|
Lääketieteellisen fysiikan ja tekniikan yhdistys (LFTY) Finnish Society for Medical Physics and Medical Engineering |
LFT-päivä 9.2.2012, Turun yliopisto/TYKS, T-sairaala
Posterikilpailuun osallistuneet
1. Ásgeir Bjarnason, Tampereen teknillinen yliopistoDevelopment of implantable pulse oxygen saturation meter for dairy-cattle respiratory monitoring
2. Elvis K. Danso, Itä-Suomen yliopisto
Influence of meniscectomy on the performance of human knee joint
3. Olli Heikkinen, Itä-Suomen yliopisto
Development and validation of an ambulatory heart rate variability measurement system
4. Jukka Hirvasniemi, Oulun yliopisto
Ultrasound evaluation of articular cartilage and subchondral bone in vitro
5. Andrei D. Jakab, Tampereen teknillinen yliopisto
Life-saving EEG system for use in emergencies
6. Jukka Kiuttu, Oulun yliopisto
Association between postexercise hypotension, autonomic nervous system and blood glucose concentration
7. Pekka Kuoppa, Itä-Suomen yliopisto
AR-mallien soveltaminen biosignaalien monimuuttuja-analyysissa
8. Tam Pham, PET-keskus
Abdominal adipose tissue glucose uptake is increased after bariatric surgery
9. Jana Podlipská, Oulun yliopisto
Quantitative ultrasound image analysis of femoral subchondral bone in knee osteoarthritis
10. Joanna Pylvänäinen, Turun yliopisto
Transnedothelial migration of breast cancer cells during cancer metastasis
11. Jari Rautiainen, Tampereen teknillinen yliopisto
Rusto-luurakenteiden kvantifiointi SWIFT-magneettikuvausmenetelmällä
12. Noora Tuovinen, Tampereen teknillinen yliopisto
Application of magnet quench analysis
13. Teemu Turunen, Aalto-yliopisto
Device for simultaneous measurement of ERG-signal from desired retinal layer and across the retina
14. Koos C.J. Zevenhoven, Aalto-yliopisto
Towards high-quality ultra-low-field MRI: novel methods for suppressing eddy currents
Postereiden abstraktit
Development of implantable pulse oxygen saturation meter for dairy-cattle respiratory monitoringÁsgeir Bjarnason, Tampereen teknillinen yliopisto
Objective: The objective of this work was to design, build and test an implantable measurement device that can be used to measure oxygen saturation (SpO2) with pulse oximetry methods. The device is intended to be incorporated into an implantable measurement device developed earlier at Tampere University of Technology. Respiratory diseases are the cause of approximately 50 % of all cattle mortality and SpO2 provides a good measurement on how the respiratory system of cattle is functioning. Implantable SpO2 recording has not been conducted before and this work aimed to solve number of problems effecting the measurements, such as low power, compact size, biocompatability issues and effects of fat tissue surrounding the implant.
Methods: The developed device measures the SpO2 with a probe consisting of two light emitting diodes (LED) and a photosensor. The device is low power and has a feedback that controls the LED supply current. In the thesis two types of coating methods were used to seal the probe, 3.5mm layer of medical grade epoxy and 15 µm layer of Parylene-C. The effect that these two coatings had on the probe components and signals were determined with measurements prior to and after the coating. A signal processing script was written in MATLAB to calculate the uncalibrated SpO2 from the raw signal. The noise level in all measurements was estimated with the standard deviation since the signal is unambiguous and it was concluded that with a moving average filter of 4- or 8-points it is possible to reduce the noise significantly.
Results: The measurement system was able to acquire uncalibrated SpO2 from external test measurements on a human subject. Parylene-C coating affected the signal acquisition less than the epoxy coating, the difference is partially explained with the fact that a thicker layer of coating creates internal scattering of light inside the coating. This light hits the photosensor before being absorbed by tissue and thus increases the DC level of the sensor. Thermal radiation of the probe was estimated with test measurements of two different LED drive currents and theoretical calculations, neither of which showed any significant increase in temperature. Tests were also conducted with a non-perfused fat tissue that indicated that signal attenuates in relation with thickness of fat tissue. It was possible to acquire signal through non-perfused tissue of thickness up to 1cm when implementing a noise reduction filter on the signal.
Conclusions: The output of the thesis is a novel device that could easily be implanted in a dairy-cow as well as other mammals. The thesis also presents new information on the effects of coating SpO2 probes and the effects of fat tissue in cattle on the SpO2 signal. Pulse oxygen saturation measurements have not been conducted with an implantable meter before in any type of animal and thus the uncertainty of measurements can only be eliminated with an implantation of a real device.
Influence of meniscectomy on the performance of human knee joint
Elvis K. Danso, Itä-Suomen yliopisto
Meniscectomy increases the probability of developing degenerative changes in articular cartilage of the human knee joint. This work investigates the effects of meniscectomy on the performance of the human knee joint.
A three-dimensional finite element model that included femur cartilage, tibia cartilage, and menisci was developed using magnetic resonance imaging and computational modelling. The menisci were modelled as isotropic linearly elastic materials and the articular cartilage a fibril reinforced poroviscoelastic material. The models were used to estimate the contact pressure, stress distribution, strain and fluid pressure in the femur cartilage and tibia cartilage. A comparison was made between a healthy tibio-femoral human knee joint and a meniscectomised tibio-femoral human knee joint.
It was observed that the contact pressure of the meniscectomised knee joint was higher as compared to the normal knee joint. Also the meniscectomised knee joint had more localised von Mises stress and logarithmic strain as compared to the normal knee joint. Furthermore, the fluid pressure was distributed in a much larger area in the normal knee joint than in the meniscectomised knee joint.
These computational results suggest that meniscectomy could lead to a degenerative joint disease. In the presented way, computational modelling in combination with imaging techniques could potentially be used as a tool for non-invasive evaluation of knee joint function and potential failure points in joints.
Development and validation of an ambulatory heart rate variability measurement system
Olli Heikkinen, Itä-Suomen yliopisto
Introduction: Validation of measurement accuracy and precision of an electrocardiography (ECG) based ambulatory heart rate variability (HRV) measurement system (eMotion HRV by Mega Electronics) were assessed in laboratory and field conditions. Also, data acquisition software for the system was developed in this work in C# for Windows. HRV can be used to assess the control of the autonomic nervous system on the heart in e.g. cardiology, diabetes care and sports science applications.
Methods: The system was validated by simultaneous measurements with an MDD and FDA approved clinical ECG device. Orthostatic, bicycle ergometer and 24 h daily activity experiments were conducted on five healthy, young persons. The accuracy of QRS detection of the system was assessed with sensitivity and positive predictivity measures. Precision of measured ECG RR intervals in the laboratory experiments was assessed with histogram of RR differences and Bland-Altman analysis. The effect of the differences in RR intervals between systems was assessed by calculating a set of time domain and frequency domain HRV measures for both systems from the orthostatic experiment data. Spectrum estimation methods Welch's periodogram and stationary AR(16) model were used. Smoothness priors was implemented to detrend the data before spectrum analysis. Reliability of the system in long-term measurements was assessed with artefact ratio analysis. All analysis methods, a QRS detection algorithm and an artefact detection algorithm were implemented in Matlab.
Results: The sensitivity and positive predictivity in the laboratory measurements were sqrs = 99.989 % and pqrs = 99.989 %. The 95 % limits of agreement from Bland-Altman analysis were (0.591 ± 0.962) ms during rest and (0.820 ± 1.308) ms during exercise. The smallest difference between systems in HRV measures was 0.001 % and the highest 0.718 %. The artefact ratios in the 24 h daily activity measurements were 0.009 % at the lowest and 0.266 % at the highest.
Conclusions: The eMotion HRV system was accurate in QRS detection and the differences in measured RR intervals and calculated HRV measures were small. Thus, the performance of the system was comparable to the clinical ECG device in laboratory measurements and the precision was better than chest strap heart rate monitors. The quality of data was good also in long-term measurements in field conditions.
Ultrasound evaluation of articular cartilage and subchondral bone in vitro
Jukka Hirvasniemi, Oulun yliopisto
Introduction: Osteoarthritis (OA) is a common joint disease causing progressive degeneration of the articular cartilage and abnormal changes in the subchondral bone. Osteochondritis dissecans (OCD) is another, quite rare, but challenging joint disease in which an osteochondral fragment may detach at worst. Typical early-stage changes of OA and OCD are poorly detected by current diagnostic techniques. Intra-articular ultrasound (IAUS) is a recently developed imaging method that is feasible for the cartilage imaging at 40 MHz frequency.
Objective: The main aims of this study were to investigate the ability of IAUS technique at 9 MHz frequency to detect typical osteoarthritic changes occurring in human articular cartilage and subchondral bone and to detect typical OCD changes occurring in human subchondral bone.
Methods: Samples from human knee joints (n = 13) were imaged with IAUS in vitro. From the ultrasound signals, ultrasound reflection coefficient (R), integrated ultrasound reflection coefficient (IRC), apparent integrated backscattering coefficient (AIB) were calculated for the cartilage and subchondral bone, and ultrasound roughness index (URI) for the cartilage surface. Previously measured cartilage histological degenerative stage (Mankin score), water content, Young’s modulus, and dynamic modulus were used as a reference. Holes with different diameters were drilled into the subchondral bone of the samples to simulate typical changes in OCD. Visual evaluation was used to detect the holes from ultrasound images.
Results: There were a statistically significant negative correlation between cartilage reflection parameters (Rcart and IRCcart) and Mankin score (rs = -0.588, p < 0.05 and rs = -0.638, p < 0.05, respectively) and positive correlation between IRCcart and Young’s modulus (r = 0.567, p < 0.05). Furthermore, there was a positive correlation between Mankin score and URI (rs = 0.550, p = 0.051). None of the ultrasound parameters correlated statistically significantly with dynamic modulus or water content of the cartilage. The holes could not be detected from the ultrasound images.
Discussion: OA changes in the cartilage could be detected with quantitatively measured ultrasound parameters. The results indicated that the IAUS method can be applied to the simultaneous imaging of the articular cartilage and subchondral bone. Further studies are needed to improve the detection of typical OCD alterations with this technique.
Life-saving EEG system for use in emergencies
Andrei D. Jakab, Tampereen teknillinen yliopisto
Objective: The World Health Organization (WHO) classifies neurological disorders as a great threat to public health since these disorders and their sequelae are estimated to affect approximately one billion people worldwide. One of the most commonly used tests for the evaluation of neurologic disorders is the electroencephalogram (EEG). Despite its proven usefulness, because traditional EEG diagnostic methods are relatively time-consuming, require specialized personnel, and the recording systems are cumbersome, the EEG remains underused in clinical practice and particularly in emergency medicine (EM), where a quick diagnosis is sought after. The objectives of this study were to design and implement two EEG recording systems that would address the specific needs of EM by enabling caregivers to quickly secure a correct diagnosis and to establish a treatment plan in the vital early stages.
Methods: Two avenues of research were pursued. The first explored the suitability of a 6-channel wireless recording system that interfaces with a portable computer. The resulting system consists of a quick-application cap that can be normally applied on the patient in under 3 min, a wireless EEG (WEEG) amplifier, which digitizes the EEG and transmits it wirelessly to a measurement computer, and recording software, which stores the digital EEG in the standard EDF+ format. The second was centred on the novel concept of allowing the EEG to be recorded using the same infrastructure that is currently used for the recording of electrocardiographic (ECG) signals. This small EEG-to-ECG adapter dynamically processes the EEG signal in order to allow it to be displayed on any available ECG monitor.
Results: A thorough validation of the systems demonstrated the recording of high-quality signals as well as reliable operation. In addition to verifying their electrical characteristics, the EEG recorded by both systems was compared to that of commercial systems and found to be equivalent. Moreover, the WEEG system successfully underwent healthy volunteer testing in an examination room at the Department of Clinical Neurophysiology and in the ICU of the Seinäjoki Central Hospital.
Conclusions: Two fully-working EEG systems suitable for recording in emergency situations with a minimal preparation time have been developed, and their quality demonstrated. These novel systems facilitate the measurement process and make it possible to expand the practice of EEG to new recording locations. Clinical studies with the WEEG system currently underway at the Seinäjoki Central Hospital will help establish the utility of portable EEG systems in other clinical specialities (e.g., polysomnography).
Association between postexercise hypotension, autonomic nervous system and blood glucose concentration
Jukka Kiuttu, Oulun yliopisto
Postexercise hypotension (PEH) is defined as a reduction in systolic arterial blood pressure (SBP) and/or diastolic arterial blood pressure (DBP) below control levels after single bout of exercise. The precise reason of this phenomenon is unclear yet but it is still known that both central and peripheral factors are responsible for it. Furthermore, it is also known that arterial baroreflex is the most important short-term mechanism for controlling blood pressure. The purpose of the study was to examine the relations between PEH and both baroreflex sensitivity (BRS) and also blood glucose concentration level which is in a closely relationship with the regulation functions of autonomic nervous system (ANS).
Blood glucose content as well as continuous ECG, blood pressure and pulmonary gas exchange were measured from 14 healthy test subjects during a rest phase (20 min), a bicycle ergometer exercise (60 min) implemented at 80 % of the ventilation threshold (Tvent), and a recovery phase (60 min). The test subjects were seated during the rest phase and the recovery phase. Heart rate variability (HRV) and blood pressure variability (BPV) were measured to determine BRS using sequence and spectral methods.
During the recovery phase, blood glucose, oxygen consumption and respiratory exchange ratio (RER), as well as BRS, were on a lower level than during the rest phase. Heart rate (HR), however, was greater at the end of the recovery phase than during the rest phase. During the recovery phase, a correlation was observed between blood pressure (BP) and BRS: SBP – a-LF (r = -0,56 and p = 0,05) and DPB – a-LF (r = -0,70 and p = 0,01). From the rest phase to the end of the recovery phase there seem to be also some trend between these variables. Correlation (r = -0,65 and p = 0,05) emerged between the changes in diastolic blood pressure and in blood glucose content between the rest phase and the end of the recovery phase. During the rest phase there were not find any significant correlations between BP and BRS or BP and blood glucose concentration level.
A significant impact was achieved with an acute endurance exercise performance upon both glucose metabolism and the functioning of the ANS. It seems that there might be some closely relations between PEH and BRS activation and also between PEH and good blood glucose concentration balance achieved by acute exercise performance.
AR-mallien soveltaminen biosignaalien monimuuttuja-analyysissa
Pekka Kuoppa, Itä-Suomen yliopisto
Tutkimuksen tavoite: Fourier-muunnokseen perustuva periodogrammi on suosituin menetelmä estimoitaessa signaalien tehospektrejä. Tässä työssä selvitetään vaihtoehtoisen autoregressiivisen monimuuttujamallin (MAR) soveltuvuus teho- ja ristispektrien estimoimiseen. Tutkimuksessa perehdyttiin satunnaismuuttujien, MAR-mallien ja spektrien teoriaan. Lisäksi tutkittiin eri menetelmiä MAR-mallin ratkasemiseksi stationaarisessa ja dynaamisessa tapauksessa sekä eri kriteerejä mallin kertaluvun valitsemiseksi.
Materiaalit ja metodit: Tutkimuksen aineistona käytettiin simuloitua dataa sekä koehenkilöistä mitattua ortostaattisen testin ja kylmäaltistuskokeen aikaisia sykevaihtelun (RR) ja systolisen verenpaineen (SBP) aikasarjoja. Simuloidulle ja ortostaattisen testin datalle suoritettiin mallin kertaluvun valinta sekä staattinen ja dynaaminen spektriestimointi. Kylmäaltistusdatalle suoritettiin vain dynaaminen spektriestimointi. Spektriestimaatteja verrattiin Fourier-muunnoksella laskettuihin periodogrammeihin.
Tulokset: MAR-mallin kertoimien estimoinnissa Least Square -menetelmä todettiin Yule-Walker -menetelmää tarkemmaksi. Eri kertaluvun valintakriteereillä saatiin hyvin yhteneviä tuloksia, mutta joidenkin menetelmien todettiin yliarvioivan kertalukua. Tämä johtui menetelmässä olevan, kertaluvun suuruudesta riippuvan, rangaistustermin liian pienestä vaikutuksesta käytetylle signaalille. MAR-mallilla lasketuista stationaarisista spektreistä tunnistettiin samat piikit kuin spektrogrammeista, mutta ne olivat kokonaisuudessaan paljon selkeämpiä tulkita. Dynaamisessa estimoinnissa MAR-mallin spektrien tehoissa näkyi suurempia muutoksia verrattuna Fourier-muunnoksella laskettuihin. Molempien menetelmien spektrien muutosten nopeus ja niiden näkyvyys todettiin kuitenkin riippuvan laskentaan ja piirtämiseen käytetyistä parametreista.
Johtopäätökset: Mallien kertaluvun estimoinnin tuloksena oli, että kaikki tutkitut valintakriteerit antavat lähes saman arvion prosessin kertaluvuksi. Jotkin menetelmät eivät kuitenkaan saa minimiään oikean kertaluvun kohdalla, mutta niiden arvot kuitenkin tasaantuvat huomattavasti ylitettäessä se. Käytännössä onkin hyvä käyttää useampaa menetelmää estimoimiseen ja arvioida niiden perusteella silmämääräisesti sopiva kertaluku. Tässä työssä keskityttiin stationaarisen mallin kertaluvun estimoimiseen ja päädyttiin käyttämään siitä saatua kertalukua myös dynaamisessa spektriestimoinnissa. Jatkossa olisi syytä tutkia käykö stationaariselle aikasarjalle valittu kertaluku myös dynaamiselle mallille.
MAR-mallien todettiin estimoivan dynaamisia spektrejä hyvin. Lisäksi mittausprotokolliin liittyvät kirjallisuudessa mainitut muutokset näkyvät osittain voimakkaammin MAR-mallien spektreissä, mutta tämä johtuu myös molempien menetelmien valituista parametreistä. AR-mallien on kuitenkin todettu toimivan spektrogrammia paremmin lyhyissä mittauksissa. Tämä on etuna AR-mallille myös dynaamisessa tapauksessa, koska epästationaaristen signaalien takia voidaan kuvitella spektrin laskemiseksi käytettävän vain lyhyt pätkä dataa. Spektrogrammin tapauksessa ajasta riippuvat spektrit lasketaankin liikkuvalla ikkunalla, jonka leveys määrää käytettävän datan pituuden. Tällöin jokaisen ajankohdan spektrin laskentaan käytetään tietty määrä signaalia ennen ja jälkeen aikapisteen. Tämä ikkuna ei voi olla liian suuri tai spektrin temporaalinen resoluutio heikkenee.
Verrattaessa MAR-mallilla saatuja spektrejä Fourier-muunnoksella saatuihin, voidaan niiden todeta olevan selkeämpiä ja helpommin tulkittavia. MAR-malli soveltuu myös dynaamisten spektrien estimoimiseen, kun mallin kertaluku ja Kalman-suotimen parametrit valitaan huolella. Parametrien alkuarvot täytyy valita tarkkaan jokaiselle signaalille erikseen, mikä tekee MAR-mallin käytöstä hieman hankalampaa Fourier-muunnokseen verrattuna. Lisäksi MAR-mallista saatavien ristispektrien avulla voidaan laskea koherenssi- ja vaihefunktio. Niitä voidaan käyttää esimerkiksi koehenkilön barorefleksiherkkyyden analysointiin RR- ja SBP-aikasarjoista.
Abdominal adipose tissue glucose uptake is increased after bariatric surgery
Tam Pham, PET-keskus
Introduction: Studies have shown that bariatric surgery improves insulin sensitivity and decrease adipose tissue masses. The aim of this study was to investigate the effect of bariatric surgery on adipose tissue regional glucose uptake.
Methods: Twenty three morbidly obese subjects (BMI =42.6± 3.5) with fifteen type 2 diabetes mellitus or IFG/IGT(diabetics) and 8 normal glucose tolerance(non diabetics). All the obese subjects were randomized for two bariatric surgery technique, either gastric sleeve(n=10) or gastric bypass(n=13). Magnetic resonance imaging (MRI) and positron emission tomography (PET) with 18F- Flourodeoxy glucose was taken during fasting and insulin clamp before and six months after the surgery. Ten healthy adult controls (BMI<27 kg/m2) were also taken MRI and PET with 18F-FDG with fasting and clamp at base-line.
Results: Bariatric surgery decreased total weight by 26.9 ± 8.3 kgs (p<0.01 ) after six months. Abdominal subcutaneous adipose tissue (SAT) mass decreased 6,6 ± 1,3 kg (p < 0,0001 ) in diabetic obese and 7,5 ± 3,9 kg (p = 0,0023) in non-diabetic obese. Abdominal visceral adipose tissue (VAT) mass was reduced by 1,8 ± 1,1 kg (p = 0,000131) and 1,2 ± 0,5 kg (p = 0,0011) respectively. During fasting, VAT and SAT regional glucose uptake in morbidly obese were higher before surgery than in normal control (p < 0.05) but not statistically different after surgery compared with normal (p>0.3). VAT and SAT regional glucose uptake remained unchanged during fasting before and after surgery (p>0.5). During insulin clamp, VAT and SAT regional glucose uptake in obese after surgery and in normal healthy control were higher than in obese before surgery (p<0.01). However, no differences in metabolic effects were observed between diabetic and non diabetic obese patients.
Conclusion: Weight lose through bariatric surgery improved insulin stimulated regional glucose uptake in both abdominal visceral and subcutaneous adipose tissue in which sleeve and gastric bypass techniques effects similarly in diabetic and non-diabetic obese.
Quantitative ultrasound image analysis of femoral subchondral bone in knee osteoarthritis
Jana Podlipská, Oulun yliopisto
Background: It has been reported that non-invasive ultrasonography (US) of the knee joint predicts better the osteoarthritic (OA) diagnostic outcome from arthroscopy than conventional radiography. However, non-invasive US has been concentrated on qualitative or semi-quantitative grading of OA changes only in articular cartilage. In this study, a potential of quantitative non-invasive knee US for detecting changes in femoral subchondral bone was investigated.
Materials and Methods: Thirty-nine patients (15 women, 24 men, mean age=52), referred to a knee arthroscopy because of knee pain, underwent the US examination. The gray-level intensity profile of medial and lateral condyle, and intercondylar notch (sulcus) of the subchondral bone was investigated from US images using custom-made Matlab script (The MathWorks Inc., Natick, MA, USA). The quantitative US findings were compared with conventional radiography using Kellgren-Lawrence (K-L) grading scale, and with arthroscopy using Noyes’ grading scale.
Results: Current in vivo results confirmed the earlier in vitro findings that the ultrasound reflection and backscattering from the subchondral bone increases in OA. This is most probably explained by the bone sclerosis, in which the bone density increases, thus also increasing the ultrasound reflection from the cartilage-bone interface. Quantitatively, statistically significant correlations were found especially between US gray-level intensity in femoral bone depth level 2 and arthroscopic or K-L grading, and between femoral bone depth level all and K-L grading. Statistically significant differences were observed between US gray-level intensity in different bone depth levels and arthroscopic grades 1 vs. 2, as well as K-L grades 0 vs. 1. In site-specific US results, most of the intensity variations were detected in medial and sulcus areas as OA progressed.
Conclusion: The findings of this study are important as it has been earlier reported that articular cartilage changes can be evaluated with non-invasive US. In the future, combined quantitative analysis of articular cartilage and subchondral bone from US data may yield even more sensitive indication for early OA changes.
Transnedothelial migration of breast cancer cells during cancer metastasis
Joanna Pylvänäinen, Turun yliopisto
Transendothelial migration (TEM) is a well known process in which a cell penetrates the endothelium of a blood or a lymphatic vessel. The most studied cell type that utilizes this mechanism is the leukocytes, which follow the chemokine gradient created by local inflammations and travel through the bloodstream towards the inflamed area. Once they reach the inflammation site, leukocytes penetrate the endothelium using one of two distinct routes, the paracellular or the transcellular route. Invasive cancer cells have shown to use similar machinery to penetrate vascular endothelium during cancer metastasis. The blood system has been considered as the main cancer metastasis route, but evidence of metastasis through lymphatic system is increasing, though the mechanism still remains unknown. The endothelium has commonly been considered as a passive mechanical barrier during TEM. However, recently it has been shown to be an active member of the cell invasion machinery, conveying signals from the invasion site into the cell and remodeling the cytoskeleton, therefore enabling TEM.
In this thesis, the invasion routes of invasive and non-invasive breast cancer cells through lymphatic and vascular endothelia were studied. Also the reorganization of endothelial cytoskeleton during TEM was followed. The main methods used for this research were confocal microscopy combined with advanced cell staining techniques.
The results suggested that in lymphatic endothelium cancer cell invasion happens predominately through paracellular route, whereas in vascular endothelia the invasion mainly occurs transcellularly. The cytoskeleton seems to play an important role in guiding the invading cancer cell to appropriate invasion site and enabling the invasion.
In conclusion, both lymphatic and vascular endothelia act as a dynamic member of the transendothelial migration machinery, although distinct endothelium specific properties remain. These properties were adapted according to the invading cancer cell type.
Rusto-luurakenteiden kvantifiointi SWIFT-magneettikuvausmenetelmällä
Jari Rautiainen, Tampereen teknillinen yliopisto
Tausta ja tavoite: Nivelrusto toimii pehmentävänä kudoksena niveltävien luiden päissä ja on täten välttämätön normaalin liikuntakyvyn ylläpitämiseksi. Nivelrikko on yleinen nivelsairaus, jonka tunnusmerkkejä ovat nivelruston rappeutuminen sekä muutokset rustonalaisessa luussa. Nivelruston kyky korjata itseään on hyvin rajallinen, jonka vuoksi sairaus olisi hyvä tunnistaa jo varhaisessa vaiheessa. Perinteiset magneettikuvaustekniikat (MK) sekä muut tämänhetkiset kliiniset menetelmät ovat liian epätarkkoja nivelrikon varhaiseen diagnosointiin tai sairauden etenemisen seurantaan. Uudet MK-tekniikat, kuten Sweep Imaging with Fourier Transformation (SWIFT) ovat lupaavimpia vaihtoehtoja. Tämän diplomityön tarkoituksena oli arvioida SWIFT-sekvenssin soveltuvuutta rusto-luurakenteiden kuvantamiseen. Rusto-luu rajapintaa ei ole aikaisemmin tutkittu yhtä laajasti SWIFT-menetelmällä.
Menetelmät: Tässä työssä tutkittiin naudan ja hevosen nivelrustonäytteitä. Nautanäytteet sisälsivät ehjää sekä entsyymein vaurioitettua nivelrustoa ja hevosnäytteet rustovaurion, jonka oli annettu parantua itsestään ennen näytteiden keräämistä. Näytteet kuvattiin kokeellisella 9.4 Teslan magneettikuvauslaitteella. Eri saturaatiotekniikoin otetuista SWIFT-kuvista mitattiin nivelruston ja rustonalaisen kovan luun paksuuksia, jonka lisäksi tutkittiin SWIFT-tekniikalla mitatun nivelruston signaali-intensiteetin vaihtelua. Vertailumenetelminä käytettiin perinteisiä MK-tekniikoita, korkean resoluution tietokonetomografiakuvia (TT) sekä histologisia menetelmiä. Työssä kehitettiin myös data-analyysiä helpottavia työkaluja.
Tulokset: Nivelruston signaali-intensiteetti SWIFT-kuvissa ei juuri vaihtele, olipa kyseessä vaurioitunut, itsestään korjautunut tai ehjä kudos. Tutkimus osoitti myös, että ruston syvistä osista sekä rustonalaisesta luusta saadaan SWIFT-tekniikalla enemmän signaalia verrattuna perinteisiin MK-menetelmiin. Nivelrusto näkyi paksumpana SWIFT-kuvissa verrattuna spin-kaiku-sekvenssillä otettuihin kuviin, kun taas gradienttikaikukuvissa rusto nähtiin saman paksuisena. Silti, SWIFT-kuvissa rustonalaisen luun signaalin määrä oli huomattavasti suurempi. Tämän työn tärkein havainto oli, että rustonalaisen kovan luun paksuusmääritykset SWIFT-kuvista olivat vertailukelpoisia TT-kuvista saatuihin tuloksiin, vaikka resoluutio TT-kuvissa on huomattavasti parempi.
Johtopäätökset: Tutkimus osoitti, että tulevaisuudessa voisi olla mahdollista määrittää rustonalaisen luun ominaisuuksia tarkasti magneettikuvista, jolloin vältytään TT-kuvauksien aiheuttamilta säteilyannoksilta. Seuraavissa tutkimuksissa olisi hyvä laskea kvantitatiivisempia parametreja, kuten esimerkiksi luu-kudossuhde, ja verrata tuloksia TT-mittauksiin. Lisäksi, jos mahdollista, tutkimuksia tulisi suorittaa myös ihmisnäytteillä tai kliinisissä tutkimuksissa.
Application of magnet quench analysis
Noora Tuovinen, Tampereen teknillinen yliopisto
Post Mortem Analysis is a software tool built for CERN for hardware commissioning and post mortem event analyzing for the Large Hadron Collider (LHC). Magnet Quench Analysis application is a part of the Post Mortem Analysis tool and serves a base for analyzing quench data. It gives a possibility to observe and analyse data collected by the quench protection system.
The goal of the thesis was to study physics phenomena of superconductivity. Superconductivity finds many applications and a closer look on biomedical and collider applications was taken with the examples from Magnetic Resonance Imaging (MRI) and the LHC with special attention to the quench phenomenon. Quench, quench protection and post quench data are studied in more detail. In addition, the thesis describes how the existing LHC magnet quench analysis was extended to the new Quench Protection System data. Application was developed using LabVIEW programming language.
Based on the observations from magnet quench analysis, users can determine such parameters as firing times of the triggers and proper operation of the quench protection system. In this work especially quench detection, energy, location, heater protection and time delays are discussed.
It was discovered that automatic quench analysis saves precious time and reduces the need for manual calculations. Biggest problems during the project were found in code modification phase between different LabVIEW versions and programmers. It is possible that further development of the code enhances the possibilities of quench analysis application. Designed methods for dipole magnets can be easily extended for different magnet types such as quadrupoles with changes in the configuration settings.
Device for simultaneous measurement of ERG-signal from desired retinal layer and across the retina
Teemu Turunen, Aalto-yliopisto
Valon absorptio verkkokalvon näköaistinsoluissa saa aikaan monivaiheisen biokemiallisten reaktioiden sarjan, jossa valon sisältämä tieto muunnetaan solutason sähköisiksi signaaleiksi. Näiden signaalien aiheuttamia muutoksia verkkokalvon soluvälitilan jänniteprofiilissa voidaan tutkia elektroretinogrammi-tekniikalla (ERG). Eristetyn verkkokalvon ERG-kokeiden avulla voidaan tehdä pitkäkestoisia, hyvin toistuvia ja farmokologisesti manipuloitavissa olevia näköaistinsolujen fysiologian tutkimuksia. Tulokset ovat vertailukelpoisia myös elävän eläimen ERG-kokeisiin (kliinisesti käytetty menetelmä). Verkkokalvon biokemiallisen signaloinnissa tärkein tutkimuskohde on näköaistinsolujen ulkojäsenten valoherkät kationivirrat, joiden tutkimiseen voidaan käyttää myös ERG-tekniikkaa suorempia mutta invasiivisempia menetelmiä. Tutkimuksen tavoitteena oli selvittää missä määrin verkkokalvon puolten väliltä rekisteröitävää ERG-signaalia (TERG) voidaan käyttää näköaistinsolujen ulkojäsenvirran tutkimuksessa. Tutkimuksessa pyrittiin kehittämään laitteisto, jolla voidaan jatkossa tutkia myös ERG-signaalin eri komponenttien alkuperää, joita ei täysin vielä tunneta.
Eristetyn verkkokalvon ERG-tekniikoissa verkkokalvo pidetään hengissä ravinneliuoksen avulla. Verkkokalvoa stimuloidaan homogeenisillä valopulsseilla ja rekisteröidään muodostuvaa valovastetta. TERG-tekniikassa verkkokalvon jänniteprofiilia rekisteröidään verkkokalvon molemmin puolin asetetuilla makroelektrodeilla ja LERG-tekniikassa käytetään mikroelektrodeja, jotka työnnetään haluttuun verkkokalvon kerrokseen. Tutkimuksessa kehitettiin laitteisto, jolla voidaan rekisteröidä valovasteita hiiren eristetystä verkkokalvosta samanaikaisesti TERG- sekä LERG-tekniikoilla. LERG-tekniikan etu TERG-tekniikkaan nähden on se, että sillä voidaan rekisteröidä verkkokalvon sähköistä signalointia mistä tahansa verkkokalvon kerroksesta. Tämän takia menetelmä on hyödyllinen ERG-signaalin komponettien alkuperän selvittämisessä. TERG-tekniikassa rajoitutaan farmakologiseen manipulointiin, jolloin signaalivälitys voidaan katkaista näköaistinsolujen synapsialueesta eteenpäin tai sulkea yksittäisiä ionikanavatyyppejä. Toisaalta TERG-tekniikka on teknisesti helpompi, stabiilimpi ja signaali-kohinasuhteeltaan huomattavati parempi.
Laitteistolla suoritettiin alustava tutkimus, jossa näköaistinsolujen valoherkkiä kationivirtoja rekisteröitiin suoraan ulkojäsenkerroksesta LERG-tekniikalla ja näitä verrattiin TERG-tekniikalla rekisteröityihin farmakologisesti erotettuihin näköaistinsolukerroksen valovasteisiin. LERG- ja TERG-vasteita vertailtaessa huomattiin, etteivät vasteiden nousunopeus, katkaisukinetiikka, nousuaika huippuarvoon tai suhteelliset amplitudit poikenneet toisistaan merkittävästi. Suurin eroavaisuus oli voimakkailla valostimuluksilla TERG-vasteisiin muodostuva nopea transientti aalto. Lisäksi todettiin, että n. 10 % näköaistinsolukerroksen ERG-signaalista muodostuu ulkojäsenkerroksen ja loput sisäjäsenkerroksen matkalla.
Towards high-quality ultra-low-field MRI: novel methods for suppressing eddy currents
Koos C.J. Zevenhoven, Aalto-yliopisto
Introduction and aim of study. While the state of the art of magnetic resonance imaging (MRI) has developed towards using multiple-tesla magnetic fields, another approach has emerged, where the signal is detected by highly sensitive SQUID sensors in a magnetic field on the order of 100 µT. Such ultra-low-field (ULF) MRI has many promising characteristics, including safety, silent operation, potentially low cost, and compatibility with other electromagnetically sensitive technology. However, ULF MRI has not yet reached its breakthrough, as the image signal-to-noise ratio (SNR) and field of view (FOV) are still insufficient for most applications. The aim of this thesis work was to improve the image quality of ULF MRI by designing and implementing upgrades to the pioneering ULF-MRI setup at UC Berkeley.
Methods. Since, in ULF MRI, the signal is proportional to a prepolarizing field Bp, which is pulsed using a designated coil, a natural approach was to increase Bp and the associated FOV by installing a larger water-cooled coil. These changes, however, directly lead also to larger unwanted transient effects, most notably, eddy currents in the conducting walls of the magnetically shielded room (MSR), which screens the system from external magnetic interference. Therefore, an MSR with weakly connected regions was designed and built to accommodate shielding currents but no harmful transients. A quantitative theory to describe eddy currents in MSR walls was presented. Based on the theory, a novel method, dynamic shielding, was introduced, where a single additional coil is pulsed with current waveforms specifically designed to kill selected spatiotemporal components of the harmful transient.
Results. The amplitude of the transient magnetic field generated by eddy currents was reduced by three orders of magnitude at roughly 15 ms after a Bp pulse, meeting the MSR design criteria. The transient eddy current patterns in the walls of the old and new MSR were also reconstructed as a function of time from a large number of magnetic-field measurements. The obtained results are, within error, identical with those of numerical calculations based on the presented theory. In computer simulations, the novel dynamic shielding method was able to eliminate selected components of the eddy-current transient with an accuracy limited by numerical precision. To show the effectiveness of the upgrades, ULF-MRIs of bacon and fruit were acquired, as human-subject approval had not yet been obtained.
Conclusions. Theory and methods were developed for tackling a family of transient-related issues that impede the progress in ULF-MR image quality. The eddy currents, being a particularly serious problem, were addressed in two ways. A new MSR was successfully designed and constructed, with an improvement corresponding to three orders of magnitude in either FOV volume or in SNR. Further, a quantitative theory of eddy currents in such an MSR was presented, which was in excellent agreement with measured eddy-current-density patterns. The dynamic shielding method was introduced, which, based on theory and simulations, gives a powerful tool for suppressing eddy currents in a more efficient and flexible way compared to any pre-existing method.