Effective Normalization Method Research Articles

Controllable multi-domain semantic artwork synthesis

We present a novel framework for the multi-domain synthesis of artworks from semantic layouts. One of the main limitations of this challenging task is the lack of publicly available segmentation datasets for art synthesis. To address this problem, we propose a dataset called ArtSem that contains 40,000 images of artwork from four different domains, with their corresponding semantic label maps. We first extracted semantic maps from landscape photography and used a conditional generative adversarial network (GAN)-based approach for generating high-quality artwork from semantic maps without requiring paired training data. Furthermore, we propose an artwork-synthesis model using domain-dependent variational encoders for high-quality multi-domain synthesis. Subsequently, the model was improved and complemented with a simple but effective normalization method based on jointly normalizing semantics and style, which we call spatially style-adaptive normalization (SSTAN). Compared to the previous methods, which only take semantic layout as the input, our model jointly learns style and semantic information representation, improving the generation quality of artistic images. These results indicate that our model learned to separate the domains in the latent space. Thus, we can perform fine-grained control of the synthesized artwork by identifying hyperplanes that separate the different domains. Moreover, by combining the proposed dataset and approach, we generated user-controllable artworks of higher quality than that of existing approaches, as corroborated by quantitative metrics and a user study.

AmpNorm: an effective style normalization method for single domain generalization

AmpNorm: an effective style normalization method for single domain generalization

Stain normalization methods for histopathology image analysis: A comprehensive review and experimental comparison

The advent of whole slide imaging has brought advanced computer-aided diagnosis via medical imaging and artificial intelligence technologies in digital pathology. The examination of tissue samples through whole slide imaging is commonly used to diagnose cancerous diseases, but the analysis of histopathology images through a decision support system is not always accurate due to variations in color caused by different scanning equipment, staining methods, and tissue reactivity. These variabilities decrease the accuracy of computer-aided diagnosis and affect the diagnosis of pathologists. In this context, an effective stain normalization method has proved as a powerful tool to standardize different color appearances and minimize color variations in histopathology images. This study reviews different stain normalization methods highlighting the main methodologies, contributions, advantages, and limitations of correlated works. The state-of-the-art methods are grouped into four distinct categories. Next, we select ten representative methods from the groups and conduct an experimental comparison to investigate the strengths and weaknesses of different methods and rank them according to selected performance accuracy measures. The quality performances of selected methods are compared in terms of quaternion structure similarity index metric, structural similarity index metric, and Pearson correlation coefficient conducting experiments on three histopathological image datasets. Our findings conclude that the structure-preserving unified transformation-based methods consistently outperform the state-of-the-art methods by improving robustness against variability and reproducibility. The comparative analysis we conducted in this paper will serve as the basis for future research, which will help to refine existing techniques and develop new approaches to address the complexities of stain normalization in complex histopathology images.

Better Generalized Few-Shot Learning Even without Base Data

This paper introduces and studies zero-base generalized few-shot learning (zero-base GFSL), which is an extreme yet practical version of few-shot learning problem. Motivated by the cases where base data is not available due to privacy or ethical issues, the goal of zero-base GFSL is to newly incorporate the knowledge of few samples of novel classes into a pretrained model without any samples of base classes. According to our analysis, we discover the fact that both mean and variance of the weight distribution of novel classes are not properly established, compared to those of base classes. The existing GFSL methods attempt to make the weight norms balanced, which we find help only the variance part, but discard the importance of mean of weights particularly for novel classes, leading to the limited performance in the GFSL problem even with base data. In this paper, we overcome this limitation by proposing a simple yet effective normalization method that can effectively control both mean and variance of the weight distribution of novel classes without using any base samples and thereby achieve a satisfactory performance on both novel and base classes. Our experimental results somewhat surprisingly show that the proposed zero-base GFSL method that does not utilize any base samples even outperforms the existing GFSL methods that make the best use of base data. Our implementation is available at: https://github.com/bigdata-inha/Zero-Base-GFSL.

Quantifying airway remodelling for research or clinical purposes: How should we normalize for airway size?

Quantifying airway remodelling for research or clinical purposes: How should we normalize for airway size?

Nailfold Microhemorrhage Segmentation with Modified U-Shape Convolutional Neural Network

Nailfold capillaroscopy is a reliable way to detect and analyze microvascular abnormalities. It is safe, simple, noninvasive, and inexpensive. Among all the capillaroscopic abnormalities, nailfold microhemorrhages are closely associated with early vascular damages and might be present in numerous diseases such as glaucoma, diabetes mellitus, and systemic sclerosis. Segmentation of nailfold microhemorrhages provides valuable pathological information that may lead to further investigations. A novel deep learning architecture named DAFM-Net is proposed for the accurate segmentation in this study. The network mainly consists of U-shape backbone, dual attention fusion module, and group normalization layer. The U-shape backbone generates rich hierarchical representations while the dual attention fusion module utilizes the captured features for fine adjustment. Group normalization is introduced as an effective normalization method to effectively improve the convergence ability of our deep neural network. The effectiveness of the proposed model is validated through ablation studies and segmentation experiments; the proposed method DAFM-Net achieves competitive performance for nailfold microhemorrhage segmentation with an IOU score of 78.03% and Dice score of 87.34% compared to the ground truth.

Statistical Assessment of Depth Normalization for Small RNA Sequencing.

PURPOSEMethods for depth normalization have been assessed primarily with simulated data or cell-line–mixture data. There is a pressing need for benchmark data enabling a more realistic and objective assessment, especially in the context of small RNA sequencing.METHODSWe collected a unique pair of microRNA sequencing data sets for the same set of tumor samples; one data set was collected with and the other without uniform handling and balanced design. The former provided a benchmark for evaluating evidence of differential expression and the latter served as a test bed for normalization. Next, we developed a data perturbation algorithm to simulate additional data set pairs. Last, we assembled a set of computational tools to visualize and quantify the assessment.RESULTSWe validated the quality of the benchmark data and showed the need for normalization of the test data. For illustration, we applied the data and tools to assess the performance of 9 existing normalization methods. Among them, trimmed mean of M-values was a better scaling method, whereas the median and the upper quartiles were consistently the worst performers; one variation of remove unwanted variation had the best chance of capturing true positives but at the cost of increased false positives. In general, these methods were, at best, moderately helpful when the level of differential expression was extensive and asymmetric.CONCLUSIONOur study (1) provides the much-needed benchmark data and computational tools for assessing depth normalization, (2) shows the dependence of normalization performance on the underlying pattern of differential expression, and (3) calls for continued research efforts to develop more effective normalization methods.

NormGAM: an R package to remove systematic biases in genome architecture mapping data

BackgroundThe genome architecture mapping (GAM) technique can capture genome-wide chromatin interactions. However, besides the known systematic biases in the raw GAM data, we have found a new type of systematic bias. It is necessary to develop and evaluate effective normalization methods to remove all systematic biases in the raw GAM data.ResultsWe have detected a new type of systematic bias, the fragment length bias, in the genome architecture mapping (GAM) data, which is significantly different from the bias of window detection frequency previously mentioned in the paper introducing the GAM method but is similar to the bias of distances between restriction sites existing in raw Hi-C data. We have found that the normalization method (a normalized variant of the linkage disequilibrium) used in the GAM paper is not able to effectively eliminate the new fragment length bias at 1 Mb resolution (slightly better at 30 kb resolution). We have developed an R package named normGAM for eliminating the new fragment length bias together with the other three biases existing in raw GAM data, which are the biases related to window detection frequency, mappability, and GC content. Five normalization methods have been implemented and included in the R package including Knight-Ruiz 2-norm (KR2, newly designed by us), normalized linkage disequilibrium (NLD), vanilla coverage (VC), sequential component normalization (SCN), and iterative correction and eigenvector decomposition (ICE).ConclusionsBased on our evaluations, the five normalization methods can eliminate the four biases existing in raw GAM data, with VC and KR2 performing better than the others. We have observed that the KR2-normalized GAM data have a higher correlation with the KR-normalized Hi-C data on the same cell samples indicating that the KR-related methods are better than the others for keeping the consistency between the GAM and Hi-C experiments. Compared with the raw GAM data, the normalized GAM data are more consistent with the normalized distances from the fluorescence in situ hybridization (FISH) experiments. The source code of normGAM can be freely downloaded from http://dna.cs.miami.edu/normGAM/.

EWAS Data Hub: a resource of DNA methylation array data and metadata.

Epigenome-Wide Association Study (EWAS) has become an effective strategy to explore epigenetic basis of complex traits. Over the past decade, a large amount of epigenetic data, especially those sourced from DNA methylation array, has been accumulated as the result of numerous EWAS projects. We present EWAS Data Hub (https://bigd.big.ac.cn/ewas/datahub), a resource for collecting and normalizing DNA methylation array data as well as archiving associated metadata. The current release of EWAS Data Hub integrates a comprehensive collection of DNA methylation array data from 75 344 samples and employs an effective normalization method to remove batch effects among different datasets. Accordingly, taking advantages of both massive high-quality DNA methylation data and standardized metadata, EWAS Data Hub provides reference DNA methylation profiles under different contexts, involving 81 tissues/cell types (that contain 25 brain parts and 25 blood cell types), six ancestry categories, and 67 diseases (including 39 cancers). In summary, EWAS Data Hub bears great promise to aid the retrieval and discovery of methylation-based biomarkers for phenotype characterization, clinical treatment and health care.

Intersubject Variability and Normalization Strategies for Spinal Cord Total Cross-Sectional and Gray Matter Areas.

The quantification of spinal cord (SC) atrophy by MRI has assumed an important role in assessment of neuroinflammatory/neurodegenerative diseases and traumatic SC injury. Recent technical advances make possible the quantification of gray matter (GM) and white matter tissues in clinical settings. However, the goal of a reliable diagnostic, prognostic or predictive marker is still elusive, in part due to large intersubject variability of SC areas. Here, we investigated the sources of this variability and explored effective strategies to reduce it. One hundred twenty-nine healthy subjects (mean age: 41.0 ± 15.9) underwent MRI on a Siemens 3T Skyra scanner. Two-dimensional PSIR at the C2-C3 vertebral level and a sagittal 1 mm3 3D T1-weighted brain acquisition extended to the upper cervical cord were acquired. Total cross-sectional area and GM area were measured at C2-C3, as well as measures of the vertebra, spinal canal and the skull. Correlations between the different metrics were explored using Pearson product-moment coefficients. The most promising metrics were used to normalize cord areas using multiple regression analyses. The most effective normalization metrics were the V-scale (from SienaX) and the product of the C2-C3 spinal canal diameters. Normalization methods based on these metrics reduced the intersubject variability of cord areas of up to 17.74%. The measured cord areas had a statistically significant sex difference, while the effect of age was moderate. The present work explored in a large cohort of healthy subjects the source of intersubject variability of SC areas and proposes effective normalization methods for its reduction.

Adaptive color deconvolution for histological WSI normalization

Background and ObjectiveColor consistency of histological images is significant for developing reliable computer-aided diagnosis (CAD) systems. However, the color appearance of digital histological images varies across different specimen preparations, staining, and scanning situations. This variability affects the diagnosis and decreases the accuracy of CAD approaches. It is important and challenging to develop effective color normalization methods for digital histological images. MethodsWe proposed a novel adaptive color deconvolution (ACD) algorithm for stain separation and color normalization of hematoxylin-eosin-stained whole slide images (WSIs). To avoid artifacts and reduce the failure rate of normalization, multiple prior knowledges of staining are considered and embedded in the ACD model. To improve the capacity of color normalization for various WSIs, an integrated optimization is designed to simultaneously estimate the parameters of the stain separation and color normalization. The solving of ACD model and application of the proposed method involves only pixel-wise operation, which makes it very efficient and applicable to WSIs. ResultsThe proposed method was evaluated on four WSI-datasets including breast, lung and cervix cancers and was compared with 6 state-of-the-art methods. The proposed method achieved the most consistent performance in color normalization according to the quantitative metrics. Through a qualitative assessment for 500 WSIs, the failure rate of normalization was 0.4% and the structure and color artifacts were effectively avoided. Applied to CAD methods, the area under receiver operating characteristic curve for cancer image classification was improved from 0.842 to 0.914. The average time of solving the ACD model is 2.97 s. ConclusionsThe proposed ACD model has prone effective for color normalization of hematoxylin-eosin-stained WSIs in various color appearances. The model is robust and can be applied to WSIs containing different lesions. The proposed model can be efficiently solved and is effective to improve the performance of cancer image recognition, which is adequate for developing automatic CAD programs and systems based on WSIs.

APA: Adaptive Pose Alignment for Pose-Invariant Face Recognition

Face alignment is a crucial step in face recognition. Through making the position of face consistent, face alignment reduces intra-class variability due to factors such as lighting, background, pose, and perspective transformation, and further facilitating the recognition tasks. In this paper, we propose a new face alignment method for pose-invariant face recognition, called adaptive pose alignment (APA), which can greatly reduce the intra-class difference and correct the noise caused by the traditional method in the alignment process, especially in unconstrained settings. Instead of aligning all faces to the pre-defined, uniform frontal shape, we adaptively learn the alignment templates according to the facial poses and then align each face of training or testing sets to its related template. To further improve face recognition performance, we propose a simple, yet effective feature normalization method which can generate more discriminative feature representation of a face or a set of faces (template) combined with the APA method. Furthermore, we introduce a pose-invariant face recognition pipeline that sequentially applies APA-based alignment, deep representation by softmax or ArcFace loss function, and the effective feature normalization procedure. We empirically show that APA-based images can accelerate the training of deep face recognition model by aligning all the images to the optimal templates. Moreover, experimental results show that the proposed method achieves the state-of-the-art performance on popular challenging face datasets including IJB-A, IJB-C, and CPLFW datasets.

GMPR: A robust normalization method for zero-inflated count data with application to microbiome sequencing data.

Normalization is the first critical step in microbiome sequencing data analysis used to account for variable library sizes. Current RNA-Seq based normalization methods that have been adapted for microbiome data fail to consider the unique characteristics of microbiome data, which contain a vast number of zeros due to the physical absence or under-sampling of the microbes. Normalization methods that specifically address the zero-inflation remain largely undeveloped. Here we propose geometric mean of pairwise ratios—a simple but effective normalization method—for zero-inflated sequencing data such as microbiome data. Simulation studies and real datasets analyses demonstrate that the proposed method is more robust than competing methods, leading to more powerful detection of differentially abundant taxa and higher reproducibility of the relative abundances of taxa.

Using the matrix-induced ion suppression method for concentration normalization in cellular metabolomics studies.

Studies of the cell metabolome greatly improve our understanding of cell biology. Currently, most cellular metabolomics studies control only cell numbers or protein content without adjusting the total metabolite concentration, mainly because of the lack of an effective concentration normalization method for cell metabolites. This study proposed a matrix-induced ion suppression (MIIS) method to measure the total amount of cellular metabolites by utilizing flow injection analysis coupled with electrospray ionization mass spectrometry (FIA-ESI-MS).We used series dilutions of HL-60 cell extracts to establish the relationship between cellular metabolite concentration and the degree of ion suppression of the ion suppression indicator, and a good correlation was obtained between 2- and 12-fold dilutions of cell extracts (R(2) = 0.999). Two lung cancer cells, CL1-0 and CL1-5, were selected as the model cell lines to evaluate the efficacy of the MIIS method and the importance of metabolite concentration normalization. Through MIIS analysis, CL1-0 cells were found to contain metabolites at a concentration 2.1 times higher than in CL1-5, and the metastatic properties of CL1-5 could only be observed after 2.1-fold dilution of CL1-0 before metabolomic analysis. Our results demonstrated that the MIIS method is an effective approach for metabolite concentration normalization and that controlling metabolite concentrations can improve data integrity in cellular metabolomics studies.

Test set bias affects reproducibility of gene signatures.

Prior to applying genomic predictors to clinical samples, the genomic data must be properly normalized to ensure that the test set data are comparable to the data upon which the predictor was trained. The most effective normalization methods depend on data from multiple patients. From a biomedical perspective, this implies that predictions for a single patient may change depending on which other patient samples they are normalized with. This test set bias will occur when any cross-sample normalization is used before clinical prediction. We demonstrate that results from existing gene signatures which rely on normalizing test data may be irreproducible when the patient population changes composition or size using a set of curated, publicly available breast cancer microarray experiments. As an alternative, we examine the use of gene signatures that rely on ranks from the data and show why signatures using rank-based features can avoid test set bias while maintaining highly accurate classification, even across platforms. The code, data and instructions necessary to reproduce our entire analysis is available at https://github.com/prpatil/testsetbias.

Abstract 5276: Normalization of miRNA-sequencing data.

Abstract microRNAs (miRNA) are endogenous, small non-coding nucleotides that negatively regulate gene expression post-transcriptionally. Through interactions with Argonaute (Ago) proteins, they form the RNA-induced silencing complex (RISC) and can recognize and bind to the 3’UTR of mRNAs in a sequence-specific manner, leading to translational inhibition or mRNA degradation. Over 30% of human protein-coding genes are predicted to be conserved targets of miRNAs. Consequently, changes in their expression are likely to be associated with the development and progression of diseases, including cancer. An increasing number of studies are utilizing high-throughput sequencing over microarrays for the expression profiling of miRNAs. Processing of the raw sequencing data usually involves filtering based on quality measures, trimming for adapters and mapping the reads to a (genome or miRNA) reference. Then, normalization of the data is crucial before any downstream analysis can be performed. Normalization is the process of removing sources of variation, which are of non-biological origins (stemming from sample handling, library preparation, imaging and so on), and can affect the measured expression levels. An effective normalization method should minimize technical and experimental bias without introducing noise; the differences that remain should be truly biological effects. Several normalization methods for miRNA-seq data have been proposed, including linear scaling, non-linear scaling, quantile normalization and variance stabilization normalization. These methods differ in terms of complexity and the assumptions made. However, no standard technique has been recommended. Read counts from each experiment are usually simply adjusted for differences in sequencing depth (library size) to reads-per-million (RPM). Unfortunately, the performance and appropriateness of any of the normalization methods cannot be assessed using real data because the true values are not known. To this end, we have used a 12x12 Latin Square design to spike in 12 different oligonucleotides with known nominal concentrations, into a complex mixture of human miRNAs. These spike-in pools were subjected to all the preparatory steps of small RNA library construction for sequencing on the Illumina HiSeq2000. Preliminary results show that the spike-in sequences can be recovered successfully from the data. Using this data set, the relative merits of different normalization procedures are being assessed based on measures of bias, variance and improved sensitivity and specificity for the detection of differentially expressed miRNAs. The goal is to identify an optimal normalization method for miRNA-seq data, which would reduce variance without increasing bias. Citation Format: Shirley Tam, Richard de Borja, Ming-Sound Tsao, John D. McPherson. Normalization of miRNA-sequencing data. [abstract]. In: Proceedings of the 104th Annual Meeting of the American Association for Cancer Research; 2013 Apr 6-10; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2013;73(8 Suppl):Abstract nr 5276. doi:10.1158/1538-7445.AM2013-5276

A Method for Application of Classification Tree Models to Map Aquatic Vegetation Using Remotely Sensed Images from Different Sensors and Dates

In previous attempts to identify aquatic vegetation from remotely-sensed images using classification trees (CT), the images used to apply CT models to different times or locations necessarily originated from the same satellite sensor as that from which the original images used in model development came, greatly limiting the application of CT. We have developed an effective normalization method to improve the robustness of CT models when applied to images originating from different sensors and dates. A total of 965 ground-truth samples of aquatic vegetation types were obtained in 2009 and 2010 in Taihu Lake, China. Using relevant spectral indices (SI) as classifiers, we manually developed a stable CT model structure and then applied a standard CT algorithm to obtain quantitative (optimal) thresholds from 2009 ground-truth data and images from Landsat7-ETM+, HJ-1B-CCD, Landsat5-TM and ALOS-AVNIR-2 sensors. Optimal CT thresholds produced average classification accuracies of 78.1%, 84.7% and 74.0% for emergent vegetation, floating-leaf vegetation and submerged vegetation, respectively. However, the optimal CT thresholds for different sensor images differed from each other, with an average relative variation (RV) of 6.40%. We developed and evaluated three new approaches to normalizing the images. The best-performing method (Method of 0.1% index scaling) normalized the SI images using tailored percentages of extreme pixel values. Using the images normalized by Method of 0.1% index scaling, CT models for a particular sensor in which thresholds were replaced by those from the models developed for images originating from other sensors provided average classification accuracies of 76.0%, 82.8% and 68.9% for emergent vegetation, floating-leaf vegetation and submerged vegetation, respectively. Applying the CT models developed for normalized 2009 images to 2010 images resulted in high classification (78.0%–93.3%) and overall (92.0%–93.1%) accuracies. Our results suggest that Method of 0.1% index scaling provides a feasible way to apply CT models directly to images from sensors or time periods that differ from those of the images used to develop the original models.

Systematic comparison of RNA-Seq normalization methods using measurement error models

Further advancement of RNA-Seq technology and its application call for the development of effective normalization methods for RNA-Seq data. Currently, different normalization methods are compared and validated by their correlations with a certain gold standard. Gene expression measurements generated by a different technology or platform such as Real-time reverse transcription polymerase chain reaction (qRT-PCR) or Microarray are usually used as the gold standard. Although the current approach is intuitive and easy to implement, it becomes statistically inadequate when the gold standard is also subject to measurement error (ME). Furthermore, the current approach is not informative, because the correlation of a normalization method with a certain gold standard does not provide much information about the exact quality of the normalized RNA-Seq measurements. We propose to use the system of ME models based on qRT-PCR, Microarray and RNA-Seq gene expression data to compare and validate RNA-Seq normalization methods. This approach does not assume the existence of a gold standard. The performance of a normalization method can be characterized by a group of parameters of the system, which are referred to as the performance parameters, and these performance parameters can be consistently estimated. Different normalization methods can thus be compared by comparing their corresponding estimated performance parameters. We applied the proposed approach to compare five existing RNA-Seq normalization methods using the gene expression data of two RNA samples from the microArray Quality Control and Sequencing Quality Control projects and gained much insight about the pros and cons of these methods.

The Effects of Anthropometric Scaling Parameters on Normalized Muscle Strength in Uninjured Baseball Pitchers

Muscle force must be normalized for between-subjects comparisons of strength to be valid. The most effective method for normalizing muscle strength has not, however, been systematically evaluated. To evaluate the effects of normalizing muscle strength using a spectrum of anthropometric parameters. Cross-sectional. Laboratory. 50 uninjured high-school-age baseball pitchers. Shoulder-rotation strength was tested at 0° and 90° abduction with a handheld dynamometer. Muscle force was normalized to parameters including subject height, weight, height × weight, body-mass index (BMI), forearm length, and forearm length × height. Statistical analysis included evaluating the coefficient of variation, skewness, and kurtosis of the nonnormalized and normalized muscle force. The most effective normalization method was determined based on the scaling factor that yielded the lowest variability for the data set and promoted the most normal distribution of the data set. Using body weight to scale muscle force was the most effective anthropometric parameter for normalizing strength values based on the group of statistical measures of variability. BMI, height × weight, and forearm length × weight as scaling factors also yielded less variable values for muscle strength compared with nonnormalized strength, but less consistently than body weight. Height and forearm length were least effective in reducing the variability of the data set relative to nonnormalized muscle force. This study provides objective support for scaling muscle strength to subject body weight. This approach to normalizing muscle strength uses methods readily accessible to clinicians and researchers and may facilitate the identification of differences in strength between individuals with diverse physical characteristics.

Metabonomic analysis of rat urine ¹H magnetic resonance spectra based on different normalization methods.

To study metabonomics in the urine of rats of different genders by magnetic resonance (MR) with 2 normalization methods. Different normalization methods such as mean-centering not scaling (Ctr) and unit variance scaling (UV) were used before orthogonal to partial least squares discriminant analysis(OPLS-DA). Distinguished metabolites in the urine of different gender rats were analyzed by calculating the correlation coefficients. The data normalized by Ctr before OPLS-DA analysis revealed high degree conception metabolites in the urine such as valine, alanine, acetate, ornithine, aminohippurate, phenylethylamine, cytosine, citrate, dimethylamine, allantoin, methylamine, fumarate and one unknown metabolite whose chemical shift was δ4.14. Data normalized by UV before OPLS-DA analysis revealed the above 12 high degree conception metabolites except citrate, and also low degree conception metabolites such as thiamine, creatinine, formate and one unknown metabolite whose chemical shift was δ2.92. Unit variance scaling is a more effective normalization method in metabonomic analysis.