RESUMO
BACKGROUND: Chronic rhinosinusitis with nasal polyps (CRSwNP) is a disease of real interest for researchers due to its heterogenicity and complex pathophysiological mechanisms. Identification of the factors that ensure success after treatment represents one of the main challenges in CRSwNP research. No consensus in this direction has been reached so far. Biomarkers for poor outcomes have been noted, but nonetheless, their prognostic value has not been extensively investigated, and needs to be sought. We aimed to evaluate the correlation between potential prognostic predictors for recalcitrant disease in patients with CRSwNP. METHODS: The study group consisted of CRSwNP patients who underwent surgical treatment and nasal polyp (NP) tissue sampling. The preoperative workup included Lund-Mackay assessment, nasal endoscopy, eosinophil blood count, asthma, and environmental allergy questionnaire. Postoperatively, in subjects with poor outcomes, imagistic osteitis severity was evaluated, and IL-33 expression was measured. RESULTS: IL-33 expression in NP was positively and significantly correlated with postoperative osteitis on CT scans (p = 0.01). Furthermore, high osteitis CT scores were related to high blood eosinophilia (p = 0.01). A positive strong correlation was found between postoperative osteitis and the Lund-Mackay preoperative score (p = 0.01), as well as the nasal endoscopy score (p = 0.01). CONCLUSIONS: Our research analyzed the levels of polyp IL-33, relative to blood eosinophilia, overall disease severity score, and osteitis severity, in patients with CRSwNP. These variables are prognostic predictors for poor outcomes and recalcitrant disease. Considering the importance of bone involvement in CRSwNP, this research aims to provide a better insight into the correlations of osteitis with clinical and biological factors.
RESUMO
The evaluation of deep-learning (DL) systems typically relies on the Area under the Receiver-Operating-Curve (AU-ROC) as a performance metric. However, AU-ROC, in its holistic form, does not sufficiently consider performance within specific ranges of sensitivity and specificity, which are critical for the intended operational context of the system. Consequently, two systems with identical AU-ROC values can exhibit significantly divergent real-world performance. This issue is particularly pronounced in the context of anomaly detection tasks, a commonly employed application of DL systems across various research domains, including medical imaging, industrial automation, manufacturing, cyber security, fraud detection, and drug research, among others. The challenge arises from the heavy class imbalance in training datasets, with the abnormality class often incurring a considerably higher misclassification cost compared to the normal class. Traditional DL systems address this by adjusting the weighting of the cost function or optimizing for specific points along the ROC curve. While these approaches yield reasonable results in many cases, they do not actively seek to maximize performance for the desired operating point. In this study, we introduce a novel technique known as AUCReshaping, designed to reshape the ROC curve exclusively within the specified sensitivity and specificity range, by optimizing sensitivity at a predetermined specificity level. This reshaping is achieved through an adaptive and iterative boosting mechanism that allows the network to focus on pertinent samples during the learning process. We primarily investigated the impact of AUCReshaping in the context of abnormality detection tasks, specifically in Chest X-Ray (CXR) analysis, followed by breast mammogram and credit card fraud detection tasks. The results reveal a substantial improvement, ranging from 2 to 40%, in sensitivity at high-specificity levels for binary classification tasks.
Assuntos
Algoritmos , Mamografia , Sensibilidade e Especificidade , Curva ROC , RadiografiaRESUMO
In this work, we report the set-up and results of the Liver Tumor Segmentation Benchmark (LiTS), which was organized in conjunction with the IEEE International Symposium on Biomedical Imaging (ISBI) 2017 and the International Conferences on Medical Image Computing and Computer-Assisted Intervention (MICCAI) 2017 and 2018. The image dataset is diverse and contains primary and secondary tumors with varied sizes and appearances with various lesion-to-background levels (hyper-/hypo-dense), created in collaboration with seven hospitals and research institutions. Seventy-five submitted liver and liver tumor segmentation algorithms were trained on a set of 131 computed tomography (CT) volumes and were tested on 70 unseen test images acquired from different patients. We found that not a single algorithm performed best for both liver and liver tumors in the three events. The best liver segmentation algorithm achieved a Dice score of 0.963, whereas, for tumor segmentation, the best algorithms achieved Dices scores of 0.674 (ISBI 2017), 0.702 (MICCAI 2017), and 0.739 (MICCAI 2018). Retrospectively, we performed additional analysis on liver tumor detection and revealed that not all top-performing segmentation algorithms worked well for tumor detection. The best liver tumor detection method achieved a lesion-wise recall of 0.458 (ISBI 2017), 0.515 (MICCAI 2017), and 0.554 (MICCAI 2018), indicating the need for further research. LiTS remains an active benchmark and resource for research, e.g., contributing the liver-related segmentation tasks in http://medicaldecathlon.com/. In addition, both data and online evaluation are accessible via https://competitions.codalab.org/competitions/17094.
Assuntos
Benchmarking , Neoplasias Hepáticas , Humanos , Estudos Retrospectivos , Neoplasias Hepáticas/diagnóstico por imagem , Neoplasias Hepáticas/patologia , Fígado/diagnóstico por imagem , Fígado/patologia , Algoritmos , Processamento de Imagem Assistida por Computador/métodosRESUMO
Purpose: Building accurate and robust artificial intelligence systems for medical image assessment requires the creation of large sets of annotated training examples. However, constructing such datasets is very costly due to the complex nature of annotation tasks, which often require expert knowledge (e.g., a radiologist). To counter this limitation, we propose a method to learn from medical images at scale in a self-supervised way. Approach: Our approach, based on contrastive learning and online feature clustering, leverages training datasets of over 100,000,000 medical images of various modalities, including radiography, computed tomography (CT), magnetic resonance (MR) imaging, and ultrasonography (US). We propose to use the learned features to guide model training in supervised and hybrid self-supervised/supervised regime on various downstream tasks. Results: We highlight a number of advantages of this strategy on challenging image assessment problems in radiography, CT, and MR: (1) significant increase in accuracy compared to the state-of-the-art (e.g., area under the curve boost of 3% to 7% for detection of abnormalities from chest radiography scans and hemorrhage detection on brain CT); (2) acceleration of model convergence during training by up to 85% compared with using no pretraining (e.g., 83% when training a model for detection of brain metastases in MR scans); and (3) increase in robustness to various image augmentations, such as intensity variations, rotations or scaling reflective of data variation seen in the field. Conclusions: The proposed approach enables large gains in accuracy and robustness on challenging image assessment problems. The improvement is significant compared with other state-of-the-art approaches trained on medical or vision images (e.g., ImageNet).
RESUMO
Knowledge regarding the influence of the microbial community in cancer promotion or protection has expanded even more through the study of bacterial metabolic products and how they can modulate cancer risk, which represents an extremely challenging approach for the relationship between intestinal microbiota and colorectal cancer (CRC). This review discusses research progress on the effect of bacterial dysbiosis from a metabolic point of view, particularly on the biochemical mechanisms of butyrate, one of the main short chain fatty acids (SCFAs) with anti-inflammatory and anti-tumor properties in CRC. Increased daily intake of omega-3 polyunsaturated fatty acids (PUFAs) significantly increases the density of bacteria that are known to produce butyrate. Omega-3 PUFAs have been proposed as a treatment to prevent gut microbiota dysregulation and lower the risk or progression of CRC.
Assuntos
Neoplasias Colorretais , Ácidos Graxos Ômega-3 , Microbioma Gastrointestinal , Butiratos/farmacologia , Neoplasias Colorretais/patologia , Disbiose , Ácidos Graxos Ômega-3/farmacologia , Microbioma Gastrointestinal/fisiologia , HumanosRESUMO
Detecting malignant pulmonary nodules at an early stage can allow medical interventions which may increase the survival rate of lung cancer patients. Using computer vision techniques to detect nodules can improve the sensitivity and the speed of interpreting chest CT for lung cancer screening. Many studies have used CNNs to detect nodule candidates. Though such approaches have been shown to outperform the conventional image processing based methods regarding the detection accuracy, CNNs are also known to be limited to generalize on under-represented samples in the training set and prone to imperceptible noise perturbations. Such limitations can not be easily addressed by scaling up the dataset or the models. In this work, we propose to add adversarial synthetic nodules and adversarial attack samples to the training data to improve the generalization and the robustness of the lung nodule detection systems. To generate hard examples of nodules from a differentiable nodule synthesizer, we use projected gradient descent (PGD) to search the latent code within a bounded neighbourhood that would generate nodules to decrease the detector response. To make the network more robust to unanticipated noise perturbations, we use PGD to search for noise patterns that can trigger the network to give over-confident mistakes. By evaluating on two different benchmark datasets containing consensus annotations from three radiologists, we show that the proposed techniques can improve the detection performance on real CT data. To understand the limitations of both the conventional networks and the proposed augmented networks, we also perform stress-tests on the false positive reduction networks by feeding different types of artificially produced patches. We show that the augmented networks are more robust to both under-represented nodules as well as resistant to noise perturbations.
Assuntos
Neoplasias Pulmonares , Nódulo Pulmonar Solitário , Detecção Precoce de Câncer , Humanos , Processamento de Imagem Assistida por Computador , Pulmão , Neoplasias Pulmonares/diagnóstico por imagem , Interpretação de Imagem Radiográfica Assistida por Computador , Nódulo Pulmonar Solitário/diagnóstico por imagem , Tomografia Computadorizada por Raios XRESUMO
Polycystic ovary syndrome (PCOS), characterized by oligo-anovulation and androgen excess is considered a high-risk condition for metabolic disorders. Herein, untargeted metabolomics analysis was applied to women with PCOS, aiming to provide deeper insights into lipidomics biomarkers signature of PCOS, for better diagnosis and management. This was a cross-sectional study in which 15 Caucasian women with PCOS and 15 Caucasian healthy, age-matched women were enrolled. Lipidomics analysis was performed using Ultra-High Performance Liquid Chromatography-Quadrupole Time of Flight Electrospray Mass Spectrometry. Partial Least Squares Discriminant Analysis retrieved the most important discriminative metabolites. Significantly increased levels of triacylglycerol (18:2/18:2/0-18:0) in addition to cholestane-3beta, 5alpha, 6beta-triol (18:0/0:0) and cholestane-5alpha (18:1/0:0) appeared as valuable variables to differentiate subjects with PCOS from controls. Acyl-carnitine 2-hydroxylauroylcarnitine was significantly elevated in PCOS in opposition to decreased phosphocholines metabolites (18:1/18:4, 18:3/18:2), to suggest a metabolic pattern linked to lipid peroxidation. A high fat intake or reduced fat energy consumption during nighttime due to diminished ability to switch to lipid oxidation during fasting time possibly contribute to hypertriglyceridemia found in PCOS. Furthermore, inflammatory mediators including metabolites of the prostaglandin (PG) E2 pathway and oxo-leukotrienes (LT) were increased in patients with PCOS. Potential lipidomics biomarkers were identified that could stratify between women with PCOS and healthy controls. The results show particular alterations in acylglycerols, PGs and LTs and phosphocholines and carnitine metabolites. The lipidomics profiles of PCOS indicate a higher risk of developing metabolic diseases.
Assuntos
Doenças Metabólicas/complicações , Síndrome do Ovário Policístico/metabolismo , Adulto , Biomarcadores/metabolismo , Cromatografia Líquida de Alta Pressão , Estudos Transversais , Feminino , Humanos , Lipidômica , Doenças Metabólicas/metabolismo , Metabolômica , Síndrome do Ovário Policístico/complicações , Síndrome do Ovário Policístico/diagnóstico , Medição de Risco , Espectrometria de Massas por Ionização por ElectrosprayRESUMO
Simultaneous segmentation of multiple organs from different medical imaging modalities is a crucial task as it can be utilized for computer-aided diagnosis, computer-assisted surgery, and therapy planning. Thanks to the recent advances in deep learning, several deep neural networks for medical image segmentation have been introduced successfully for this purpose. In this paper, we focus on learning a deep multi-organ segmentation network that labels voxels. In particular, we examine the critical choice of a loss function in order to handle the notorious imbalance problem that plagues both the input and output of a learning model. The input imbalance refers to the class-imbalance in the input training samples (i.e., small foreground objects embedded in an abundance of background voxels, as well as organs of varying sizes). The output imbalance refers to the imbalance between the false positives and false negatives of the inference model. In order to tackle both types of imbalance during training and inference, we introduce a new curriculum learning based loss function. Specifically, we leverage Dice similarity coefficient to deter model parameters from being held at bad local minima and at the same time gradually learn better model parameters by penalizing for false positives/negatives using a cross entropy term. We evaluated the proposed loss function on three datasets: whole body positron emission tomography (PET) scans with 5 target organs, magnetic resonance imaging (MRI) prostate scans, and ultrasound echocardigraphy images with a single target organ i.e., left ventricular. We show that a simple network architecture with the proposed integrative loss function can outperform state-of-the-art methods and results of the competing methods can be improved when our proposed loss is used.
Assuntos
Interpretação de Imagem Assistida por Computador , Processamento de Imagem Assistida por Computador/métodos , Algoritmos , Currículo , Aprendizado Profundo , Educação Médica , Eletrocardiografia , Humanos , Redes Neurais de Computação , Tomografia por Emissão de Pósitrons , Tomografia Computadorizada por Raios X , UltrassonografiaRESUMO
Background: Metabolomics-the novel science that evaluates the multitude of low-molecular-weight metabolites in a biological system, provides new data on pathogenic mechanisms of diseases, including endocrine tumors. Although development of metabolomic profiling in pituitary disorders is at an early stage, it seems to be a promising approach in the near future in identifying specific disease biomarkers and understanding cellular signaling networks. Objectives: To review the metabolomic profile and the contributions of metabolomics in pituitary adenomas (PA). Methods: A systematic review was conducted via PubMed, Web of Science Core Collection and Scopus databases, summarizing studies that have described metabolomic aspects of PA. Results: Liquid chromatography tandem mass spectrometry (LC-MS/MS) and nuclear magnetic resonance (NMR) spectrometry, which are traditional techniques employed in metabolomics, suggest amino acids metabolism appears to be primarily altered in PA. N-acetyl aspartate, choline-containing compounds and creatine appear as highly effective in differentiating PA from healthy tissue. Deoxycholic and 4-pyridoxic acids, 3-methyladipate, short chain fatty acids and glucose-6-phosphate unveil metabolite biomarkers in patients with Cushing's disease. Phosphoethanolamine, N-acetyl aspartate and myo-inositol are down regulated in prolactinoma, whereas aspartate, glutamate and glutamine are up regulated. Phosphoethanolamine, taurine, alanine, choline-containing compounds, homocysteine, and methionine were up regulated in unclassified PA across studies. Intraoperative use of ultra high mass resolution matrix-assisted laser desorption/ionization mass spectrometry imaging (MALDI-MSI), which allows localization and delineation between functional PA and healthy pituitary tissue, may contribute to achievement of complete tumor resection in addition to preservation of pituitary cell lines and vasopressin secretory cells, thus avoiding postoperative diabetes insipidus. Conclusion: Implementation of ultra high performance metabolomics analysis techniques in the study of PA will significantly improve diagnosis and, potentially, the therapeutic approach, by identifying highly specific disease biomarkers in addition to novel molecular pathogenic mechanisms. Ultra high mass resolution MALDI-MSI emerges as a helpful clinical tool in the neurosurgical treatment of pituitary tumors. Therefore, metabolomics appears to be a science with a promising prospect in the sphere of PA, and a starting point in pituitary care.
RESUMO
Antineoplastic targeted therapies, such as EGFR inhibitors, tyrosine kinase inhibitors and BRAF inhibitors, frequently lead to systemic and cutaneous side effects, significantly affecting patient's quality of life. Patients with new targeted therapies have an increased risk of developing skin reactions. The new molecular target therapies developed in the last decades can induce severe skin reactions, which may require dose reduction or discontinuation of treatment and consequently, a decrease in patient's quality of life. The present paper describes toxic cutaneous reactions associated with the most frequently used molecular therapies (epidermal growth factor receptor inhibitors, tyrosine kinase inhibitors, BRAF-inhibitors), frequency of occurrence and methods of diagnosis and treatment, in order to offer a clinically efficient management for maintaining a good quality of life, with compliance to treatment and good therapeutic efficacy. Knowledge of cutaneous adverse reactions in new therapies is mandatory in order to have a proper management of oncologic patients. Recognizing target therapy toxicities by both oncologists and dermatologists, understanding therapeutic mechanisms and choosing optimum treatments for oncologic patients are critical. A correct evaluation of skin toxicity can allow for an adequate decision regarding treatment dose or discontinuation, impacting therapy response and patient survival.
RESUMO
Medical images constitute a source of information essential for disease diagnosis, treatment and follow-up. In addition, due to its patient-specific nature, imaging information represents a critical component required for advancing precision medicine into clinical practice. This manuscript describes recently developed technologies for better handling of image information: photorealistic visualization of medical images with Cinematic Rendering, artificial agents for in-depth image understanding, support for minimally invasive procedures, and patient-specific computational models with enhanced predictive power. Throughout the manuscript we will analyze the capabilities of such technologies and extrapolate on their potential impact to advance the quality of medical care, while reducing its cost.
Assuntos
Diagnóstico por Imagem/tendências , Medicina de Precisão/tendências , Algoritmos , Inteligência Artificial , Sistemas de Apoio a Decisões Clínicas , Diagnóstico por Imagem/economia , Humanos , Procedimentos Cirúrgicos Minimamente InvasivosRESUMO
Treatment of mitral valve (MV) diseases requires comprehensive clinical evaluation and therapy personalization to optimize outcomes. Finite-element models (FEMs) of MV physiology have been proposed to study the biomechanical impact of MV repair, but their translation into the clinics remains challenging. As a step towards this goal, we present an integrated framework for finite-element modeling of the MV closure based on patient-specific anatomies and boundary conditions. Starting from temporal medical images, we estimate a comprehensive model of the MV apparatus dynamics, including papillary tips, using a machine-learning approach. A detailed model of the open MV at end-diastole is then computed, which is finally closed according to a FEM of MV biomechanics. The motion of the mitral annulus and papillary tips are constrained from the image data for increased accuracy. A sensitivity analysis of our system shows that chordae rest length and boundary conditions have a significant influence upon the simulation results. We quantitatively test the generalization of our framework on 25 consecutive patients. Comparisons between the simulated closed valve and ground truth show encouraging results (average point-to-mesh distance: 1.49 ± 0.62 mm) but also the need for personalization of tissue properties, as illustrated in three patients. Finally, the predictive power of our model is tested on one patient who underwent MitralClip by comparing the simulated intervention with the real outcome in terms of MV closure, yielding promising prediction. By providing an integrated way to perform MV simulation, our framework may constitute a surrogate tool for model validation and therapy planning.
Assuntos
Anuloplastia da Valva Mitral/instrumentação , Insuficiência da Valva Mitral/fisiopatologia , Insuficiência da Valva Mitral/cirurgia , Valva Mitral/fisiopatologia , Valva Mitral/cirurgia , Modelos Cardiovasculares , Instrumentos Cirúrgicos , Cateteres Cardíacos , Simulação por Computador , Análise de Falha de Equipamento , Análise de Elementos Finitos , Humanos , Insuficiência da Valva Mitral/diagnóstico , Desenho de Prótese , Ajuste de Prótese , Cirurgia Assistida por Computador/instrumentação , Cirurgia Assistida por Computador/métodos , Integração de Sistemas , Resultado do TratamentoRESUMO
Minimal invasive procedures such as transcatheter valve interventions are substituting conventional surgical techniques. Thus, novel operating rooms have been designed to augment traditional surgical equipment with advanced imaging systems to guide the procedures. We propose a novel method to fuse pre-operative and intra-operative information by jointly estimating anatomical models from multiple image modalities. Thereby high-quality patient-specific models are integrated into the imaging environment of operating rooms to guide cardiac interventions. Robust and fast machine learning techniques are utilized to guide the estimation process. Our method integrates both the redundant and complementary multimodal information to achieve a comprehensive modeling and simultaneously reduce the estimation uncertainty. Experiments performed on 28 patients with pairs of multimodal volumetric data are used to demonstrate high quality intra-operative patient-specific modeling of the aortic valve with a precision of 1.09mm in TEE and 1.73mm in 3D C-arm CT. Within a processing time of 10 seconds we additionally obtain model sensitive mapping between the pre- and intraoperative images.
Assuntos
Valva Aórtica/patologia , Imageamento Tridimensional/métodos , Procedimentos Cirúrgicos Minimamente Invasivos/métodos , Tomografia Computadorizada por Raios X/métodos , Algoritmos , Aorta/patologia , Valva Aórtica/cirurgia , Inteligência Artificial , Cateterismo , Simulação por Computador , Desenho de Equipamento , Humanos , Processamento de Imagem Assistida por Computador/métodos , Modelos Estatísticos , Probabilidade , Reprodutibilidade dos Testes , Propriedades de SuperfícieRESUMO
C-arm CT is an emerging imaging technique in transcatheter aortic valve implantation (TAVI) surgery. Automatic aorta segmentation and valve landmark detection in a C-arm CT volume has important applications in TAVI by providing valuable 3D measurements for surgery planning. Overlaying 3D segmentation onto 2D real time fluoroscopic images also provides critical visual guidance during the surgery. In this paper, we present a part-based aorta segmentation approach, which can handle aorta structure variation in case that the aortic arch and descending aorta are missing in the volume. The whole aorta model is split into four parts: aortic root, ascending aorta, aortic arch, and descending aorta. Discriminative learning is applied to train a detector for each part separately to exploit the rich domain knowledge embedded in an expert-annotated dataset. Eight important aortic valve landmarks (three aortic hinge points, three commissure points, and two coronary ostia) are also detected automatically in our system. Under the guidance of the detected landmarks, the physicians can deploy the prosthetic valve properly. Our approach is robust under variations of contrast agent. Taking about 1.4 seconds to process one volume, it is also computationally efficient.
Assuntos
Aortografia/métodos , Doenças das Valvas Cardíacas/diagnóstico por imagem , Doenças das Valvas Cardíacas/cirurgia , Implante de Prótese de Valva Cardíaca/métodos , Reconhecimento Automatizado de Padrão/métodos , Cirurgia Assistida por Computador/métodos , Tomografia Computadorizada por Raios X/métodos , Aorta/cirurgia , Humanos , Interpretação de Imagem Radiográfica Assistida por Computador/métodos , Reprodutibilidade dos Testes , Sensibilidade e EspecificidadeRESUMO
Pulmonary valve disease affects a significant portion of the global population and often occurs in conjunction with other heart dysfunctions. Emerging interventional methods enable percutaneous pulmonary valve implantation, which constitute an alternative to open heart surgery. As minimal invasive procedures become common practice, imaging and non-invasive assessment techniques turn into key clinical tools. In this paper, we propose a novel approach for intervention planning as well as morphological and functional quantification of the pulmonary trunk and valve. An abstraction of the anatomic structures is represented through a four-dimensional, physiological model able to capture large pathological variation. A hierarchical estimation, based on robust learning methods, is applied to identify the patient-specific model parameters from volumetric CT scans. The algorithm involves detection of piecewise affine parameters, fast centre-line computation and local surface delineation. The estimated personalized model enables for efficient and precise quantification of function and morphology. This ability may have impact on the assessment and surgical interventions of the pulmonary valve and trunk. Experiments performed on 50 cardiac computer tomography sequences demonstrated the average speed of 202 seconds and accuracy of 2.2mm for the proposed approach. An initial clinical validation yielded a significant correlation between model-based and expert measurements. To the best of our knowledge this is the first dynamic model of the pulmonary trunk and right ventricle outflow track estimated from CT data.
Assuntos
Doenças das Valvas Cardíacas/diagnóstico por imagem , Doenças das Valvas Cardíacas/cirurgia , Implante de Prótese de Valva Cardíaca/métodos , Modelos Cardiovasculares , Valva Pulmonar/diagnóstico por imagem , Valva Pulmonar/cirurgia , Cirurgia Assistida por Computador/métodos , Tomografia Computadorizada por Raios X/métodos , Angiografia/métodos , Simulação por Computador , Humanos , Cuidados Pré-Operatórios/métodos , Interpretação de Imagem Radiográfica Assistida por Computador/métodosRESUMO
In many subspecialties of pathology, the intrinsic complexity of rendering accurate diagnostic decisions is compounded by a lack of definitive criteria for detecting and characterizing diseases and their corresponding histological features. In some cases, there exists a striking disparity between the diagnoses rendered by recognized authorities and those provided by non-experts. We previously reported the development of an Image Guided Decision Support (IGDS) system, which was shown to reliably discriminate among malignant lymphomas and leukemia that are sometimes confused with one another during routine microscopic evaluation. As an extension of those efforts, we report here a web-based intelligent archiving subsystem that can automatically detect, image, and index new cells into distributed ground-truth databases. Systematic experiments showed that through the use of robust texture descriptors and density estimation based fusion the reliability and performance of the governing classifications of the system were improved significantly while simultaneously reducing the dimensionality of the feature space.