hypothesis that implies that a hit to the head when first sensing the aura symptoms will stop the migraine attack. An investigation was made to ensure that the blow does not cause a head injury. The prescale pressure film from Fujifilm is a sensor that measures pressure and the pressure distribution. It is supplied with charts to determine the pressure when purchased. The pressure film was used to measure the reached loading impact from the blow. The new chart for the pressure film would in that case be used to create simulations to study the potential brain damage a hit to the head can result in. This paper examines whether loading rates has an influence on a prescale pressure film from Fujifilm, in order to create new pressure charts if such dependency is found. The testing was made with the help of a loading machine where the desired force could be set. The results showed a lower color density in the test results when comparing with the provided charts. The reason for this is unknown, although it could be a consequence of the environmental conditions that the pressure film was held in. The conclusion made by studying the results was that no loading rate dependency exists.

Hip osteoarthritis is a common disease associated with everyday pain and limited mobility. This causes a change in gait characteristics during walking, which tends to give a lower energy efficiency than normal. For patients with hip osteoarthritis, hip arthroplasty is considered a last option due to the related risks of a surgery. To investigate whether the patients show signs of improvement after the surgery, through an energy efficiency standpoint, data have been collected through a gait analysis. In this study, gait analysis was performed on a healthy reference group to obtain reference data. Data from the patient group, which consisted of patients with hip osteoarthritis pre- and post surgery using total hip arthroplasty, THA, was given by Karolinska Institutet. To determine if THA is an appropriate action, data from the patients were compared to the reference data. By performing and collecting data through a gait analysis on a healthy reference group and on patients with hip osteoarthritis pre- and post surgery using total hip arthroplasty, THA, this study has investigated whether the patients show signs of improvement after the surgery and also if the patients differentiate from the reference group through an energy efficiency standpoint. This is used to determine whether THA is an appropriate action.  The study showed a difference in energy efficiency pre- and post surgery. An obvious improvement after THA was shown in most of the patients (p = 0,0094). However, 36 % showed in at least one case, decreased- or an unchanged value in energy efficiency. Although, through an average standpoint only two patients showed decreased efficiency, but the patients after THA only reached an energy efficiency that was approximately half the efficiency of the reference group. The results in this study indicates that further studies are necessary. Considerable alternatives to THA, especially surface hip arthroplasty, SRA, as well as parameters such as pain and mobility should be examined with respect to energy efficiency. Until then, hip arthroplasty should still be considered a last option for patients. / Höftartros är en vanligt förekommande sjukdom som ger daglig smärta och begränsar den fysiska förmågan. Detta leder till en förändrad gångstil vid gång och följaktligen en lägre energieffektivitet än normalt. För höftartrospatienter ses en höftledsplastik som en sista åtgärd på grund av dess möjliga komplikationer. För att undersöka huruvida patienter visar en förbättring efter operation, gällande återhämtning i energieffektivitet, har datainsamling via gånganalys utförts. I denna studie samlades data in från en frisk referensgrupp. Insamlade data från höftartrospatienter före- och efter total höftledsplastik, THA, har erhållits från Karolinska Institutet. För att avgöra om THA är en lämplig åtgärd jämfördes patientdata med referensgruppens data. Studien visade att det är skillnad före- gentemot efter operation där en tydlig förbättring i energieffektivitet uppträdde efter operation hos majoriteten av patienterna (p = 0,0094). 36 % av patienterna indikerar dock minst ett fall utan förbättring efter operation, men vid medelvärden av förändringen för varje patient visar endast två på försämring efter. Patienter efter THA nådde endast upp till drygt hälften av referensgruppens effektivitet, sett ur ett medelvärdesperspektiv. Utifrån resultaten föreslås det att vidare studier genomförs. Alternativa operationsmetoder, i synnerhet ytersättningsplastik, SRA, samt fler parametrar så som smärta och rörlighet, bör undersökas med avseende på energieffektivitet vid gång. Tills dess bör THA fortsatt ses som en sista utväg för patienter.

Projektet syftade till att undersöka skillnaden på tre olika utformningar av fotledsortoser när det kommer till muskel-aktivitet i underbenet vid gång. Det genomfördes på grund av att det finns få studier på hur olika fotledsortoser förhåller sig till varandra, därmed är kunskapen begränsad. Tillvägagångssättet var att skriva ett matlabskript som manipulerade och analyserade all indata och levererade jämförbara resultat för de enskilda ortoserna. Analysen skedde i fyra övergripande moment där det första var att manipulera EMG:ts rådata för att förenkla analysen av den, det andra momentet var att definiera stegtider utifrån fotsuletrycksdata. Tredje momentet innefattade att manipulera referensdata på samma sätt som i moment ett och det fjärde var att uttrycka EMG-data i procent av referensdatan.Efter att EMG-data analyserats så jämfördes de olika ortoserna genom att medelvärden från samtliga försökspersoner beräknades i Excel. Resultaten visade att muskelaktiviteten i "Triceps Surae" överlag verkar minska med ökad plantarflexion. Slutsatsen fick delas upp muskel för muskel och har ingen genomgående trend utan slutsatsen blev olika för varje muskel. / The aim of this project is to investigate the differences in muscle activity of the lower leg between three types of ankle foot orthoses. The reason behind the project is that there is little knowledge about differences between different types of orthoses. The method used to tackle this task was to write a Matlab script that manipulated and anlyzed all the data required to yield a result for each individual orthoses that then could be used to compare them to each other. This was done in four major steps, the first of which was to manipulate the EMG-data of the chosen file so that it became easier to analyze, the second step was to use the data from a pressure sole to determine when steps started and ended. Thirdly, the reference data was manipulated in the same way as in step one. The fourth and final step was to express the chosen file in percent of the reference data. Finally, Excel was used to compare the different orthoses by calculating mean values from the test subjects data. The results showed that the muscel actvity in "Triceps Surae" seems to decrease with increased plantarflexion.The conclussion had to be divided into one conclussion per muscle, the different conlussions differed for every muscle.

Kostnader i samband med arbetsrelaterade skador är ett problem i samhället. Ett sätt att bedöma arbetsbelastning är genom att mäta hjärtfrekvensen, vilket kan göras med hjälp av olika mätmetoder. Syftet med denna studie var att jämföra mätning av hjärtfrekvens för två metoder – aktivitetsarmband (Fitbit Charge 2) och smarta kläder – med mätningar från ett pulsband (Zephyr HxM). Fysiska tester gjordes på 12 personer (tre kvinnor och nio män) med varierande åldrar och innebar simulering av fem olika yrkesgrupper: kontorsarbetare, målare, brevbärare, styckare och byggarbetare. Därtill bedömdes även arbetsaktiviteternas arbetsbelastning utifrån mätvärdena från pulsbandet.Resultatet visade på god korrelation och överrensstämmelse mellan smarta kläder och pulsbandet, medan aktivitetsarmbandets korrelation och överensstämmelse varierade med aktiviteternas intensitet där bäst resultat påträffades vid låg intensitet samt vid aktiviteter som inte involverade armrörelse. Arbetsbelastningen för kontorsarbete bedömdes som lätt arbete, måleri och styckning bedömdes som lätt till måttligt arbete medan brevbäring och byggarbete varierade från lätt till tungt arbete beroende på testperson.Slutligen konstaterades att aktivitetsarmbandet inte duger i syfte att bedöma arbetsbelastningen på arbetsplatser utan kompletterande metoder. Smarta kläder har däremot potential att på längre sikt kunna förebygga skador.

One of the most recent challenges faced worldwide is the so-called “ageing population” phenomenon. Besides the disabilities and illnesses that the elderly is prone to, traumatic brain injuries represent a common and potentially fatal factor that greatly affects this population. Other subjects are however also likely to experience traumatic brain injuries i.e. people affected by some kinds of diseases and disorders such as epilepsy and dementia that are known to have a high risk of falling. Furthermore, sports accidents can also lead to traumatic brain injuries, for example cycling. Current solutions and devices for head injury prevention are limited and from a design point of view there is space for improvements. The focus of this master thesis is to model an innovative headband meant to be worn around the scalp during various daily life activities and in multiple situations. Its function is to prevent the wearer from traumatic brain injuries. The headband model is a finite element model, created in LS-PrePost. Subsequently, simulations are performed in LS-Dyna to replicate real life head impact scenarios. Different parameters and features of both the impacting condition and the headband are tested for creating the optimal headband structure for head injury prevention.

Finite element models have the potential to accurately represent the pediatric body, both from the anatomical and topological point of view. They can describe changes in size and shape as well as changes in the biomechanical properties. Starting from the PIPER human body model, whose baseline represents the anatomy of a 6 years old child, a family of five models between the age of 2 and 6 was created with the purpose of investigating how the body reacts to a traumatic impact. To create these models, a detailed knowledge of pediatric biomechanics was needed, so a deep literature research was performed to characterize all the human body tissues with age-related material properties. Then, an environment model was chosen to investigate how injuries are related to the age of the subject. Since car crashes are a leading cause of death among children, a car accident was simulated to reach the aim of the project. The anatomical and biomechanical scaling process, as well as the positioning of the child in the environment model, were performed with the PIPER tools, while simulations were run with Ls-Dyna. From the literature, age-dependent material properties were found for almost all the tissues of the human body, allowing the development of a detailed pediatric FE model. Then, biomechanical injury predictors, such as the brain strain, the skull acceleration, the chest displacement, the lung pressure and the Von Mises stress in the limbs, were extracted from the simulations to evaluate how injuries changes with the age. The head showed to be the body segment most affected by the age, with an increasing injury severity with the decreasing of the age. Moreover, it was observed that the probability of bone fractures increases for higher bone stiffness.

Increased workload on our healthcare system calls for more automation, this too in the pre-operative planning of total knee arthroplasty (TKA) surgery. This should keep in mind however that success in TKA is measured through having proper limb alignment. This thesis project implemented a virtual positioning assessment software that provides an initial position based on its bone and implant inputs with their specific landmarks. Positions can be furtherly adapted and evaluated based on flexion facets, obtaining decreased pre-operative workloads. The software was validated through an analysis comparison of the femur positioning of nine cases with the former used analysis tool of the university Hospital of Ghent. The newly implemented software was concluded to approximate the patients’ pre-operative alignment better for all translational and rotational parameters, except anteroposterior translation and internal/external rotation of the femur.

Lung cancer has been a major cause of death among types of cancers in the world. In the early stages, lung nodules can be detected by the aid of imaging modalities such as Computed Tomography (CT). In this stage, radiologists look for irregular rounded-shaped nodules in the lung which are normally less than 3 centimeters in diameter. Recent advancements in image analysis have proven that images contain more information than regular parameters such as intensity, histogram and morphological details. Therefore, in this project we have focused on extracting quantitative, hand-crafted features from nearly 1400 lung CT images to train a variety of classifiers based on them. In the first experiment, in total 424 Radiomics features per image has been used to train classifiers such as: Random Forest (RF), Support Vector Machine (SVM), Decision Tree (DT), Naive Bayes (NB), Linear Discriminant Analysis (LDA) and Multi-Layer Perceptron (MLP). In the second experiment, we evaluate each feature category separately with our classifiers. The third experiment includes wrapper feature selection methods (Forward/Backward/Recursive) and filter-based feature selection methods (Fisher score, Gini Index and Mutual information). They have been implemented to find the most relevant feature set in model construction. Performance of each learning method has been evaluated by accuracy score, wherewe achieved the highest accuracy of 78% with Random Forest classifier (74% in 5-fold average) and 0.82 Area Under the Receiver Operating Characteristics (AUROC) curve. After RF, NB and MLP showed the best average accuracy of 71.4% and 71% respectively.

Clinical ultrasound imaging techniques can be greatly improved by the use of ultrasound contrast agents (UCAs). While microbubbles (MBs) without shell are unstable and cannot be used for practical applications,a shell produced from biocompatible polyvinylalcohol (PVA) significantly improves chemical versatility and stability. The oscillation characteristics of a UCA are strongly dependent on concentration, applied pressure and viscoelastic parameters of the shell. Modifications in the shell as incorporation of antibodies or targeted molecules affect the bubble oscillation and resonance frequency of the MB suspension. In this presented work a tool for systematic characterization of UCAs is developed. Linear acoustic behaviour of PVA shelled MBs is examined. The acoustic driving pressure is kept below 100 kPa. The MB concentration is 1·10^{6} ml^{-1}. Attenuation and phase velocity profiles of ultrasound waves propagating through the UCA are measured using six narrow-band single crystal transducers that cover a frequency range between 1 and 15 MHz. The oscillation of a single bubble is modeled as a linear oscillator adapting HOFF’s model suitable for allshell thicknesses. The suspension is modeled through superposition of single bubbles. Knowing all parameters the resonance frequency of a MB suspension can be predicted. The model is fitted to experimental data to determine the viscoelastic shell parameters. The shell thickness is challenging to determine exactly and assumed to be either proportional to the outer shell radius or constant. Assuming a proportional shell thickness the calculated resulting shell parameters were shear modulus G_s = 14.5 MPa, shear viscosity η_s = 0.322 Pa·s and shell thickness d_s = 16 % of the outer radius. When instead assuming a constant shell thickness the determined parameters were in similar order of magnitude. Resonance frequency of the suspension was determined to 11.6 MHz. The developed tool can be used to characterize MBs with a modified shell independently of shell thickness and to predict resonance frequency of gas or air filled UCAs with known shell parameters.

Brain-Computer Interface (BCI) offers the opportunity to paralyzed patients to control their movements without any neuromuscular activity. Signal processing of neuronal activity enables to decode movement intentions. Ability for patient to control an effector is closely linked to this decoding performance. In this study, I tackle a recent way to decode neuronal activity: Deep learning. The study is based on public data extracted by Schalk et al. for BCI Competition IV. Electrocorticogram (ECoG) data from three epileptic patients were recorded. During the experiment setup, the team asked subjects to move their fingers and recorded finger movements thanks to a data glove. An artificial neural network (ANN) was built based on a common BCI feature extraction pipeline made of successive convolutional layers. This network firstly mimics a spatial filtering with a spatial reduction of sources. Then, it realizes a time-frequency analysis and performs a log power extraction of the band-pass filtered signals. The first investigation was on the optimization of the network. Then, the same architecture was used on each subject and the decoding performances were computed for a 6-class classification. I especially investigated the spatial and temporal filtering. Finally, a preliminary study was conducted on prediction of finger movement. This study demonstrated that deep learning could be an effective way to decode brain signal. For 6-class classification, results stressed similar performances as traditional decoding algorithm. As spatial or temporal weights after training are slightly described in the literature, we especially worked on interpretation of weights after training. The spatial weight study demonstrated that the network is able to select specific ECoG channels notified in the literature as the most informative. Moreover, the network is able to converge to the same spatial solution, independently to the initialization. Finally, a preliminary study was conducted on prediction of movement position and gives encouraging results.