Search

Skin cancer is the most common cancer, and its incidence is increasing each year. Its diagnosis is based on a histopathological inspection by pathologists. Other imaging techniques are currently being actively researched to assist pathologists in improving diagnostic accuracy. Fourier Transform Infrared (FTIR) spectroscopy is gaining in popularity among researchers due to its ability to provide spatially resolved biochemical and biological information while also being compatible with histopathological samples. Furthermore, facilitating FTIR spectroscopic data with various machine learning methodologies for automating classification has become a popular practice. These methodologies have achieved high accuracy in discriminating cancerous samples from healthy ones, but limited work has been done to determine the metastatic potential of cells and optimize the FTIR spectroscopy protocol for achieving the best classification result. This thesis reviewed all skin cancer studies that used FTIR spectroscopy as a potential diagnostic tool to assess its effectiveness in various aspects of skin cancer research and to analyse the performance of different methods used for differentiating diseased samples from healthy ones. In addition, a model was developed to examine whether different malignant melanoma cell lines (primary and metastatic) can be discriminated using the spectral data acquired from the samples. Furthermore, the effects of different modes (transmission and transflection) and sample processing methods (deparaffinized and paraffinized) on the acquired spectra were investigated, and the combination providing the best classification result was determined. According to the results of this thesis, FTIR spectroscopy can differentiate different types of skin cancers from healthy samples as well as between different skin cancers (melanoma and non-melanoma) and the most common non-melanoma cancers. Melanoma has received the most attention in skin cancer research du, Tiivistelmä Ihosyövät ovat maailmanlaajuisesti yleisimpiä syöpiä, ja niiden ilmaantuvuus kasvaa joka vuosi. Ihosyöpien diagnoosi perustuu patologin tekemään histopatologiseen tutkimukseen. Kuvantamistekniikoita tutkitaan aktiivisesti, jotta ihosyöpien diagnostiikkaa voitaisiin parantaa. Infrapuna (IR)-spektroskopia on kasvattanut suosiota tutkijoiden keskuudessa, koska sillä voidaan tutkia kudosleikkeistä alueellisia eroja kudoksen koostumuksessa. Kudosten IR-spektrien luokittamisessa käytetään usein koneoppimiseen perustuvia automaattisia menetelmiä. Näillä menetelmillä on kyetty erottamaan tarkasti syöpäsolut terveistä soluista. Toisaalta syöpäsolujen metastasoimispotentiaalin määrittämiseksi IR-spektreistä on tehty vain vähän tutkimuksia, ja mittausprotokollia ei ole vielä optimoitu parhaimman luokitustuloksen saavuttamiseksi. Tässä väitöskirjatyössä tehtiin kirjallisuuskatsaus kaikista IR-spektroskopiaa hyödyntävistä ihosyöpätutkimuksista. Katsauksessa arvioitiin IR-spektroskopian soveltuvuutta ihosyöpätutkimuksen eri osa-alueilla ja analysoitiin eri mittaus- ja data-analyysimenetelmien kykyä erottaa sairaat ja terveet ihonäytteet toisistaan. Lisäksi tässä työssä kehitettiin malli, jolla pyrittiin luokittelemaan pahanlaatuisia melanoomasolulinjoja niiden IR-spektrien perusteella. Työssä tutkittiin myös eri mittausmoodien (transmissio ja transflektio) ja näytteenkäsittelymenetelmien (ei-parafinoitu ja parafinoitu) vaikutuksia mitattuihin IR-spektreihin ja määritettiin näiden yhdistelmä, jolla saadaan paras tulos melanoomasolujen luokittelemisessa. Väitöskirjan tulosten perusteella voidaan sanoa, että IR-spektroskopialla pystytään erottamaan eri ihosyövät terveestä ihokudoksesta ja tunnistamaan ihosyövän tyyppi. Melanooma on saanut eniten huomiota ihosyöpätutkimuksessa, koska se pystyy muodostamaan etäpesäkkeitä lähellä oleviin imusolmukkeisiin ja muihin elintärkeisiin elimiin lyhyessä ajassa. IR-spektroskopialla voidaan erottaa eri pahanlaatuiset melanoomasolulin

Osteoarthritis (OA) is a serious joint disease affecting millions of people globally. Early detection of the disease allows for slowing down its progression, leading to reduced healthcare costs. Furthermore, dental pathological conditions are highly prevalent and often lead to more extensive infection, comorbidities and costly late-stage interventions when left untreated. In this doctoral study, new methods are proposed for OA research and clinical computed tomography (CT) of mineralized tissues using laboratory micro-computed tomography (µCT) imaging and machine learning. First, automated software was developed to assess the histopathologic grade of human osteochondral samples using machine learning and contrast-enhanced µCT. The automated grading performed adequately on independent data and, with further development, could be useful for the development of new OA drugs requiring objective assessment of the tissue changes. In the second sub-study, deep-learning based segmentation and morphological analysis were used to assess the calcified cartilage thickness in New Zealand White rabbit knees. Analysis of µCT reconstructions and histological images could differentiate anatomical surfaces with distinct biomechanical environments. The developed technique is useful in understanding the role of osteochondral junction during mechanical load, aging and disease. Finally, deep-learning based super-resolution models were trained on ex vivo µCT images of human knee samples and extracted teeth. The models were assessed rigorously with independent CT test data for technical image quality, ability to predict trabecular bone microstructure, as well as clinical image quality on the knee, wrist, ankle and maxillofacial region. The super-resolution techniques have the potential for earlier and more accurate diagnostics, treatment planning and dose reductions. In conclusion, the first two sub-studies facilitate a better understanding of OA etiology in basic research. The thi, Tiivistelmä Nivelrikko on maailman yleisin nivelsairaus, joka vaikeuttaa miljoonien ihmisten elämää maailmanlaajuisesti. Sairauden varhainen havaitseminen mahdollistaisi sen etenemisen hidastamisen, mikä edelleen johtaisi terveydenhuollon kustannusten alenemiseen. Hampaiston sairaudet ovat myös erittäin yleisiä ja johtavat hoitamattomina laajempiin tulehduksiin, liitännäissairauksiin ja kalliisiin myöhäisvaiheen toimenpiteisiin. Tässä väitöskirjassa kehitettiin uusia menetelmiä nivelrikon tutkimukseen ja mineralisoituneiden kudosten kliiniseen tietokonetomografiaan (TT) mikrotietokonetomografia (µTT) -laboratoriokuvantamisen ja koneoppimisen avulla. Ensimmäisessä osatyössä kehitettiin automaattinen tietokoneohjelmisto ihmisen luu-rustonäytteiden histopatologiseen arviointiin koneoppimisen ja kontrastitehostetun µTT:n avulla. Automaattiarvioinnin suorituskyky osoittautui hyväksi, ja jatkokehityksen jälkeen menetelmää voitaisiin tulevaisuudessa käyttää esimerkiksi uusien nivelrikkolääkkeiden kehityksessä, missä vaaditaan objektiivista arviota nivelrikon kudosmuutoksille. Toisessa osatyössä kehitettiin syväoppiva kuva-alueen rajausmenetelmä kalkkeutuneen nivelruston paksuuden määrittämiseen jäniksen polvinivelessä. Kehitetyn menetelmän avulla pystyttiin erottamaan anatomiset nivelpinnat, joilla on erilainen biomekaaninen ympäristö. Menetelmä voi auttaa ymmärtämään luun ja ruston liitoskohdan merkitystä mekaanisen rasituksen, ikääntymisen ja nivelrikon aikana. Kolmannessa osatyössä koulutettiin syväoppivia superresoluutiomalleja ihmisen polvinäytteiden ja poistettujen hampaiden µTT-kuville. Mallien tarkkuutta arvioitiin riippumattomalla testiaineistolla teknisen kuvanlaadun, luun mikrorakenteiden ennustamiskyvyn sekä polven, ranteen, nilkan ja hampaiston kliinisen kuvanlaadun määrittämiseksi. Tällaisilla superresoluutiomalleilla tulee mahdollisesti olemaan paljon sovelluksia mm. tehokkaammassa diagnostiikassa ja hoidossa sekä potilaiden säteilyannosten alentamisessa. Yh

Osteoarthritis is globally a highly prevalent joint disorder leading to major disabilities among affected patients. With increasing obesity and prolonging life expectancy of the world’s population, the burden of this disease will be even more substantial in the future. However, little is known about the onset and the earliest events of osteoarthritis on the microstructural level—especially regarding the soft tissues of the joint. Therefore, being able to image and quantify them volumetrically on a micrometer level might help us gain a better understanding of osteoarthritis and its early events. The main objective of this thesis was to develop a hexamethyldisilazane (HMDS)-based micro-computed tomography (micro-CT) method that enables the visualization of knee joint soft tissues. The developed methodology was validated by morphological and microstructural quantitative analysis of articular cartilage and meniscus specimens from osteoarthritic patients and osteoarthritis-free tissue donors. Conventional histology was used as a reference methodology. It was shown that with HMDS-based micro-CT, similar scale structural features of articular cartilage and meniscus can be visualized in 3D as with conventional 2D histology. Importantly, the proposed technique enabled quantitative analysis of the studied tissues. In articular cartilage, morphological analyses showed that chondrons were larger and more elongated in osteoarthritic samples than in intact tissue. In meniscus, microstructural local orientation analysis revealed increased disorganization in the medial menisci of medial compartment knee osteoarthritis patients compared to the lateral menisci of the same patients and medial and lateral menisci of osteoarthritis-free tissue donors. From the analysis of meniscal calcifications, it seemed that there are two patterns of meniscal degeneration: one with diminutive calcifications and other with large to widespread calcifications. To conclude, the HMDS-based micro, Tiivistelmä Nivelrikko on maailman yleisin nivelsairaus, joka lopulta johtaa liikuntakyvyn menetykseen. Ylipainon yleistyessä ja elinajanodotteen pidentyessä tämän sairauden aiheuttama yhteiskunnallinen taakka tulee olemaan entistäkin merkittävämpi tulevaisuudessa. Nivelrikon alkuvaiheen mikrorakennetason muutoksista ei kuitenkaan vielä tiedetä kovin paljoa etenkään nivelen pehmytkudosten osalta. Näiden kudosten kolmiulotteinen kuvantaminen mikrometrien tarkkuudella voisi auttaa ymmärtämään nivelrikkoa sekä sen alkuvaiheen muutoksia paremmin. Tämän väitöskirjan päätavoitteena oli kehittää näytteiden kuivaamiseen heksametyylidisilatsaanilla (HMDS) perustuva mikrotietokonetomografiamenetelmä (mikro-TT-menetelmä) polvinivelen pehmytkudosten kuvantamiseen. Lisäksi kyseistä menetelmää validoitiin nivelruston ja -kierukan morfologiaa ja mikrorakenteita määrittävillä kvantitatiivisilla analyyseillä käyttäen nivelrikkopotilailta ja nivelrikkoa sairastamattomilta kudoslahjoittajilta saatuja näytteitä. Perinteinen histologia toimi verrokkimenetelmänä. HMDS:ään perustuvan mikro-TT:n osoitettiin kuvaavan saman kokoluokan rakenteellisia piirteitä nivelrustossa ja -kierukassa kolmiulotteisesti kuin perinteinen histologia kaksiulotteisesti. Lisäksi menetelmä mahdollisti kvantitatiivisten analyysien käytön: nivelrustossa morfologiset analyysit osoittivat, että kondronit olivat suurempia ja pitkänomaisempia nivelrikkoisissa näytteissä verrattuna terveisiin näytteisiin. Nivelkierukan orientaatioanalyysissä havaittiin, että mediaalipuolen nivelkierukan mikrorakenteet ovat epäjärjestäytyneempiä mediaalipuolen nivelrikkoa sairastavilla potilailla verrattuna samojen potilaiden lateraalipuolen näytteisiin ja terveiden kudoslahjoittajien mediaali- ja lateraalipuolen näytteisiin. Nivelkierukoiden kalkkeuma-analyyseissä havaitsimme kahdentyyppistä nivelkierukan rappeutumista: ensimmäisessä tyypissä oli erittäin vähäinen määrä kalkkeumia ja toisessa tyypissä oli suuria ja/tai laajasti levinne

The collagen network is the highly organized backbone of articular cartilage providing tissue tensile stiffness and restricting proteoglycan bleaching out of the tissue. Osteoarthritis (OA) diminishes proper collagen network adaptation. Our aim was to provide quantitative three-dimensional (3D) information of the cartilage collagen network adaptation in early osteoarthritis using high resolution micro-computed tomography (µCT)-imaging. Osteochondral samples from the femoral condyles were collected from healthy (N = 8, both legs) and experimental OA rabbit model with anterior cruciate ligament transection (N = 14, single leg). Samples were processed for cartilage µCT-imaging and histological evaluation with polarized light microscopy (PLM). Structure tensor analysis was used to analyse the collagen fibre orientation and anisotropy of the µCT-images, and PLM was used as a validation for structural changes. Depth-wise comparison of collagen fibre orientation acquired with µCT-imaging and PLM correlated well, but the values obtained with PLM were systematically greater than those measured with µCT-imaging. Structure tensor analysis allowed for 3D quantification of collagen network anisotropy. Finally, µCT-imaging revealed only minor differences between the control and experimental groups.

This paper addresses the challenge of grading visual features in lumbar spine MRI using Deep Learning. Such a method is essential for the automatic quantification of structural changes in the spine, which is valuable for understanding low back pain. Multiple recent studies investigated different architecture designs, and the most recent success has been attributed to the use of transformer architectures. In this work, we argue that with a well-tuned three-stage pipeline comprising semantic segmentation, localization, and classification, convolutional networks outperform the state-of-the-art approaches. We conducted an ablation study of the existing methods in a population cohort, and report performance generalization across various subgroups. Our code is publicly available to advance research on disc degeneration and low back pain.

Mid-infrared spectroscopy (MIR), near-infrared spectroscopy (NIR), and Raman spectroscopy are all well-established analytical techniques in biomedical applications. Since they provide complementary chemical information, we aimed to determine whether combining them amplifies their strengths and mitigates their weaknesses. This study investigates the feasibility of the fusion of MIR, NIR, and Raman spectroscopic data for characterising articular cartilage integrity. Osteochondral specimens from bovine patellae were subjected to mechanical and enzymatic damage, and then MIR, NIR, and Raman data were acquired from the damaged and control specimens. We assessed the capacity of individual spectroscopic methods to classify the samples into damage or control groups using Partial Least Squares Discriminant Analysis (PLS-DA). Multi-block PLS-DA was carried out to assess the potential of data fusion by combining the dataset by applying two-block (MIR and NIR, MIR and Raman, NIR and Raman) and three-block approaches (MIR, NIR, and Raman). The results of the one-block models show a higher classification accuracy for NIR (93%) and MIR (92%) than for Raman (76%) spectroscopy. In contrast, we observed the highest classification efficiency of 94% and 93% for the two-block (MIR and NIR) and three-block models, respectively. The detailed correlative analysis of the spectral features contributing to the discrimination in the three-block models adds considerably more insight into the molecular origin of cartilage damage.

Traditionally, osteoarthritis (OA) is diagnosed with the clinical examination supplemented by the conventional radiography (CR). In the research literature, the role of ultrasound (US) imaging in the diagnostics of OA has risen steadily during the last two decades. US imaging is cheap and globally widely available often already in primary healthcare. Here, we reviewed the most essential US literature focusing on OA diagnostics and progression prediction using the various search engines. Starting from the year 2000, our search provided 1 445 journal articles. After reviewing the abstracts, 89 articles were finally included. Most of the reviewed articles focused on the imaging of knee and hand OA, whereas only a minority dealt with the imaging of hip, ankle, midfoot, acromioclavicular, and temporomandibular joints. Overall, during the last 20 years, the use of US imaging for OA assessment has increased in the scientific literature. In knee and hand joints, US imaging has been reported to be a promising tool to evaluate OA changes. Furthermore, the reproducibility of US as well as its association to MRI findings are excellent. Importantly, US seems to even outperform CR in certain aspects, such as detection of osteophytes, joint inflammation, meniscus protrusion, and localized cartilage damage (especially at the medial femoral condyle and sulcus area). Based on the reviewed literature, US can be truly considered as a complementary tool to CR in the clinical setup for OA diagnostics. New technical developments may even enhance the diagnostic value of the US in the future.

The relationships between structure and function in human knee femoral cartilage are not well-known at different stages of osteoarthritis. Thus, our aim was to characterize the depth-dependent composition and structure (proteoglycan content, collagen network organization and collagen content) of normal and osteoarthritic human femoral condyle cartilage (n = 47) and relate them to their viscoelastic and constituent-specific mechanical properties that are obtained through dynamic sinusoidal testing and fibril-reinforced poroelastic material modeling of stress-relaxation testing, respectively. We characterized the proteoglycan content using digital densitometry, collagen network organization (orientation angle and anisotropy) using polarized light microscopy and collagen content using Fourier transform infrared spectroscopy. In the superficial cartilage (0–10 % of thickness), the collagen network disorganization and proteoglycan loss were associated with the smaller initial fibril network modulus — a parameter representing the pretension of the collagen network. Furthermore, the proteoglycan loss was associated with the greater strain-dependent fibril network modulus — a measure of nonlinear mechanical behavior. The proteoglycan loss was also associated with greater cartilage viscosity at a low loading frequency (0.005 Hz), while the collagen network disorganization was associated with greater cartilage viscosity at a high loading frequency (1 Hz). Our results suggest that proteoglycan loss and collagen network disorganization reduce the pretension of the collagen network while proteoglycan degradation also increases the nonlinear mechanical behavior of the collagen network. Further, the results also highlight that proteoglycan loss and collagen disorganization increase the viscosity of femoral cartilage, but their contribution to increased viscosity occurs in completely different loading frequencies.

Objective: To investigate the value of ultrasonographic data in predicting total knee replacement (TKR). Design: Data from the Musculoskeletal Pain in Ullensaker study (MUST) was linked to the Norwegian Arthroplasty Register to form a 5–7 year prospective cohort study of 630 persons (69% women, mean (SD) age 64 (8.7) years). We examined the predictive power of ultrasound (US) features, i.e. osteophytes, meniscal extrusion, synovitis in the suprapatellar recess, femoral cartilage thickness, and quality for future knee osteoarthritis (OA) surgery. We investigated 4 main settings for multivariate predictive modeling: 1) clinical predictors (age, sex, body mass index, knee injury, familial OA and workload), 2) radiographic data (assessed by the Kellgren Lawrence grade, KL) with clinical predictors, 3) US features and clinical predictors. Finally, we also considered an ensemble of models 2) and 3) and used it as our fifth model. All models were compared using the Average Precision (AP) and the Area Under Receiver Operating Characteristic Curve (AUC) metrics. Results: Clinical predictors yielded AP of 0.11 (95% confidence interval [CI] 0.05–0.23) and AUC of 0.69 (0.58–0.79). Clinical predictors with KL grade yielded AP of 0.20 (0.12–0.33) and AUC of 0.81 (0.67–0.90). The clinical variables with ultrasound yielded AP of 0.17 (0.08–0.30) and AUC of 0.79 (0.69–0.86). Conclusions: Ultrasonographic examination of the knee may provide added value to basic clinical and demographic descriptors when predicting TKR. While it does not achieve the same predictive performance as radiography, it can provide additional value to the radiographic examination.

Preclassification of raw infrared spectra has often been neglected in scientific literature. Separating spectra of low spectral quality, due to low signal-to-noise ratio, presence of artifacts, and low analyte presence, is crucial for accurate model development. Furthermore, it is very important for sparse data, where it becomes challenging to visually inspect spectra of different natures. Hence, a preclassification approach to separate infrared spectra for sparse data is needed. In this study, we propose a preclassification approach based on Multiplicative Signal Correction (MSC). The MSC approach was applied on human and the bovine knee cartilage broadband Fourier Transform Infrared (FTIR) spectra and on a sparse data subset comprising of only seven wavelengths. The goal of the preclassification was to separate spectra with analyte-rich signals (i.e., cartilage) from spectra with analyte-poor (and high-matrix) signals (i.e., water). The human datasets 1 and 2 contained 814 and 815 spectra, while the bovine dataset contained 396 spectra. A pure water spectrum was used as a reference spectrum in the MSC approach. A threshold for the root mean square error (RMSE) was used to separate cartilage from water spectra for broadband and the sparse spectral data. Additionally, standard noise-to-ratio and principle component analysis were applied on broadband spectra. The fully automated MSC preclassification approach, using water as reference spectrum, performed as well as the manual visual inspection. Moreover, it enabled not only separation of cartilage from water spectra in broadband spectral datasets, but also in sparse datasets where manual visual inspection cannot be applied.

Attenuated total reflection Fourier-transform infrared (ATR-FTIR) spectroscopy coupled with machine learning-based partial least squares discriminant analysis (PLS-DA) was applied to study if severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) could be detected from nasopharyngeal swab samples originally collected for polymerase chain reaction (PCR) analysis. Our retrospective study included 558 positive and 558 negative samples collected from Northern Finland. Overall, we found moderate diagnostic performance for ATR-FTIR when PCR analysis was used as the gold standard: the average area under the receiver operating characteristics curve (AUROC) was 0.67–0.68 (min. 0.65, max. 0.69) with 20, 10 and 5 k-fold cross validations. Mean accuracy, sensitivity and specificity was 0.62–0.63 (min. 0.60, max. 0.65), 0.61 (min. 0.58, max. 0.65) and 0.64 (min. 0.59, max. 0.67) with 20, 10 and 5 k-fold cross validations. As a conclusion, our study with relatively large sample set clearly indicate that measured ATR-FTIR spectrum contains specific information for SARS-CoV-2 infection (P < 0.001 for AUROC in label permutation test). However, the diagnostic performance of ATR-FTIR remained only moderate, potentially due to low concentration of viral particles in the transport medium. Further studies are needed before ATR-FTIR can be recommended for fast screening of SARS-CoV-2 from nasopharyngeal swab samples.

Infotech Oulu Focus Institute promotes research in the University of Oulu’s strategic focus area of “Digitalization and smart society”. Being one of the University of Oulu’s four focus institutes, Infotech Oulu promotes multidisciplinary and internationally significant research, cross-faculty research cooperation, and researcher interaction and community spirit. Infotech Oulu Focus Institute’s periodic report from 2018–2021 summarizes the results of the Institute’s twelve strategic spearhead projects and five emerging projects including their bibliometric analysis. These projects have accomplished 300+ publications. During the four-year project duration, 29 doctoral theses have been finalized and more is expected in 2023. In addition, the report gives an overview of other activities within the focus area, such as events organized during the period.

Injuries to the ligaments of the knee commonly impact vulnerable and physically active individuals. These injuries can lead to the development of degenerative diseases such as post-traumatic osteoarthritis (PTOA). Non-invasive optical modalities, such as infrared and Raman spectroscopy, provide means for quantitative evaluation of knee joint tissues and have been proposed as potential quantitative diagnostic tools for arthroscopy. In this study, we evaluate Raman spectroscopy as a viable tool for estimating functional properties of collateral ligaments. Artificial trauma was induced by anterior cruciate ligament transection (ACLT) in the left or right knee joint of skeletally mature New Zealand rabbits. The corresponding contralateral (CL) samples were extracted from healthy unoperated joints along with a separate group of control (CNTRL) animals. The rabbits were sacrificed at 8 weeks after ACLT. The ligaments were then harvested and measured using Raman spectroscopy. A uniaxial tensile stress-relaxation testing protocol was adopted for determining several biomechanical properties of the samples. Partial least squares (PLS) regression models were then employed to correlate the spectral data with the biomechanical properties. Results show that the capacity of Raman spectroscopy for estimating the biomechanical properties of the ligament samples varies depending on the target property, with prediction error ranging from 15.78% for tissue cross-sectional area to 30.39% for stiffness. The hysteresis under cyclic loading at 2 Hz (RMSE = 6.22%, Normalized RMSE = 22.24%) can be accurately estimated from the Raman data which describes the viscous damping properties of the tissue. We conclude that Raman spectroscopy has the potential for non-destructively estimating ligament biomechanical properties in health and disease, thus enhancing the diagnostic value of optical arthroscopic evaluations of ligament integrity.

Ligaments of the knee provide stability and prevent excessive motions of the joint. Rupture of the anterior cruciate ligament (ACL), a common sports injury, results in an altered loading environment for other tissues in the joint, likely leading to their mechanical adaptation. In the collateral ligaments, the patterns and mechanisms of biomechanical adaptation following ACL transection (ACLT) remain unknown. We aimed to characterize the adaptation of elastic and viscoelastic properties of the lateral and medial collateral ligaments eight weeks after ACLT. Unilateral ACLT was performed in six rabbits, and collateral ligaments were harvested from transected and contralateral knee joints after eight weeks, and from an intact control group (eight knees from four animals). The cross-sectional areas were measured with micro-computed tomography. Stepwise tensile stress-relaxation testing was conducted up to 6% final strain, and the elastic and viscoelastic properties were characterized with a fibril-reinforced poroviscoelastic material model. We found that the cross-sectional area of the collateral ligaments in the ACL transected knees increased, the nonlinear elastic collagen network modulus of the LCL decreased, and the amount of fast relaxation in the MCL decreased. Our results indicate that rupture of the ACL leads to an early adaptation of the elastic and viscoelastic properties of the collagen fibrillar network in the collateral ligaments. These adaptations may be important to consider when evaluating whole knee joint mechanics after ACL rupture, and the results aid in understanding the consequences of ACL rupture on other tissues.

Accurate prediction of knee osteoarthritis (KOA) progression from structural MRI has a potential to enhance disease understanding and support clinical trials. Prior art focused on manually designed imaging biomarkers, which may not fully exploit all disease-related information present in MRI scan. In contrast, our method learns relevant representations from raw data end-to-end using Deep Learning, and uses them for progression prediction. The method employs a 2D CNN to process the data slice-wise and aggregate the extracted features using a Transformer. Evaluated on a large cohort (n=4,866), the proposed method outperforms conventional 2D and 3D CNN-based models and achieves average precision of 0.58 ± 0.03 and ROC AUC of 0.78 ± 0.01. This paper sets a baseline on end-to-end KOA progression prediction from structural MRI. Our code is publicly available at https://github.com/MIPT-Oulu/OAProgressionMR.

In medical applications, deep learning methods are designed to automate diagnostic tasks. However, a clinically relevant question that practitioners usually face, is how to predict the future trajectory of a disease (prognosis). Current methods for such a problem often require domain knowledge, and are complicated to apply. In this paper, we formulate the prognosis prediction problem as a one-to-many forecasting problem from multimodal data. Inspired by a clinical decision-making process with two agents — a radiologist and a general practitioner, we model a prognosis prediction problem with two transformer-based components that share information between each other. The first block in this model aims to analyze the imaging data, and the second block leverages the internal representations of the first one as inputs, also fusing them with auxiliary patient data. We show the effectiveness of our method in predicting the development of structural knee osteoarthritis changes over time. Our results show that the proposed method outperforms the state-of-the-art baselines in terms of various performance metrics. In addition, we empirically show that the existence of the multi-agent transformers with depths of 2 is sufficient to achieve good performances. Our code is publicly available at https://github.com/MIPT-Oulu/CLIMAT.

Objectives: To determine the diagnostic performance of ultrasonography (US) for evaluation of the ankle joint osteoarthritic (OA) changes. Cone-beam computed tomography (CT) was used as the gold standard and US performance was compared with conventional radiography (CR). As a secondary aim, associations between the imaging findings and ankle symptoms were assessed. Methods: US was performed to 51 patients with ankle OA. Every patient had prior ankle CR and underwent cone-beam CT during the same day as US examination. On US, effusion/synovitis, osteophytes, talar cartilage damage, and tenosynovitis were evaluated. Comparison to respective imaging findings on CR and cone-beam CT was then performed. Single radiologist blinded to other modalities assessed all the imaging studies. Symptoms questionnaire, the Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC), was available for 48 patients. Results: US detected effusion/synovitis of the talocrural joint with 45% sensitivity and 90% specificity. For the detection of anterior talocrural osteophytes, US sensitivity was 78% and specificity 79%. For the medial talocrural osteophytes, they were 39 and 83%, and for the lateral talocrural osteophytes 54 and 100%, respectively. Considering cartilage damage of the talus, US yielded a low sensitivity of 18% and high specificity of 97%. Overall, the performance of US was only moderate and comparable to CR. The imaging findings showed only weak associations with ankle symptoms. Conclusions: The ability of US to detect ankle OA is only moderate. Interestingly, performance of CR also remained moderate. The associations between imaging findings and WOMAC score seem to be weak in ankle OA.

The aim of the study was to optimize preprocessing of sparse infrared spectral data. The sparse data were obtained by reducing broadband Fourier transform infrared attenuated total reflectance spectra of bovine and human cartilage, as well as of simulated spectral data, comprising several thousand spectral variables into datasets comprising only seven spectral variables. Different preprocessing approaches were compared, including simple baseline correction and normalization procedures, and model-based preprocessing, such as multiplicative signal correction (MSC). The optimal preprocessing was selected based on the quality of classification models established by partial least squares discriminant analysis for discriminating healthy and damaged cartilage samples. The best results for the sparse data were obtained by preprocessing using a baseline offset correction at 1800 cm⁻¹, followed by peak normalization at 850 cm⁻¹ and preprocessing by MSC.

Finite element (FE) modeling is becoming an increasingly popular method for analyzing knee joint mechanics and biomechanical mechanisms leading to osteoarthritis (OA). The most common and widely available imaging method for knee OA diagnostics is planar X-ray imaging, while more sophisticated imaging methods, e.g., magnetic resonance imaging (MRI) and computed tomography (CT), are seldom used. Hence, the capability to produce accurate biomechanical knee joint models directly from X-ray imaging would bring FE modeling closer to clinical use. Here, we extend our atlas-based framework by generating FE knee models from X-ray images (N = 28). Based on measured anatomical landmarks from X-ray and MRI, knee joint templates were selected from the atlas library. The cartilage stresses and strains of the X-ray-based model were then compared with the MRI-based model during the stance phase of the gait. The biomechanical responses were statistically not different between MRI- vs. X-ray-based models when the template obtained from X-ray imaging was the same as the MRI template. However, if this was not the case, the peak values of biomechanical responses were statistically different between X-ray and MRI models. The developed X-ray-based framework may pave the way for a clinically feasible approach for knee joint FE modeling.

Correction to: Jahangir, S., Mohammadi, A., Mononen, M.E. et al. Rapid X-Ray-Based 3-D Finite Element Modeling of Medial Knee Joint Cartilage Biomechanics During Walking. Ann Biomed Eng 50, 666–679 (2022). https://doi.org/10.1007/s10439-022-02941-0 Rinnakkaistallennettu versio / Self-archived version

Objective: To assess the ability of texture features for detecting radiographic patellofemoral osteoarthritis (PFOA) from knee lateral view radiographs. Design: We used lateral view knee radiographs from The Multicenter Osteoarthritis Study (MOST) public use datasets (n = 5507 knees). Patellar region-of-interest (ROI) was automatically detected using landmark detection tool (BoneFinder), and subsequently, these anatomical landmarks were used to extract three different texture ROIs. Hand-crafted features, based on Local Binary Patterns (LBP), were then extracted to describe the patellar texture. First, a machine learning model (Gradient Boosting Machine) was trained to detect radiographic PFOA from the LBP features. Furthermore, we used end-to-end trained deep convolutional neural networks (CNNs) directly on the texture patches for detecting the PFOA. The proposed classification models were eventually compared with more conventional reference models that use clinical assessments and participant characteristics such as age, sex, body mass index (BMI), the total Western Ontario and McMaster Universities Arthritis Index (WOMAC) score, and tibiofemoral Kellgren-Lawrence (KL) grade. Atlas-guided visual assessment of PFOA status by expert readers provided in the MOST public use datasets was used as a classification outcome for the models. Performance of prediction models was assessed using the area under the receiver operating characteristic curve (ROC AUC), the area under the precision-recall (PR) curve -average precision (AP)-, and Brier score in the stratified 5-fold cross validation setting. Results: Of the 5507 knees, 953 (17.3%) had PFOA. AUC and AP for the strongest reference model including age, sex, BMI, WOMAC score, and tibiofemoral KL grade to predict PFOA were 0.817 and 0.487, respectively. Textural ROI classification using CNN significantly improved the prediction performance (ROC AUC = 0.889, AP = 0.714). Conclusions: We present the first study that

Morphological changes in knee cartilage subregions are valuable imaging-based biomarkers for understanding progression of osteoarthritis, and they are typically detected from magnetic resonance imaging (MRI). So far, accurate segmentation of cartilage has been done manually. Deep learning approaches show high promise in automating the task; however, they lack clinically relevant evaluation. We introduce a fully automatic method for segmentation and subregional assessment of articular cartilage, and evaluate its predictive power in context of radiographic osteoarthritis progression. Two data sets of 3D double-echo steady-state (DESS) MRI derived from the Osteoarthritis Initiative were used: first, n = 88; second, n = 600, 0-/12-/24-month visits. Our method performed deep learning-based segmentation of knee cartilage tissues, their subregional division via multi-atlas registration, and extraction of subregional volume and thickness. The segmentation model was developed and assessed on the first data set. Subsequently, on the second data set, the morphological measurements from our and the prior methods were analyzed in correlation and agreement, and, eventually, by their discriminative power of radiographic osteoarthritis progression over 12 and 24 months, retrospectively. The segmentation model showed very high correlation (r > 0.934) and agreement (mean difference < 116 mm³) in volumetric measurements with the reference segmentations. Comparison of our and manual segmentation methods yielded r = 0.845–0.973 and mean differences = 262–501 mm³ for weight-bearing cartilage volume, and r = 0.770–0.962 and mean differences = 0.513–1.138 mm for subregional cartilage thickness. With regard to osteoarthritis progression, our method found most of the significant associations identified using the manual segmentation method, for both 12- and 24-month subregional cartilage changes. The method may be effectively applied in osteoarthritis progression studies to extra

Osteoporosis significantly increases the risk of bone fracture. Falls are the leading cause of osteoporotic fracture among elderly people. The incidence of low-energy acetabular fracture among the elderly population has been increasing in developed countries in recent decades. Low-energy acetabular fractures mostly happen due to falling from a standing height. While the mechanism of proximal femur fracture due to low-energy falls has been investigated widely owing to higher incidence and the associated morbidities and mortalities, the biomechanical mechanism responsible for acetabular fracture has remained underexplored. Through developing comprehensive finite element models as a reliable and cost-effective method of study, this thesis aims to evaluate the effects of various biomechanical factors contributing to the occurrence, severity, and type of acetabular fracture during low-energy sideways falls. The finite element models were based on abdominal computed tomography (CT) images of a median male without acetabular fracture history, retrieved from a large clinical dataset. While the bony parts and trochanteric soft tissue were reconstructed directly from the CT images, cartilages and ligaments were added manually according to anatomy and literature. The material models able to simulate the behavior of tissues during the impact loading condition were considered. Initial/boundary conditions, loads, contacts, interactions, and controlling parameters were applied according to the literature. By varying the corresponding variables in the model, the effects of body configuration, impact velocity, flooring material, trochanteric soft tissue stiffness, and bone loss on the risk the acetabular fracture were investigated. Results of the current study showed that the effects of body configuration, impact velocity, and bone loss on the risk of acetabular fracture are substantial. However, the effects of flooring material and trochanteric soft tissue stiffness remain, Tiivistelmä Luukato eli osteoporoosi lisää merkittävästi luunmurtuman riskiä. Kaatuminen on tärkein osteoporoottisten murtumien syy ikääntyvillä. Matalaenergisten lonkkamaljan murtumien esiintyvyys on kasvanut kehittyneissä maissa viime vuosikymmeninä. Useimmiten matalaenergiset lonkkamaljan murtumat ovat seurausta kaatumisesta. Kaatumisen aiheuttamien lonkkamurtumien mekamismia on tutkittu runsaasti, koska niitä esiintyy enemmän ja niihin liittyy korkea sairastavuus ja kuolleisuus. Sen sijaan lonkkamaljan biomekaanista mekanismia on toistaiseksi tutkittu vain vähän. Tämän väitöstutkimuksen tavoitteena oli arvioida erilaisten biomekaanisten tekijöiden vaikutusta matalaenergisen kaatumisen aiheuttaman lonkkamaljan murtuman esiintymiseen, vakavuuteen ja murtumatyyppiin, kehittäen siihen soveltuvia luotettavia, kustannustehokkaita elementtimalleja. Elementtimallit perustuivat tietokonetomografiakuviin (TT), jotka oli valikoitu laajasta kliinisestä aineistosta edustamaan sellaisten miesten mediaania, joilla ei ole lonkkamaljan murtumaa. Luut ja pehmytkudos rekonstruoitiin suoraan TT-kuvista. Rustokudos ja nivelsiteet lisättiin manuaalisesti niiden anatomian ja kirjallisuuden perusteella. Materiaalimallit valittiin niin, että niillä voidaan simuloida kudosten käyttäytymistä törmäyskuormitustilanteessa. Alku- ja reunaehdot, kuormitukset, rajakontaktit, vuorovaikutukset ja säätöparametrit valittiin kirjallisuuden perusteella. Kehon asennon, törmäysnopeuden, lattiamateriaalin, pehmytkudoksen jäykkyyden ja luukadon vaikutusta lonkkamaljan murtumariskiin tutkittiin vaihtelemalla kyseisiä muuttujia. Tutkimuksen tulokset osoittivat, että kehon asento, törmäysnopeus ja luukato vaikuttavat oleellisesti lonkkamaljan murtumariskiin. Sen sijaan lattiamateriaalin ja pehmytkudoksen jäykkyyden vaikutus jäi vähäiseksi. Tutkimuksen perusteella törmäyshetkellä vaakasuorassa olevat vartalo ja reisi, yli 3,17 m/s törmäysnopeus sekä merkittävä luukato lisäävät huomattavasti lonkkamaljan murt

Dental caries is the most common oral disease that causes demineralization of the enamel and later of the dentin. Depth-wise assessment of the demineralization process could be used to help in treatment planning. In this study, we aimed to provide baseline information for the development of a Raman probe by characterizing the mineral composition of the dental tissues from large composition maps (6 × 3 mm² with 15 μm step size) using Raman microspectroscopy. Ten human wisdom teeth with different stages of dental caries lesions were examined. All of the teeth were cut in half at representative locations of the caries lesions and then imaged with a Raman imaging microscope. The pre-processed spectral maps were combined into a single data matrix, and the spectra of the enamel, dentin, and caries were identified by K-means cluster analysis. Our results showed that unsupervised identification of dental caries is possible with the K-means clustering. The compositional analysis revealed that the carious lesions are less mineralized than the healthy enamel, and when the lesions extend into the dentin, they are even less mineralized. Furthermore, there were more carbonate imperfections in the mineral crystal lattice of the caries tissues than in healthy tissues. Interestingly, we observed gradients in the sound enamel showing higher mineralization and greater mineral crystal perfection towards the tooth surface. To conclude, our results provide a baseline for the methodological development aimed at clinical diagnostics for the early detection of active caries lesions.

Collagen XIII is a conserved transmembrane collagen mainly expressed in mesenchymal tissues. Previously, we have shown that collagen XIII modulates tissue development and homeostasis. Integrins are a family of receptors that mediate signals from the environment into the cells and vice versa. Integrin α11β1 is a collagen receptor known to recognize the GFOGER (O=hydroxyproline) sequence in collagens. Interestingly, collagen XIII and integrin α11β1 both have a role in the regulation of bone homeostasis. To study whether α11β1 is a receptor for collagen XIII, we utilized C2C12 cells transfected to express α11β1 as their only collagen receptor. The interaction between collagen XIII and integrin α11β1 was also confirmed by surface plasmon resonance and pull-down assays. We discovered that integrin α11β1 mediates cell adhesion to two collagenous motifs, namely GPKGER and GF(S)QGEK, that were shown to act as the recognition sites for the integrin α11-I domain. Furthermore, we studied the in vivo significance of the α11β1-collagen XIII interaction by crossbreeding α11 null mice (Itga11−/−) with mice overexpressing Col13a1 (Col13a1oe). When we evaluated the bone morphology by microcomputed tomography, Col13a1oe mice had a drastic bone overgrowth followed by severe osteoporosis, whereas the double mutant mouse line showed a much milder bone phenotype. To conclude, our data identifies integrin α11β1 as a new collagen XIII receptor and demonstrates that this ligand-receptor pair has a role in the maintenance of bone homeostasis.

Osteoarthritis (OA) degrades articular cartilage and weakens its function. Modern fibril-reinforced poroelastic (FRPE) computational models can distinguish the mechanical properties of main cartilage constituents, namely collagen, proteoglycans, and fluid, thus, they can precisely characterize the complex mechanical behavior of the tissue. However, these properties are not known for human femoral condyle cartilage. Therefore, we aimed to characterize them from human subjects undergoing knee replacement and from deceased donors without known OA. Multi-step stress-relaxation measurements coupled with sample-specific finite element analyses were conducted to obtain the FRPE material properties. Samples were graded using OARSI scoring to determine the severity of histopathological cartilage degradation. The results suggest that alterations in the FRPE properties are not evident in the moderate stages of cartilage degradation (OARSI 2-3) as compared with normal tissue (OARSI 0-1). Drastic deterioration of the FRPE properties was observed in severely degraded cartilage (OARSI 4). We also found that the FRPE properties of femoral condyle cartilage related to the collagen network (initial fibril-network modulus) and proteoglycan matrix (non-fibrillar matrix modulus) were greater compared to tibial and patellar cartilage in OA. These findings may inform cartilage tissue-engineering efforts and help to improve the accuracy of cartilage representations in computational knee joint models.

Structural dynamics of calcified cartilage (CC) are poorly understood. Conventionally, CC structure is analyzed using histological sections. Micro-computed tomography (µCT) allows for three-dimensional (3D) imaging of mineralized tissues; however, the segmentation between bone and mineralized cartilage is challenging. Here, we present state-of-the-art deep learning segmentation for µCT images to assess 3D CC morphology. The sample includes 16 knees from 12 New Zealand White rabbits dissected into osteochondral samples from six anatomical regions: lateral and medial femoral condyles, lateral and medial tibial plateaus, femoral groove, and patella (n = 96). The samples were imaged with µCT and processed for conventional histology. Manually segmented CC from the images was used to train segmentation models with different encoder–decoder architectures. The models with the greatest out-of-fold evaluation Dice score were selected. CC thickness was compared across 24 regions, co-registered between the imaging modalities using Pearson correlation and Bland–Altman analyses. Finally, the anatomical CC thickness variation was assessed via a Linear Mixed Model analysis. The best segmentation models yielded average Dice of 0.891 and 0.807 for histology and µCT segmentation, respectively. The correlation between the co-registered regions was strong (r = 0.897, bias = 21.9 µm, standard deviation = 21.5 µm). Finally, both methods could separate the CC thickness between the patella, femoral, and tibial regions (p < 0.001). As a conclusion, the proposed µCT analysis allows for ex vivo 3D assessment of CC morphology. We demonstrated the biomedical relevance of the method by quantifying CC thickness in different anatomical regions with a varying mean thickness. CC was thickest in the patella and thinnest in the tibial plateau. Our method is relatively straightforward to implement into standard µCT analysis pipelines, allowing the analysis of CC morphology. In future researc

Raman spectroscopy is promising in vivo tool in various biomedical applications; moreover, in recent years, its use for characterizing articular cartilage degeneration has been developing. It has also shown potential for scoring the severity of cartilage lesions, which could be useful in determining the optimal treatment strategy during cartilage repair surgery. However, the effect of different cartilage injury types on Raman spectra is unknown. This study aims to investigate the potential of Raman spectroscopy for detecting changes in cartilage due to different injury types. Artificial injuries were induced in cartilage samples using established mechanical and enzymatic approaches to mimic trauma‐induced and natural degeneration. Mechanical damage was induced using surface abrasion (ABR, n = 12) or impact loading (IMP, n = 12), while enzymatic damage was induced using three different treatments: 30 min trypsin digestion (T30, n = 12), 90 min collagenase digestion (C90, n = 12), and 24 h collagenase digestion (C24, n = 12). Raman spectra were obtained from all specimens, and partial least squares discriminant analysis (PLS‐DA) was used to distinguish cartilage injury types from their respective controls. PLS‐DA cross‐validation accuracies were higher for C24 (88%) and IMP (79%) than for C90 (67%), T30 (63%), and ABR (58%) groups. This study indicates that Raman spectroscopy, combined with multivariate analysis, can discern different cartilage injury types. This knowledge could be useful in clinical decision‐making, for example, selecting the optimal treatment remedy during cartilage repair surgery.

While the incidence of hip fractures has declined during the last decades, the incidence of acetabular fractures resulting from low-energy sideways falls has increased, and the mechanisms responsible for this trend remain unknown. Previous studies have suggested that body configuration during the impact plays an important role in a hip fracture. Thus, the aim of this study was to investigate the effect of body configuration angles (trunk tilt angle, trunk flexion angle, femur horizontal rotation angle, and femur diaphysis angle) on low-energy acetabular fractures via a parametric analysis. A computed tomography–based (CT) finite element model of the ground–proximal femur–pelvis complex was created, and strain magnitude, time-history response, and distribution within the acetabulum were evaluated. Results showed that while the trunk tilt angle and femur diaphysis angle have the greatest effect on strain magnitude, the direction of the fall (lateral vs. posterolateral) contributes to strain distribution within the acetabulum. The results also suggest that strain level and distribution within the proximal femur and acetabulum resulting from a sideways fall are not similar and, in some cases, even opposite. Taken together, our simulations suggest that a more horizontal trunk and femoral shaft at the impact phase can increase the risk of low-energy acetabular fractures.

A low-energy acetabular fracture, as a result of falling from standing height, is common among elderly patients and the number of cases is increasing rapidly in developed countries. Several biomechanical factors contribute to the incidence, severity, and type of acetabular fractures, such as body configuration at the impact moment or bone and soft-tissue quality. The current parametric study developed a comprehensive finite element model of the pelvic girdle and simple representation of the whole body and investigated the effects of impact velocity, conventional indoor/outdoor flooring material, and trochanteric soft-tissue stiffness on an acetabular fracture. Our results show that whereas the impact velocity has a substantial influence on the incidence and type of acetabular fracture, the effects of conventional flooring materials and trochanteric soft-tissue quality are not remarkable. It seems that other factors such as the quality of bone (healthy vs. osteoporotic), the thickness of trochanteric soft-tissue, and body configuration at the impact are more critical in the occurrence and type of the acetabular fracture. These results can be valuable in the prevention of acetabular fractures and the design of protective measures such as hip pads or novel flooring materials.

The incidence of low-energy acetabular fractures has increased. However, the structural factors for these fractures remain unclear. The objective of this study was to extract trabecular bone architecture and proximal femur geometry (PFG) measures from clinical computed tomography (CT) images to (1) identify possible structural risk factors of acetabular fractures, and (2) to discriminate fracture cases from controls using machine learning methods. CT images of 107 acetabular fracture subjects (25 females, 82 males) and 107 age-gender matched controls were examined. Three volumes of interest, one at the acetabulum and two at the femoral head, were extracted to calculate bone volume fraction (BV/TV), gray-level co-occurrence matrix and histogram of the gray values (GV). The PFG was defined by neck shaft angle and femoral neck axis length. Relationships between the variables were assessed by statistical mean comparisons and correlation analyses. Bayesian logistic regression and Elastic net machine learning models were implemented for classification. We found lower BV/TV at the femoral head (0.51 vs. 0.55, p = 0.012) and lower mean GV at both the acetabulum (98.81 vs. 115.33, p < 0.001) and femoral head (150.63 vs. 163.47, p = 0.005) of fracture subjects when compared to their matched controls. The trabeculae within the femoral heads of the acetabular fracture sides differed in structure, density and texture from the corresponding control sides of the fracture subjects. Moreover, the PFG and trabecular architectural variables, alone and in combination, were able to discriminate fracture cases from controls (area under the receiver operating characteristics curve 0.70 to 0.79). In conclusion, lower density in the acetabulum and femoral head with abnormal trabecular structure and texture at the femoral head, appear to be risk factors for low-energy acetabular fractures.

The temporomandibular joint (TMJ) enables complex movements and is mechanically loaded during function. The mandibular condylar cartilage (MCC), which is categorized as fibrocartilage, is different from general hyaline cartilage and plays a crucial role as a stress absorber during function. The MCC is an avascular tissue due to no direct blood supply, and is composed of chondrocytes surrounded by extracellular matrix (ECM) that mainly contains collagens and proteoglycans. TMJ degenerative disease (TMJ-DD) is characterized by degeneration of soft and hard tissues of the TMJ. The etiology of TMJ-DD is multifactorial. Loading, estrogen, and aging have been presented as potential risk factors for TMJ-DD. The aim of the current study was to investigate the effects of loading, estrogen level, and aging on the histomorphology of rat MCC and on the expression of various biomolecules (HIF-1α, VEGF, MMP-3, MMP-8, type I, II, and X collagens) of the condylar cartilage. Loading and aging were found to be associated with histomorphological changes in the MCC, being the main factors for the thickness changes of the anterior and central parts of the MCC. Aging was also the main factor for these changes in the posterior part of the MCC. Increased loading upregulated the expression of HIF-1α, VEGF, MMP-3, as well as the expression of type II and X collagens in the condylar cartilage. Estrogen deficiency was associated with lower expression of MMP-8, type II and X collagens. Aging increased the expression of type I collagen but decreased the expression of MMP-8. The results of this study show that sufficient loading and physiological estrogen level are beneficial for the development and remodeling of the MCC, while aging is likely to increase susceptibility to reduced capacity for adaptation of the MCC., Tiivistelmä Leukanivel mahdollistaa alaleuan monitahoiset liikkeet samalla kun nivel on toiminnan aikana jatkuvasti kuormituksen alaisena. Alaleuan nivelpään rusto, joka luetaan fibroottisiin rustoihin, poikkeaa rakenteeltaan useiden muiden luiden hyaliinirustosta. Nivelpään ruston keskeinen tehtävä on tasata painevaihteluita leukanivelen toiminnan aikana. Nivelpään rustossa ei ole omaa verenkiertoa. Rusto koostuu kondrosyyttisoluista ja niitä ympäröivästä ekstrasellulaarimatriksista, jonka pääasiallisen rakenteet muodostuvat kollageeneistä ja proteoglygaaneista. Leukanivelen rakenteen tuhoutumiseen liittyvissä tautitiloissa sekä nivelen pehmyt- että kovakudosten hajoaminen on luonteenomaista ja tautitilojen etiologia on monisyinen. Riskitekijöiksi on esitetty varsinaisten nivelsairauksien lisäksi nivelen lisääntynyttä kuormitusta, estrogeenitason muutoksia ja korkeaa ikää. Tämän tutkimuksen tarkoituksena oli tutkia kuormituksen, estrogeenitason ja ikääntymisen vaikutuksia nivelpään rustoon sekä useiden biomolekyylien (HIF-1α, VEGF, MMP-3, MMP-8, ja tyypin I, II ja X kollageenit) aktiivisuutta naarasrottien nivelpään rustossa. Kuormituksen ja ikääntymisen havaittiin olevan yhteydessä nivelpään ruston histomorfologisiin muutoksiin ja ne olivat tärkeimpiä tekijöitä, jotka vaikuttivat nivelpään ruston paksuuteen etummaisessa ja keskimmäisessä osassa nivelpään rustoa. Kuormituksen lisäämisen havaittiin lisäävän HIF-1α, VEGF, MMP-3 molekyylien sekä tyyppien II ja X kollageenien immunohistokemiallista aktiivisuutta nivelpään rustossa. Estrogeenin puute oli yhteydessä MMP-8:n ja tyyppien II and X kollageenien alhaisempaan aktiivisuuteen. Ikääntyminen oli yhteydessä tyypin I kollageenin voimakkaampaan immunohistokemialliseen aktiivisuuteen ja MMP-8-aktiivisuuden vähenemiseen. Tämän tutkimuksen tulokset osoittavat, että alaleuan nivelpäähän kohdistuva riittävä kuormitus ja fysiologinen estrogeenitaso ovat tarpeellisia alaleuan nivelpään ruston kehitykselle ja muovautumi

The incidence of low-energy acetabular fractures has increased. However, the structural factors for these fractures remain unclear. The objective of this study was to extract trabecular bone architecture and proximal femur geometry (PFG) measures from clinical computed tomography (CT) images to (1) identify possible structural risk factors of acetabular fractures, and (2) to discriminate fracture cases from controls using machine learning methods. CT images of 107 acetabular fracture subjects (25 females, 82 males) and 107 age-gender matched controls were examined. Three volumes of interest, one at the acetabulum and two at the femoral head, were extracted to calculate bone volume fraction (BV/TV), gray-level co-occurrence matrix and histogram of the gray values (GV). The PFG was defined by neck shaft angle and femoral neck axis length. Relationships between

Knee osteoarthritis (OA) is the most common musculoskeletal disease in the world. In primary healthcare, knee OA is diagnosed using clinical examination and radiographic assessment. Osteoarthritis Research Society International (OARSI) atlas of OA radiographic features allows performing independent assessment of knee osteophytes, joint space narrowing and other knee features. This provides a fine-grained OA severity assessment of the knee, compared to the gold standard and most commonly used Kellgren–Lawrence (KL) composite score. In this study, we developed an automatic method to predict KL and OARSI grades from knee radiographs. Our method is based on Deep Learning and leverages an ensemble of residual networks with 50 layers. We used transfer learning from ImageNet with a fine-tuning on the Osteoarthritis Initiative (OAI) dataset. An independent testing of our model was performed on the Multicenter Osteoarthritis Study (MOST) dataset. Our method yielded Cohen’s kappa coefficients of 0.82 for KL-grade and 0.79, 0.84, 0.94, 0.83, 0.84 and 0.90 for femoral osteophytes, tibial osteophytes and joint space narrowing for lateral and medial compartments, respectively. Furthermore, our method yielded area under the ROC curve of 0.98 and average precision of 0.98 for detecting the presence of radiographic OA, which is better than the current state-of-the-art.

This study evaluates the feasibility of near infrared (NIR) spectroscopy to distinguish between different cartilage injury types associated with post-traumatic osteoarthritis and idiopathic osteoarthritis (OA) induced by mechanical and enzymatic damages. Bovine osteochondral samples (n = 72) were subjected to mechanical (n = 24) and enzymatic (n = 36) damage; NIR spectral measurements were acquired from each sample before and after damage, and from a separate control group (n = 12). Biomechanical measurements were then conducted to determine the functional integrity of the samples. NIR spectral variations resulting from different damage types were investigated and the samples classified using partial least squares discriminant analysis (PLS-DA). Partial least squares regression (PLSR) was then employed to investigate the relationship between the NIR spectra and biomechanical properties of the samples. Results of the study demonstrate that substantial spectral changes occur in the region of 1700–2200 nm due to tissue damages, while differences between enzymatically and mechanically induced damages can be observed mainly in the region of 1780–1810 nm. We conclude that NIR spectroscopy, combined with multivariate analysis, is capable of discriminating between cartilage injuries that mimic idiopathic OA and traumatic injuries based on specific spectral features. This information could be useful in determining the optimal treatment strategy during cartilage repair in arthroscopy.

Osteoarthritis (OA) is the most common musculoskeletal disorder in the world, affecting hand, hip, and knee joints. At the final stage, OA leads to joint replacement, causing an immense burden at the individual and societal levels. Multiple risk factors that can lead to OA are known; however, the etiology of OA and the underlying mechanisms of OA progression are not currently known. OA is currently diagnosed by a clinical examination and, when necessary, confirmed by imaging — a radiographic evaluation. However, these conventional tools are not sensitive to detect the early stages of OA, which makes the development of preventive measures for further disease progression difficult. Therefore, there is a need for other methods that could allow for the early diagnosis of OA. As such, computer vision-based techniques provide quantitative biomarkers that allow for an automatic and systematic assessment of OA severity from images. In recent years, the rapid development of computer vision and machine learning methods have merged into a new field — deep learning (DL). DL allows for one to formulate the problems of computer vision and other fields in a machine learning fashion. In the medical field, DL has made a tremendous impact and allowed to approach for human-level decision-making accuracy in diagnostic and prognostic tasks compared with the traditional computer vision-based methods. The focus of this thesis is on the development of DL-based methods for fully automatic knee OA severity diagnosis and the prediction of its progression. Multiple new methods for localizing the region of interest, landmark localization, knee OA severity assessment, and OA progression prediction are proposed. The results exceeded the state-of-the-art or formed completely new benchmarks for the evaluation of diagnostic and predictive model performance in OA. The main conclusion is that DL yields excellent performance in the diagnostics of OA and in the prediction of its progression. Al, Tiivistelmä Nivelrikko on maailman yleisin käden, lonkan ja polven niveliin vaikuttava liikuntaelinsairaus. Viimekädessä nivelrikko johtaa tekonivelleikkauksiin, aiheuttaen merkittävää rasitetta niin yksilö- kuin yhteiskunnallisella tasolla. Monia nivelrikolle altistavia tekijöitä on jo tunnistettu, mutta kaikkia nivelrikon syitä ja vaikutusmekanismeja nivelrikon etenemisessä ei tunneta. Nivelrikko diagnosoidaan kliinisellä tutkimuksella ja vahvistetaan/varmistetaan tarvittaessa tehtävällä kuvantamistutkimuksella — tekemällä radiografinen arviointi. Nämä perinteiset työkalut eivät kuitenkaan ole riittävän herkkiä nivelrikon varhaisten vaiheiden havaitsemiseen, ja tämä hankaloittaa sairauden kehittymistä ehkäisevien toimenpiteiden kehittämistä. Näistä syistä johtuen tarvitaan muita menetelmiä, jotka mahdollistavat nivelrikon varhaisen diagnosoinnin. Konenäkömenetelmät sellaisenaan tuottavat kvantitatiivisia biologisia indikaattoreita jotka mahdollistavat automaattisen ja järjestelmällisen nivelrikon vakavuusarvion tekemisen kuvamateriaalista. Viime vuosina konenäkö- ja koneoppimismenetelmien nopea kehitys on synnyttänyt uuden syväoppimisen haaran. Syväoppiminen mahdollistaa konenäkö- ja muiden ongelmien määrittelyn koneoppimisongelman tavoin. Verrattuna perinteisiin lääketieteessä käytettyihin tietokonenäkömenetelmiin, syväoppiminen on mahdollistanut ihmisen suorituskykyä lähestyvät toteutukset lääketieteen diagnostisissa ja prognostisissa tehtävissä ja niiden vaikutus alan kehitykselle on ollut merkittävä. Tämän väitöskirja keskittyy kehittämään syväoppimismenetelmiä täysautomaattiseen polven nivelrikon vakavuuden diagnosointiin ja taudin kehittymisen ennustamiseen. Työssä ehdotetaan/esitetään useita uusia menetelmiä kohdealueen paikallistamiseen, maamerkkien paikallistamiseen, polven nivelrikon vakavuuden arviointiin ja nivelrikon etenemisen ennustamiseen. Työn tulokset ylittävät viimeisintä tekniikkaa edustavat ratkaisut tai muodostavat täysin uuden mittarin diagn