Revealing Tumor Habitats from Texture Heterogeneity Analysis for Classification of Lung Cancer Malignancy and Aggressiveness

Cherezov, Dmitry; Goldgof, Dmitry B.; Hall, Lawrence O.; Gillies, Robert J.; Schabath, Matthew B.; Müller, Henning; Depeursinge, Adrien

doi:10.1038/s41598-019-38831-0

Cited by 38 publications

(30 citation statements)

References 48 publications

(33 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The COBRIT study was a phase 3, double-blind, randomized clinical trial conducted over a span of 4 years to investigate the effects of citicoline compared to placebo on patients with TBI [11]. The study sample consisted of 1213 non-penetrating TBI patients (ages 18-70 years) with diverse severity levels according to GCS scores (3)(4)(5)(6)(7)(8)(9)(10)(11)(12)(13)(14)(15). It includes baseline data on demographics, injury information, and metabolic, liver and hematologic functions.…”

Section: Input Features (Phenotypes) Extractionmentioning

confidence: 99%

“…Some subjects may experience single or repetitive concussion (mild TBI) [3]. Sorting out the heterogeneity present in clinical data, though challenging, has the potential to reveal insights that could aid clinicians [4]. This study investigates an effective framework to characterize heterogeneous clinical data, specifically TBI, using unsupervised learning methods guided by domain expert knowledge and supplemented by statistical analyses to aid in applicability to prognostic and diagnostic analysis.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

An Explainable and Statistically Validated Ensemble Clustering Model Applied to the Identification of Traumatic Brain Injury Subgroups

et al. 2020

View full text Add to dashboard Cite

We present a framework for an explainable and statistically validated ensemble clustering model applied to Traumatic Brain Injury (TBI). The objective of our analysis is to identify patient injury severity subgroups and key phenotypes that delineate these subgroups using varied clinical and computed tomography data. Explainable and statistically-validated models are essential because a datadriven identification of subgroups is an inherently multidisciplinary undertaking. In our case, this procedure yielded six distinct patient subgroups with respect to mechanism of injury, severity of presentation, anatomy, psychometric, and functional outcome. This framework for ensemble cluster analysis fully integrates statistical methods at several stages of analysis to enhance the quality and the explainability of results. This methodology is applicable to other clinical data sets that exhibit significant heterogeneity as well as other diverse data science applications in biomedicine and elsewhere.

show abstract

Section: Input Features (Phenotypes) Extractionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

An Explainable and Statistically Validated Ensemble Clustering Model Applied to the Identification of Traumatic Brain Injury Subgroups

et al. 2020

View full text Add to dashboard Cite

show abstract

“…Finally, Cherezov et al 11proposed a method for revealing tumor habitats using texture features based on the well-known concept that tumors are heterogeneous, and the level of heterogeneity may help to identify the malignancy and aggressiveness of tumors. Those results demonstrated an AUC of 0.9 and an accuracy of 85% to discriminate long-term and short-term survival rates among patients with lung cancer (11).…”

Section: Mediastinal Lymph Node Evaluationmentioning

confidence: 94%

Radiomics in Lung Cancer from Basic to Advanced: Current Status and Future Directions

et al. 2020

View full text Add to dashboard Cite

Ideally, radiomics features and radiomics signatures can be used as imaging biomarkers for diagnosis, staging, prognosis, and prediction of tumor response. Thus, the number of published radiomics studies is increasing exponentially, leading to a myriad of new radiomics-based evidence for lung cancer. Consequently, it is challenging for radiologists to keep up with the development of radiomics features and their clinical applications. In this article, we review the basics to advanced radiomics in lung cancer to guide young researchers who are eager to start exploring radiomics investigations. In addition, we also include technical issues of radiomics, because knowledge of the technical aspects of radiomics supports a well-informed interpretation of the use of radiomics in lung cancer.

show abstract

“…9,10 In the domain of lung cancer, radiomics-based models have been demonstrated to predict overall survival, 11 response to therapy, [12][13][14][15] tumor characterization 16 and malignancy identification. [17][18][19][20][21] Of the radiomics-based applications proposed in the literature to classify benign from malignant lesions in lung cancer [17][18][19][20][21] however, few have been externally validated to evaluate their generalizability to datasets independent from the ones used for training. 22 External validation is important in demonstrating the feature robustness 23 and predictive performance of the model on independent datasets, 24,25 as these are critical determinants to clinical adoption.…”

Section: Introductionmentioning

confidence: 99%

“…In this context, the radiomics concept of extracting features describing tumor characteristics such as intensity, shape, and heterogeneity from medical imaging data to identify those that correlate with clinically useful outcomes, has gained prominence 9,10 . In the domain of lung cancer, radiomics‐based models have been demonstrated to predict overall survival, 11 response to therapy, 12–15 tumor characterization 16 and malignancy identification 17–21 . Of the radiomics‐based applications proposed in the literature to classify benign from malignant lesions in lung cancer 17–21 however, few have been externally validated to evaluate their generalizability to datasets independent from the ones used for training 22 .…”

Section: Introductionmentioning

confidence: 99%

External validation of radiomics‐based predictive models in low‐dose CT screening for early lung cancer diagnosis

et al. 2020

View full text Add to dashboard Cite

Purpose Low‐dose CT screening allows early lung cancer detection, but is affected by frequent false positive results, inter/intra observer variation and uncertain diagnoses of lung nodules. Radiomics‐based models have recently been introduced to overcome these issues, but limitations in demonstrating their generalizability on independent datasets are slowing their introduction to clinic. The aim of this study is to evaluate two radiomics‐based models to classify malignant pulmonary nodules in low‐dose CT screening, and to externally validate them on an independent cohort. The effect of a radiomics features harmonization technique is also investigated to evaluate its impact on the classification of lung nodules from a multicenter data. Methods Pulmonary nodules from two independent cohorts were considered in this study; the first cohort (110 subjects, 113 nodules) was used to train prediction models, and the second cohort (72 nodules) to externally validate them. Literature‐based radiomics features were extracted and, after feature selection, used as predictive variables in models for malignancy identification. An in‐house prediction model based on artificial neural network (ANN) was implemented and evaluated, along with an alternative model from the literature, based on a support vector machine (SVM) classifier coupled with a least absolute shrinkage and selection operator (LASSO). External validation was performed on the second cohort to evaluate models’ generalization ability. Additionally, the impact of the Combat harmonization method was investigated to compensate for multicenter datasets variabilities. A new training of the models based on harmonized features was performed on the first cohort, then tested separately on the harmonized and non‐harmonized features of the second cohort. Results Preliminary results showed a good accuracy of the investigated models in distinguishing benign from malignant pulmonary nodules with both sets of radiomics features (i.e., non‐harmonized and harmonized). The performance of the models, quantified in terms of Area Under the Curve (AUC), was > 0.89 in the training set and > 0.82 in the external validation set for all the investigated scenarios, outperforming the clinical standard (AUC of 0.76). Slightly higher performance was observed for the SVM‐LASSO model than the ANN in the external dataset, although they did not result significantly different. For both harmonized and non‐harmonized features, no statistical difference was found between Receiver operating characteristic (ROC) curves related to training and test set for both models. Conclusions Although no significant improvements were observed when applying the Combat harmonization method, both in‐house and literature‐based models were able to classify lung nodules with good generalization to an independent dataset, thus showing their potential as tools for clinical decision‐making in lung cancer screening.

show abstract

Revealing Tumor Habitats from Texture Heterogeneity Analysis for Classification of Lung Cancer Malignancy and Aggressiveness

Cited by 38 publications

References 48 publications

An Explainable and Statistically Validated Ensemble Clustering Model Applied to the Identification of Traumatic Brain Injury Subgroups

An Explainable and Statistically Validated Ensemble Clustering Model Applied to the Identification of Traumatic Brain Injury Subgroups

Radiomics in Lung Cancer from Basic to Advanced: Current Status and Future Directions

External validation of radiomics‐based predictive models in low‐dose CT screening for early lung cancer diagnosis

Contact Info

Product

Resources

About