A human body can carry out many physiological complex processes which can be mechanical, electrical or bio-chemical. Each mechanical activity generates a signal that describes the characteristics of the particular action in the form of pressure or temperature. Any irregularity in the process changes the usual functioning thus affecting the performance of the system. Several techniques were introduced to evaluate these muscular signals in order to get a deeper understanding of the medical abnormalities. Displacement sensors, laser optics, electrodes, accelerometers and microphones are some of the widely used devices in measuring the electrical and mechanical activities produced in the muscles. The aim of this thesis project was to find and implement a simple non-contact optical method to measure and monitor the displacements caused on the surface of the skin due to muscular movements. In this study, a device was developed using photo electric sensors that can record surface changes caused on the skin due to the movements forearm muscles. / Människokroppens aktiviteter genererar olika mätbara signaler som kan vara biokemiska, elektriska, mekaniska. Naturligtvis är det viktigt att kunna mäta dessa signaler för att kunna veta om kroppens olika organ fungerar som de ska göra eller inte. När det gäller rent mekaniska aktiviteter genereras signaler av olika typer som beskriver denna aktivitet, såsom tryck, temperatur och förflyttning. Om en sådan process avviker från det normala fallet, kommer kroppssystemets prestanda att försämras. Ett antal tekniker utvecklades för att kunna mäta dessa signaler och uppnå djupare förståelse av möjliga icke-normala medicinska konsekvenser. Förflyttningssensorer, laser optik, elektroder, accelerometrar och mikrofoner är exempel på mättekniker som används för att studera elektrisk och mekanisk aktivitet i muskelvävnader. Målet med detta arbete är att hitta, utveckla och implementera en enkel, användarvänlig, beröringsfri, optisk teknik för att mäta och studera de ytliga förflyttningar som förändrar hudytans landskap och resulterar från muskelaktiviteter och rörelser. Detta projekt resulterade i en enkel prototyp för ett mätinstrument som ser ut som ett armband med två fotoelektriska sensorer som används för att mäta hudytans förändringar på grund av olika arm- och handrörelser.

Healthcare-associated infections is the most common healthcare related injury and affect almost every tenth patient. With the purpose of reducing these infections Infektionsverktyget, The Anti-Infection Tool, was developed for registration and feedback of infection data. The tool is now used in all Swedish county councils resulting in a wealth of data. The purpose of this thesis was thus to investigate how data mining can be applied to describe patterns in this data and predict patient outcomes regarding healthcare-associated infections that need to be treated with antibiotics. Data mining was performed with Microsoft SQL Server 2008 in which models based on six different data mining algorithms with different parameter settings were developed. They used the attributes gender, age and previous diagnoses and medical actions as inputs and antibiotic treated healthcare-associated infection outcome as output. The predictive performance of the models was evaluated using 5-fold cross validation and macro averaged measures of recall, precision and F-measure. Patterns generated by selected models were extracted. Models based on the Naive Bayes algorithm showed the highest predictive capabilities with respect to recall and models based on the Decision Trees algorithm with low pruning had the highest precision. Although, none were considered to perform sufficiently well and several areas of improvement were identified. The most important factor in the inadequate performance is believed to be the relatively rare occurrences of infections in the dataset. Extracted patterns based on the Association Rules algorithm were considered the easiest to interpret. Patterns included clinically valid and invalid as well as trivial relationships. Future studies should be focused on further model improvements and gathering of more patient data. The idea is that data mining in Infektionsverktyget in the future could be used both to provide ideas for further medical research and to identify risk patients and prevent healthcare-associated infections in daily clinical work. / Vårdrelaterade infektioner är den vanligaste vårdskadan och drabbar nästan var tionde patient. Med syfte att minska antalet vårdrelaterade infektioner utvecklades Infektionsverktyget för registrering och återkoppling av infektionsdata. Verktyget används nu i alla Sveriges landsting vilket resulterar i stora mängder data. Syftet med detta examensarbete var därför att undersöka hur data mining kan användas för att beskriva mönster i denna data och för att förutsäga om patienter kommer att drabbas av en vårdrelaterad infektion som behöver antibiotikabehandlas. Data mining genomfördes med Microsoft SQL Server 2008 där modeller baserade på sex olika data mining-algoritmer med olika parameterinställningar utvecklades. De hade inputattributen kön, ålder och tidigare diagnoser och medicinska åtgärder, och outputattributet utfall av antibiotikabehandlad vårdrelaterad infektion. Förutsägelseförmågan hos modellerna utvärderades med 5-delad korsvalidering och makrogenomsnitt av måtten recall, precision och F-measure. Fyra modeller användes även för att ta fram mönster ur datamängden. Modeller baserade på Naive Bayes-algoritmen hade den bästa förutsägelseförmågan med avseende på recall och modeller baserade på Decision Trees-algoritmen med en låg beskärningsnivå uppnådde bäst precision. Trots detta ansågs ingen av modellerna prestera tillräckligt bra och flera möjliga förbättringsområden hittades. Den viktigaste anledningen till den otillräckliga förutsägelseförmågan tros vara att infektioner är relativt ovanliga i datamängden. Mönster som tagits fram med Association Rules-algoritmen ansågs vara lättast att tolka. Mönstren innehöll både kliniskt relevanta och irrelevanta såväl som triviala samband. Framtida studier bör fokuseras på att förbättra modellerna ytterligare och att samla in mer patientdata. Idén är att data mining i Infektionsverktyget i framtiden skulle kunna användas för att ge uppslag till medicinsk forskning och för att identifiera riskpatienter och därmed förebygga vårdrelaterade infektioner i den dagliga kliniska verksamheten.

Image-guided neurosurgery interventions are becoming sur- gical procedure routines. We suggest a novel method for automatic marker localization in X-ray images for Leksell SurgiPlan® which is an image-based neurosergical treat- ment planning software provided by Elekta Instrument AB. We implemented an algorithm for fiducial marker localiza- tion based on feature detection, classification and prior geo- metrical knowledge of the markers. Automatic localization ca help to decrease the human error associated with manual registration of these fiducial markers which is the current applied method for X-ray images in Leksell SurgiPlan®.

Previous studies have shown the positive effect of increased physical activity for cancer patients during treatments of chemotherapy and stem cell transplantation. Moderate exercise has shown to cause significantly less loss of muscle mass, less symptoms of cancer related fatigue, less need for platelet transfusions during treatment time and shorter hospitalisation. Inactivity at hospital clinics is though still a major concern and it seems like lack of motivation plays a big roll. It has been shown that an overview of activity level, personal goal setting and education on the importance of physical activity can work as a motivation towards increased physical activity. This project aimed to make a prototype that can quantify physical activity of hospitalised cancer patients and represent it in a motivational and informative way. An accelerometer was used to collect activity data; the data was processed and used to train a support vector machine for classification of activities. Activities recognised by the prototype are the postures lying down, sitting and standing as well as recognising when the user is active. Over 90% accuracy was obtained in activity recognition for specific training sets. The prototype was tested on patients at the haematology clinic at the Karolinska hospital in Huddinge. Test subjects rated the classification accuracy and the motivational value of the prototype on a scale of 1-5. The accuracy was rated 4.2 out of 5 and the motivational value 3.25 out of 5. A pilot study to further test the feasibility of the product will be performed in the summer of 2015.

The field of medicine is always evolving and one step in this evolution is the use of decision support systems like artificial intelligence. These systems open the possibility to minimize human error in diagnostics as practitioners can use objective measurements and analysis to assist with the diagnosis. In this study the focus has been to explore the possibility of using deep learning models to classify stroke, vestibular neuritis and control groups based on datafrom a video head impulse test (vHIT). This was done by pre-processing data from vHIT into features that could be used as input to an artificial neural network. Three different modelswere designed, where the first two used mean motion data describing the motion of the head and eyes and their standard deviations, and the last model used extracted parameters. The models were trained from vHIT-data from 76 control cases, 37 vestibular neuritis cases and 46 stroke cases. To get a better grasp of the differences between the groups, a comparison was made between the parameters and the mean curves. The resulting models performed to a varying degree with the first model correctly classified 77.8 % of the control cases, 55.6 % of the stroke cases and 80 % of the vestibular neuritis cases. The second model correctly classified 100 % of the control cases, 11.1 % of the stroke cases and 80.0 % of thevestibular neuritis cases. Lastly the third model correctly classified 77.8 % of the control cases, 22.2 % of the stroke cases and 100 % of the vestibular neuritis cases. The results are still insufficient when it comes to clinical use, as the stroke classification requires a higher sensitivity. This means that the cases are correctly classified and gets the urgent care they need. However, with more data and research, these methods could improve further and then provide a valuable service as decision support systems.

Inledning: Vårdrelaterade infektioner (VRI) är ett stort samhällsproblem, inte bara i Sverige utan även internationellt. Vårdrelaterade infektioner orsakar ett stort lidande för patienter är en stor belastning på sjukvård och hela välfärdssystemet. Patienter blir sjukare och kräver i högre grad specialiserad vård vilket leder till längre sjukhusvistelser. Radiologiska avdelningar tar emot patienter med olika sjukdomsbilder från alla samhällsgrupper, kön och åldrar. Patienter kommer både hemifrån samt från sjukhusens alla avdelningar. Radiologin har fått en allt centralare roll inom vården och allt fler diagnoser ställs med hjälp av radiografin. Ett allt högre patientflöde passerar därför röntgen. Syfte: Syftet med denna litteraturstudie är att beskriva åtgärder, som främjar sjukvårdspersonals handhygien, för att minska VRI. Metod: Denna studie utfördes som allmän litteraturöversikt. Databaserna Cinahl och PubMed användes till att finna artiklar som kvalitetsgranskades samt analyserades. Studien innehåller tio vetenskapliga kvantitativa artiklar som svarade på studiens syfte. Resultat: Genom dataanalysen skapades tre beskrivande kategorier: tekniska hjälpmedel och feedback, tvätteknik samt utbildning. Under varje kategori presenterades olika faktorer som kan främja vårdpersonals handhygien. Slutsats: Vårdpersonals handhygien går att förbättra med enkla åtgärder så som utbildning, information, handtvätt och feedback. God handhygienen kan konstateras vara en färskvara som frekvent måste underhållas för att vidhålla goda nivåer.

Artificial hearts are an essential solution for patients suffering from end-stage heart failure. The precise control of these devices is critical for replicating the natural heart’s behavior and ensuring optimal patient health. This thesis presents the development and evaluation of control algorithms for a Total Artificial Heart (TAH). Our research initially considered the Proportional Integral Derivative (PID), Fuzzy-PID, and Artificial Neural Networks (ANN)-PID controllers. Through an iterative process of development and testing, two controllers emerged as the most effective: a Proportional (P) controller and a fuzzy Proportional Derivative (FPD) controller. These controllers were designed and simulated, followed by the generation of C code for implementation on an embedded system. An iterative approach was employed to design and test the controllers. First, the controllers were tested in a simulated environment, and then the validated designs were implemented and evaluated in a physical mock-loop system that mimicked the human circulatory system. The results demonstrated that both the P and FPD controllers were able to regulate the TAH operation. Notably, the FPD controller performs better based on the settling time, overshoot, and rise time in the simulation environment and stability in the physical environment. This thesis contributes to ongoing research in the field of TAH by providing a continuation that can advance the field’s development. These advancements could potentially improve the quality of life of patients awaiting heart transplants. Future work will include refining the FPD controller and conducting extensive physical testing and tuning.

Respiratory diseases, such as asthma and chronic obstructive pulmonary disease (COPD), are pathological conditions affecting the airways of the respiratory system. Currently more than 90 million people are suffering from respiratory diseases, and COPD is predicted to become the third leading cause of death in the world by 2030. Inhalation devices are commonly used in the treatment of respiratory diseases, where an aerosolised medication is delivered to the lungs of the patient via inhalation. The inspiratory flow rate is one of the main factors affecting the drug deposition in the lung, but is currently not controlled in most inhalation systems. The purpose of this master thesis is to design, manufacture and characterize a passive flow regulator device for portable inhalation systems. The designed prototype utilises the principle of a Venturi nozzle and membrane deflection to create a variable flow constriction, which acts as a negative feedback loop for the flow rate regulation. The flow regulator is based on a previously working device used for controlling exhalation flow rate in the range of 3 L/min. Experimental results are evaluated and compared to an analytical solution of the classical Venturi design. Additionally, membrane deflection is measured to analyse the physical behaviour of the membrane within the device. The flow regulating device is scaled up for inhalation flow rate ranges (>30 L/min) and a flow rate sensor is constructed for the measurements. The passive control of flow rates using a deflecting membrane is deemed promising. However, further improvements of the scaled up model used for inhalation are necessary.

Sammanfattning För den dubbelstakande längdåkaren är överansträngda muskler ett pågående problem. På uppdrag av Johnny Nilsson – lektor vid Gymnastik och Idrottshögskolan (GIH) i Stockholm – har en prototyp tagits fram för att övervaka överansträngning hos musculus trapezius (m. trapezius) med hjälp av elektromyografi (EMG). Prototypen kravspecificerades och projektgruppen fick uppdraget att med fria händer utveckla en rörelsereferens med mål att validera var i rörelsen som längdåkaren överanstränger muskeln. För att bygga elektromyografen användes den öppna hårdvaran Arduino som med EMG-modul kunde samla in data genom ytelektroder fästa på m. trapezius. För att uppfylla de funktionskrav som ställdes på rörelsereferensen byggdes en elektrogoniometer där en vridpotentiometer fästes på armbågsleden och mätte dess flexions- och extensionsvinkel. Arduinon programmerades genom dess egen programmeringsmiljö och ett SD-kort installerades på mikrokontrollerkortet för insamling av rådata; efterbehandlingen och presentation av data skedde sedan i Matlab. Med hjälp av Peter Arfert vid medicinsk bildteknik KTH designades ett 3D-utskrivet hölje till EMG-prototypen. Slutligen fick projektgruppen möjlighet att besöka LIVI-laboratoriet i Falun där längdåkning utövades på rullband. Studiebesöket gjorde det möjligt att praktiskt utföra testförsök av prototypen i dess tilltänkta testmiljö och samla in rådata. / Overwork of muscle can be a problem for the double poling cross country skier potentially resultingin lower efficiency. An assignment was established – on behalf of Johnny Nilsson at Gymnastik- ochIdrottshögskolan (GIH) in Stockholm – in order to build a prototype able to monitor and record datafrom musculus trapezius (m. trapezius) through the use of electromyography (EMG). The EMG wasmade using the open source hardware Arduino. The prototype was able to record bilateralmeasurements with the use of EMG-shields, where surface-electrodes were attached to m. trapezius.By creating a prototype based on a rotary potentiometer attached to the elbow joint a reference ofmovement was established by measuring the extension and flexion angle of the elbow. Arduino’s ownIDE was used to program the hardware of the prototype and data was post-processed and presented inMatlab. Data was transferred with the use of an SD-card reader installed on the microcontroller. Withthe help of Peter Arfert at KTH, a 3D-printed model was made for the prototype. The final prototypewas attached to an elite level cross-country skier and tested on a professional treadmill at the LIVIlaboratory in Falun, Sweden. Raw-data was successfully recorded during these trials.

With a strong and continuous development in medical technology products, the need to connect the right data to the right person is increasing, a form of patient identification. When the identification is not working it could lead to catastrophic accidents like the wrong person being operated. The problem could be solved with a mobile application. After desktop research and interview a prototype was made that would work as a patient identifier. The prototype is made as a mobile application that uses the NFC technique to transfer data. After, among other things, interviews with staff from health care the conclusion was that only bigger errors within patient identification were documented. It was hard to know how many errors were made in total. Therefore, the solution to the problem could be a mobile application that reads ID bands. This could be one step in the patient identification chain to reduce even small errors. / Med en stark och kontinuerlig utveckling av medicintekniska produkter ökar behovet att koppla rätt data till rätt person, en form av patientidentifiering. När det inte blir rätt kan det leda till katastrofala olyckor som exempelvis att fel patient opereras. Detta skulle då kunna lösas med hjälp av en mobilapplikation. Efter skrivbordsundersökning och intervju framställdes en prototyp som ska fungera som patientidentifiering. Prototypen är gjord som en mobilapplikation som använder NFC-teknik för att överföra data. Efter bland annat intervjuer med anställda inom vården har slutsatsen dragits att enbart större felhandlingar kring felidentifiering har dokumenterats och att det har varit svårt att veta hur många felidentifieringar som totalt uppstår i vården. Därför skulle en lösning kunna vara en mobilapplikation som avläser ID-band vara ett steg i identifieringskedjan för att även minska små fel inom patientidentifiering.