首页 | 本学科首页   官方微博 | 高级检索  
相似文献
 共查询到20条相似文献，搜索用时 15 毫秒
1.
    
PurposeHighlighting the risk of biases in radiomics-based models will help improve their quality and increase usage as decision support systems in the clinic. In this study we use machine learning-based methods to identify the presence of volume-confounding effects in radiomics features.Methods841 radiomics features were extracted from two retrospective publicly available datasets of lung and head neck cancers using open source software. Unsupervised hierarchical clustering and principal component analysis (PCA) identified relations between radiomics and clinical outcomes (overall survival). Bootstrapping techniques with logistic regression verified features’ prognostic power and robustness.ResultsOver 80% of the features had large pairwise correlations. Nearly 30% of the features presented strong correlations with tumor volume. Using volume-independent features for clustering and PCA did not allow risk stratification of patients. Clinical predictors outperformed radiomics features in bootstrapping and logistic regression.ConclusionsThe adoption of safeguards in radiomics is imperative to improve the quality of radiomics studies. We proposed machine learning (ML) – based methods for robust radiomics signatures development.  相似文献   

2.
    
  相似文献   

3.
    
Accurate identification of compound–protein interactions(CPIs) in silico may deepen our understanding of the underlying mechanisms of drug action and thus remarkably facilitate drug discovery and development.Conventional similarity-or docking-based computational methods for predicting CPIs rarely exploit latent features from currently available large-scale unlabeled compound and protein data and often limit their usage to relatively small-scale datasets.In the present study,we propose Deep CPI,a novel general and scalable computational framework that combines effective feature embedding(a technique of representation learning) with powerful deep learning methods to accurately predict CPIs at a large scale.Deep CPI automatically learns the implicit yet expressive low-dimensional features of compounds and proteins from a massive amount of unlabeled data.Evaluations of the measured CPIs in large-scale databases,such as Ch EMBL and Binding DB,as well as of the known drug–target interactions from Drug Bank,demonstrated the superior predictive performance of Deep CPI.Furthermore,several interactions among smallmolecule compounds and three G protein-coupled receptor targets(glucagon-like peptide-1 receptor,glucagon receptor,and vasoactive intestinal peptide receptor) predicted using Deep CPI were experimentally validated.The present study suggests that Deep CPI is a useful and powerful tool for drug discovery and repositioning.The source code of Deep CPI can be downloaded from https://github.com/Fangping Wan/Deep CPI.  相似文献   

4.
    
Optical microscopy has emerged as a key driver of fundamental research since it provides the ability to probe into imperceptible structures in the biomedical world. For the detailed investigation of samples, a high-resolution image with enhanced contrast and minimal damage is preferred. To achieve this, an automated image analysis method is preferable over manual analysis in terms of both speed of acquisition and reduced error accumulation. In this regard, deep learning (DL)-based image processing can be highly beneficial. The review summarises and critiques the use of DL in image processing for the data collected using various optical microscopic techniques. In tandem with optical microscopy, DL has already found applications in various problems related to image classification and segmentation. It has also performed well in enhancing image resolution in smartphone-based microscopy, which in turn enablse crucial medical assistance in remote places.Graphical abstract  相似文献   

5.
    
Artificial intelligence (AI) has recently become a very popular buzzword, as a consequence of disruptive technical advances and impressive experimental results, notably in the field of image analysis and processing. In medicine, specialties where images are central, like radiology, pathology or oncology, have seized the opportunity and considerable efforts in research and development have been deployed to transfer the potential of AI to clinical applications. With AI becoming a more mainstream tool for typical medical imaging analysis tasks, such as diagnosis, segmentation, or classification, the key for a safe and efficient use of clinical AI applications relies, in part, on informed practitioners. The aim of this review is to present the basic technological pillars of AI, together with the state-of-the-art machine learning methods and their application to medical imaging. In addition, we discuss the new trends and future research directions. This will help the reader to understand how AI methods are now becoming an ubiquitous tool in any medical image analysis workflow and pave the way for the clinical implementation of AI-based solutions.  相似文献   

6.
    
IntroductionDeep learning (DL) is used to classify, detect, and quantify gold nanoparticles (AuNPs) in a human-sized phantom with a clinical MDCT scanner.MethodsAuNPs were imaged at concentrations between 0.0274 and 200 mgAu/mL in a 33 cm phantom. 1 mm-thick CT image slices were acquired at 120 kVp with a CTDIvol of 23.6 mGy. A convolutional neural network (CNN) was trained on 544 images to classify 17 different tissue types and AuNP concentrations. A second set of 544 images was then used for testing.ResultsAuNPs were classified with 95% accuracy at 0.1095 mgAu/mL and 97% accuracy at 0.2189 mgAu/mL. Both these concentrations are lower than what humans can visually perceive (0.3–1.4 mgAu/mL). AuNP concentrations were also classified with 95% accuracy at 150 and 200 mgAu/mL. These high concentrations result in CT numbers that are at or above the 12-bit limit for CT’s dynamic range where extended Hounsfield scales are otherwise required for measuring differences in contrast.ConclusionsWe have shown that DL can be used to detect AuNPs at concentrations lower than what humans can visually perceive and can also quantify very high AuNP concentrations that exceed the typical 12-bit dynamic range of clinical MDCT scanners. This second finding is possible due to inhomogeneous AuNP distributions and characteristic streak artifacts. It may even be possible to extend this approach beyond AuNP imaging in CT for quantifying high density objects without extended Hounsfield scales.  相似文献   

7.
    
AimWe conducted a study to validate the MDASI-HN based nomogram, which is used to predict the acute toxicities in head and neck cancer patients undergoing radiation therapy with or without chemotherapy.BackgroundTolerance to radiation varies from patient to patient and also depends on various other factors like tumor volume, dose of radiation, chemotherapy. Predicting the toxicities allow us to identify potential candidates who are likely to have a higher toxicity and, in addition, evaluates the nomogram when done on an independent group of patients.Materials and MethodsSixty biopsy confirmed head and neck cancer patients undergoing radiation were the subjects of the study. The patients completed patient reported outcome instrument (PRO) MDASI-HN questionnaire at the beginning and at the fifth week of radiation. The baseline score obtained was used to obtain the predicted score using nomogram. The nomogram was also externally validated as per the TRIPOD guidelines.ResultsThe mean baseline, predicted and score at the fifth week were 27.28 ± 11.04, 73.33 ± 15.51 and 82.62 ± 17.67, respectively, for all sub-sites. A positive, significant correlation (p < 0.01) between the predicted score and the score at the fifth week was seen across all sub sites such as Oral cavity (p = 0.05), Oropharynx (p = 0.02), Hypo pharynx (p = 0.02) and Larynx (p = 0.02).ConclusionThe MDASI-HN questionnaire based nomogram is simple, easily doable and takes into consideration the initial symptoms as well the treatment details; thereby, it is able to predict the toxicities accurately.  相似文献   

8.
    
PurposeArtificial intelligence (AI) models are playing an increasing role in biomedical research and healthcare services. This review focuses on challenges points to be clarified about how to develop AI applications as clinical decision support systems in the real-world context.MethodsA narrative review has been performed including a critical assessment of articles published between 1989 and 2021 that guided challenging sections.ResultsWe first illustrate the architectural characteristics of machine learning (ML)/radiomics and deep learning (DL) approaches. For ML/radiomics, the phases of feature selection and of training, validation, and testing are described. DL models are presented as multi-layered artificial/convolutional neural networks, allowing us to directly process images. The data curation section includes technical steps such as image labelling, image annotation (with segmentation as a crucial step in radiomics), data harmonization (enabling compensation for differences in imaging protocols that typically generate noise in non-AI imaging studies) and federated learning. Thereafter, we dedicate specific sections to: sample size calculation, considering multiple testing in AI approaches; procedures for data augmentation to work with limited and unbalanced datasets; and the interpretability of AI models (the so-called black box issue). Pros and cons for choosing ML versus DL to implement AI applications to medical imaging are finally presented in a synoptic way.ConclusionsBiomedicine and healthcare systems are one of the most important fields for AI applications and medical imaging is probably the most suitable and promising domain. Clarification of specific challenging points facilitates the development of such systems and their translation to clinical practice.  相似文献   

9.
Summary The therapeutic effects of intravenous recombinant human erythropoietin (r-hEPO) administration on anemia induced by radiation therapy (3 cases), chemotherapy (18 cases) and combined therapies (5 cases) in patients with head and neck malignancies were examined. The effectiveness was evaluated by the changes in the hemoglobin concentration examined before and after the r-hEPO administration. The r-hEPO administration combined with anticancer therapies improved anemia induced by all three treatments. The therapeutic effectiveness of r-hEPO injection was also noted on anemia induced by all of four different chemotherapeutic regimens that have been ordinarily used for head and neck malignancies. Furthermore, the efficacy of the different dose schedules, 3000 IU (12 cases) or 6000 IU (14 cases), three times a week, was compared. Both of the r-hEPO dose schedules were effective for anemia, but the efficacy of 6000 IU was superior to that of 3000 IU. No significant changes were observed in the number of white blood cells and platelets and the results of biochemical examinations after the r-hEPO injection. There were no objective side-effects related to the r-hEPO administration. These results suggest that anemia induced by chemotherapy and/or radiotherapy could be prevented by r-hEPO administration. The addition of r-hEPO to anticancer therapies would make it possible to pursue the planned therapeutic schedules, prevent the decrease of immunity after allogeneic blood transfusion and bring about an improvement in the prognosis of patients with malignancies.  相似文献   

10.
    
PurposeTo perform a systematic review on the research on the application of artificial intelligence (AI) to imaging published in Italy and identify its fields of application, methods and results.Materials and MethodsA Pubmed search was conducted using terms Artificial Intelligence, Machine Learning, Deep learning, imaging, and Italy as affiliation, excluding reviews and papers outside time interval 2015–2020. In a second phase, participants of the working group AI4MP on Artificial Intelligence of the Italian Association of Physics in Medicine (AIFM) searched for papers on AI in imaging.ResultsThe Pubmed search produced 794 results. 168 studies were selected, of which 122 were from Pubmed search and 46 from the working group. The most used imaging modality was MRI (44%) followed by CT(12%) ad radiography/mammography (11%). The most common clinical indication were neurological diseases (29%) and diagnosis of cancer (25%). Classification was the most common task for AI (57%) followed by segmentation (16%). 65% of studies used machine learning and 35% used deep learning. We observed a rapid increase of research in Italy on artificial intelligence in the last 5 years, peaking at 155% from 2018 to 2019.ConclusionsWe are witnessing an unprecedented interest in AI applied to imaging in Italy, in a diversity of fields and imaging techniques. Further initiatives are needed to build common frameworks and databases, collaborations among different types of institutions, and guidelines for research on AI.  相似文献   

11.
Head and neck cancer represents a challenging disease. Despite recent treatment advances, which have improved functional outcomes, the long-term survival of head and neck cancer patients has remained unchanged for the past 25 years. One of the goals of adjuvant cancer therapy is to eradicate local regional microscopic and micrometastatic disease with minimal toxicity to surrounding normal cells. In this respect, antigen-specific immunotherapy is an attractive therapeutic approach. With the advances in molecular genetics and fundamental immunology, antigen-specific immunotherapy is being actively explored using DNA, bacterial vector, viral vector, peptide, protein, dendritic cell, and tumor-cell based vaccines. Early phase clinical trials have demonstrated the safety and feasibility of these novel therapies and the emphasis is now shifting towards the development of strategies, which can increase the potency of these vaccines. As the field of immunotherapy matures and as our understanding of the complex interaction between tumor and host develops, we get closer to realizing the potential of immunotherapy as an adjunctive method to control head and neck cancer and improve long-term survival in this patient population.  相似文献   

12.
目前临床普遍采用功能与分子影像检测手段能来评价头颈部肿瘤的放射治疗计划和疗效,可指导个体化治疗从而提高疗效。文章概述了功能与分子影像技术CT,MRI,PET-CT,超声检测技术在头颈部肿瘤放射治疗计划制定和疗效评价中的应用进展。结果显示,不同分子影像检测方法如在检查时机的选择、诊断和鉴别诊断的价值、观察放射治疗后肿瘤的残存和复发、预测放射治疗效果、指导后续治疗等方面均可起到重要作用。采用图像融合技术进行联合应用,如PET-CT和MRI-CT等,可提高检测的准确率。临床医生需在常规影像学手段的基础上,根据头颈部肿瘤患者病情和治疗方法的不同选用正确的功能和分子影像检测手段,更好地指导制定放射治疗计划及综合评价放射治疗后的疗效。  相似文献   

13.
    
Papermaking wastewater accounts for a large proportion of industrial wastewater, and it is essential to obtain accurate and reliable effluent indices in real-time. Considering the complexity, nonlinearity, and time variability of wastewater treatment processes, a dynamic kernel extreme learning machine (DKELM) method is proposed to predict the key quality indices of effluent chemical oxygen demand (COD). A time lag coefficient is introduced and a kernel function is embedded into the extreme learning machine (ELM) to extract dynamic information and obtain better prediction accuracy. A case study for modeling a wastewater treatment process is demonstrated to evaluate the performance of the proposed DKELM. The results illustrate that both training and prediction accuracy of the DKELM model is superior to other models. For the prediction of the quality indices of effluent COD, the determinate coefficient of the DKELM model is increased by 27.52 %, 21.36 %, 10.42 %, and 10.81 %, compared with partial least squares, ELM, dynamic ELM, and kernel ELM, respectively.  相似文献   

14.
    
This review presents a modern perspective on dynamical systems in the context of current goals and open challenges. In particular, our review focuses on the key challenges of discovering dynamics from data and finding data-driven representations that make nonlinear systems amenable to linear analysis. We explore various challenges in modern dynamical systems, along with emerging techniques in data science and machine learning to tackle them. The two chief challenges are (1) nonlinear dynamics and (2) unknown or partially known dynamics. Machine learning is providing new and powerful techniques for both challenges. Dimensionality reduction methods are used for projecting dynamical methods in reduced form, and these methods perform computational efficiency on real-world data. Data-driven models drive to discover the governing equations and give laws of physics. The identification of dynamical systems through deep learning techniques succeeds in inferring physical systems. Machine learning provides advanced new and powerful algorithms for nonlinear dynamics. Advanced deep learning methods like autoencoders, recurrent neural networks, convolutional neural networks, and reinforcement learning are used in modeling of dynamical systems.  相似文献   

15.
    
Intrinsically disordered or unstructured proteins (or regions in proteins) have been found to be important in a wide range of biological functions and implicated in many diseases. Due to the high cost and low efficiency of experimental determination of intrinsic disorder and the exponential increase of unannotated protein sequences, developing complementary computational prediction methods has been an active area of research for several decades. Here, we employed an ensemble of deep Squeeze-and-Excitation residual inception and long short-term memory (LSTM) networks for predicting protein intrinsic disorder with input from evolutionary information and predicted one-dimensional structural properties. The method, called SPOT-Disorder2, offers substantial and consistent improvement not only over our previous technique based on LSTM networks alone, but also over other state-of-the-art techniques in three independent tests with different ratios of disordered to ordered amino acid residues, and for sequences with either rich or limited evolutionary information. More importantly, semi-disordered regions predicted in SPOT-Disorder2 are more accurate in identifying molecular recognition features (MoRFs) than methods directly designed for MoRFs prediction. SPOT-Disorder2 is available as a web server and as a standalone program at https://sparks-lab.org/server/spot-disorder2/.  相似文献   

16.
    
Prediction of protein structure from sequence has been intensely studied for many decades, owing to the problem's importance and its uniquely well-defined physical and computational bases. While progress has historically ebbed and flowed, the past two years saw dramatic advances driven by the increasing “neuralization” of structure prediction pipelines, whereby computations previously based on energy models and sampling procedures are replaced by neural networks. The extraction of physical contacts from the evolutionary record; the distillation of sequence–structure patterns from known structures; the incorporation of templates from homologs in the Protein Databank; and the refinement of coarsely predicted structures into finely resolved ones have all been reformulated using neural networks. Cumulatively, this transformation has resulted in algorithms that can now predict single protein domains with a median accuracy of 2.1 Å, setting the stage for a foundational reconfiguration of the role of biomolecular modeling within the life sciences.  相似文献   

17.
目的通过对传统培养法和PCR法在假丝酵母菌感染检出率的比较,拟探索一种能够早期、快速、高效检测头颈部放疗患者假丝酵母菌感染的方法。方法收集120名头颈部放疗患者唾液,分别应用假丝酵母菌显色培养基（CHROMagar）进行分离、培养和鉴定;同时提取基因组DNA,通过假丝酵母菌通用引物、特异性引物、改良引物进行PCR扩增,结果与假丝酵母菌表型进行对比。结果与传统培养法相比,PCR法检出率更高（χ2=47.672,P=0.000）;改良特异性引物D扩增的检出率为77%,高于通用引物B（χ2=7.702,P=0.006）和特异性引物C（χ2=12.522,P=0.001）。结论本研究证实PCR技术耗时短,阳性检出率高,可用于头颈部肿瘤放疗患者假丝酵母菌感染的快速检测。  相似文献   

18.
    
BackgroundThere is no clinically applicable tumor marker for head and neck cancers. Telomerase is detected in approximately 90% of all malignant tumors, it may predict poor or favorable outcomes, thus being both a highly attractive biomarker and a target for the development of molecular-based cancer diagnostics, prognostics, and therapeuticsAimPrimary aim was to detect a change of telomerase activity before and after curative treatment.Materials and MethodsPatients with biopsy proven head and neck squamous cell carcinoma, stage I-IVB treated with a curative intent, performance status 0–2 and malignancy at one primary site were included in the study. Telomerase levels were tested in tissue biopsy. Plasma telomerase levels were tested at baseline, 5 days and at 3 months after treatment using ELISA.ResultsRaised plasma telomerase activity was seen in all the patients with cancer at baseline. The mean plasma telomerase level at baseline was 861.4522 ng/ml, at 5 days after completion of curative treatment was 928.92 ng/ml and at 3 months of follow up was 898.87 ng/ml. The mean tissue biopsy telomerase level was 19768.53 ng/mg. There was a significant increase in baseline telomerase levels in cancer patients compared to normals (volunteers) (t = −3.52, p = 0.001).There was a significant increase in plasma levels of telomerase at 3 months compared to baseline values (z = −1.98, p = 0.04). The increase in telomerase level did not correlate with the response of the treatment.ConclusionIn patients with head and neck squamous cell carcinomas treated with a curative intent, the change in levels of telomerase correlates neither with the disease status nor with prognostic factors.  相似文献   

19.
    
ObjectiveTo investigate potential associations between body mass index (BMI) and head and neck cancer (HNC) risk in an East Asian population.MethodsWe conducted a hospital-based multicenter case-control study in East Asia including 921 cases and 806 controls. We estimated the odds ratios (ORs) and 95% confidence intervals (95% CI) for HNC risks by using logistic regression, adjusting on potential confounders.ResultsCompared to normal BMI at interview (18.5–<25 kg/m2), being underweight (BMI < 18.5 kg/m2) was associated with a higher HNC risk (OR = 2.71, 95% CI 1.40–5.26). Additionally, obesity (BMI > 30 kg/m2) was associated with a lower HNC risk (OR = 0.30, 95% CI 0.16–0.57). Being underweight at age 20 was also associated with an increased risk of HNC. However, being underweight at 5 years or 2 years before interview was not associated with a higher risk of HNC.ConclusionWe observed an inverse association between BMI and HNC risk, which is consistent with previous studies in other geographic regions. Being underweight at age 20 was also associated with a higher risk of HNC, suggesting that reverse causality was not the main source of the association.  相似文献   

20.
    
PurposeTo investigate the changes in quality of the volumetric modulated arc therapy (VMAT) plans with couch-shift between arcs by half of a multi-leaf collimator (MLC) leaf width.MethodsA total of 22 patients with head-and-neck cancer were retrospectively selected. Since the smallest MLC leaf width was 5 mm in this study, the couch was shifted by 2.5 mm in the longitudinal-direction between arcs to increase the resolution of fluence map. A total of three types of VMAT plans were generated for each patient; the three types of plans were a two-full-arc plan without couch-shift (NS plan), a two-half-arc-pair plan with couch-shift (HAS plan), and a two-full-arc pair plan with couch-shift (FAS plan). Changes in the dose-volumetric parameters were investigated.ResultsThe FAS plan showed the best plan quality for the target volumes and organs at risk compared to the NS and HAS plans. However, the magnitudes of differences among the three types of plans were minimal, and every plan was clinically acceptable. The average integral doses of the NS, HAS, and FAS plans were 160,549 ± 37,600 Gy-cc, 147,828 ± 33,343 Gy-cc, and 156,030 ± 36,263 Gy-cc, respectively. The average monitor unit of the NS, HAS, and FAS plans were 717 ± 120 MU, 648 ± 100 MU, and 763 ± 158 MU, respectively.ConclusionsThe HAS plan was better than the others in terms of normal tissue sparing and plan efficiency. By shifting the couch by half of the MLC leaf width in the longitudinal direction between arcs, the VMAT plan quality could be improved.  相似文献   

设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号