Standardization of imaging methods for machine learning in neuro-oncology

Li, Xiaotian; Huang, Raymond Y.

doi:10.1093/noajnl/vdaa054

Cited by 21 publications

(16 citation statements)

References 66 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Moreover, a vast body of literature exists dealing with radiomics, deep learning and machine learning with special emphasis on ( 18 F-FET) PET and hybrid imaging in neurooncology (43)(44)(45)(46)(47)(48)(49), not just for the differentiation of treatmentrelated changes from real progression (44,50,51), but also for the predication of prognostically relevant mutations such as the IDH-mutation (52). Hence, it needs to be evaluated, if further PET-based analyses with the extraction of radiomic features may add value to the conventional image analysis in order to noninvasively identify the TERTp-mutational status.…”

Section: Discussionmentioning

confidence: 99%

TERT-Promoter Mutational Status in Glioblastoma – Is There an Association With Amino Acid Uptake on Dynamic 18F-FET PET?

Ruf¹,

Rohr

Suchorska

et al. 2021

Front. Oncol.

View full text Add to dashboard Cite

ObjectiveThe mutation of the ‘telomerase reverse transcriptase gene promoter’ (TERTp) has been identified as an important factor for individual prognostication and tumorigenesis and will be implemented in upcoming glioma classifications. Uptake characteristics on dynamic 18F-FET PET have been shown to serve as additional imaging biomarker for prognosis. However, data on the correlation of TERTp-mutational status and amino acid uptake on dynamic 18F-FET PET are missing. Therefore, we aimed to analyze whether static and dynamic 18F-FET PET parameters are associated with the TERTp-mutational status in de-novo IDH-wildtype glioblastoma and whether a TERTp-mutation can be predicted by dynamic 18F-FET PET.MethodsPatients with de-novo IDH-wildtype glioblastoma, WHO grade IV, available TERTp-mutational status and dynamic 18F-FET PET scan prior to any therapy were included. Here, established clinical parameters maximal and mean tumor-to-background-ratios (TBRmax/TBRmean), the biological-tumor-volume (BTV) and minimal-time-to-peak (TTPmin) on dynamic PET were analyzed and correlated with the TERTp-mutational status.ResultsOne hundred IDH-wildtype glioblastoma patients were evaluated; 85/100 of the analyzed tumors showed a TERTp-mutation (C228T or C250T), 15/100 were classified as TERTp-wildtype. None of the static PET parameters was associated with the TERTp-mutational status (median TBRmax 3.41 vs. 3.32 (p=0.362), TBRmean 2.09 vs. 2.02 (p=0.349) and BTV 26.1 vs. 22.4 ml (p=0.377)). Also, the dynamic PET parameter TTPmin did not differ in both groups (12.5 vs. 12.5 min, p=0.411). Within the TERTp-mutant subgroups (i.e., C228T (n=23) & C250T (n=62)), the median TBRmax (3.33 vs. 3.69, p=0.095), TBRmean (2.08 vs. 2.09, p=0.352), BTV (25.4 vs. 30.0 ml, p=0.130) and TTPmin (12.5 vs. 12.5 min, p=0.190) were comparable, too.ConclusionUptake characteristics on dynamic 18F-FET PET are not associated with the TERTp-mutational status in glioblastoma However, as both, dynamic 18F-FET PET parameters as well as the TERTp-mutation status are well-known prognostic biomarkers, future studies should investigate the complementary and independent prognostic value of both factors in order to further stratify patients into risk groups.

show abstract

Section: Discussionmentioning

confidence: 99%

TERT-Promoter Mutational Status in Glioblastoma – Is There an Association With Amino Acid Uptake on Dynamic 18F-FET PET?

Ruf¹,

Rohr

Suchorska

et al. 2021

Front. Oncol.

View full text Add to dashboard Cite

show abstract

“…Thus, establishment of TCIA allows researchers to train and validate their prediction models on imaging data acquired from other clinical sites to help researchers develop more accurate and robust models that can eventually be translated to the clinic. Additionally, developing and implementing image pre-processing algorithms to effectively standardize or normalize images acquired from different machines or clinic sites ( 146 , 147 ) have also attracted research interest and effort, which may also need before AI-based models can be adopted on a widescale clinical level.…”

Section: Discussion – Outlook and Challengesmentioning

confidence: 99%

Applying artificial intelligence technology to assist with breast cancer diagnosis and prognosis prediction

et al. 2022

View full text Add to dashboard Cite

Breast cancer remains the most diagnosed cancer in women. Advances in medical imaging modalities and technologies have greatly aided in the early detection of breast cancer and the decline of patient mortality rates. However, reading and interpreting breast images remains difficult due to the high heterogeneity of breast tumors and fibro-glandular tissue, which results in lower cancer detection sensitivity and specificity and large inter-reader variability. In order to help overcome these clinical challenges, researchers have made great efforts to develop computer-aided detection and/or diagnosis (CAD) schemes of breast images to provide radiologists with decision-making support tools. Recent rapid advances in high throughput data analysis methods and artificial intelligence (AI) technologies, particularly radiomics and deep learning techniques, have led to an exponential increase in the development of new AI-based models of breast images that cover a broad range of application topics. In this review paper, we focus on reviewing recent advances in better understanding the association between radiomics features and tumor microenvironment and the progress in developing new AI-based quantitative image feature analysis models in three realms of breast cancer: predicting breast cancer risk, the likelihood of tumor malignancy, and tumor response to treatment. The outlook and three major challenges of applying new AI-based models of breast images to clinical practice are also discussed. Through this review we conclude that although developing new AI-based models of breast images has achieved significant progress and promising results, several obstacles to applying these new AI-based models to clinical practice remain. Therefore, more research effort is needed in future studies.

show abstract

“…One reason is the current overstandardization of methods. This limits the use of developed radiomics models across universities and clinical contexts [49].…”

Section: Limitationmentioning

confidence: 99%

Artificial intelligence for radiomics; diagnostic biomarkers for neuro-oncology

Vahedifard¹,

Hassani²,

Afrasiabi³

et al. 2022

World J. Adv. Res. Rev.

View full text Add to dashboard Cite

Recent advances in medical image analysis have been made to improve our understanding of how disease develops, behaves, and responds to treatment. Magnetic resonance imaging (MRI) and positron emission tomography (PET) advanced imaging strategies provide structural and functional phenotypic biomarkers that correlate with key disease processes. Through radiomics and radiogenomics, ML-medical imaging has opened up new perspectives in high-grade glioma diagnosis. As a result, non-invasive and in vivo biomarkers for patient survival, tumor recurrence, and genomics are identified. Tumor genomic imaging signatures can help identify patients who benefit from targeted therapies. Molecular characterization of gliomas and prediction of their evolution would allow treatment optimization. Radiomics-based biomarkers allow for a more in-depth analysis of pathophysiologic processes and insights into diagnosing better, classifying, stratifying, and prognosticating brain tumors and assessing their response to therapy. Radiomics is a new data-driven approach that can help answer clinical questions like diagnosis, prognosis, and treatment response. With encouraging outcomes in brain tumor patients, radiomics and deep learning are still not widely used in clinical practice, requiring more extensive and practical clinical studies.

show abstract

Standardization of imaging methods for machine learning in neuro-oncology

Cited by 21 publications

References 66 publications

TERT-Promoter Mutational Status in Glioblastoma – Is There an Association With Amino Acid Uptake on Dynamic 18F-FET PET?

TERT-Promoter Mutational Status in Glioblastoma – Is There an Association With Amino Acid Uptake on Dynamic 18F-FET PET?

Applying artificial intelligence technology to assist with breast cancer diagnosis and prognosis prediction

Artificial intelligence for radiomics; diagnostic biomarkers for neuro-oncology

Contact Info

Product

Resources

About