Pathology diagnostics relies on the assessment of morphology by trained experts, which remains subjective and qualitative. Here we developed a framework for large-scale histomorphometry (FLASH) performing deep learning-based semantic segmentation and subsequent large-scale extraction of interpretable, quantitative, morphometric features in non-tumour kidney histology. We use two internal and three external, multi-centre cohorts to analyse over 1000 kidney biopsies and nephrectomies. By associating morphometric features with clinical parameters, we confirm previous concepts and reveal unexpected relations. We show that the extracted features are independent predictors of long-term clinical outcomes in IgA-nephropathy. We introduce single-structure morphometric analysis by applying techniques from single-cell transcriptomics, identifying distinct glomerular populations and morphometric phenotypes along a trajectory of disease progression. Our study provides a concept for Next-generation Morphometry (NGM), enabling comprehensive quantitative pathology data mining, i.e., pathomics.
- MeSH
- Kidney Glomerulus * pathology MeSH
- Kidney * pathology MeSH
- Publication type
- Journal Article MeSH
- Research Support, Non-U.S. Gov't MeSH
The anatomy of the biceps brachii muscle has been a subject of interest to many researchers. In particular, the presence of one or more accessory heads has been reported to be the most common variation of the biceps brachii muscle. In fact, contemporary knowledge is quite inconsistent and lacks a definitive summary. Taking this into account, the present study aims to investigate the overall prevalence of the accessory heads and related questions, such as their broad morphological features, population variance, sexual dimorphism, side distribution or the changes in prevalence rates over time. A literature search of major scientific databases was conducted and produced 78 eligible articles, including 10,603 upper limbs, for our study. Relevant data were extracted and consequently analysed with the use of random-effects meta-analysis. As a result, the accessory heads occur with an overall prevalence of 9.6% (95% CI 8-11%) and by far the most common is the presence of a single accessory head (8.4%; 95% CI 7-10%). Additional sub-analyses revealed that accessory heads appear more frequently unilaterally and in males. Differences between the occurrence on the right or left side were not significant. Moreover, a decreasing trend in prevalence rates over time was observed, pointing towards an evolutionary adaptation. The innervation and blood supply of the accessory heads are nearly identical to that of the normal biceps brachii muscle. Although the accessory heads are usually asymptomatic, their potential presence must be kept in mind while interpreting various conditions. In addition, their direct clinical importance is speculated and is thought that they might be conductive to several pathological processes around the shoulder girdle and brachial region. While many of the morphological and morphometric aspects of the accessory heads are well documented, their functional value is still a matter for future investigations.
- MeSH
- Upper Extremity MeSH
- Muscle, Skeletal * anatomy & histology MeSH
- Humans MeSH
- Organogenesis MeSH
- Arm * MeSH
- Check Tag
- Humans MeSH
- Male MeSH
- Publication type
- Journal Article MeSH
- Meta-Analysis MeSH
- Systematic Review MeSH
This work explores the design and implementation of an algorithm for the classification of magnetic resonance imaging data for computer-aided diagnosis of schizophrenia. Features for classification were first extracted using two morphometric methods: voxel-based morphometry (VBM) and deformation-based morphometry (DBM). These features were then transformed into a wavelet domain using the discrete wavelet transform with various numbers of decomposition levels. The number of features was then reduced by thresholding and subsequent selection by: Fisher's Discrimination Ratio (FDR), Bhattacharyya Distance, and Variances (Var.). A Support Vector Machine with a linear kernel was used for classification. The evaluation strategy was based on leave-one-out cross-validation.
- MeSH
- Humans MeSH
- Magnetic Resonance Imaging MeSH
- Schizophrenia * MeSH
- Support Vector Machine MeSH
- Wavelet Analysis MeSH
- Check Tag
- Humans MeSH
- Publication type
- Journal Article MeSH
This paper introduces an automated method for estimating sex from cranial sex diagnostic traits by extracting and evaluating specialized morphometric features from the glabella, the supraorbital ridge, the occipital protuberance, and the mastoid process. The proposed method was developed and evaluated using two European population samples, a Czech sample comprising 170 crania reconstructed from anonymized CT scans and a Greek sample of 156 crania from the Athens Collection. It is based on a fully automatic algorithm applied on 3D models for extracting sex diagnostic morphometric features which are further processed by computer vision and machine learning algorithms. Classification accuracy was evaluated in a population specific and a population generic 2-way cross-validation scheme. Population-specific accuracy for individual morphometric features ranged from 78.5 to 96.7%, whereas population generic correct classification ranged from 71.7 to 90.8%. Combining all sex diagnostic traits in multi-feature sex estimation yielded correct classification performance in excess of 91% for the entire sample, whereas the sex of about three fourths of the sample could be determined with 100% accuracy according to posterior probability estimates. The proposed method provides an efficient and reliable way to estimate sex from cranial remains, and it offers significant advantages over existing methods. The proposed method can be readily implemented with the skullanalyzer computer program and the estimate_sex.m GNU Octave function, which are freely available under a suitable license.
- MeSH
- Algorithms * MeSH
- Adult MeSH
- Cephalometry * MeSH
- Skull anatomy & histology MeSH
- Middle Aged MeSH
- Humans MeSH
- Sex Characteristics MeSH
- Aged, 80 and over MeSH
- Aged MeSH
- Software MeSH
- Forensic Anthropology methods MeSH
- Sex Determination by Skeleton methods MeSH
- Imaging, Three-Dimensional MeSH
- Check Tag
- Adult MeSH
- Middle Aged MeSH
- Humans MeSH
- Male MeSH
- Aged, 80 and over MeSH
- Aged MeSH
- Female MeSH
- Publication type
- Journal Article MeSH
- Geographicals
- Czech Republic MeSH
- Greece MeSH
Correlations between facial bony structures and soft facial features are fundamental for facial approximation methods The purpose of this study was to assess the strength of the association between craniofacial shape and the shape of the soft-tissue profile and to determine the extent to which it might be possible to predict the latter from the former. Soft-tissue and skeletal facial profile curves were extracted from 86 lateral head cephalograms of a recent Central European population (52 males and 34 females, aged between 19 and 43 years), divided into five parts, segmented automatically and evaluated using geometric morphometrics. The magnitude of the hard-soft shape association was assessed by principal component analysis and subsequent multiple linear regression (Halazonetis, 2007), by partial least square analysis (PLS) (Rohlf and Corti, 2000) and the RV coefficient (Klingenberg, 2009). The greatest amount of association between the skeletal contour and overlying soft tissues was exhibited by the region of the forehead (predictive power: 95.1%, RV=0.91, correlation for PLS1 r=0,96), followed by the region of the nasal root (predictive power: 40.2%, RV=0.42, rPLS1=0.72) and the lower lip and chin (predictive power: 37.3%, RV=0.41, rPLS1=0.65). The smallest statistically significant covariation was displayed by the upper lip and the maxilla (predictive power: 9.6%, RV=0.14, rPLS1=0.43). The shape covariation between the nasal bridge and the tip and lateral border of the nasal aperture was found to be statistically insignificant (predictive power: 5.8%, RV=0.05, rPLS1=0.26). Shape covariation was visualized and described by thin-plate spine grids. These findings correspond with the observation that the shape of the nasal profile and the upper lip contour are difficult to reconstruct or predict reliably in facial approximations. It seems that the shape of soft tissues might not follow underlying structures as closely as expected.
- MeSH
- Principal Component Analysis MeSH
- Anatomic Landmarks MeSH
- Adult MeSH
- Cephalometry methods MeSH
- Skull anatomy & histology diagnostic imaging MeSH
- Humans MeSH
- Linear Models MeSH
- Least-Squares Analysis MeSH
- Young Adult MeSH
- Face anatomy & histology MeSH
- Radiography MeSH
- Software * MeSH
- Forensic Anthropology MeSH
- Check Tag
- Adult MeSH
- Humans MeSH
- Young Adult MeSH
- Male MeSH
- Female MeSH
- Publication type
- Journal Article MeSH
The process of manual species identification is a daunting task, so much so that the number of taxonomists is seen to be declining. In order to assist taxonomists, many methods and algorithms have been proposed to develop semi-automated and fully automated systems for species identification. While semi-automated tools would require manual intervention by a domain expert, fully automated tools are assumed to be not as reliable as manual or semiautomated identification tools. Hence, in this study we investigate the accuracy of fully automated and semi-automated models for species identification. We have built fully automated and semi-automated species classification models using the monogenean species image dataset. With respect to monogeneans' morphology, they are differentiated based on the morphological characteristics of haptoral bars, anchors, marginal hooks and reproductive organs (male and female copulatory organs). Landmarks (in the semi-automated model) and shape morphometric features (in the fully automated model) were extracted from four monogenean species images, which were then classified using k-nearest neighbour and artificial neural network. In semi-automated models, a classification accuracy of 96.67 % was obtained using the k-nearest neighbour and 97.5 % using the artificial neural network, whereas in fully automated models, a classification accuracy of 90 % was obtained using the k-nearest neighbour and 98.8 % using the artificial neural network. As for the crossvalidation, semi-automated models performed at 91.2 %, whereas fully automated models performed slightly higher at 93.75 %.
