Efficacy and safety of sirolimus early conversion protocol in liver transplant patients with hepatocellular carcinoma: A single-arm, multicenter, prospective study.

Mammalian target of rapamycin (mTOR) inhibitor as an attractive drug target with promising antitumor effects has been widely investigated. High quality clinical trial has been conducted in liver transplant (LT) recipients in Western countries. However, the pertinent studies in Eastern world are paucity. Therefore, we designed a clinical trial to test whether sirolimus can improve recurrence-free survival (RFS) in hepatocellular carcinoma (HCC) patients beyond the Milan criteria after LT. This is an open-labeled, single-arm, prospective, multicenter, and real-world study aiming to evaluate the clinical outcomes of early switch to sirolimus-based regimens in HCC patients after LT. Patients with a histologically proven HCC and beyond the Milan criteria will be enrolled. The initial immunosuppressant regimens are center-specific for the first 4-6 weeks. The following regimens integrated sirolimus into the regimens as a combination therapy with reduced calcineurin inhibitors based on the condition of patients and centers. The study is planned for 4 years in total with a 2-year enrollment period and a 2-year follow-up. We predict that sirolimus conversion regimen will provide survival benefits for patients particular in the key indicator RFS as well as better quality of life. If the trial is conducted successfully, we will have a continued monitoring over a longer follow-up time to estimate indicator of overall survival. We hope that the outcome will provide better evidence for clinical decision-making and revising treatment guidelines based on Chinese population data. Trial register: Trial registered at http://www.chictr.org.cn: ChiCTR2100042869.

Effect of increased positive end-expiratory pressure on intracranial pressure and cerebral oxygenation: impact of respiratory mechanics and hypovolemia.

BACKGROUND: To evaluate the impact of positive end-expiratory pressure (PEEP) on intracranial pressure (ICP) in animals with different respiratory mechanics, baseline ICP and volume status. METHODS: A total of 50 male adult Bama miniature pigs were involved in four different protocols (n = 20, 12, 12, and 6, respectively). Under the monitoring of ICP, brain tissue oxygen tension and hemodynamical parameters, PEEP was applied in increments of 5 cm H2O from 5 to 25 cm H2O. Measurements were taken in pigs with normal ICP and normovolemia (Series I), or with intracranial hypertension (via inflating intracranial balloon catheter) and normovolemia (Series II), or with intracranial hypertension and hypovolemia (via exsanguination) (Series III). Pigs randomized to the control group received only hydrochloride instillation while the intervention group received additional chest wall strapping. Common carotid arterial blood flow before and after exsanguination at each PEEP level was measured in pigs with intracranial hypertension and chest wall strapping (Series IV). RESULTS: ICP was elevated by increased PEEP in both normal ICP and intracranial hypertension conditions in animals with normal blood volume, while resulted in decreased ICP with PEEP increments in animals with hypovolemia. Increasing PEEP resulted in a decrease in brain tissue oxygen tension in both normovolemic and hypovolemic conditions. The impacts of PEEP on hemodynamical parameters, ICP and brain tissue oxygen tension became more evident with increased chest wall elastance. Compare to normovolemic condition, common carotid arterial blood flow was further lowered when PEEP was raised in the condition of hypovolemia. CONCLUSIONS: The impacts of PEEP on ICP and cerebral oxygenation are determined by both volume status and respiratory mechanics. Potential conditions that may increase chest wall elastance should also be ruled out to avoid the deleterious effects of PEEP.

Advanced zoomed diffusion-weighted imaging vs. full-field-of-view diffusion-weighted imaging in prostate cancer detection: a radiomic features study.

OBJECTIVES: We aimed to compare the efficiency of prostate cancer (PCa) detection using a radiomics signature based on advanced zoomed diffusion-weighted imaging and conventional full-field-of-view DWI. METHODS: A total of 136 patients, including 73 patients with PCa and 63 without PCa, underwent multi-parametric magnetic resonance imaging (mp-MRI). Radiomic features were extracted from prostate lesion areas segmented on full-field-of-view DWI with b-value = 1500 s/mm2 (f-DWIb1500), advanced zoomed DWI images with b-value = 1500 s/mm2 (z-DWIb1500), calculated zoomed DWI with b-value = 2000 s/mm2 (z-calDWIb2000), and apparent diffusion coefficient (ADC) maps derived from both sequences (f-ADC and z-ADC). Single-imaging modality radiomics signature, mp-MRI radiomics signature, and a mixed model based on mp-MRI and clinically independent risk factors were built to predict PCa probability. The diagnostic efficacy and the potential net benefits of each model were evaluated. RESULTS: Both z-DWIb1500 and z-calDWIb2000 had significantly better predictive performance than f-DWIb1500 (z-DWIb1500 vs. f-DWIb1500: p = 0.048; z-calDWIb2000 vs. f-DWIb1500: p = 0.014). z-ADC had a slightly higher area under the curve (AUC) value compared with f-ADC value but was not significantly different (p = 0.127). For predicting the presence of PCa, the AUCs of clinical independent risk factors model, mp-MRI model, and mixed model were 0.81, 0.93, and 0.94 in training sets, and 0.74, 0.92, and 0.93 in validation sets, respectively. CONCLUSION: Radiomics signatures based on the z-DWI technology had better diagnostic accuracy for PCa than that based on the f-DWI technology. The mixed model was better at diagnosing PCa and guiding clinical interventions for patients with suspected PCa compared with mp-MRI signatures and clinically independent risk factors. KEY POINTS: • Advanced zoomed DWI technology can improve the diagnostic accuracy of radiomics signatures for PCa. • Radiomics signatures based on z-calDWIb2000 have the best diagnostic performance among individual imaging modalities. • Compared with the independent clinical risk factors and the mp-MRI model, the mixed model has the best diagnostic efficiency.

[The X-Ray Fluorescence Spectrometer Based on Pyroelectric Effect].

Pyroelectric X-ray generator is implemented, and an X-ray fluorescence spectrometer is accomplished by combining the pyroelectric X-ray generator with a high energy resolution silicon drift detector. Firstly, the parameters of the X-ray generator are decided by analyzing and calculating the influence of the thickness of the pyroelectriccrystal and the thickness of the target on emitted X-ray. Secondly, the emitted X-ray is measured. The energy of emitted X-ray is from 1 to 27 keV, containing the characteristic X-ray of Cu and Ta, and the max counting rate is more than 3 000 per second. The measurement also proves that the detector of the spectrometer has a high energy resolution which the FWMH is 210 eV at 8. 05 keV. Lastly, samples of Fe, Ti, Cr and high-Ti basalt are analyzed using the spectrometer, and the results are agreed with the elements of the samples. It shows that the spectrometer consisting of a pyroelectric X-ray generator and a silicon drift detector is effective for element analysis. Additionally, because each part of the spectrometer has a small volume, it can be easily modified to a portable one which is suitable for non-destructive, on-site and quick element analysis.

Structural characterization of enzymatic products in the dTDP-d-Qui4NFo biosynthetic pathway using electrospray ionization tandem mass spectrometry.

RATIONALE: Structural characterization of biosynthetic precursors is very important in assigning enzymatic function to proteins that have been identified as functional homologs on the basis of sequence homology alone. The objective of this study is to demonstrate the use of electrospray ionization tandem mass spectrometry (ESI-MS/MS) as a powerful technique for the characterization of enzymatic products in the biosynthetic pathway of deoxythymidine 5'-diphosphate-4-formamido-4,6-dideoxy-D-glucose (dTDP-D-Qui4NFo) in Providencia alcalifaciens O30. METHODS: The glucose-1-phosphate thymidyltransferase (RmlA), dTDP-d-glucose 4,6-dehydratase (RmlB), dTDP-4-keto-6-deoxy-d-glucose aminotransferase (VioA), and formyltransferase (VioF) catalyzed reactions were directly monitored by ESI-MS, followed by a detailed structural characterization of the final enzymatic products using ESI-MS/MS in the negative-ion mode after minimal cleanup. RESULTS: The biosynthetic pathway of dTDP-D-Qui4NFo, beginning from α-D-glucose-1-phosphate in four reaction steps catalyzed by RmlA, RmlB, VioA and VioF, was characterized solely by ESI-MS/MS. The results obtained were in good agreement with that of traditional high-performance liquid chromatography (HPLC) monitoring and preparation, as well as nuclear magnetic resonance (NMR) and ESI-MS structural characterization. CONCLUSIONS: MS provides efficient and simple characterization of important unusual dTDP-sugar biosynthetic pathways in the O-chains of bacterial lipopolysaccharides.

[Study on composing prescription laws of treating aplastic anemia by Chinese medicine using applying data mining technique].

OBJECTIVE: To explore composing prescription laws of treating aplastic anemia (AA) by Chinese medicine (CM). METHODS: The literatures on treating AA by CM were recruited from various medical periodicals at home from 1979 to 2009 including China National Knowledge Infrastructure (CNKI), VIP information network, and Wangfang data knowledge service platform. The database correlated to CM features was established using the technique of computer data bank. The data mining (DM) technique was applied to analyze drugs sorts, frequency of drug application, and association degree. RESULTS: Three hundred and eleven pertinent literatures including 677 prescriptions and 254 Chinese herbs (CHs) were screened. There were 69 CHs for invigorating deficiency, 42 for heat clearing, 20 for promoting blood circulation and removing blood stasis, 16 for arresting bleeding, and 16 for relieving exterior syndrome, which occupied the top 5. The frequency of drug application of 254 CHs amounted to 7 547, in which the frequency of drug application of Mongolian milkvetch root, Rehmannia root, Suberect spatholobus stem, Hairyvein agrimonia herb, and Chinese thorowax root were 379, 248, 167, 85, and 13 respectively, and they occupied the first place of CHs for invigorating deficiency, heat clearing, promoting blood circulation and removing blood stasis, arresting bleeding, and relieving exterior syndrome, respectively. The number of the prescriptions containing 12, 10, and 11 CHs was occupied the top 3. The coverage rate of the prescription including Mongolian milkvetch root and Chinese angelica was 60%, and thus 4 core drugs groups were established covering invigorating qi and enriching the blood, reinforcing Shen and supporting yang, replenishing yin to tonify Shen, tonifying Shen to replenish essence, and invigorating qi and enriching blood respectively. Summarized were six potential composing prescription laws covering invigorating qi and enriching blood, reinforcing Shen and supporting yang, replenishing yin to tonify Shen, strengthening Pi and harmonizing Wei, tonifying the blood and promoting blood circulation, clearing away heat and toxic materials, and removing heat from the blood to stop bleeding. CONCLUSIONS: Applying DM technique, the fundamental core drugs groups consisting of Mongolian milkvetch root and Chinese angelica were discovered. The 4 core drugs groups established were in accordance with the realization of modern CM for the pathomechanism of AA. The 6 composing prescription laws summarized revealed the rules of drug application.

Few-Shot Class-Incremental Learning by Sampling Multi-Phase Tasks.

New classes arise frequently in our ever-changing world, e.g., emerging topics in social media and new types of products in e-commerce. A model should recognize new classes and meanwhile maintain discriminability over old classes. Under severe circumstances, only limited novel instances are available to incrementally update the model. The task of recognizing few-shot new classes without forgetting old classes is called few-shot class-incremental learning (FSCIL). In this work, we propose a new paradigm for FSCIL based on meta-learning by LearnIng Multi-phase Incremental Tasks (Limit), which synthesizes fake FSCIL tasks from the base dataset. The data format of fake tasks is consistent with the 'real' incremental tasks, and we can build a generalizable feature space for the unseen tasks through meta-learning. Besides, Limit also constructs a calibration module based on transformer, which calibrates the old class classifiers and new class prototypes into the same scale and fills in the semantic gap. The calibration module also adaptively contextualizes the instance-specific embedding with a set-to-set function. Limit efficiently adapts to new classes and meanwhile resists forgetting over old classes. Experiments on three benchmark datasets (CIFAR100, miniImageNet, and CUB200) and large-scale dataset, i.e., ImageNet ILSVRC2012 validate that Limit achieves state-of-the-art performance.

mRNA therapeutics for disease therapy: principles, delivery, and clinical translation.

Messenger RNA (mRNA) has become a key focus in the development of therapeutic agents, showing significant potential in preventing and treating a wide range of diseases. The COVID-19 pandemic in 2020 has accelerated the development of mRNA nucleic therapeutics and attracted significant investment from global biopharmaceutical companies. These therapeutics deliver genetic information into cells without altering the host genome, making them a promising treatment option. However, their clinical applications have been limited by issues such as instability, inefficient in vivo delivery, and low translational efficiency. Recent advances in molecular design and nanotechnology have helped overcome these challenges, and several mRNA formulations have demonstrated promising results in both animal and human testing against infectious diseases and cancer. This review provides an overview of the latest research progress in structural optimization strategies and delivery systems, and discusses key considerations for their future clinical use.

[ESI fragmentation studies of six unusual nucleotide sugars].

Unusual dTDP-sugars are key intermediate in many pathogenic bacteria. In this study, negative-ion electrospray tandem mass spectrometry (ESI-MS-MS) with collision-induced dissociation (CID) was used to study the fragmentation characteristics of six unusual nucleotide diphosphate sugars. The results indicated the major fragment of the six unusual nucleoside sugars observed in the ESI-MS-MS spectra resulted from cleavage of diphosphate moiety and their characteristic fragment ions at m/z 401, 383, and 321, correspond to [TDP-H] together with fragment ions resulting from the loss of water and phosphate moiety, respectively. Furthermore, 4-position substituted change of unusual sugar rings affected the stability of two important characteristic fragment ions of [glycosyl-1"-PO3](-) and [glycosyl-1"-P2O6](-).

Learning to Classify With Incremental New Class.

New class detection and effective model expansion are of great importance in incremental data mining. In open incremental data environments, data often come with novel classes, e.g., the emergence of new classes in image classification or new topics in opinion monitoring, and is denoted as class-incremental learning (C-IL) in literature. There are two main challenges in C-IL: how to conduct novelty detection and how to update the model with few novel class instances. Most previous methods pay much attention to the former challenge while ignoring the problem of efficiently updating models. To solve this problem, we propose a novel framework to handle the incremental new class, named learning to classify with incremental new class (LC-INC), which can process these two challenges automatically in one unified framework. In detail, LC-INC utilizes a novel structure network to consider the prototype information between class centers of known classes and newly incoming instances, which can dynamically combine the prediction information with structure information to detect novel class instances efficiently. On the other hand, the proposed structure network can also act as a meta-network, which can learn to expand the model much faster and more efficiently with inadequate novel class instances. Experiments on synthetic and real-world datasets successfully validate the effectiveness of our proposed method.

Adversarial training for prostate cancer classification using magnetic resonance imaging.

Background: To use adversarial training to increase the generalizability and diagnostic accuracy of deep learning models for prostate cancer diagnosis. Methods: This multicenter study retrospectively included 396 prostate cancer patients who underwent magnetic resonance imaging (development set, 297 patients from Shanghai Jiao Tong University Affiliated Sixth People's Hospital and Eighth People's Hospital; test set, 99 patients from Renmin Hospital of Wuhan University). Two binary classification deep learning models for clinically significant prostate cancer classification [PM1, pretraining Visual Geometry Group network (VGGNet)-16-based model 1; PM2, pretraining residual network (ResNet)-50-based model 2] and two multiclass classification deep learning models for prostate cancer grading (PM3, pretraining VGGNet-16-based model 3; PM4: pretraining ResNet-50-based model 4) were built using apparent diffusion coefficient and T2-weighted images. These models were then retrained with adversarial examples starting from the initial random model parameters (AM1, adversarial training VGGNet-16 model 1; AM2, adversarial training ResNet-50 model 2; AM3, adversarial training VGGNet-16 model 3; AM4, adversarial training ResNet-50 model 4, respectively). To verify whether adversarial training can improve the diagnostic model's effectiveness, we compared the diagnostic performance of the deep learning methods before and after adversarial training. Receiver operating characteristic curve analysis was performed to evaluate significant prostate cancer classification models. Differences in areas under the curve (AUCs) were compared using Delong's tests. The quadratic weighted kappa score was used to verify the PCa grading models. Results: AM1 and AM2 had significantly higher AUCs than PM1 and PM2 in the internal validation dataset (0.84 vs. 0.89 and 0.83 vs. 0.87) and test dataset (0.73 vs. 0.86 and 0.72 vs. 0.82). AM3 and AM4 showed higher κ values than PM3 and PM4 in the internal validation dataset {0.266 [95% confidence interval (CI): 0.152-0.379] vs. 0.292 (95% CI: 0.178-0.405) and 0.254 (95% CI: 0.159-0.390) vs. 0.279 (95% CI: 0.163-0.396)} and test set [0.196 (95% CI: 0.029-0.362) vs. 0.268 (95% CI: 0.109-0.427) and 0.183 (95% CI: 0.015-0.351) vs. 0.228 (95% CI: 0.068-0.389)]. Conclusions: Using adversarial examples to train prostate cancer classification deep learning models can improve their generalizability and classification abilities.

[Estradiol stimulates proliferation of prostatic smooth muscle cells via estrogen receptor alpha and IGF1].

OBJECTIVE: To investigate the effects of estrogen receptor alpha (ERa) and insulin-like growth factor 1 (IGF1) on the proliferation of prostatic smooth muscle cells (PSMCs) in vitro. METHODS: The ERalpha shRNA expression frame was subcloned to the pGSadeno adenovirus vector by homologous recombination technology to construct the pGSaaeno-ERalpha vector. After the mouse PSMCs were transfected in vitro by pGSaaeno-ERalpha, the mRNA and protein expression levels of ERalpha were detected by RT-PCR and Western blot respectively. The expression of IGF1 in the ERa-reduced cells was determined by Western blot 6 hours after treatment with 17beta-estradiol (E2) at 10(-8) mol/L. The post-transfection activity of estrogen or exogenous IGF1 in the proliferation of PSMCs was evaluated by MTT chlormetric analysis. RESULTS: After treatment with E2, the proliferation of PSMCs and the expression of the IGF1 gene were significantly increased in the normal control group (P <0.05), but not obviously changed in the ERalpha-siRNA group (P> 0.05). And exogenous IGF1 failed to induce the proliferation of the ERalpha-reduced PSMCs. CONCLUSION: E2 induces the expression of IGF1 via ERalpha, and IGFl, with the interaction of ERalpha, promotes the proliferation of PSMCs.

Calculation of Apparent Diffusion Coefficients in Prostate Cancer Using Deep Learning Algorithms: A Pilot Study.

BACKGROUND: Apparent diffusion coefficients (ADCs) obtained with diffusion-weighted imaging (DWI) are highly valuable for the detection and staging of prostate cancer and for assessing the response to treatment. However, DWI suffers from significant anatomic distortions and susceptibility artifacts, resulting in reduced accuracy and reproducibility of the ADC calculations. The current methods for improving the DWI quality are heavily dependent on software, hardware, and additional scan time. Therefore, their clinical application is limited. An accelerated ADC generation method that maintains calculation accuracy and repeatability without heavy dependence on magnetic resonance imaging scanners is of great clinical value. OBJECTIVES: We aimed to establish and evaluate a supervised learning framework for synthesizing ADC images using generative adversarial networks. METHODS: This prospective study included 200 patients with suspected prostate cancer (training set: 150 patients; test set #1: 50 patients) and 10 healthy volunteers (test set #2) who underwent both full field-of-view (FOV) diffusion-weighted imaging (f-DWI) and zoomed-FOV DWI (z-DWI) with b-values of 50, 1,000, and 1,500 s/mm2. ADC values based on f-DWI and z-DWI (f-ADC and z-ADC) were calculated. Herein we propose an ADC synthesis method based on generative adversarial networks that uses f-DWI with a single b-value to generate synthesized ADC (s-ADC) values using z-ADC as a reference. The image quality of the s-ADC sets was evaluated using the peak signal-to-noise ratio (PSNR), root mean squared error (RMSE), structural similarity (SSIM), and feature similarity (FSIM). The distortions of each ADC set were evaluated using the T2-weighted image reference. The calculation reproducibility of the different ADC sets was compared using the intraclass correlation coefficient. The tumor detection and classification abilities of each ADC set were evaluated using a receiver operating characteristic curve analysis and a Spearman correlation coefficient. RESULTS: The s-ADCb1000 had a significantly lower RMSE score and higher PSNR, SSIM, and FSIM scores than the s-ADCb50 and s-ADCb1500 (all P < 0.001). Both z-ADC and s-ADCb1000 had less distortion and better quantitative ADC value reproducibility for all the evaluated tissues, and they demonstrated better tumor detection and classification performance than f-ADC. CONCLUSION: The deep learning algorithm might be a feasible method for generating ADC maps, as an alternative to z-ADC maps, without depending on hardware systems and additional scan time requirements.

Synthesizing High-b-Value Diffusion-weighted Imaging of the Prostate Using Generative Adversarial Networks.

PURPOSE: To develop and evaluate a diffusion-weighted imaging (DWI) deep learning framework based on the generative adversarial network (GAN) to generate synthetic high-b-value (b =1500 sec/mm2) DWI (SYNb1500) sets from acquired standard-b-value (b = 800 sec/mm2) DWI (ACQb800) and acquired standard-b-value (b = 1000 sec/mm2) DWI (ACQb1000) sets. MATERIALS AND METHODS: This retrospective multicenter study included 395 patients who underwent prostate multiparametric MRI. This cohort was split into internal training (96 patients) and external testing (299 patients) datasets. To create SYNb1500 sets from ACQb800 and ACQb1000 sets, a deep learning model based on GAN (M0) was developed by using the internal dataset. M0 was trained and compared with a conventional model based on the cycle GAN (Mcyc). M0 was further optimized by using denoising and edge-enhancement techniques (optimized version of the M0 [Opt-M0]). The SYNb1500 sets were synthesized by using the M0 and the Opt-M0 were synthesized by using ACQb800 and ACQb1000 sets from the external testing dataset. For comparison, traditional calculated (b =1500 sec/mm2) DWI (CALb1500) sets were also obtained. Reader ratings for image quality and prostate cancer detection were performed on the acquired high-b-value (b = 1500 sec/mm2) DWI (ACQb1500), CALb1500, and SYNb1500 sets and the SYNb1500 set generated by the Opt-M0 (Opt-SYNb1500). Wilcoxon signed rank tests were used to compare the readers' scores. A multiple-reader multiple-case receiver operating characteristic curve was used to compare the diagnostic utility of each DWI set. RESULTS: When compared with the Mcyc, the M0 yielded a lower mean squared difference and higher mean scores for the peak signal-to-noise ratio, structural similarity, and feature similarity (P < .001 for all). Opt-SYNb1500 resulted in significantly better image quality (P ≤ .001 for all) and a higher mean area under the curve than ACQb1500 and CALb1500 (P ≤ .042 for all). CONCLUSION: A deep learning framework based on GAN is a promising method to synthesize realistic high-b-value DWI sets with good image quality and accuracy in prostate cancer detection.Keywords: Prostate Cancer, Abdomen/GI, Diffusion-weighted Imaging, Deep Learning Framework, High b Value, Generative Adversarial Networks© RSNA, 2021 Supplemental material is available for this article.

Study of phase coherence degradation induced by a tapered semiconductor amplifier with frequency-modulated continuous-wave and pulsed seed lasers.

We investigate the phase coherence between a seed laser and a laser amplified by a tapered semiconductor amplifier (TSA) when the seed laser is either continuous wave (CW) or pulsed. The phase fluctuations in the time domain are employed to describe the degradation of phase coherence induced by a TSA. The amplified laser is measured to be approximately 99.98% coherent with the seed, when the CW or pulsed laser is seeded, at different supplying currents of the TSA. Furthermore, the phase coherence is measured when the seed laser is modulated. The results reveal that the phase coherence degradations induced by the TSA remain the same for a seed laser with and without modulation, when different supplying currents of the TSA are applied.

[Liquid chromatography frontal analysis of the protein binding of glimepiride].

AIM: To study the protein binding of glimepiride. METHODS: An HPLC-FA method is performed by using Pinkerton GFF II-S5-80 internal-surface reversed-phase silica support (150 mm x 4.6 mm ID, 5 microm) at pH 7.4 in a 67 mmol x L(-1) isotonic sodium phosphate buffer at 37 degree C. Other conditions included flow rate of 0.2 mL x min(-1), UV detection at wavelength 230 nm and injection volume 900 microL. RESULTS: Nonlinear regression parameter estimation was used for the association constant measurement of glimepiride to both primary and secondary sites, which were 5.1 (micromol x L(-1)-1 and 1 for K1 and n1, and 0.017 (micromol x L(-1))-1 and 7 for K2 and n2, respectively. CONCLUSION: The method is shown to be suitable for investigation of protein binding of glimepiride.

Random forests-based differential analysis of gene sets for gene expression data.

In DNA microarray studies, gene-set analysis (GSA) has become the focus of gene expression data analysis. GSA utilizes the gene expression profiles of functionally related gene sets in Gene Ontology (GO) categories or priori-defined biological classes to assess the significance of gene sets associated with clinical outcomes or phenotypes. Many statistical approaches have been proposed to determine whether such functionally related gene sets express differentially (enrichment and/or deletion) in variations of phenotypes. However, little attention has been given to the discriminatory power of gene sets and classification of patients. In this study, we propose a method of gene set analysis, in which gene sets are used to develop classifications of patients based on the Random Forest (RF) algorithm. The corresponding empirical p-value of an observed out-of-bag (OOB) error rate of the classifier is introduced to identify differentially expressed gene sets using an adequate resampling method. In addition, we discuss the impacts and correlations of genes within each gene set based on the measures of variable importance in the RF algorithm. Significant classifications are reported and visualized together with the underlying gene sets and their contribution to the phenotypes of interest. Numerical studies using both synthesized data and a series of publicly available gene expression data sets are conducted to evaluate the performance of the proposed methods. Compared with other hypothesis testing approaches, our proposed methods are reliable and successful in identifying enriched gene sets and in discovering the contributions of genes within a gene set. The classification results of identified gene sets can provide an valuable alternative to gene set testing to reveal the unknown, biologically relevant classes of samples or patients. In summary, our proposed method allows one to simultaneously assess the discriminatory ability of gene sets and the importance of genes for interpretation of data in complex biological systems. The classifications of biologically defined gene sets can reveal the underlying interactions of gene sets associated with the phenotypes, and provide an insightful complement to conventional gene set analyses.