The Image Biomarker Standardization Initiative: Standardized Quantitative Radiomics for High-Throughput Image-based Phenotyping

Radiology, Journal Year: 2020, Volume and Issue: 295(2), P. 328 - 338

Published: March 10, 2020

The image biomarker standardisation initiative (IBSI) is an independent international collaboration which works towards standardising the extraction of biomarkers from acquired imaging for purpose high-throughput quantitative analysis (radiomics). Lack reproducibility and validation studies considered to be a major challenge field. Part this lies in scantiness consensus-based guidelines definitions process translating into biomarkers. IBSI therefore seeks provide nomenclature definitions, benchmark data sets, values verify processing calculations, as well reporting guidelines, analysis.

Language: Английский

---

Introduction to Radiomics

Journal of Nuclear Medicine, Journal Year: 2020, Volume and Issue: 61(4), P. 488 - 495

Published: Feb. 14, 2020

Radiomics is a rapidly evolving field of research concerned with the extraction quantitative metrics—the so-called radiomic features—within medical images. Radiomic features capture tissue and lesion characteristics such as heterogeneity shape may, alone or in combination demographic, histologic, genomic, proteomic data, be used for clinical problem solving. The goal this continuing education article to provide an introduction field, covering basic radiomics workflow: feature calculation selection, dimensionality reduction, data processing. Potential applications nuclear medicine that include PET radiomics-based prediction treatment response survival will discussed. Current limitations radiomics, sensitivity acquisition parameter variations, common pitfalls also covered.

Language: Английский

---

Radiomics in medical imaging—“how-to” guide and critical reflection

Insights into Imaging, Journal Year: 2020, Volume and Issue: 11(1)

Published: Aug. 12, 2020

Abstract Radiomics is a quantitative approach to medical imaging, which aims at enhancing the existing data available clinicians by means of advanced mathematical analysis. Through extraction spatial distribution signal intensities and pixel interrelationships, radiomics quantifies textural information using analysis methods from field artificial intelligence. Various studies different fields in imaging have been published so far, highlighting potential enhance clinical decision-making. However, faces several important challenges, are mainly caused various technical factors influencing extracted radiomic features. The aim present review twofold: first, we typical workflow deliver practical “how-to” guide for Second, discuss current limitations radiomics, suggest improvements, summarize relevant literature on subject.

Language: Английский

---

CheckList for EvaluAtion of Radiomics research (CLEAR): a step-by-step reporting guideline for authors and reviewers endorsed by ESR and EuSoMII

Insights into Imaging, Journal Year: 2023, Volume and Issue: 14(1)

Published: May 4, 2023

Even though radiomics can hold great potential for supporting clinical decision-making, its current use is mostly limited to academic research, without applications in routine practice. The workflow of complex due several methodological steps and nuances, which often leads inadequate reporting evaluation, poor reproducibility. Available guidelines checklists artificial intelligence predictive modeling include relevant good practices, but they are not tailored radiomic research. There a clear need complete checklist study planning, manuscript writing, evaluation during the review process facilitate repeatability reproducibility studies. We here present documentation standard research that guide authors reviewers. Our motivation improve quality reliability and, turn, name CLEAR (CheckList EvaluAtion Radiomics research), convey idea being more transparent. With 58 items, should be considered standardization tool providing minimum requirements presenting In addition dynamic online version checklist, public repository has also been set up allow community comment on items adapt future versions. Prepared revised by an international group experts using modified Delphi method, we hope will serve well as single scientific reviewers literature.

Language: Английский

---

Artificial intelligence: Deep learning in oncological radiomics and challenges of interpretability and data harmonization

Physica Medica, Journal Year: 2021, Volume and Issue: 83, P. 108 - 121

Published: March 1, 2021

Over the last decade there has been an extensive evolution in Artificial Intelligence (AI) field. Modern radiation oncology is based on exploitation of advanced computational methods aiming to personalization and high diagnostic therapeutic precision. The quantity available imaging data increased developments Machine Learning (ML), particularly Deep (DL), triggered research uncovering "hidden" biomarkers quantitative features from anatomical functional medical images. Neural Networks (DNN) have achieved outstanding performance broad implementation image processing tasks. Lately, DNNs considered for radiomics their potentials explainable AI (XAI) may help classification prediction clinical practice. However, most them are using limited datasets lack generalized applicability. In this study we review basics feature extraction, analysis, major interpretability that enable AI. Furthermore, discuss crucial requirement multicenter recruitment large datasets, increasing variability, so as establish potential value development robust models.

Language: Английский

---

METhodological RadiomICs Score (METRICS): a quality scoring tool for radiomics research endorsed by EuSoMII

Insights into Imaging, Journal Year: 2024, Volume and Issue: 15(1)

Published: Jan. 17, 2024

Abstract Purpose To propose a new quality scoring tool, METhodological RadiomICs Score (METRICS), to assess and improve research of radiomics studies. Methods We conducted an online modified Delphi study with group international experts. It was performed in three consecutive stages: Stage#1, item preparation; Stage#2, panel discussion among EuSoMII Auditing Group members identify the items be voted; Stage#3, four rounds exercise by panelists determine eligible for METRICS their weights. The consensus threshold 75%. Based on median ranks derived from expert opinion rank-sum based conversion importance scores, category weights were calculated. Result In total, 59 19 countries participated selection ranking categories. Final tool included 30 within 9 According weights, categories descending order importance: design, imaging data, image processing feature extraction, metrics comparison, testing, processing, preparation modeling, segmentation, open science. A web application repository developed streamline calculation score collect feedback community. Conclusion this work, we assessing methodological research, large protocol. With its conditional format cover variations, it provides well-constructed framework key concepts radiomic papers. Critical relevance statement assessment is made available domain experts, transparent methodology, aiming at evaluating improving machine learning. Key points • METRICS, proposed presents opinion-based methodology first time. accounts varying use cases, handcrafted entirely deep learning-based pipelines. has been help ( https://metricsscore.github.io/metrics/METRICS.html ) created community https://github.com/metricsscore/metrics ). Graphical

Language: Английский

---

Joint EANM/SNMMI guideline on radiomics in nuclear medicine

European Journal of Nuclear Medicine and Molecular Imaging, Journal Year: 2022, Volume and Issue: 50(2), P. 352 - 375

Published: Nov. 3, 2022

The purpose of this guideline is to provide comprehensive information on best practices for robust radiomics analyses both hand-crafted and deep learning-based approaches.

Language: Английский

---

Artificial intelligence in pancreatic cancer

Theranostics, Journal Year: 2022, Volume and Issue: 12(16), P. 6931 - 6954

Published: Jan. 1, 2022

Pancreatic cancer is the deadliest disease, with a five-year overall survival rate of just 11%.The pancreatic patients diagnosed early screening have median nearly ten years, compared 1.5 years for those not screening.Therefore, diagnosis and treatment are particularly critical.However, as rare general cost high, accuracy existing tumor markers enough, efficacy methods exact.In terms diagnosis, artificial intelligence technology can quickly locate high-risk groups through medical images, pathological examination, biomarkers, other aspects, then lesions early.At same time, algorithm also be used to predict recurrence risk, metastasis, therapy response which could affect prognosis.In addition, widely in health records, estimating imaging parameters, developing computer-aided systems, etc. Advances AI applications will require concerted effort among clinicians, basic scientists, statisticians, engineers.Although it has some limitations, play an essential role overcoming foreseeable future due its mighty computing power.

Language: Английский

---

Impact of feature harmonization on radiogenomics analysis: Prediction of EGFR and KRAS mutations from non-small cell lung cancer PET/CT images

Computers in Biology and Medicine, Journal Year: 2022, Volume and Issue: 142, P. 105230 - 105230

Published: Jan. 11, 2022

To investigate the impact of harmonization on performance CT, PET, and fused PET/CT radiomic features toward prediction mutations status, for epidermal growth factor receptor (EGFR) Kirsten rat sarcoma viral oncogene (KRAS) genes in non-small cell lung cancer (NSCLC) patients.Radiomic were extracted from tumors delineated wavelet images obtained 136 histologically proven NSCLC patients. Univariate multivariate predictive models developed using before after ComBat to predict EGFR KRAS mutation statuses. Multivariate built minimum redundancy maximum relevance feature selection random forest classifier. We utilized 70/30% splitting patient datasets training/testing, respectively, repeated procedure 10 times. The area under receiver operator characteristic curve (AUC), accuracy, sensitivity, specificity used assess model performance. (univariate multivariate), was compared statistical analyses.While most univariate modeling significantly improved prediction, did not show any significant difference prediction. Average AUCs all both (q-value < 0.05) following harmonization. mean ranges increased 0.87-0.90 0.92-0.94 EGFR, 0.85-0.90 0.91-0.94 KRAS. highest achieved by harmonized F_R0.66_W0.75 with AUC 0.94, 0.93 KRAS, respectively.Our results demonstrated that regarding modelling, while had generally a better status its effect is feature-dependent. Hence, no systematic observed. Regarding models, radiomics more successful statuses Thus, eliminating batch multi-centric sets, promising tool developing robust reproducible vast variant datasets.

Language: Английский

---

The Image Biomarker Standardization Initiative: Standardized Convolutional Filters for Reproducible Radiomics and Enhanced Clinical Insights

Radiology, Journal Year: 2024, Volume and Issue: 310(2)

Published: Feb. 1, 2024

Filters are commonly used to enhance specific structures and patterns in images, such as vessels or peritumoral regions, enable clinical insights beyond the visible image using radiomics. However, their lack of standardization restricts reproducibility translation radiomics decision support tools. In this special report, teams researchers who developed software participated a three-phase study (September 2020 December 2022) establish standardized set filters. The first two phases focused on finding reference filtered images feature values for convolutional filters: mean, Laplacian Gaussian, Laws Gabor kernels, separable nonseparable wavelets (including decomposed forms), Riesz transformations. phase, 15 digital phantoms 33 36 filter configurations. phase 2, 11 chest CT derive 323 396 features computed from 22 processing Reference transformations were not established. Reproducibility filters was validated public data multimodal imaging (CT, fluorodeoxyglucose PET, T1-weighted MRI) 51 patients with soft-tissue sarcoma. At validation, 486 nine configurations × three modalities assessed lower bounds 95% CIs intraclass correlation coefficients. Out features, 458 found be reproducible across coefficients greater than 0.75. conclusion, eight types verifying calibrating packages. A web-based tool is available compliance checking. © RSNA, 2024 Supplemental material article. See also editorial by Huisman D'Antonoli issue.

Language: Английский

---