High performance computing and networking in medical imaging applications.
Explore the source record for details and available documents.
Algorithms↗
SEARCH · PubMed Health
Results for “Performance benchmarking”
Explore indexed PubMed citations for clinical trials, systematic reviews and public health research. Read source abstracts and follow each citation to its original PubMed record.
Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.
Identifying wrist fracture patients with high accuracy by automatic categorization of X-ray reports.
The authors performed this study to determine the accuracy of several text classification methods to categorize wrist x-ray reports. We randomly sampled 751 textual wrist x-ray reports. Two expert reviewers rated the presence (n = 301) or absence (n = 450) of an acute fracture of wrist. We developed two information retrieval (IR) text classification methods and a machine learning method using a support vector machine (TC-1). In cross-validation on the derivation set (n = 493), TC-1 outperformed the two IR based methods and six benchmark classifiers, including Naive Bayes and a Neural Network. In the validation set (n = 258), TC-1 demonstrated consistent performance with 93.8% accuracy; 95.5% sensitivity; 92.9% specificity; and 87.5% positive predictive value. TC-1 was easy to implement and superior in performance to the other classification methods.
Artificial Intelligence↗
Systematic evaluation of metatranscriptomic differential gene expression in silico, in vitro, and in vivo enables elucidation of inter-species cross-feeding.
Metatranscriptomic (MTX) sequencing quantifies gene expression from the collective genomes of microbial communities (microbiomes), enabling assessment of functional activity rather than functional potential. While differential expression testing is instrumental to RNA-sequencing analysis, current metatranscriptomic approaches have been benchmarked only on simulated data and not under real operating conditions, resulting in a lack of standard practices. Here, we evaluate the performance of statistical differential expression methods on both simulated datasets and data collected from real bacterial 'mock communities' designed for this purpose. We assess the robustness of individual methods to organisms' low relative abundance, differential abundance, low prevalence, and transcription rate changes, showing that no existing methods perform adequately across all confounding conditions. We then apply the same approaches to metatranscriptomic datasets generated from gnotobiotic mice colonized with defined consortia of human bacterial strains and show that the method nominated by our mock community comparisons successfully inferred cross-feeding dynamics which were validated in vitro. We conclude that MTX method benchmarking on real, not simulated, datasets can and should optimize model implementation, enabling inference and validation of cross-feeding and other inter-species and host-microbe dynamics from in vivo studies.
Journal Article↗
Benchmarking: the key to influencing physicians.
Managing physicians to achieve cost reductions can seem impossible, especially when managed care penetration is low. Physicians feel little pressure to change when asked merely to cut costs, especially when their boat is not rocking. But physicians will respond to benchmarking data on CPT-coded procedures that are directly comparable to their own practices. When surgeons see that others take less time to perform a procedure and/or use fewer and lower cost supplies, their competitive spirits are aroused. They become inquisitive about why this is so and then are eager to change by trying new methods and improving their techniques. Science is the key motivator, not savings. When benchmarking recommendations are implemented in a facility, better practice and substantial cost savings are the positive results.
Benchmarking↗
Outcome of first-trimester chorionic villus sampling for genetic investigation in multiple pregnancy.
OBJECTIVES: To evaluate the efficacy of and risk associated with chorionic villus sampling for genetic investigations in multiple pregnancies, and to evaluate the accuracy of the ultrasonographic detection of chorionicity during the first trimester. PATIENTS AND METHODS: A total of 198 sets of twins and nine sets of triplets from 10 087 consecutive first-trimester pregnancies undergoing chorionic villus sampling were considered. Gestational age ranged from 7 to 12.6 weeks. Assessment and confirmation of chorionicity was based on a multiplicity of features. Dichorionicity was established in 169 sets of twins (85.3%) and trichorionicity in all triplet cases, while 29 twins were considered monochorionic. Chorionic villus sampling was performed transabdominally in all but one case, and identification of the placental insertion of the umbilical cord was the main benchmark for sampling. Sampling risks were evaluated by comparing clinical outcome with that of a control population of 63 dichorionic twin pregnancies which underwent no invasive procedure. RESULTS: Determination of the presence or absence of the lambda sign led to a correct assignment of chorionicity in all cases, while the presence of a membrane thickness of 2 mm or more reflected a 100% specificity with a 22% false negative rate. Sampling was successfully performed in all cases and in only four cases (1.0%) were two needle insertions needed. At follow-up no evidence of incorrect sampling was reported. Karyotyping was provided to all patients, and in 94.1% of cases both short and long-term culture methods were carried out. No difference in fetal and perinatal losses between the study and control populations was found, but a higher rate of deliveries before 37 weeks and of low birth weight babies was noted amongst controls. CONCLUSIONS: Chorionicity in twin pregnancy can be determined with certainty between 7 and 12 weeks of gestation; in cases of confluent placentas reliability is provided by determining the presence or absence of the lambda sign. This study indicates that first-trimester transabdominal chorionic villus sampling is a highly efficient, reliable, and relatively safe approach for genetic diagnosis in twin pregnancies. Although a precise evaluation of the relative risks of chorionic villus sampling and mid-trimester amniocentesis in twins must await randomized control studies, the advantages of a first-trimester diagnosis to enable early decision-making about selective fetal reduction are obvious.
Birth Weight↗
Objective and subjective impairment from often-used sedative/analgesic combinations in ambulatory surgery, using alcohol as a benchmark.
Impairment caused by different sedative/analgesic combinations commonly used in ambulatory settings was compared to that of alcohol at blood alcohol concentrations (BACs) higher than or equal to 0.10%. Impairment was measured via subjective (mood) and objective (psychomotor performance) assays. Twelve healthy human volunteers (10 males and 2 females; age range 21-34 yr) participated in this prospective, double-blind, randomized, cross-over study. Each subject was exposed to five drug conditions across 5 wk. Each of the following drug conditions were adjusted for body weight (per 70 kg):fentanyl 50 micrograms and propofol 35 mg (FP), fentanyl 50 micrograms and midazolam 2 mg (FM), fentanyl 50 micrograms, midazolam 2 mg, and propofol 35 mg (FMP), alcohol 56 g (orally administered), and placebo (PLC). With the exception of alcohol, the other drugs were administered via the intravenous route. Tests for psychomotor performance, subjective effects, and short-term memory were done at baseline, and at different intervals until 240 min postinjection. Psychomotor impairment caused by alcohol at 15 min postingestion (at a BAC of 0.11% +/- 0.03% [mean +/- SE]) was used as a benchmark with which impairment caused by other sedative/analgesic combinations was compared. All the study drug combinations produced impairment (i.e., impairment greater than that seen with PLC), similar to that observed with alcohol at a BAC of 0.11%. We have demonstrated that some sedative/analgesic drug combinations used in anesthesia for ambulatory procedures produce impairment similar to or greater than that observed with a large dose of alcohol.
Adult↗
Numerically controlled oscillators with hybrid function generators.
Numerically controlled oscillators (NCOs), with a hybrid scheme of both look-up tables (LUT) and coordinate transformation digital computer (CORDIC) algorithms for a hardware efficient, high performance sine/cosine function generation are investigated. This scheme combines fast access and power efficiency of reasonably sized LUTs, and arbitrary precision obtainable from a rigorous iteration algorithm. Systematic studies using hardware description language (HDL) models and synthesis lead to optimum LUT/CORDIC ratios, which minimize power consumption and silicon area for a given operating clock frequency. First order error models are presented as guidelines for choosing internal NCO parameters. The NCO accuracy is tested with HDL simulations for all algorithmic states to limit output errors to 1 least significant bit (LSB) and by spectra derived from discrete Fourier transform (DFT) for typical frequency inputs f, resulting in a signal to noise ratio (SNR) of better than 100 dB for an amplitude word length AW of 16 Bit. Two benchmark designs were adopted for the two clock frequencies 200 MHz and 20 MHz, as "high" and "moderate" performance, respectively. The NCO models are synthesized in a 0.35 microm CMOS standard cell target technology and optimized to actually achieve after layout maximum clock frequencies exceeding 310 MHz, i.e., signal frequencies of up to 100 MHz.
Journal Article↗
Priorities and benchmarks for new faculty in physical therapist education programs: perceptions of program directors.
The purpose of this study was to establish development priorities, ideal workload and performance levels, and preferred faculty development activities for new faculty in professional-level physical therapist education programs. A preliminary questionnaire was sent to 183 directors of these programs to identify new faculty and their program directors. Eighty-five new faculty and 79 of their program directors were identified and sent comparable surveys with questions that pertained to the current and ideal teaching, scholarship, and service activities of the new faculty. Results indicated that both new faculty and program directors agreed on ideal workload levels and that workloads for new faculty should shift from teaching to scholarship. New faculty set performance levels that are significantly higher than those identified by their program directors. Development priorities for the new faculty varied from knowledge and skills in instruction as expressed by program directors to scholarship as expressed by new faculty. Individual consultations were the highest preferred faculty development activity reported by both groups. These findings provide direction for the content and delivery method for faculty development activities for new faculty in these and similar education programs.
Attitude↗
Dietary Polyphenol Acteoside-Related Molecular Signatures in Clear Cell Renal Cell Carcinoma: Multi-Omics Profiling and Functional Validation of IMPDH1.
Clear cell renal cell carcinoma (ccRCC) is characterized by substantial metabolic and molecular heterogeneity, but the disease-relevant programs associated with acteoside, a dietary polyphenol, remain poorly understood. We integrated predicted acteoside targets with bulk, single-cell, and spatial transcriptomic data from ccRCC and combined molecular subtyping with cross-cohort machine-learning analysis. Acteoside-related signatures were preferentially enriched in malignant compartments and increased with tumor grade and stage. Consensus clustering identified two molecular subtypes with distinct biological and clinical features. C1 was associated with immune activation, metabolic activity, and more favorable survival, whereas C2 showed greater genomic instability, reduced renal epithelial differentiation, and poorer outcomes. We further benchmarked multiple machine-learning strategies and established a 10-gene prognostic model that retained predictive performance across independent cohorts, with IMPDH1 emerging as the strongest risk-associated feature. Functional experiments confirmed the biological relevance of IMPDH1: its knockdown suppressed ccRCC cell proliferation, DNA synthesis, colony formation, and migration, whereas overexpression produced the opposite effects. Together, these findings indicate that acteoside-related molecular signatures capture clinically relevant heterogeneity in ccRCC and provide a framework for linking dietary-polyphenol-related molecular space with tumor biology. The identification and functional validation of IMPDH1 further highlight its potential importance in ccRCC progression.
IMPDH1↗
GBFN: A gated bimodal fusion network leveraging foundation model embeddings for cancer drug sensitivity prediction.
Despite recent progress in deep learning for cancer drug sensitivity prediction, many existing models still rely on task-specific representation learning or relatively simple multimodal fusion, which may limit their ability to capture complex drug-cell interactions. To address this issue, we developed GBFN, a gated bimodal fusion network for continuous IC50 prediction that integrates pretrained drug and cell-line representations. Specifically, drug embeddings were obtained from SMI-TED, whereas cell-line embeddings were derived from transcriptomic profiles using BulkFormer. These two modalities were then combined through a dimension-wise gated fusion module and used to predict IC50 values in matched drug-cell line pairs. On the CCLE-based benchmark, GBFN outperformed representative neural baselines, including GraphDRP, TGSA, and TransEDRP, and achieved the best overall performance, with an R² of 0.8714 and an RMSE of 0.8938. Moreover, ablation analysis showed that the model using drug features and cell-line expression data with gated fusion performed better than the corresponding model using direct concatenation, indicating that the improvement was associated with the fusion strategy rather than with the input modalities alone. In addition, cell-line expression data were more informative than mutation data in the present setting, and adding mutation data to the model using drug features and expression data did not further improve performance. Across major cancer types, GBFN maintained generally high cell-line-level predictive performance, and perturbation-based attribution identified biologically relevant transcriptomic programs in selected drug-cell line settings. Together, these findings support GBFN as a compact and effective framework for continuous drug response prediction.
Humans↗
Rectal cancer.
Rectal cancer is an important tumour from an epidemiological point of view and represents the benchmark for an optimal use of integrated treatments (surgery, radiotherapy and chemotherapy) in the oncological practice. Performing radio-chemotherapy (best if preoperatively), medical and radiation oncologists are now able to increase survival, to decrease the occurrence of pelvic recurrence and to ameliorate the quality of life of patients. Updated recommendations for the management of these patients are here reported.
Antineoplastic Combined Chemotherapy Protocols↗
Self-organizing molecular field analysis: a tool for structure-activity studies.
Self-organizing molecular field analysis (SOMFA) is a novel technique for three-dimensional quantitative structure-activity relations (3D-QSAR). It is simple and intuitive in concept and avoids the complex statistical tools and variable selection procedures favored by other methods. Our calculations show the method to be as predictive as the best 3D-QSAR methods available. Importantly, steric and electrostatic maps can be produced to aid the molecular design process by highlighting important molecular features. The simplicity of the technique leaves scope for further development, particularly with regard to handling molecular alignment and conformation selection. Here, the method has been used to predict the corticosteroid-binding globulin binding affinity of the "benchmark" steroids, expanded from the usual 31 compounds to 43 compounds. Test predictions have also been performed on a set of sulfonamide endothelin inhibitors.
Drug Design↗
Utilizing a disease management approach to improve ESRD patient outcomes.
In this era of processes and systems to improve quality, disease management is one methodology to improve care delivery and outcomes for patients with chronic kidney disease (CKD). In most disease management systems a senior renal nurse coordinates all aspects of the patient's care and ensures that the prescribed and necessary care is delivered for both CKD-related and comorbid conditions. The nurse also continually monitors outcomes on quality indicators and key performance measures. These outcome data are then aggregated and analyzed, are compared with local and national benchmarks, and drive the continuous quality improvement (CQI) process. Such a system attempts to centralize the currently fragmented care delivery system, continually improve patient outcomes, and conserve scarce economic resources. Early data suggest a disease management approach may improve both the morbidity and mortality of CKD patients.
Disease Management↗
An HMM posterior decoder for sequence feature prediction that includes homology information.
MOTIVATION: When predicting sequence features like transmembrane topology, signal peptides, coil-coil structures, protein secondary structure or genes, extra support can be gained from homologs. RESULTS: We present here a general hidden Markov model (HMM) decoding algorithm that combines probabilities for sequence features of homologs by considering the average of the posterior label probability of each position in a global sequence alignment. The algorithm is an extension of the previously described 'optimal accuracy' decoder, allowing homology information to be used. It was benchmarked using an HMM for transmembrane topology and signal peptide prediction, Phobius. We found that the performance was substantially increased when incorporating information from homologs. AVAILABILITY: A prediction server for transmembrane topology and signal peptides that uses the algorithm is available at http://phobius.cgb.ki.se/poly.html. An implementation of the algorithm is available on request from the authors.
Algorithms↗
LS Bound based gene selection for DNA microarray data.
MOTIVATION: One problem with discriminant analysis of DNA microarray data is that each sample is represented by quite a large number of genes, and many of them are irrelevant, insignificant or redundant to the discriminant problem at hand. Methods for selecting important genes are, therefore, of much significance in microarray data analysis. In the present study, a new criterion, called LS Bound measure, is proposed to address the gene selection problem. The LS Bound measure is derived from leave-one-out procedure of LS-SVMs (least squares support vector machines), and as the upper bound for leave-one-out classification results it reflects to some extent the generalization performance of gene subsets. RESULTS: We applied this LS Bound measure for gene selection on two benchmark microarray datasets: colon cancer and leukemia. We also compared the LS Bound measure with other evaluation criteria, including the well-known Fisher's ratio and Mahalanobis class separability measure, and other published gene selection algorithms, including Weighting factor and SVM Recursive Feature Elimination. The strength of the LS Bound measure is that it provides gene subsets leading to more accurate classification results than the filter method while its computational complexity is at the level of the filter method. AVAILABILITY: A companion website can be accessed at http://www.ntu.edu.sg/home5/pg02776030/lsbound/. The website contains: (1) the source code of the gene selection algorithm; (2) the complete set of tables and figures regarding the experimental study; (3) proof of the inequality (9). CONTACT: ekzmao@ntu.edu.sg.
Biomarkers, Tumor↗
A Protein Classification Benchmark collection for machine learning.
Protein classification by machine learning algorithms is now widely used in structural and functional annotation of proteins. The Protein Classification Benchmark collection (http://hydra.icgeb.trieste.it/benchmark) was created in order to provide standard datasets on which the performance of machine learning methods can be compared. It is primarily meant for method developers and users interested in comparing methods under standardized conditions. The collection contains datasets of sequences and structures, and each set is subdivided into positive/negative, training/test sets in several ways. There is a total of 6405 classification tasks, 3297 on protein sequences, 3095 on protein structures and 10 on protein coding regions in DNA. Typical tasks include the classification of structural domains in the SCOP and CATH databases based on their sequences or structures, as well as various functional and taxonomic classification problems. In the case of hierarchical classification schemes, the classification tasks can be defined at various levels of the hierarchy (such as classes, folds, superfamilies, etc.). For each dataset there are distance matrices available that contain all vs. all comparison of the data, based on various sequence or structure comparison methods, as well as a set of classification performance measures computed with various classifier algorithms.
Algorithms↗
Comparison of thermal neutron distributions within shield materials obtained by experiments, SN and Monte Carlo code calculations.
Benchmark experiments of thermal neutron distributions within the shield materials, graphite pile and pure water, were performed by using 252Cf fission neutrons and gold foil activation detectors, and, to these results, the estimates obtained by using the discrete ordinate code ANISN and the Monte Carlo code MCNP5 with two different cross-section libraries, ENDF/B-VI and the Japanese new version of JENDL-3.3, were compared. The results revealed that the MCNP5 calculations with the two libraries closely agree with the experiments and that there are slight differences between the MCNP5 and the ANISN calculations. The differences are caused mainly by the overestimation of the thermal neutron absorption cross sections constructed in NJOY99. The ANISN calculations with the modified absorption cross sections reproduced the results of the MCNP5 fairly well.
Computer Simulation↗
Surgeon contribution to hospital bottom line: not all are created equal.
OBJECTIVE: We hypothesized that surgeon productivity is directly related to hospital operating margin, but significant variation in margin contribution exists between specialties. SUMMARY BACKGROUND DATA: As the independent practitioner becomes an endangered species, it is critical to better understand the surgeon's importance to a hospital's bottom line. An appreciation of surgeon contribution to hospital profitability may prove useful in negotiations relating to full-time employment or other models. METHODS: Surgeon total relative value units (RVUs), a measure of productivity, were collected from operating room (OR) logs. Annual hospital margin per specialty was provided by hospital finance. Hospital margin data were normalized by dividing by a constant such that the highest relative hospital margin (RHM) in fiscal year 2004 expressed as margin units (mu) was 1 million mu. For each specialty, data analyzed included RHM/OR HR, RHM/case, and RHM/RVU. RESULTS: Thoracic (34.55 mu/RVU) and transplant (25.13 mu/RVU) were the biggest contributors to hospital margin. Plastics (-0.57 mu/RVU), maxillofacial (1.41 mu/RVU), and gynecology (1.66 mu/RVU) contributed least to hospital margin. Relative hospital margin per OR HR for transplant slightly exceeded thoracic (275.74 mu vs 233.94 mu) at the top and plastics and maxillofacial contributed the least (-3.83 mu/OR HR vs 9.36 mu/OR HR). CONCLUSIONS: Surgeons contribute significantly to hospital margin with certain specialties being more profitable than others. Payer mix, the penetration of managed care, and negotiated contracts as well as a number of other factors all have an impact on an individual hospital's margin. Surgeons should be fully cognizant of their significant influence in the marketplace.
Benchmarking↗