Independent Training Datasets Research Articles

Frontoparietal and salience network synchronizations during nonsymbolic magnitude processing predict brain age and mathematical performance in youth.

The development and refinement of functional brain circuits crucial to human cognition is a continuous process that spans from childhood to adulthood. Research increasingly focuses on mapping these evolving configurations, with the aim to identify markers for functional impairments and atypical development. Among human cognitive systems, nonsymbolic magnitude representations serve as a foundational building block for future success in mathematical learning and achievement for individuals. Using task-based frontoparietal (FPN) and salience network (SN) features during nonsymbolic magnitude processing alongside machine learning algorithms, we developed a framework to construct brain age prediction models for participants aged 7-30. Our study revealed differential developmental profiles in the synchronization within and between FPN and SN networks. Specifically, we observed a linear increase in FPN connectivity, concomitant with a decline in SN connectivity across the age span. A nonlinear U-shaped trajectory in the connectivity between the FPN and SN was discerned, revealing reduced FPN-SN synchronization among adolescents compared to both pediatric and adult cohorts. Leveraging the Gradient Boosting machine learning algorithm and nested fivefold stratified cross-validation with independent training datasets, we demonstrated that functional connectivity measures of the FPN and SN nodes predict chronological age, with a correlation coefficient of .727 and a mean absolute error of 2.944 between actual and predicted ages. Notably, connectivity within the FPN emerged as the most contributing feature for age prediction. Critically, a more matured brain age estimate is associated with better arithmetic performance. Our findings shed light on the intricate developmental changes occurring in the neural networks supporting magnitude representations. We emphasize brain age estimation as a potent tool for understanding cognitive development and its relationship to mathematical abilities across the critical developmental period of youth. PRACTITIONER POINTS: This study investigated the prolonged changes in the brain's architecture across childhood, adolescence, and adulthood, with a focus on task-state frontoparietal and salience networks. Distinct developmental pathways were identified: frontoparietal synchronization strengthens consistently throughout development, while salience network connectivity diminishes with age. Furthermore, adolescents show a unique dip in connectivity between these networks. Leveraging advanced machine learning methods, we accurately predicted individuals' ages based on these brain circuits, with a more mature estimated brain age correlating with better math skills.

Trained neural networking framework based skin cancer diagnosis and categorization using grey wolf optimization

Skin Cancer is caused due to the mutational differences in epidermis hormones and patch appearances. Many studies are focused on the design and development of effective approaches in diagnosis and categorization of skin cancer. The decisions are made on independent training dataset under limited editions and scenarios. In this research, the kaggle based datasets are optimized and categorized into a labeled data array towards indexing using Federated learning (FL). The technique is developed on grey wolf optimization algorithm to assure the dataset attribute dependencies are extracted and dimensional mapping is processed. The threshold value validation of the dimensional mapping datasets is effectively optimized and trained under the neural networking framework further expanded via federated learning standards. The technique has demonstrated 95.82% accuracy under GWO technique and 94.9% on inter-combination of Trained Neural Networking (TNN) framework and Recessive Learning (RL) in accuracy.

Egocentric vision-based detection of surfaces: towards context-aware free-living digital biomarkers for gait and fall risk assessment

BackgroundFalls in older adults are a critical public health problem. As a means to assess fall risks, free-living digital biomarkers (FLDBs), including spatiotemporal gait measures, drawn from wearable inertial measurement unit (IMU) data have been investigated to identify those at high risk. Although gait-related FLDBs can be impacted by intrinsic (e.g., gait impairment) and/or environmental (e.g., walking surfaces) factors, their respective impacts have not been differentiated by the majority of free-living fall risk assessment methods. This may lead to the ambiguous interpretation of the subsequent FLDBs, and therefore, less precise intervention strategies to prevent falls.MethodsWith the aim of improving the interpretability of gait-related FLDBs and investigating the impact of environment on older adults’ gait, a vision-based framework was proposed to automatically detect the most common level walking surfaces. Using a belt-mounted camera and IMUs worn by fallers and non-fallers (mean age 73.6 yrs), a unique dataset (i.e., Multimodal Ambulatory Gait and Fall Risk Assessment in the Wild (MAGFRA-W)) was acquired. The frames and image patches attributed to nine participants’ gait were annotated: (a) outdoor terrains: pavement (asphalt, cement, outdoor bricks/tiles), gravel, grass/foliage, soil, snow/slush; and (b) indoor terrains: high-friction materials (e.g., carpet, laminated floor), wood, and tiles. A series of ConvNets were developed: EgoPlaceNet categorizes frames into indoor and outdoor; and EgoTerrainNet (with outdoor and indoor versions) detects the enclosed terrain type in patches. To improve the framework’s generalizability, an independent training dataset with 9,424 samples was curated from different databases including GTOS and MINC-2500, and used for pretrained models’ (e.g., MobileNetV2) fine-tuning.ResultsEgoPlaceNet detected outdoor and indoor scenes in MAGFRA-W with 97.36% and 95.59% (leave-one-subject-out) accuracies, respectively. EgoTerrainNet-Indoor and -Outdoor achieved high detection accuracies for pavement (87.63%), foliage (91.24%), gravel (95.12%), and high-friction materials (95.02%), which indicate the models’ high generalizabiliy.ConclusionsEncouraging results suggest that the integration of wearable cameras and deep learning approaches can provide objective contextual information in an automated manner, towards context-aware FLDBs for gait and fall risk assessment in the wild.

A novel NGS kit solution for multi-cancer early detection using circulating cell free DNA based methylation analysis.

10542 Background: Cancer patient survival can be greatly improved by early detection. Cell-free DNA (cfDNA)-based cancer detection using targeted methylation sequencing (TMS) holds great promise. Here, we report the performance of the first cfDNA TMS kit for early cancer detection. Methods: We developed a comprehensive multi-cancer TMS panel, named “Omni1” (900kb) containing ̃3000 cancer specific hypermethylation markers for the detection of the 16 most prevalent solid tumor cancer types in the US. These selected biomarkers all have high signals in cancer, low signals in normal tissue and close to zero background in plasma. The TMS analysis is empowered by Point-n-Seq DNA enrichment and soft bisulfite sequencing technology. Across the different cancer types, the Point-n-Seq Methylation Index (PMI) derived from the Omni1 panel showed over 10 times the difference in the genomic DNA of tumor compared to normal tissues. 8 to 32 early-stage plasma samples from each of the 7 most prevalent cancer types were used for assay testing. cfDNA extracted from each donor was subjected to our Point-n-Seq TMS analysis with the Omni1 panel. The TMS workflow took around five hours, from cfDNA to a sequencing ready library. 6M reads were assigned to each sample resulting in ̃1400X average raw coverage across the target regions. Results: A computational model to calculate the methylation index and the thresholds were established on independent training data sets for each cancer type. For the testing, specificity was set at 89% (CI = 0.81, 0.97, n = 55 self-claimed healthy individuals). The detection rate for stage I (n = 55) and stage II (n = 56) of all 7 cancer types combined is 65% (CI = 0.53,0.78) and 75% (CI = 0.64, 0.86) respectively. Even within this small sample size, it was clear that the detection sensitivity varied among different cancer types, and it increased from stage I to II: colorectal (65%, n = 20; 75%, n = 12), lung (75% n = 8; 92%, n = 12), pancreas (80%, n = 10; 100%, n = 10), ovary (60%, n = 5; 80%, n = 5), breast (40%, n = 7; 100%, n = 3), and prostate cancer (0%, n = 2; 22% n = 9) for stage I and II respectively. A decrease in sensitivity from stage I to II in liver (100%, n = 3; 60%, n = 5) is possibly due to variability from the small sample size. For the cases of CRC, the detection rates obtained from Omni1 are comparable to the stand-alone CRC560 panel (170 kb). Conclusions: In this study we demonstrated that by leveraging an economy-size panel, a robust methylation sequencing kit, and a straightforward analysis methodology, early cancer detection for multiple cancer types can be done with just 1ml plasma. The kit strategy provides an easy-to-use and distributable solution, without the complications of developing a lab-specific test that uses a large methylation panel and a high sequencing cost.

Discrimination of Stressed and Non-Stressed Food-Related Bacteria Using Raman-Microspectroscopy.

As the identification of microorganisms becomes more significant in industry, so does the utilization of microspectroscopy and the development of effective chemometric models for data analysis and classification. Since only microorganisms cultivated under laboratory conditions can be identified, but they are exposed to a variety of stress factors, such as temperature differences, there is a demand for a method that can take these stress factors and the associated reactions of the bacteria into account. Therefore, bacterial stress reactions to lifetime conditions (regular treatment, 25 °C, HCl, 2-propanol, NaOH) and sampling conditions (cold sampling, desiccation, heat drying) were induced to explore the effects on Raman spectra in order to improve the chemometric models. As a result, in this study nine food-relevant bacteria were exposed to seven stress conditions in addition to routine cultivation as a control. Spectral alterations in lipids, polysaccharides, nucleic acids, and proteins were observed when compared to normal growth circumstances without stresses. Regardless of the involvement of several stress factors and storage times, a model for differentiating the analyzed microorganisms from genus down to strain level was developed. Classification of the independent training dataset at genus and species level for Escherichia coli and at strain level for the other food relevant microorganisms showed a classification rate of 97.6%.

Estimating Aggregate Environmental Risk Score in Psychiatry: The Exposome Score for Schizophrenia.

To understand the role of environment in the pathoetiology of psychosis spectrum disorders, research has thus far mainly investigated the effects of single exposures in isolation, such as the association between cannabis use and schizophrenia. However, this approach fails to acknowledge the complexity of the exposome, which represents the totality of the environment involving many exposures over an individual's lifetime. Therefore, contemporary research adopting the exposome paradigm has aimed at capturing the combined effect of different environmental exposures by utilizing an aggregate environmental vulnerability score for schizophrenia: the exposome score for schizophrenia. Here, we attempt to provide a comprehensive overview of studies applying the exposome score for schizophrenia. First, we describe several approaches estimating exposomic vulnerability for schizophrenia, which falls into three categories: simple environmental sum scores (sum of dichotomized exposures), meta-analysis-based environmental risk score (sum scores weighted by estimates from meta-analyses), and the exposome score (sum score weighted by estimates from an analysis in an independent training dataset). Studies show that the exposome score for schizophrenia that assumes interdependency of exposures performs better than scores that assume independence of exposures, such as the environmental sum score and the meta-analysis-based environmental risk score. Second, we discuss findings on the pluripotency of the exposome score for schizophrenia and summarize findings from gene-environment studies using the exposome score for schizophrenia. Finally, we discuss possible scientific, clinical, and population-based applications of exposome score for schizophrenia, as well as limitations and future directions for exposome research to understand the etiology of psychosis spectrum disorders.

SHMnet: Condition assessment of bolted connection with beyond human-level performance

Deep learning algorithms are transforming a variety of research areas with accuracy levels that the traditional methods cannot compete with. Recently, increasingly more research efforts have been put into the structural health monitoring domain. In this work, we propose a new deep convolutional neural network, namely SHMnet, for a challenging structural condition identification case, that is, steel frame with bolted connection damage. We perform systematic studies on the optimisation of network architecture and the preparation of the training data. In the laboratory, repeated impact hammer tests are conducted on a steel frame with different bolted connection damage scenarios, as small as one bolt loosened. The time-domain monitoring data from a single accelerometer are used for training. We conduct parametric studies on different layer numbers, different sensor locations, the quantity of the training datasets and noise levels. The results show that the proposed SHMnet is effective and reliable with at least four independent training datasets and by avoiding vibration node points as sensor locations. Under up to 60% additive Gaussian noise, the average identification accuracy is over 98%. In comparison, the traditional methods based on the identified modal parameters inevitably fail due to the unnoticeable changes of identified natural frequencies and mode shapes. The results provide confidence in using the developed method as an effective structural condition identification framework. It has the potential to transform the structural health monitoring practice. The code and relevant information can be found at https://github.com/capepoint/SHMnet.

Utility of a Clinical Decision Support System in Weight Loss Prediction After Head and Neck Cancer Radiotherapy.

To evaluate the utility of a clinical decision support system (CDSS) using a weight loss prediction model. A prediction model for significant weight loss (loss of greater than or equal to 7.5% of body mass at 3-month post radiotherapy) was created with clinical, dosimetric, and radiomics predictors from 63 patients in an independent training data set (accuracy, 0.78; area under the curve [AUC], 0.81) using least absolute shrinkage and selection operator logistic regression. Four physicians with varying experience levels were then recruited to evaluate 100 patients in an independent validation data set of head and neck cancer twice (ie, a pre-post design): first without and then with the aid of a CDSS derived from the prediction model. At both evaluations, physicians were asked to predict the development (yes/no) and probability of significant weight loss for each patient on the basis of patient characteristics, including pretreatment dysphagia and weight loss and information from the treatment plan. At the second evaluation, physicians were also provided with the prediction model's results for weight loss probability. Physicians' predictions were compared with actual weight loss, and accuracy and AUC were investigated between the two evaluations. The mean accuracy of the physicians' ability to identify patients who will experience significant weight loss (yes/no) increased from 0.58 (range, 0.47 to 0.63) to 0.63 (range, 0.58 to 0.72) with the CDSS ( P = .06). The AUC of weight loss probability predicted by physicians significantly increased from 0.56 (range, 0.46 to 0.64) to 0.69 (range, 0.63 to 0.73) with the aid of the CDSS ( P < .05). Specifically, more improvement was observed among less-experienced physicians ( P < .01). Our preliminary results demonstrate that physicians' decisions may be improved by a weight loss CDSS model, especially among less-experienced physicians. Additional study with a larger cohort of patients and more participating physicians is thus warranted for understanding the usefulness of CDSSs.

Using Deep Learning and Transfer Learning to Accurately Diagnose Early-Onset Glaucoma From Macular Optical Coherence Tomography Images

We sought to construct and evaluate a deep learning (DL) model to diagnose early glaucoma from spectral-domain optical coherence tomography (OCT) images. Artificial intelligence diagnostic tool development, evaluation, and comparison. This multi-institution study included pretraining data of 4316 OCT images (RS3000) from 1371 eyes with open angle glaucoma (OAG) regardless of the stage of glaucoma and 193 normal eyes. Training data included OCT-1000/2000 images from 94 eyes of 94 patients with early OAG (mean deviation >-5.0 dB) and 84 eyes of 84 normal subjects. Testing data included OCT-1000/2000 from 114 eyes of 114 patients with early OAG (mean deviation >-5.0 dB) and 82 eyes of 82 normal subjects. A DL (convolutional neural network) classifier was trained using a pretraining dataset, followed by a second round of training using an independent training dataset. The DL model input features were the 8× 8 grid macular retinal nerve fiber layer thickness and ganglion cell complex layer thickness from spectral-domain OCT. Diagnostic accuracy was investigated in the testing dataset. For comparison, diagnostic accuracy was also evaluated using the random forests and support vector machine models. The primary outcome measure was the area under the receiver operating characteristic curve (AROC). The AROC with the DL model was 93.7%. The AROC significantly decreased to between 76.6% and 78.8% without the pretraining process. Significantly smaller AROCs were obtained with random forests and support vector machine models (82.0% and 67.4%, respectively). A DL model for glaucoma using spectral-domain OCT offers a substantive increase in diagnostic performance.

A genome-by-environment interaction classifier for precision medicine: personal transcriptome response to rhinovirus identifies children prone to asthma exacerbations.

ObjectiveTo introduce a disease prognosis framework enabled by a robust classification scheme derived from patient-specific transcriptomic response to stimulation.Materials and MethodsWithin an illustrative case study to predict asthma exacerbation, we designed a stimulation assay that reveals individualized transcriptomic response to human rhinovirus. Gene expression from peripheral blood mononuclear cells was quantified from 23 pediatric asthmatic patients and stimulated in vitro with human rhinovirus. Responses were obtained via the single-subject gene set testing methodology “N-of-1-pathways.” The classifier was trained on a related independent training dataset (n = 19). Novel visualizations of personal transcriptomic responses are provided.ResultsOf the 23 pediatric asthmatic patients, 12 experienced recurrent exacerbations. Our classifier, using individualized responses and trained on an independent dataset, obtained 74% accuracy (area under the receiver operating curve of 71%; 2-sided P = .039). Conventional classifiers using messenger RNA (mRNA) expression within the viral-exposed samples were unsuccessful (all patients predicted to have recurrent exacerbations; accuracy of 52%).DiscussionPrognosis based on single time point, static mRNA expression alone neglects the importance of dynamic genome-by-environment interplay in phenotypic presentation. Individualized transcriptomic response quantified at the pathway (gene sets) level reveals interpretable signals related to clinical outcomes.ConclusionThe proposed framework provides an innovative approach to precision medicine. We show that quantifying personal pathway–level transcriptomic response to a disease-relevant environmental challenge predicts disease progression. This genome-by-environment interaction assay offers a noninvasive opportunity to translate omics data to clinical practice by improving the ability to predict disease exacerbation and increasing the potential to produce more effective treatment decisions.

MP1-04 EVALUATION OF GENOMIC SIGNATURES IN INTERMEDIATE TO HIGH RISK MEN TREATED WITH RADICAL PROSTATECTOMY BUT WITHOUT ADDITIONAL THERAPY UPON PSA RISE

You have accessJournal of UrologyProstate Cancer: Markers I1 Apr 2015MP1-04 EVALUATION OF GENOMIC SIGNATURES IN INTERMEDIATE TO HIGH RISK MEN TREATED WITH RADICAL PROSTATECTOMY BUT WITHOUT ADDITIONAL THERAPY UPON PSA RISE Michael Johnson, Kasra Yousefi, Nicholas Erho, Voleak Choeurng, Lucia Lam, Helen L. Fedor, Elai Davicioni, Edward Schaeffer, and Ashley Ross Michael JohnsonMichael Johnson More articles by this author , Kasra YousefiKasra Yousefi More articles by this author , Nicholas ErhoNicholas Erho More articles by this author , Voleak ChoeurngVoleak Choeurng More articles by this author , Lucia LamLucia Lam More articles by this author , Helen L. FedorHelen L. Fedor More articles by this author , Elai DavicioniElai Davicioni More articles by this author , Edward SchaefferEdward Schaeffer More articles by this author , and Ashley RossAshley Ross More articles by this author View All Author Informationhttps://doi.org/10.1016/j.juro.2015.02.167AboutPDF ToolsAdd to favoritesDownload CitationsTrack CitationsPermissionsReprints ShareFacebookTwitterLinked InEmail INTRODUCTION AND OBJECTIVES Radical prostatectomy (RP) is a primary treatment option for men with intermediate and high risk prostate cancer. Though many will be effectively cured with local therapy alone, these men are by definition at higher risk of adverse pathologic findings and clinical disease recurrence. Several genomic tests are currently on the market for use as risk stratification and prognostic aids. Here we evaluate four previously validated genomic signatures in a natural history cohort of at risk men who received no additional treatment until the time of metastatic progression. METHODS Men with NCCN intermediate or high risk localized prostate cancer treated with RP at the Johns Hopkins Medical Institute (1992–2010) with at least 5 years of post-operative follow up were identified. Only men with initial undetectable PSA after surgery and who received no therapy prior to metastasis detection were included using a case-cohort design (n=337). Random forest modeling in an independent training dataset (n=359) was used to derive risk scores for Decipher, CCP, GPS and Penney genomic signatures as previously described by Erho et al., 2013 and compared to the CAPRA-S risk scores for predicting metastasis 10 years post-RP. RESULTS Risk scores were computed for 260 patients with complete follow up data. Overall, the four signatures had moderate correlations except for the Decipher and Penney signature that had a strong correlation of 0.66. ROC analysis for the survival data for metastasis 10 years post-RP was 0.76, 0.74, 0.62 and 0.62 for Decipher, Penney, CCP and GPS, respectively. In univariable analysis, all four gene signatures were significant predictors of metastatic progression, but in multivariable analysis adjusting for CAPRA-S score, only Decipher (p<0.0001), Penney (p<0.0001) and CCP (p=0.001) retained significance. CONCLUSIONS We evaluated four previously validated prognostic signatures in a natural history after RP cohort of intermediate to high-risk men. Decipher and Penney signatures were the best predictors of metastasis, although all signatures demonstrated some measure of prognostic significance. Future studies aimed at comparing the risk models and their comparative-effectiveness for postoperative decision-making are warranted. © 2015 by American Urological Association Education and Research, Inc.FiguresReferencesRelatedDetails Volume 193Issue 4SApril 2015Page: e2 Advertisement Copyright & Permissions© 2015 by American Urological Association Education and Research, Inc.MetricsAuthor Information Michael Johnson More articles by this author Kasra Yousefi More articles by this author Nicholas Erho More articles by this author Voleak Choeurng More articles by this author Lucia Lam More articles by this author Helen L. Fedor More articles by this author Elai Davicioni More articles by this author Edward Schaeffer More articles by this author Ashley Ross More articles by this author Expand All Advertisement Advertisement PDF downloadLoading ...

Partially Supervised Speaker Clustering

Content-based multimedia indexing, retrieval, and processing as well as multimedia databases demand the structuring of the media content (image, audio, video, text, etc.), one significant goal being to associate the identity of the content to the individual segments of the signals. In this paper, we specifically address the problem of speaker clustering, the task of assigning every speech utterance in an audio stream to its speaker. We offer a complete treatment to the idea of partially supervised speaker clustering, which refers to the use of our prior knowledge of speakers in general to assist the unsupervised speaker clustering process. By means of an independent training data set, we encode the prior knowledge at the various stages of the speaker clustering pipeline via 1) learning a speaker-discriminative acoustic feature transformation, 2) learning a universal speaker prior model, and 3) learning a discriminative speaker subspace, or equivalently, a speaker-discriminative distance metric. We study the directional scattering property of the Gaussian mixture model (GMM) mean supervector representation of utterances in the high-dimensional space, and advocate exploiting this property by using the cosine distance metric instead of the euclidean distance metric for speaker clustering in the GMM mean supervector space. We propose to perform discriminant analysis based on the cosine distance metric, which leads to a novel distance metric learning algorithm—linear spherical discriminant analysis (LSDA). We show that the proposed LSDA formulation can be systematically solved within the elegant graph embedding general dimensionality reduction framework. Our speaker clustering experiments on the GALE database clearly indicate that 1) our speaker clustering methods based on the GMM mean supervector representation and vector-based distance metrics outperform traditional speaker clustering methods based on the “bag of acoustic features” representation and statistical model-based distance metrics, 2) our advocated use of the cosine distance metric yields consistent increases in the speaker clustering performance as compared to the commonly used euclidean distance metric, 3) our partially supervised speaker clustering concept and strategies significantly improve the speaker clustering performance over the baselines, and 4) our proposed LSDA algorithm further leads to state-of-the-art speaker clustering performance.

A mutual information‐based metric for evaluation of fMRI data‐processing approaches

We propose a novel approach for evaluating the performance of activation detection in real (experimental) datasets using a new mutual information (MI)-based metric and compare its sensitivity to several existing performance metrics in both simulated and real datasets. The proposed approach is based on measuring the approximate MI between the fMRI time-series of a validation dataset and a calculated activation map (thresholded label map or continuous map) from an independent training dataset. The MI metric is used to measure the amount of information preserved during the extraction of an activation map from experimentally related fMRI time-series. The processing method that preserves maximal information between the maps and related time-series is proposed to be superior. The results on simulation datasets for multiple analysis models are consistent with the results of ROC curves, but are shown to have lower information content than for real datasets, limiting their generalizability. In real datasets for group analyses using the general linear model (GLM; FSL4 and SPM5), we show that MI values are (1) larger for groups of 15 versus 10 subjects and (2) more sensitive measures than reproducibility (for continuous maps) or Jaccard overlap metrics (for thresholded maps). We also show that (1) for an increasing fraction of nominally active voxels, both MI and false discovery rate (FDR) increase, and (2) at a fixed FDR, GLM using FSL4 tends to extract more voxels and more information than SPM5 using the default processing techniques in each package.