Machine learning-based multiparametric MRI radiomics for predicting the aggressiveness of papillary thyroid carcinoma

et al. 2021

BMC Med Imaging

Self Cite

Background To determine the predictive capability of MRI-based radiomics for extrathyroidal extension detection in papillary thyroid cancer (PTC) pre-surgically. Methods The present retrospective trial assessed individuals with thyroid nodules examined by multiparametric MRI and subsequently administered thyroid surgery. Diagnosis and extrathyroidal extension (ETE) feature of PTC were based on pathological assessment. The thyroid tumors underwent manual segmentation, for radiomic feature extraction. Participants were randomized to the training and testing cohorts, at a ratio of 7:3. The mRMR (maximum correlation minimum redundancy) algorithm and the least absolute shrinkage and selection operator were utilized for radiomics feature selection. Then, a radiomics predictive model was generated via a linear combination of the features. The model’s performance in distinguishing the ETE feature of PTC was assessed by analyzing the receiver operating characteristic curve. Results Totally 132 patients were assessed in this study, including 92 and 40 in the training and test cohorts, respectively). Next, the 16 top-performing features, including 4, 7 and 5 from diffusion weighted (DWI), T2-weighted (T2 WI), and contrast-enhanced T1-weighted (CE-T1WI) images, respectively, were finally retained to construct the radiomics signature. There were 8 RLM, 5 CM, 2 shape, and 1 SZM features. The radiomics prediction model achieved AUCs of 0.96 and 0.87 in the training and testing sets, respectively. Conclusions Our study indicated that MRI radiomics approach had the potential to stratify patients based on ETE in PTCs preoperatively.

Section: Discussionmentioning

confidence: 99%

Radiomics based on multiparametric MRI for extrathyroidal extension feature prediction in papillary thyroid cancer

Wei

et al. 2021

BMC Med Imaging

Self Cite

“…The LASSO is considered a promising technique for optimal feature selection, and could combine these radiomic features to generate a radiomic signature [35,36]. A previous study [37] assessed many feature selection techniques, and LASSO showed an optimal performance.…”

Section: Discussionmentioning

confidence: 99%

Radiomics Based on Multiparametric MRI for Extrathyroidal Extension Feature Prediction in Papillary Thyroid Cancer

Wei

et al. 2020

Preprint

Self Cite

Purpose: To determine the predictive capability of MRI-based radiomics for extrathyroidal extension detection in papillary thyroid cancer (PTC) pre-surgically.Methods: The present retrospective trial assessed individuals with thyroid nodules examined by multiparametric MRI and subsequently administered thyroid surgery. Diagnosis and extrathyroidal extension (ETE) feature of PTC were based on pathological assessment. The thyroid tumors underwent manual segmentation, for radiomic feature extraction. Participants were randomized to the training and testing cohorts, at a ratio of 7:3. The mRMR (maximum correlation minimum redundancy) algorithm and the least absolute shrinkage and selection operator (LASSO) were utilized for radiomics feature selection. Then, a radiomics predictive model was generated via a linear combination of the features. The model’s performance in distinguishing the ETE feature of PTC was assessed by analyzing the receiver operating characteristic (ROC) curve. Results: Totally 132 patients were assessed in this study, including 92 and 40 in the training and test cohorts, respectively). Next, the 16 top-performing features, including 4, 7 and 5 from diffusion weighted (DWI), T2-weighted (T2 WI), and contrast-enhanced T1-weighted (CE-T1WI) images, respectively, were finally retained to construct the radiomics signature. There were 8 RLM, 5 CM, 2 shape, and 1 SZM features. The radiomics prediction model achieved AUCs of 0.96 and 0.87 in the training and testing sets, respectively.Conclusions: Our study indicated that MRI radiomics approach had the potential to stratify patients based on ETE in PTCs preoperatively.

“…Radiomics transforms medical images into quantitative indexes through high-throughput extraction by data-assessment algorithms for predicting important clinical outcomes [12,25]. However, there are few published reports applying radiomics to explore the functional outcomes of ischemic stroke cases, leaving a gap in knowledge.…”

Section: Discussionmentioning

confidence: 99%

A clinical-radiomics nomogram for functional outcome predictions in ischemic stroke

Sun

Wu³

et al. 2020

Preprint

Self Cite

Background: Stroke remains the reading cause of death and disability worldwide. Effective and prompt prognostic evaluation is vital for management strategy choice. Radiomics is an emerging non-invasive method used to identify the quantitative imaging indicators for predicting important clinical outcomes. This study was to investigate and validate a radiomics nomogram for predicting ischemic stroke prognosis using the modified Rankin scale (mRS).Methods: Totally 598 consecutive patients with subacute infarction confirmed by diffusion-weighted imaging (DWI) in 2018-2019 were retrospectively assessed. They were assigned to the good (mRS ≤ 2) and poor (mRS > 2) functional outcome groups, respectively. Then, 399 and 199 patients examined by MR scanners 1 and 2, respectively, were assigned to the training and validation cohorts, respectively. Infarction lesions underwent manual segmentation on DWI, extracting 402 radiomic properties, including infarction volume. A radiomics nomogram encompassing patient characteristics and the radiomics signature was built using a multivariate logistic regression model. The performance of the nomogram was evaluated in the training and validation cohorts. Ultimately, decision curve analysis was implemented to assess the clinical value of the nomogram.Results: Stroke lesion volume showed average performance with an area under the curve (AUC) of 0.678. The radiomics signature, including 11 radiomics features, exhibited good prediction performance. Meanwhile, the radiomics nomogram, encompassing clinical characteristics (age, hemorrhage and 24 h National Institutes of Health Stroke Scale score) and the radiomics signature, presented good discriminatory potential in the training cohort (AUC=0.80; 95% confidence interval [CI] 0.75–0.86), which was validated in the validation cohort (AUC=0.73; 95%CI 0.63–0.82). In addition, it demonstrated good calibration in the training (p=0.55) and validation (p=0.21) cohorts. Decision curve analysis confirmed the clinical value of this nomogram.Conclusions: The novel noninvasive clinical-radiomics nomogram shows good performance in predicting ischemic stroke prognosis.