Systemic and Local Strategies for Primary Prevention of Breast Cancer.

One in eight women will develop breast cancer in the US. For women with moderate (15-20%) to average (12.5%) risk of breast cancer, there are few options available for risk reduction. For high-risk (>20%) women, such as BRCA mutation carriers, primary prevention strategies are limited to evidence-based surgical removal of breasts and/or ovaries and anti-estrogen treatment. Despite their effectiveness in risk reduction, not many high-risk individuals opt for surgical or hormonal interventions due to severe side effects and potentially life-changing outcomes as key deterrents. Thus, better communication about the benefits of existing strategies and the development of new strategies with minimal side effects are needed to offer women adequate risk-reducing interventions. We extensively review and discuss innovative investigational strategies for primary prevention. Most of these investigational strategies are at the pre-clinical stage, but some are already being evaluated in clinical trials and others are expected to lead to first-in-human clinical trials within 5 years. Likely, these strategies would be initially tested in high-risk individuals but may be applicable to lower-risk women, if shown to decrease risk at a similar rate to existing strategies, but with minimal side effects.

Shape Anisotropy-Governed High-Performance Nanomagnetosol for In Vivo Magnetic Particle Imaging of Lungs.

Caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), coronavirus disease 2019 (COVID-19) has shown extensive lung manifestations in vulnerable individuals, putting lung imaging and monitoring at the forefront of early detection and treatment. Magnetic particle imaging (MPI) is an imaging modality, which can bring excellent contrast, sensitivity, and signal-to-noise ratios to lung imaging for the development of new theranostic approaches for respiratory diseases. Advances in MPI tracers would offer additional improvements and increase the potential for clinical translation of MPI. Here, a high-performance nanotracer based on shape anisotropy of magnetic nanoparticles is developed and its use in MPI imaging of the lung is demonstrated. Shape anisotropy proves to be a critical parameter for increasing signal intensity and resolution and exceeding those properties of conventional spherical nanoparticles. The 0D nanoparticles exhibit a 2-fold increase, while the 1D nanorods have a > 5-fold increase in signal intensity when compared to VivoTrax. Newly designed 1D nanorods displayed high signal intensities and excellent resolution in lung images. A spatiotemporal lung imaging study in mice revealed that this tracer offers new opportunities for monitoring disease and guiding intervention.

MicroRNA-210-3p Regulates Endometriotic Lesion Development by Targeting IGFBP3 in Baboons and Women with Endometriosis.

MicroRNAs (miRs) play an important role in the pathophysiology of endometriosis; however, the role of miR-210 in endometriosis remains unclear. This study explores the role of miR-210 and its targets, IGFBP3 and COL8A1, in ectopic lesion growth and development. Matched eutopic (EuE) and ectopic (EcE) endometrial samples were obtained for analysis from baboons and women with endometriosis. Immortalized human ectopic endometriotic epithelial cells (12Z cells) were utilized for functional assays. Endometriosis was experimentally induced in female baboons (n = 5). Human matched endometrial and endometriotic tissues were obtained from women (n = 9, 18-45 years old) with regular menstrual cycles. Quantitative reverse transcript polymerase chain reaction (RT-qPCR) analysis was performed for in vivo characterization of miR-210, IGFBP3, and COL8A1. In situ hybridization and immunohistochemical analysis were performed for cell-specific localization. Immortalized endometriotic epithelial cell lines (12Z) were utilized for in vitro functional assays. MiR-210 expression was decreased in EcE, while IGFBP3 and COL8A1 expression was increased in EcE. MiR-210 was expressed in the glandular epithelium of EuE but attenuated in those of EcE. IGFBP3 and COL8A1 were expressed in the glandular epithelium of EuE and were increased compared to EcE. MiR-210 overexpression in 12Z cells suppressed IGFBP3 expression and attenuated cell proliferation and migration. MiR-210 repression and subsequent unopposed IGFBP3 expression may contribute to endometriotic lesion development by increasing cell proliferation and migration.

The transcription factor VAX1 in VIP neurons of the suprachiasmatic nucleus impacts circadian rhythm generation, depressive-like behavior, and the reproductive axis in a sex-specific manner in mice.

Background: The suprachiasmatic nucleus (SCN) within the hypothalamus is a key brain structure required to relay light information to the body and synchronize cell and tissue level rhythms and hormone release. Specific subpopulations of SCN neurons, defined by their peptide expression, regulate defined SCN output. Here we focus on the vasoactive intestinal peptide (VIP) expressing neurons of the SCN. SCN VIP neurons are known to regulate circadian rhythms and reproductive function. Methods: To specifically study SCN VIP neurons, we generated a novel knock out mouse line by conditionally deleting the SCN enriched transcription factor, Ventral Anterior Homeobox 1 (Vax1), in VIP neurons (Vax1Vip; Vax1fl/fl:VipCre). Results: We found that Vax1Vip females presented with lengthened estrous cycles, reduced circulating estrogen, and increased depressive-like behavior. Further, Vax1Vip males and females presented with a shortened circadian period in locomotor activity and ex vivo SCN circadian period. On a molecular level, the shortening of the SCN period was driven, at least partially, by a direct regulatory role of VAX1 on the circadian clock genes Bmal1 and Per2. Interestingly, Vax1Vip females presented with increased expression of arginine vasopressin (Avp) in the paraventricular nucleus, which resulted in increased circulating corticosterone. SCN VIP and AVP neurons regulate the reproductive gonadotropin-releasing hormone (GnRH) and kisspeptin neurons. To determine how the reproductive neuroendocrine network was impacted in Vax1Vip mice, we assessed GnRH sensitivity to a kisspeptin challenge in vivo. We found that GnRH neurons in Vax1Vip females, but not males, had an increased sensitivity to kisspeptin, leading to increased luteinizing hormone release. Interestingly, Vax1Vip males showed a small, but significant increase in total sperm and a modest delay in pubertal onset. Both male and female Vax1Vip mice were fertile and generated litters comparable in size and frequency to controls. Conclusion: Together, these data identify VAX1 in SCN VIP neurons as a neurological overlap between circadian timekeeping, female reproduction, and depressive-like symptoms in mice, and provide novel insight into the role of SCN VIP neurons.

X-Ray Visualization of Intraductal Ethanol-Based Ablative Treatment for Prevention of Breast Cancer in Rat Models.

There are still a limited number of primary interventions for prevention of breast cancer. For women at a high risk of developing breast cancer, the most effective intervention is prophylactic mastectomy. This is a drastic surgical procedure in which the mammary epithelial cells that can give rise to breast cancer are completely removed along with the surrounding tissue. The goal of this protocol is to demonstrate the feasibility of a minimally invasive intraductal procedure that could become a new primary intervention for breast cancer prevention. This local procedure would preferentially ablate mammary epithelial cells before they can become malignant. Intraductal methods to deliver solutions directly to these epithelial cells in rodent models of breast cancer have been developed at Michigan State University and elsewhere. The rat mammary gland consists of a single ductal tree that has a simpler and more linear architecture compared to the human breast. However, chemically induced rat models of breast cancer offer valuable tools for proof-of-concept studies of new preventive interventions and scalability from mouse models to humans. Here, a procedure for intraductal delivery of an ethanol-based ablative solution containing tantalum oxide nanoparticles as X-ray contrast agent and ethyl cellulose as gelling agent into the rat mammary ductal tree is described. Delivery of aqueous reagents (e.g., cytotoxic compounds, siRNAs, AdCre) by intraductal injection has been described previously in mouse and rat models. This protocol description emphasizes methodological changes and steps that pertain uniquely to delivering an ablative solution, formulation consideration to minimize local and systemic side effects of the ablative solution, and X-ray imaging for in vivo assessment of ductal tree filling. Fluoroscopy and micro-CT techniques enable to determine the success of ablative solution delivery and the extent of ductal tree filling thanks to compatibility with the tantalum-containing contrast agent.

Intraductal Delivery and X-ray Visualization of Ethanol-Based Ablative Solution for Prevention and Local Treatment of Breast Cancer in Mouse Models.

Breast cancer is the most prevalent cancer and the second-leading cause of cancer-related death for women in the USA. For high-risk women, prophylactic mastectomy is the most effective primary prevention strategy. Prophylactic mastectomy is an aggressive surgical procedure that completely removes the mammary epithelial cells from which breast cancer arises along with the surrounding tissue. We seek to develop a minimally invasive intraductal procedure as an alternative to prophylactic mastectomy to locally ablate the mammary epithelial cells before they can become malignant. We and others have developed an intraductal delivery procedure to reach and treat these epithelial cells in rodent models of breast cancer. While the mouse mammary gland with a single non-anastomosed ductal tree opening at the nipple has a much less complex and tortuous architecture than the human breast, chemically induced and genetically engineered mouse models of breast cancer are valuable to produce proof-of-concept studies of new preventative strategies. Here, we describe a procedure for intraductal delivery of an ethanol-based ablative solution containing micro-CT/X-ray tantalum-based contrast agent within the mouse mammary ductal tree for the therapeutic purpose of primary prevention of breast cancer. Intraductal delivery of aqueous reagents (e.g., cytotoxic compounds, siRNAs, AdCre) has been previously described in mouse models. Thus, we focus our protocol description on methodological modifications and unique experimental considerations for optimizing delivery of ethanol, for minimizing local and systemic side effects of ethanol administration, and for in vivo visualization of ductal tree filling via micro-CT/fluoroscopy imaging. Visualization of the ductal tree immediately after injection of a contrast-containing solution allows for confirmation of complete filling or unsuccessful outcomes such as underfilling or overfilling. This procedure can be applied for delivery and imaging of other ablative compounds aimed at either preventing tumor formation or locally treating early-stage tumors accessible via the ductal tree.

Clinical Applications of Short Non-Coding RNA-Based Therapies in the Era of Precision Medicine.

Traditional targeted therapeutic agents have relied on small synthetic molecules or large proteins, such as monoclonal antibodies. These agents leave a lot of therapeutic targets undruggable because of the lack or inaccessibility of active sites and/or pockets in their three-dimensional structure that can be chemically engaged. RNA presents an attractive, transformative opportunity to reach any genetic target with therapeutic intent. RNA therapeutic design is amenable to modularity and tunability and is based on a computational blueprint presented by the genetic code. Here, we will focus on short non-coding RNAs (sncRNAs) as a promising therapeutic modality because of their potency and versatility. We review recent progress towards clinical application of small interfering RNAs (siRNAs) for single-target therapy and microRNA (miRNA) activity modulators for multi-target therapy. siRNAs derive their potency from the fact that the underlying RNA interference (RNAi) mechanism is catalytic and reliant on post-transcriptional mRNA degradation. Therapeutic siRNAs can be designed against virtually any mRNA sequence in the transcriptome and specifically target a disease-causing mRNA variant. Two main classes of microRNA activity modulators exist to increase (miRNA mimics) or decrease (anti-miRNA inhibitors) the function of a specific microRNA. Since a single microRNA regulates the expression of multiple target genes, a miRNA activity modulator can have a more profound effect on global gene expression and protein output than siRNAs do. Both types of sncRNA-based drugs have been investigated in clinical trials and some siRNAs have already been granted FDA approval for the treatment of genetic, cardiometabolic, and infectious diseases. Here, we detail clinical results using siRNA and miRNA therapeutics and present an outlook for the potential of these sncRNAs in medicine.

Role of non-coding RNAs in tumor progression and metastasis in pancreatic cancer.

Pancreatic ductal adenocarcinoma (PDAC) is one of the most lethal types of cancer with an overall 5-year survival rate of less than 10%. The 1-year survival rate of patients with locally advanced or metastatic disease is abysmal. The aggressive nature of cancer cells, hypovascularization, extensive desmoplastic stroma, and immunosuppressive tumor microenvironment (TME) endows PDAC tumors with multiple mechanisms of drug resistance. With no obvious genetic mutation(s) driving tumor progression or metastatic transition, the challenges for understanding the biological mechanism(s) of these processes are paramount. A better understanding of the molecular and cellular mechanisms of these processes could lead to new diagnostic tools for patient management and new targets for therapeutic intervention. microRNAs (miRNAs) are an evolutionarily conserved gene class of short non-coding regulatory RNAs. miRNAs are an extensive regulatory layer that controls gene expression at the posttranscriptional level. This review focuses on preclinical models that functionally dissect miRNA activity in tumor progression or metastatic processes in PDAC. Collectively, these studies suggest an influence of miRNAs and RNA-RNA networks in the processes of epithelial to mesenchymal cell transition and cancer cell stemness. At a cell-type level, some miRNAs mainly influence cancer cell-intrinsic processes and pathways, whereas other miRNAs predominantly act in distinct cellular compartments of the TME to regulate fibroblast and immune cell functions and/or influence other cell types' function via cell-to-cell communications by transfer of extracellular vesicles. At a molecular level, the influence of miRNA-mediated regulation often converges in core signaling pathways, including TGF-ß, JAK/STAT, PI3K/AKT, and NF-κB.

microRNA-based diagnostic and therapeutic applications in cancer medicine.

It has been almost two decades since the first link between microRNAs and cancer was established. In the ensuing years, this abundant class of short noncoding regulatory RNAs has been studied in virtually all cancer types. This tremendously large body of research has generated innovative technological advances for detection of microRNAs in tissue and bodily fluids, identified the diagnostic, prognostic, and/or predictive value of individual microRNAs or microRNA signatures as potential biomarkers for patient management, shed light on regulatory mechanisms of RNA-RNA interactions that modulate gene expression, uncovered cell-autonomous and cell-to-cell communication roles of specific microRNAs, and developed a battery of viral and nonviral delivery approaches for therapeutic intervention. Despite these intense and prolific research efforts in preclinical and clinical settings, there are a limited number of microRNA-based applications that have been incorporated into clinical practice. We review recent literature and ongoing clinical trials that highlight most promising approaches and standing challenges to translate these findings into viable microRNA-based clinical tools for cancer medicine. This article is categorized under: RNA in Disease and Development > RNA in Disease.

Automated Five-Color Multiplex Co-detection of MicroRNA and Protein Expression in Fixed Tissue Specimens.

microRNAs are an important class of noncoding regulatory RNAs with functional roles in development, physiology, and disease. Visualization of microRNA expression at a single-cell level has contributed to a better understanding of their biological function in animal models and their etiological contribution to human diseases. In addition, several microRNAs have been highlighted as potential biomarkers carrying diagnostic and prognostic information. Co-detection of microRNA expression with that of cell-type-specific proteins can enhance the interpretative power of expression changes during development or altered expression in pathological conditions. Here, we describe an automated fluorescence-based five-color multiplex assay for co-detection of microRNA (e.g., miR-10b, miR-21, miR-205), noncoding RNA (e.g., snRNA U6, 18S rRNA), and protein expression (e.g., cytokeratin 19, vimentin, collagen I) in paraffin-embedded formalin-fixed tissue slides on a Leica Bond Rx staining station. While this protocol uses mainly mouse tissues to demonstrate multiplex detection, it can be generally applied to single-cell expression analysis of other animal models and clinical specimens.

Loss of microRNA-21 leads to profound stromal remodeling and short survival in K-Ras-driven mouse models of pancreatic cancer.

The microenvironment of pancreatic cancer adenocarcinoma (PDAC) is highly desmoplastic with distinct tumor-restraining and tumor-promoting fibroblast subpopulations. Re-education rather than indiscriminate elimination of these fibroblasts has emerged as a new strategy for combination therapy. Here, we studied the effects of global loss of profibrotic noncoding regulatory microRNA-21 (miR-21) in K-Ras-driven p53-deleted genetically engineered mouse models of PDAC. Strikingly, loss of miR-21 accelerated tumor initiation via mucinous cystic neoplastic lesions and progression to locally advanced invasive carcinoma from which animals precipitously succumbed at an early age. The absence of tumor-restraining myofibroblasts and a massive infiltrate of immune cells were salient phenotypic features of global miR-21 loss. Stromal miR-21 activity was required for induction of tumor-restraining myofibroblasts in in vivo isograft transplantation experiments. Low miR-21 expression negatively correlated with a fibroblast gene expression signature and positively with an immune cell gene expression signature in The Cancer Genome Atlas PDAC data set (n = 156) mirroring findings in the mouse models. Our results exposed an overall tumor-suppressive function of miR-21 in in vivo PDAC models. These results have important clinical implications for anti-miR-21-based inhibitory therapeutic approaches under consideration for PDAC and other cancer types. Mechanistic dissection of the cell-intrinsic role of miR-21 in cancer-associated fibroblasts and other cell types will be needed to inform best strategies for pharmacological modulation of miR-21 activity to remodel the tumor microenvironment and enhance treatment response in PDAC.

Tantalum oxide nanoparticles as versatile contrast agents for X-ray computed tomography.

Here, we describe the synthesis, characterization and in vitro and in vivo performance of a series of tantalum oxide (TaOx) based nanoparticles (NPs) for computed tomography (CT). Five distinct versions of 9-12 nm diameter silane coated TaOx nanocrystals (NCs) were fabricated by a sol-gel method with varying degrees of hydrophilicity and with or without fluorescence, with the highest reported Ta content to date (78%). Highly hydrophilic NCs were left bare and were evaluated in vivo in mice for micro-CT of full body vasculature, where following intravenous injection, TaOx NCs demonstrate high vascular CT contrast, circulation in blood for ∼3 h, and eventual accumulation in RES organs; and following injection locally in the mammary gland, where the full ductal tree structure can be clearly delineated. Partially hydrophilic NCs were encapsulated within mesoporous silica nanoparticles (MSNPs; TaOx@MSNPs) and hydrophobic NCs were encapsulated within poly(lactic-co-glycolic acid) (PLGA; TaOx@PLGA) NPs, serving as potential CT-imagable drug delivery vehicles. Bolus intramuscular injections of TaOx@PLGA NPs and TaOx@MSNPs to mimic the accumulation of NPs at a tumor site produce high signal enhancement in mice. In vitro studies on bare NCs and formulated NPs demonstrate high cytocompatibility and low dissolution of TaOx. This work solidifies that TaOx-based NPs are versatile contrast agents for CT.

Ductal tree ablation by local delivery of ethanol prevents tumor formation in an aggressive mouse model of breast cancer.

BACKGROUND: Prophylactic mastectomy is the most effective intervention to prevent breast cancer. However, this major surgery has life-changing consequences at the physical, emotional, psychological, and social levels. Therefore, only high-risk individuals consider this aggressive procedure, which completely removes the mammary epithelial cells from which breast cancer arises along with surrounding tissue. Here, we seek to develop a minimally invasive procedure as an alternative to prophylactic mastectomy by intraductal (ID) delivery of a cell-killing solution that locally ablates the mammary epithelial cells before they become malignant. METHODS: After ID injection of a 70% ethanol-containing solution in FVB/NJ female animals, ex vivo dual stained whole-mount tissue analysis and in vivo X-ray microcomputed tomography imaging were used to visualize ductal tree filling, and histological and multiplex immunohistochemical assays were used to characterize ablative effects and quantitate the number of intact epithelial cells and stroma. After ID injection of 70% ethanol or other solutions in cancer-prone FVB-Tg-C3(1)-TAg female animals, mammary glands were palpated weekly to establish tumor latency and examined after necropsy to record tumor incidence. Statistical difference in median tumor latency and tumor incidence between experimental groups was analyzed by log-rank test and logistic mixed-effects model, respectively. RESULTS: We report that ID injection of 70% ethanol effectively ablates the mammary epithelia with limited collateral damage to surrounding stroma and vasculature in the murine ductal tree. ID injection of 70% ethanol into the mammary glands of the C3(1)-TAg multifocal breast cancer model significantly delayed tumor formation (median latency of 150 days in the untreated control group [n = 25] vs. 217 days in the ethanol-treated group [n = 13], p value < 0.0001) and reduced tumor incidence (34% of glands with tumors [85 of 250] in the untreated control group vs. 7.3% of glands with tumor [7 of 95] in the ethanol-treated group, risk ratio = 4.76 [95% CI 1.89 to 11.97, p value < 0.0001]). CONCLUSIONS: This preclinical study demonstrates the feasibility of local ductal tree ablation as a novel strategy for primary prevention of breast cancer. Given the existing clinical uses of ethanol, ethanol-based ablation protocols could be readily implemented in first-in-human clinical trials for high-risk individuals.

MicroRNA-155 Expression Is Enhanced by T-cell Receptor Stimulation Strength and Correlates with Improved Tumor Control in Melanoma.

microRNAs are short noncoding RNAs that regulate protein expression posttranscriptionally. We previously showed that miR-155 promotes effector CD8+ T-cell responses. However, little is known about the regulation of miR-155 expression. Here, we report that antigen affinity and dose determine miR-155 expression in CD8+ T cells. In B16 tumors expressing a low-affinity antigen ligand, tumor-specific infiltrating CD8+ T cells showed variable miR-155 expression, whereby high miR-155 expression was associated with more cytokine-producing cells and tumor control. Moreover, anti-PD-1 treatment led to both increased miR-155 expression and tumor control by specific CD8+ T cells. In addition, miR-155 overexpression enhanced exhausted CD8+ T-cell persistence in the LCMV cl13 chronic viral infection model. In agreement with these observations in mouse models, miR-155 expression in human effector memory CD8+ T cells positively correlated with their frequencies in tumor-infiltrated lymph nodes of melanoma patients. Low miR-155 target gene signature in tumors was associated with prolonged overall survival in melanoma patients. Altogether, these results raise the possibility that high miR-155 expression in CD8+ tumor-infiltrating T cells may be a surrogate marker of the relative potency of in situ antigen-specific CD8+ T-cell responses.

Celebrating 25 Years of MicroRNA Research: From Discovery to Clinical Application.

In 1993, the Ambros lab reported the cloning and developmental function of lin-4, the first microRNA [...].

Exosomal MicroRNAs in Breast Cancer towards Diagnostic and Therapeutic Applications.

Soon after the discovery of microRNAs over 15 years ago, a myriad of research groups around the world sought to develop clinical applications in breast cancer for these short, noncoding, regulatory RNAs. While little of this knowledge has translated into the clinic, the recent research explosion on cell-to-cell communication via exosomes and other extracellular vesicles has rekindled interest in microRNA-based clinical applications. microRNAs appear to be a preferential and important cargo of exosomes in mediating biological effects in recipient cells. This review highlights recent studies on the biology of exosomal microRNAs (exo-miRNAs) and discusses potential clinical applications. From a diagnostic perspective, circulating exo-miRNAs may represent breast cancer cell content and/or tumor microenvironmental reactions to cancer cell growth. Thus, serum or plasma analysis of exo-miRNAs could be useful for early disease detection or for monitoring treatment response and disease progression. From a therapeutic perspective, exo-miRNAs derived from different cell types have been implicated in supporting or restraining tumor growth, conferring drug resistance, and preparing the metastatic niche. Strategies to interfere with the loading or delivery of tumor-promoting exo-miRNAs or to replenish tumor-suppressive miRNAs via exosomal delivery are under investigation. These recent studies provide new hope and opportunities, but study design limitations and technical challenges will need to be overcome before seriously considering clinical application of exo-miRNAs.

A Uniform System for the Annotation of Vertebrate microRNA Genes and the Evolution of the Human microRNAome.

Although microRNAs (miRNAs) are among the most intensively studied molecules of the past 20 years, determining what is and what is not a miRNA has not been straightforward. Here, we present a uniform system for the annotation and nomenclature of miRNA genes. We show that less than a third of the 1,881 human miRBase entries, and only approximately 16% of the 7,095 metazoan miRBase entries, are robustly supported as miRNA genes. Furthermore, we show that the human repertoire of miRNAs has been shaped by periods of intense miRNA innovation and that mature gene products show a very different tempo and mode of sequence evolution than star products. We establish a new open access database--MirGeneDB ( http://mirgenedb.org )--to catalog this set of miRNAs, which complements the efforts of miRBase but differs from it by annotating the mature versus star products and by imposing an evolutionary hierarchy upon this curated and consistently named repertoire.

Segment and fit thresholding: a new method for image analysis applied to microarray and immunofluorescence data.

Experiments involving the high-throughput quantification of image data require algorithms for automation. A challenge in the development of such algorithms is to properly interpret signals over a broad range of image characteristics, without the need for manual adjustment of parameters. Here we present a new approach for locating signals in image data, called Segment and Fit Thresholding (SFT). The method assesses statistical characteristics of small segments of the image and determines the best-fit trends between the statistics. Based on the relationships, SFT identifies segments belonging to background regions; analyzes the background to determine optimal thresholds; and analyzes all segments to identify signal pixels. We optimized the initial settings for locating background and signal in antibody microarray and immunofluorescence data and found that SFT performed well over multiple, diverse image characteristics without readjustment of settings. When used for the automated analysis of multicolor, tissue-microarray images, SFT correctly found the overlap of markers with known subcellular localization, and it performed better than a fixed threshold and Otsu's method for selected images. SFT promises to advance the goal of full automation in image analysis.