Santeri Rytky - Academia.edu (original) (raw)
Papers by Santeri Rytky
Osteoarthritis and Cartilage, 2021
Purpose: Osteoarthritis (OA) affects millions of people worldwide. In hand OA, the thumb base is ... more Purpose: Osteoarthritis (OA) affects millions of people worldwide. In hand OA, the thumb base is the most affected single joint. The reported radiographic prevalence ranges from 0 to 100%, making the true radiographic prevalence unclear. Hence, we conducted a meta-analysis on the prevalence of radiographic thumb base OA. Methods: We performed a search in Embase, Medline Ovid, Web of Science Core Collection, Cochrane Central Register of Trials, and Google Scholar. We included studies of the general population that reported thumb base OA for males and females separately based on a hand radiograph and reported the age of these groups. Using meta-regression, we estimated the odds ratio (OR) of having radiographic thumb base OA for age and sex, while adjusting for within-study correlation. Results: The initial search yielded 4,278 articles; we finally included 16 studies that reported the age-and sex-stratified prevalence of 104 subgroups. The prevalence of radiographic OA for the 50-year-old male and female patients was 5.8% and 7.3%, respectively, while the respective prevalence for 80-year-old male and female patients was 33.1% and 39.0%. We found an OR for having radiographic OA of 1.06 (95%CI[1.055-1.065], p < 0.001) per increasing year of age, and 1.30 (95%CI: 1.05-1.61], p¼0.014) for women.
Journal of Anatomy, 2021
This is an open access article under the terms of the Creative Commons Attribution License, which... more This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited.
PurposeOnly little is known how calcified cartilage (CC) structure changes during exercise, aging... more PurposeOnly little is known how calcified cartilage (CC) structure changes during exercise, aging and disease. CC thickness (CC.Th) can be analyzed using conventional histological sections. Micro-computed tomography (μCT) allows for three-dimensional (3D) imaging of mineralized tissues, however, the segmentation between bone and CC is challenging. Here, we present state-of-the-art deep learning segmentation for μCT images to enable assessment of CC morphology.MethodsSixteen knees from twelve New Zealand White rabbits were dissected into osteochondral samples from six anatomical regions: lateral and medial femoral condyles, lateral and medial tibial plateaus, femoral groove and patella (n = 96). Samples were imaged with μCT and processed for conventional histology. Manually segmented CC from the histology and reconstructed μCT images was used as the gold standard to train segmentation models with different encoder-decoder architectures. The models with the greatest out-of-fold evalua...
Osteoarthritis and Cartilage, 2020
quantitative semi-automatic assessments of cartilage morphology by Chondrometrics GmbH (CHM). The... more quantitative semi-automatic assessments of cartilage morphology by Chondrometrics GmbH (CHM). These measurements were done for 16 subregions proposed by Wirth et al. (interior, central, and exterior subregions of central medial and central lateral femoral cartilage-[i,c,e] cMF, [i,c,e]cLF; anterior, posterior, interior, exterior, and central subregions of medial and lateral tibial cartilage-[a,p,i,e,c]MT, [a,p,i,e,c]LT). We approached the problem of sub-regional assessment via DL-based segmentation and multi-atlas-based registration (Figure 1). Firstly, a DL-based model was used to segment all the cartilage tissues from IMO. We employed a pre-trained VGG19 encoder in an adapted UNet architecture. This model was developed on the training subset of IMO in a 5fold cross-validation setting. Secondly, we applied rigid multi-atlas registration to divide the segmented tissues into the subregions. We used five scans from IMO to build the multi-atlas and annotated them according to the CHM sub-regions. The segmented tissues were divided voxel-wise based on the proximity to the multi-atlas subregions. Finally, the following sub-regional morphological features were measured: cartilage volume via voxel-wise integration (.VC) and average cartilage thickness (.ThC) via local thickness algorithm. We evaluated our segmentation on the test set of IMO. Here, correlation (r 2) and agreement (Bland-Altman analysis) with the reference segmentations were computed for sub-regional VCs and ThCs. On FBC data, we performed the same comparison between the results of our method and the available CHM measurements. Finally, both methods were analyzed in terms of discriminative power of radiographic OA progression defined by odds ratio (OR). For each subject, we calculated differences (baseline-24-month) in sub-regional VC and ThC features. Then, for each method, logistic regression (LR) models were fit to discriminate between two subject groups based on a single differential feature, yielding the final ORs. The described LR protocol was used in two analyses: "NP versus RP", "NPþPP versus RPþRPP".
Annals of Biomedical Engineering, 2019
The aim of this study was to quantify sub-resolution trabecular bone morphometrics, which are als... more The aim of this study was to quantify sub-resolution trabecular bone morphometrics, which are also related to osteoarthritis (OA), from clinical resolution cone beam computed tomography (CBCT). Samples (n = 53) were harvested from human tibiae (N = 4) and femora (N = 7). Grey-level co-occurrence matrix (GLCM) texture and histogram-based parameters were calculated from CBCT imaged trabecular bone data, and compared with the morphometric parameters quantified from micro-computed tomography. As a reference for OA severity, histological sections were subjected to OARSI histopathological grading. GLCM and histogram parameters were correlated to bone morphometrics and OARSI individually. Furthermore, a statistical model of combined GLCM/histogram parameters was generated to estimate the bone morphometrics. Several individual histogram and GLCM parameters had strong associations with various bone morphometrics (|r| > 0.7). The most prominent correlation was observed between the histogra...
ObjectiveTo develop and validate a machine learning (ML) approach for automatic three-dimensional... more ObjectiveTo develop and validate a machine learning (ML) approach for automatic three-dimensional (3D) histopathological grading of osteochondral samples imaged with contrast-enhanced micro-computed tomography (CEμCT).DesignOsteochondral cores from 24 total knee arthroplasty patients and 2 asymptomatic cadavers (n = 34, Ø = 2 mm; n = 45, Ø = 4 mm) were imaged using CEμCT with phosphotungstic acid-staining. Volumes-of-interest (VOI) in surface (SZ), deep (DZ) and calcified (CZ) zones were extracted depthwise and subjected to dimensionally reduced Local Binary Pattern-textural feature analysis. Regularized Ridge and Logistic regression (LR) models were trained zone-wise against the manually assessed semi-quantitative histopathological CEμCT grades (Ø = 2 mm samples). Models were validated using nested leave-one-out cross-validation and an independent test set (Ø = 4 mm samples). The performance was assessed using Spearman’s correlation, Average Precision (AP) and Area under the Receiv...
Osteoarthritis and Cartilage, 2021
Purpose: Osteoarthritis (OA) affects millions of people worldwide. In hand OA, the thumb base is ... more Purpose: Osteoarthritis (OA) affects millions of people worldwide. In hand OA, the thumb base is the most affected single joint. The reported radiographic prevalence ranges from 0 to 100%, making the true radiographic prevalence unclear. Hence, we conducted a meta-analysis on the prevalence of radiographic thumb base OA. Methods: We performed a search in Embase, Medline Ovid, Web of Science Core Collection, Cochrane Central Register of Trials, and Google Scholar. We included studies of the general population that reported thumb base OA for males and females separately based on a hand radiograph and reported the age of these groups. Using meta-regression, we estimated the odds ratio (OR) of having radiographic thumb base OA for age and sex, while adjusting for within-study correlation. Results: The initial search yielded 4,278 articles; we finally included 16 studies that reported the age-and sex-stratified prevalence of 104 subgroups. The prevalence of radiographic OA for the 50-year-old male and female patients was 5.8% and 7.3%, respectively, while the respective prevalence for 80-year-old male and female patients was 33.1% and 39.0%. We found an OR for having radiographic OA of 1.06 (95%CI[1.055-1.065], p < 0.001) per increasing year of age, and 1.30 (95%CI: 1.05-1.61], p¼0.014) for women.
Journal of Anatomy, 2021
This is an open access article under the terms of the Creative Commons Attribution License, which... more This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited.
PurposeOnly little is known how calcified cartilage (CC) structure changes during exercise, aging... more PurposeOnly little is known how calcified cartilage (CC) structure changes during exercise, aging and disease. CC thickness (CC.Th) can be analyzed using conventional histological sections. Micro-computed tomography (μCT) allows for three-dimensional (3D) imaging of mineralized tissues, however, the segmentation between bone and CC is challenging. Here, we present state-of-the-art deep learning segmentation for μCT images to enable assessment of CC morphology.MethodsSixteen knees from twelve New Zealand White rabbits were dissected into osteochondral samples from six anatomical regions: lateral and medial femoral condyles, lateral and medial tibial plateaus, femoral groove and patella (n = 96). Samples were imaged with μCT and processed for conventional histology. Manually segmented CC from the histology and reconstructed μCT images was used as the gold standard to train segmentation models with different encoder-decoder architectures. The models with the greatest out-of-fold evalua...
Osteoarthritis and Cartilage, 2020
quantitative semi-automatic assessments of cartilage morphology by Chondrometrics GmbH (CHM). The... more quantitative semi-automatic assessments of cartilage morphology by Chondrometrics GmbH (CHM). These measurements were done for 16 subregions proposed by Wirth et al. (interior, central, and exterior subregions of central medial and central lateral femoral cartilage-[i,c,e] cMF, [i,c,e]cLF; anterior, posterior, interior, exterior, and central subregions of medial and lateral tibial cartilage-[a,p,i,e,c]MT, [a,p,i,e,c]LT). We approached the problem of sub-regional assessment via DL-based segmentation and multi-atlas-based registration (Figure 1). Firstly, a DL-based model was used to segment all the cartilage tissues from IMO. We employed a pre-trained VGG19 encoder in an adapted UNet architecture. This model was developed on the training subset of IMO in a 5fold cross-validation setting. Secondly, we applied rigid multi-atlas registration to divide the segmented tissues into the subregions. We used five scans from IMO to build the multi-atlas and annotated them according to the CHM sub-regions. The segmented tissues were divided voxel-wise based on the proximity to the multi-atlas subregions. Finally, the following sub-regional morphological features were measured: cartilage volume via voxel-wise integration (.VC) and average cartilage thickness (.ThC) via local thickness algorithm. We evaluated our segmentation on the test set of IMO. Here, correlation (r 2) and agreement (Bland-Altman analysis) with the reference segmentations were computed for sub-regional VCs and ThCs. On FBC data, we performed the same comparison between the results of our method and the available CHM measurements. Finally, both methods were analyzed in terms of discriminative power of radiographic OA progression defined by odds ratio (OR). For each subject, we calculated differences (baseline-24-month) in sub-regional VC and ThC features. Then, for each method, logistic regression (LR) models were fit to discriminate between two subject groups based on a single differential feature, yielding the final ORs. The described LR protocol was used in two analyses: "NP versus RP", "NPþPP versus RPþRPP".
Annals of Biomedical Engineering, 2019
The aim of this study was to quantify sub-resolution trabecular bone morphometrics, which are als... more The aim of this study was to quantify sub-resolution trabecular bone morphometrics, which are also related to osteoarthritis (OA), from clinical resolution cone beam computed tomography (CBCT). Samples (n = 53) were harvested from human tibiae (N = 4) and femora (N = 7). Grey-level co-occurrence matrix (GLCM) texture and histogram-based parameters were calculated from CBCT imaged trabecular bone data, and compared with the morphometric parameters quantified from micro-computed tomography. As a reference for OA severity, histological sections were subjected to OARSI histopathological grading. GLCM and histogram parameters were correlated to bone morphometrics and OARSI individually. Furthermore, a statistical model of combined GLCM/histogram parameters was generated to estimate the bone morphometrics. Several individual histogram and GLCM parameters had strong associations with various bone morphometrics (|r| > 0.7). The most prominent correlation was observed between the histogra...
ObjectiveTo develop and validate a machine learning (ML) approach for automatic three-dimensional... more ObjectiveTo develop and validate a machine learning (ML) approach for automatic three-dimensional (3D) histopathological grading of osteochondral samples imaged with contrast-enhanced micro-computed tomography (CEμCT).DesignOsteochondral cores from 24 total knee arthroplasty patients and 2 asymptomatic cadavers (n = 34, Ø = 2 mm; n = 45, Ø = 4 mm) were imaged using CEμCT with phosphotungstic acid-staining. Volumes-of-interest (VOI) in surface (SZ), deep (DZ) and calcified (CZ) zones were extracted depthwise and subjected to dimensionally reduced Local Binary Pattern-textural feature analysis. Regularized Ridge and Logistic regression (LR) models were trained zone-wise against the manually assessed semi-quantitative histopathological CEμCT grades (Ø = 2 mm samples). Models were validated using nested leave-one-out cross-validation and an independent test set (Ø = 4 mm samples). The performance was assessed using Spearman’s correlation, Average Precision (AP) and Area under the Receiv...