A review in radiomics: Making personalized medicine a reality via routine imaging

Guiot, Julien; Vaidyanathan, Akshayaa; Deprez, Louis; Zerka, Fadila; Danthine, Denis; Frix, Anne‐Noëlle; Lambin, Philippe; Bottari, Fabio; Tsoutzidis, Nathan; Miraglio, Benjamin; Walsh, Seán; Vos, Wim; Hustinx, Roland; Ferreira, Marta S.; Lovinfosse, Pierre; Leijenaar, Ralph T.H.

doi:10.1002/med.21846

Cited by 152 publications

(85 citation statements)

References 147 publications

(245 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Furthermore, in this study, three radiomic features were selected for the EOAD-YC groups-namely, kurtosis, coarseness, and informational measure of correlation 1 (IMC1). Kurtosis measures the degree of histogram sharpness, coarseness reflects the spatial rate of changes in gray-level intensities, and IMC1 captures the spatial relationships of pairs of pixels (Guiot et al, 2022). Concerning the LOAD-OC groups, kurtosis, skewness, inverse difference moment normalized (IDMN), and dependence entropy were filtered.…”

Section: Discussionmentioning

confidence: 99%

Radiomic Features of the Hippocampus for Diagnosing Early-Onset and Late-Onset Alzheimer’s Disease

Zhang

Fang

et al. 2022

Front. Aging Neurosci.

View full text Add to dashboard Cite

Background: Late-onset Alzheimer’s disease (LOAD) and early-onset Alzheimer’s disease (EOAD) are different subtypes of AD. This study aimed to build and validate radiomics models of the hippocampus for EOAD and young controls (YCs), LOAD and old controls (OCs), as well as EOAD and LOAD.Methods: Thirty-six EOAD patients, 36 LOAD patients, 36 YCs, and 36 OCs from the Alzheimer’s Disease Neuroimaging Initiative (ADNI) database were enrolled and allocated to training and test sets of the EOAD-YC groups, LOAD-OC groups, and EOAD-LOAD groups. Independent external validation sets including 15 EOAD patients, 15 LOAD patients, 15 YCs, and 15 OCs from Shanghai Mental Health Center were constructed, respectively. Bilateral hippocampal segmentation and feature extraction were performed for each subject, and the least absolute shrinkage and selection operator (LASSO) method was used to select radiomic features. Support vector machine (SVM) models were constructed based on the identified features to distinguish EOAD from YC subjects, LOAD from OC subjects, and EOAD from LOAD subjects. The areas under the receiver operating characteristic curves (AUCs) were used to evaluate the performance of the models.Results: Three, three, and four features were selected for EOAD and YC subjects, LOAD and OC subjects, and EOAD and LOAD subjects, respectively. The AUC and accuracy of the SVM model were 0.90 and 0.77 in the test set and 0.91 and 0.87 in the validation set for EOAD and YC subjects, respectively; for LOAD and OC subjects, the AUC and accuracy were 0.94 and 0.86 in the test set and 0.92 and 0.78 in the validation set, respectively. For the SVM model of EOAD and LOAD subjects, the AUC was 0.87 and the accuracy was 0.79 in the test set; additionally, the AUC was 0.86 and the accuracy was 0.77 in the validation set.Conclusion: The findings of this study provide insights into the potential of hippocampal radiomic features as biomarkers to diagnose EOAD and LOAD. This study is the first to show that SVM classification analysis based on hippocampal radiomic features is a valuable method for clinical applications in EOAD.

show abstract

Section: Discussionmentioning

confidence: 99%

Radiomic Features of the Hippocampus for Diagnosing Early-Onset and Late-Onset Alzheimer’s Disease

Zhang

Fang

et al. 2022

Front. Aging Neurosci.

View full text Add to dashboard Cite

show abstract

“…It can be challenging to identify obstructive airflow limitation in patients suffering from ILDs due to the reduction of total lung capacity (TLC). Whereas HRCT is nowadays the major way to explore and diagnose ILDs [ 31 , 32 ], it is of low interest in order to identify O-ILD patients. Facing a significant vital forced capacity (FVC) reduction in the ILDs, this can, by definition, reduce the capability of clinicians to confirm obstructive airflow limitation.…”

Section: Introductionmentioning

confidence: 99%

Combined obstructive airflow limitation associated with interstitial lung diseases (O-ILD): the bad phenotype ?

Guiot

Henket²,

Frix³

et al. 2022

Respir Res

Self Cite

View full text Add to dashboard Cite

Background Patients suffering from combined obstructive and interstitial lung disease (O-ILD) represent a pathological entity which still has to be well clinically described. The aim of this descriptive and explorative study was to describe the phenotype and functional characteristics of a cohort of patients suffering from functional obstruction in a population of ILD patients in order to raise the need of dedicated prospective observational studies and the evaluation of the impact of anti-fibrotic therapies. Methods The current authors conducted a retrospective study including 557 ILD patients, with either obstructive (O-ILD, n = 82) or non-obstructive (non O-ILD, n = 475) pattern. Patients included were mainly males (54%) with a mean age of 62 years. Results Patients with O-ILD exhibited a characteristic functional profile with reduced percent predicted forced expired volume in 1 s (FEV1) [65% (53–77) vs 83% (71–96), p < 0.00001], small airway involvement assessed by maximum expiratory flow (MEF) 25/75 [29% (20–41) vs 81% (64–108), p < 0.00001], reduced sGaw [60% (42–75) vs 87% (59–119), p < 0.01] and sub-normal functional residual capacity (FRC) [113% (93–134) vs 92% (75–109), p < 0.00001] with no impaired of carbon monoxide diffusing capacity of the lung (DLCO) compared to those without obstruction. Total lung capacity (TLC) was increased in O-ILD patients [93% (82–107) vs 79% (69–91), p < 0.00001]. Of interest, DLCO sharply dropped in O-ILD patients over a 5-year follow-up. We did not identify a significant increase in mortality in patients with O-ILD. Interestingly, the global mortality was increased in the specific sub-group of patients with O-ILD and no progressive fibrosing ILD phenotype and in those with connective tissue disease associated ILD especially in case of rheumatoid arthritis. Conclusions The authors individualized a specific functional-based pattern of ILD patients with obstructive lung disease, who are at risk of increased mortality and rapid DLCO decline over time. As classically those patients are excluded from clinical trials, a dedicated prospective study would be of interest in order to define more precisely treatment response of those patients.

show abstract

“…chemotherapy, radiotherapy, immunotherapy), treatment response, drug usage, surgery, and genetic mutations. The rise in popularity of radiomics has resulted in a large number of papers and a wide variety of methods (Yip and Aerts, 2016;Rizzo et al, 2018;Traverso et al, 2018;Sollini et al, 2019;Afshar et al, 2019;Parekh and Jacobs, 2019;Bodalal et al, 2019;Song et al, 2020;Avanzo et al, 2020;Aerts, 2016;Guiot et al, 2021). In a new radiomics application, finding the optimal method out of the wide range of available options has to be done manually through a heuristic trial-and-error process.…”

Section: Introductionmentioning

confidence: 99%

Reproducible radiomics through automated machine learning validated on twelve clinical applications

Starmans¹,

Voort²,

Phil³

et al. 2021

Preprint

View full text Add to dashboard Cite

Radiomics uses quantitative medical imaging features to predict clinical outcomes. While many radiomics methods have been described in the literature, these are generally designed for a single application. The aim of this study is to generalize radiomics across applications by proposing a framework to automatically construct and optimize the radiomics workflow per application. To this end, we formulate radiomics as a modular workflow, consisting of several components: image and segmentation preprocessing, feature extraction, feature and sample preprocessing, and machine learning. For each component, a collection of common algorithms is included. To optimize the workflow per application, we employ automated machine learning using a random search and ensembling. We evaluate our method in twelve different clinical applications, resulting in the following area under the curves: 1) liposarcoma (0.83); 2) desmoid-type fibromatosis (0.82); 3) primary liver tumors (0.81); 4) gastrointestinal stromal tumors (0.77); 5) colorectal liver metastases (0.68); 6) melanoma metastases (0.51); 7) hepatocellular carcinoma (0.75); 8) mesenteric fibrosis (0.81); 9) prostate cancer (0.72); 10) glioma (0.70); 11) Alzheimer's disease (0.87); and 12) head and neck cancer (0.84). Concluding, our method fully automatically constructs and optimizes the radiomics workflow, thereby streamlining the search for radiomics biomarkers in new applications. To facilitate reproducibility and future research, we publicly release six datasets, the software implementation of our framework (open-source), and the code to reproduce this study.

show abstract

A review in radiomics: Making personalized medicine a reality via routine imaging

Cited by 152 publications

References 147 publications

Radiomic Features of the Hippocampus for Diagnosing Early-Onset and Late-Onset Alzheimer’s Disease

Radiomic Features of the Hippocampus for Diagnosing Early-Onset and Late-Onset Alzheimer’s Disease

Combined obstructive airflow limitation associated with interstitial lung diseases (O-ILD): the bad phenotype ?

Reproducible radiomics through automated machine learning validated on twelve clinical applications

Contact Info

Product

Resources

About