Automatic recognition of different tissue types in histological images is an essential part in the digital pathology toolbox. Texture analysis is commonly used to address this problem; mainly in the context of estimating the tumour/stroma ratio on histological samples. However, although histological images typically contain more than two tissue types, only few studies have addressed the multi-class problem. For colorectal cancer, one of the most prevalent tumour types, there are in fact no published results on multiclass texture separation. In this paper we present a new dataset of 5,000 histological images of human colorectal cancer including eight different types of tissue. We used this set to assess the classification performance of a wide range of texture descriptors and classifiers. As a result, we found an optimal classification strategy that markedly outperformed traditional methods, improving the state of the art for tumour-stroma separation from 96.9% to 98.6% accuracy and setting a new standard for multiclass tissue separation (87.4% accuracy for eight classes). We make our dataset of histological images publicly available under a Creative Commons license and encourage other researchers to use it as a benchmark for their studies.
BackgroundMyocardial T1-mapping recently emerged as a promising quantitative method for non-invasive tissue characterization in numerous cardiomyopathies. Commonly performed with an inversion-recovery (IR) magnetization preparation at 1.5T, the application at 3T has gained due to increased quantification precision. Alternatively, saturation-recovery (SR) T1-mapping has recently been introduced at 1.5T for improved accuracy.Thus, the purpose of this study is to investigate the robustness and precision of SR T1-mapping at 3T and to establish accurate reference values for native T1-times and extracellular volume fraction (ECV) of healthy myocardium.MethodsBalanced Steady-State Free-Precession (bSSFP) Saturation-Pulse Prepared Heart-rate independent Inversion-REcovery (SAPPHIRE) and Saturation-recovery Single-SHot Acquisition (SASHA) T1-mapping were compared with the Modified Look-Locker inversion recovery (MOLLI) sequence at 3T. Accuracy and precision were studied in phantom. Native and post-contrast T1-times and regional ECV were determined in 20 healthy subjects (10 men, 27 ± 5 years). Subjective image quality, susceptibility artifact rating, in-vivo precision and reproducibility were analyzed.ResultsSR T1-mapping showed <4 % deviation from the spin-echo reference in phantom in the range of T1 = 100–2300 ms. The average quality and artifact scores of the T1-mapping methods were: MOLLI:3.4/3.6, SAPPHIRE:3.1/3.4, SASHA:2.9/3.2; (1: poor - 4: excellent/1: strong - 4: none). SAPPHIRE and SASHA yielded significantly higher T1-times (SAPPHIRE: 1578 ± 42 ms, SASHA: 1523 ± 46 ms), in-vivo T1-time variation (SAPPHIRE: 60.1 ± 8.7 ms, SASHA: 70.0 ± 9.3 ms) and lower ECV-values (SAPPHIRE: 0.20 ± 0.02, SASHA: 0.21 ± 0.03) compared with MOLLI (T1: 1181 ± 47 ms, ECV: 0.26 ± 0.03, Precision: 53.7 ± 8.1 ms). No significant difference was found in the inter-subject variability of T1-times or ECV-values (T1: p = 0.90, ECV: p = 0.78), the observer agreement (inter: p > 0.19; intra: p > 0.09) or consistency (inter: p > 0.07; intra: p > 0.17) between the three methods.ConclusionsSaturation-recovery T1-mapping at 3T yields higher accuracy, comparable inter-subject, inter- and intra-observer variability and less than 30 % precision-loss compared to MOLLI.Electronic supplementary materialThe online version of this article (doi:10.1186/s12968-016-0302-x) contains supplementary material, which is available to authorized users.
To develop a generic Open Source MRI perfusion analysis tool for quantitative parameter mapping to be used in a clinical workflow and methods for quality management of perfusion data. We implemented a classic, pixel-by-pixel deconvolution approach to quantify T1-weighted contrastenhanced dynamic MR imaging (DCE-MRI) perfusion data as an OsiriX plug-in. It features parallel computing capabilities and an automated reporting scheme for quality management. Furthermore, by our implementation design, it could be easily extendable to other perfusion algorithms. Obtained results are saved as DICOM objects and directly added to the patient study. The plug-in was evaluated on ten MR perfusion data sets of the prostate and a calibration data set by comparing obtained parametric maps (plasma flow, volume of distribution, and mean transit time) to a widely used reference implementation in IDL. For all data, parametric maps could be calculated and the plug-in worked correctly and stable. On average, a deviation of 0.032±0.02 ml/100 ml/ min for the plasma flow, 0.004±0.0007 ml/100 ml for the volume of distribution, and 0.037±0.03 s for the mean transit time between our implementation and a reference implementation was observed. By using computer hardware with eight CPU cores, calculation time could be reduced by a factor of 2.5. We developed successfully an Open Source OsiriX plugin for T1-DCE-MRI perfusion analysis in a routine quality managed clinical environment. Using model-free deconvolution, it allows for perfusion analysis in various clinical applications. By our plug-in, information about measured physiological processes can be obtained and transferred into clinical practice.
Machine learning by means of SVM in combination with whole-tumor rCBV histogram analysis can be used to identify early patient survival in aggressive gliomas. The SVM model returned higher diagnostic accuracy values than an expert reader, and the model appears to be insensitive to patient, observer, and institutional variations.
ObjectivesTo establish arterial spin labelling (ASL) for quantitative renal perfusion measurements in a rat model at 3 Tesla and to test the diagnostic significance of ASL and dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) in a model of acute kidney injury (AKI).Material and MethodsASL and DCE-MRI were consecutively employed on six Lewis rats, five of which had a unilateral ischaemic AKI. All measurements in this study were performed on a 3 Tesla MR scanner using a FAIR True-FISP approach and a TWIST sequence for ASL and DCE-MRI, respectively. Perfusion maps were calculated for both methods and the cortical perfusion of healthy and diseased kidneys was inter- and intramethodically compared using a region-of-interest based analysis.Results/SignificanceBoth methods produce significantly different values for the healthy and the diseased kidneys (P<0.01). The mean difference was 147±47 ml/100 g/min and 141±46 ml/100 g/min for ASL and DCE-MRI, respectively. ASL measurements yielded a mean cortical perfusion of 416±124 ml/100 g/min for the healthy and 316±102 ml/100 g/min for the diseased kidneys. The DCE-MRI values were systematically higher and the mean cortical renal blood flow (RBF) was found to be 542±85 ml/100 g/min (healthy) and 407±119 ml/100 g/min (AKI).ConclusionBoth methods are equally able to detect abnormal perfusion in diseased (AKI) kidneys. This shows that ASL is a capable alternative to DCE-MRI regarding the detection of abnormal renal blood flow. Regarding absolute perfusion values, nontrivial differences and variations remain when comparing the two methods.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.