A transfer learning approach for predictive modeling of bioprocesses using small data

Rogers, Alexander W.; Vega-Ramon, Fernando; Yan, Jiangtao; Rio‐Chanona, Ehecatl Antonio del; Jing, Keju; Zhang, Dongda

doi:10.1002/bit.27980

Cited by 34 publications

(10 citation statements)

References 32 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…They argued that the performance gains obtained through transferring learning techniques can improve the application of computer‐aided diagnoses in various conditions. Generally, the performance of CNN models improves as the number of datasets increases 23 . However, in the field of medical imaging, annotating a large number of medical images is still challenging and time‐consuming.…”

Section: Discussionmentioning

confidence: 99%

“…Generally, the performance of CNN models improves as the number of datasets increases. 23 However, in the field of medical imaging, annotating a large number of medical images is still challenging and time-consuming. Although dozens or hundreds of GBM datasets comprising patient information and complete labels provide significant amounts of data, in terms of CNN model training, they are still considered small, which may lead to overfitting.…”

Section: Discussionmentioning

confidence: 99%

See 1 more Smart Citation

Deep Learning Radiomics for the Assessment of Telomerase Reverse Transcriptase Promoter Mutation Status in Patients With Glioblastoma Using Multiparametric MRI

Zhang

et al. 2023

Magnetic Resonance Imaging

View full text Add to dashboard Cite

BackgroundStudies have shown that magnetic resonance imaging (MRI)‐based deep learning radiomics (DLR) has the potential to assess glioma grade; however, its role in predicting telomerase reverse transcriptase (TERT) promoter mutation status in patients with glioblastoma (GBM) remains unclear.PurposeTo evaluate the value of deep learning (DL) in multiparametric MRI‐based radiomics in identifying TERT promoter mutations in patients with GBM preoperatively.Study TypeRetrospective.PopulationA total of 274 patients with isocitrate dehydrogenase‐wildtype GBM were included in the study. The training and external validation cohorts included 156 (54.3 ± 12.7 years; 96 males) and 118 (54 .2 ± 13.4 years; 73 males) patients, respectively.Field Strength/SequenceAxial contrast‐enhanced T1‐weighted spin‐echo inversion recovery sequence (T1CE), T1‐weighted spin‐echo inversion recovery sequence (T1WI), and T2‐weighted spin‐echo inversion recovery sequence (T2WI) on 1.5‐T and 3.0‐T scanners were used in this study.AssessmentOverall tumor area regions (the tumor core and edema) were segmented, and the radiomics and DL features were extracted from preprocessed multiparameter preoperative brain MRI images—T1WI, T1CE, and T2WI. A model based on the DLR signature, clinical signature, and clinical DLR (CDLR) nomogram was developed and validated to identify TERT promoter mutation status.Statistical TestsThe Mann–Whitney U test, Pearson test, least absolute shrinkage and selection operator, and logistic regression analysis were applied for feature selection and construction of radiomics and DL signatures. Results were considered statistically significant at P‐value <0.05.ResultsThe DLR signature showed the best discriminative power for predicting TERT promoter mutations, yielding an AUC of 0.990 and 0.890 in the training and external validation cohorts, respectively. Furthermore, the DLR signature outperformed CDLR nomogram (P = 0.670) and significantly outperformed clinical models in the validation cohort.Data ConclusionThe multiparameter MRI‐based DLR signature exhibited a promising performance for the assessment of TERT promoter mutations in patients with GBM, which could provide information for individualized treatment.Level of Evidence3Technical EfficacyStage 2

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Discussionmentioning

confidence: 99%

Deep Learning Radiomics for the Assessment of Telomerase Reverse Transcriptase Promoter Mutation Status in Patients With Glioblastoma Using Multiparametric MRI

Zhang

et al. 2023

Magnetic Resonance Imaging

View full text Add to dashboard Cite

show abstract

“…In another study of 105 gliomas by Peng et al (2021), the AUC and accuracy of classifying IDH status by the Lasso method were 0.770 and 0.823. In general, the performance of the model increases with the size of the dataset (Rogers et al, 2022). However, even in a large cohort, the classification results obtained with a single algorithm are still unsatisfactory.…”

Section: Discussionmentioning

confidence: 99%

“…Yan et al. investigated 357 glioma patients by using Bayesian neural networks and integrated radiomics features from Gd‐T1 and ADC; the AUC, accuracy, sensitivity, and specificity for predicting TERT mutations were 0.598, 0.685, 0.976, and 0.290, respectively (Rogers et al., 2022 ). In this regard, it is very important to use appropriate algorithms to build the effectiveness model.…”

Section: Discussionmentioning

confidence: 99%

MRI radiomics model for predicting TERT promoter mutation status in glioblastoma

Chen,

et al. 2023

Brain and Behavior

View full text Add to dashboard Cite

Background and purposeThe presence of TERT promoter mutations has been associated with worse prognosis and resistance to therapy for patients with glioblastoma (GBM). This study aimed to determine whether the combination model of different feature selections and classification algorithms based on multiparameter MRI can be used to predict TERT subtype in GBM patients.MethodsA total of 143 patients were included in our retrospective study, and 2553 features were obtained. The datasets were randomly divided into training and test sets in a ratio of 7:3. The synthetic minority oversampling technique was used to achieve data balance. The Pearson correlation coefficients were used for dimension reduction. Three feature selections and five classification algorithms were used to model the selected features. Finally, 10‐fold cross validation was applied to the training dataset.ResultsA model with eight features generated by recursive feature elimination (RFE) and linear discriminant analysis (LDA) showed the greatest diagnostic performance (area under the curve values for the training, validation, and testing sets: 0.983, 0.964, and 0.926, respectively), followed by relief and random forest (RF), analysis of variance and RF. Furthermore, the relief was the optimal feature selection for separately evaluating those five classification algorithms, and RF was the most preferable algorithm for separately assessing the three feature selectors. ADC entropy was the parameter that made the greatest contribution to the discrimination of TERT mutations.ConclusionsRadiomics model generated by RFE and LDA mainly based on ADC entropy showed good performance in predicting TERT promoter mutations in GBM.

show abstract

“…It would enable the utilization of biological data published in the literature and reduce the demand for data collection when investigating new processes. Rogers et al ( 55 ) showed that transfer learning is a highly data-efficient technique to predict the modeling of bioprocesses, a typically low-dimensional, small-data problem, but its usage in the biomedical field is scarce. In the present study, we will extract generalizable knowledge from well-understood published biological data, capturing essential process mechanisms to set a source model.…”

Section: Discussionmentioning

confidence: 99%

Mechanism of Qihuang needle therapy in the management of tic disorders: a clinical trial protocol

Tang

et al. 2023

Front. Neurol.

View full text Add to dashboard Cite

BackgroundQihuang needle therapy is a newly developed acupuncture therapy to treat tic disorders in clinical practice. However, the mechanism to reduce tic severity remains unknown. Changes in intestinal flora and circulation metabolites are perhaps the potential pathogenesis of tic disorders. As a result, we present a protocol for a controlled clinical trial using multi-omics analysis to probe the mechanism of the Qihuang needle in managing tic disorders.MethodsThis is a matched-pairs design, controlled, clinical trial for patients with tic disorders. Participants will be allocated to either an experimental group or a healthy control group. The main acupoints are Baihui (GV20), Yintang (EX-HN3), and Jueyinshu (BL14). The experimental group will receive Qihuang needle therapy for a month, while the control group will receive no interventions.Expected outcomesThe change in the severity of the tic disorder is set as the main outcome. Secondary outcomes include gastrointestinal severity index and recurrence rate, which will be calculated after a 12-week follow-up. Gut microbiota, measured by 16S rRNA gene sequencing; serum metabolomics, assessed via LC/MS; and serum zonulin, assessed by enzyme-linked immunosorbent assay (ELISA), will be used as biological specimen analysis outcomes. The present study will investigate the possible interactions between intestinal flora and serum metabolites and the improvement of clinical profiles, which may elucidate the mechanism of Qihuang needle therapy for tic disorders.Trial registrationThis trial is registered at the Chinese Clinical Trial Registry (http://www.chictr.org.cn/). Registration number: ChiCTR2200057723, Date: 2022-04-14.

show abstract

A transfer learning approach for predictive modeling of bioprocesses using small data

Cited by 34 publications

References 32 publications

Deep Learning Radiomics for the Assessment of Telomerase Reverse Transcriptase Promoter Mutation Status in Patients With Glioblastoma Using Multiparametric MRI

Deep Learning Radiomics for the Assessment of Telomerase Reverse Transcriptase Promoter Mutation Status in Patients With Glioblastoma Using Multiparametric MRI

MRI radiomics model for predicting TERT promoter mutation status in glioblastoma

Mechanism of Qihuang needle therapy in the management of tic disorders: a clinical trial protocol

Contact Info

Product

Resources

About