Radiomic features on MRI enable risk categorization of prostate cancer patients on active surveillance: Preliminary findings

Algohary, Ahmad; Viswanath, Satish; Shiradkar, Rakesh; Ghose, Soumya; Pahwa, Shivani; Moses, Daniel; Jambor, Ivan; Shnier, Ron; Böhm, Maret; Haynes, Anne Maree; Brenner, Phillip; Delprado, Warick; Thompson, James; Pulbrock, Marley; Purysko, Andrei S.; Verma, Sadhna; Ponsky, Lee; Stricker, Phillip D.; Madabhushi, Anant

doi:10.1002/jmri.25983

Cited by 94 publications

(79 citation statements)

References 42 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…The voxels were tagged as cancerous or noncancerous using a support vector machine (SVM) classifier . In Algohary et al, the prostate was segmented into areas according to the aggressiveness between malignant and normal regions in the training groups. A voxel‐wise random forest model (RF) with a conditional random field spatial regulation was used to classify the voxels in multimodal MRI (T1, Contrast — Enhanced (CE) T1, T2, and FLAIR) of the brain of glioblastoma multiforme (GBM) patients into five classes: nontumor region and four tumor subregions including necrosis, edema, nonenhancing area, and enhancing area area .…”

Section: Overview Of Research and Clinical Applications Of Cancer Radmentioning

confidence: 99%

Machine and deep learning methods for radiomics

et al. 2020

View full text Add to dashboard Cite

Radiomics is an emerging area in quantitative image analysis that aims to relate large‐scale extracted imaging information to clinical and biological endpoints. The development of quantitative imaging methods along with machine learning has enabled the opportunity to move data science research towards translation for more personalized cancer treatments. Accumulating evidence has indeed demonstrated that noninvasive advanced imaging analytics, that is, radiomics, can reveal key components of tumor phenotype for multiple three‐dimensional lesions at multiple time points over and beyond the course of treatment. These developments in the use of CT, PET, US, and MR imaging could augment patient stratification and prognostication buttressing emerging targeted therapeutic approaches. In recent years, deep learning architectures have demonstrated their tremendous potential for image segmentation, reconstruction, recognition, and classification. Many powerful open‐source and commercial platforms are currently available to embark in new research areas of radiomics. Quantitative imaging research, however, is complex and key statistical principles should be followed to realize its full potential. The field of radiomics, in particular, requires a renewed focus on optimal study design/reporting practices and standardization of image acquisition, feature calculation, and rigorous statistical analysis for the field to move forward. In this article, the role of machine and deep learning as a major computational vehicle for advanced model building of radiomics‐based signatures or classifiers, and diverse clinical applications, working principles, research opportunities, and available computational platforms for radiomics will be reviewed with examples drawn primarily from oncology. We also address issues related to common applications in medical physics, such as standardization, feature extraction, model building, and validation.

show abstract

Section: Overview Of Research and Clinical Applications Of Cancer Radmentioning

confidence: 99%

Machine and deep learning methods for radiomics

et al. 2020

View full text Add to dashboard Cite

show abstract

“…Quantitative assessment of the change in diffusion-weighted imaging (DWI) over time is of interest as a way to detect who develop clinically significant disease whilst on surveillance [4,[6][7][8]. This technique measures the diffusion of water molecules within the extracellular space by the calculation of a quantitative parameter called apparent diffusion coefficient (ADC) [9].…”

mentioning

confidence: 99%

DWI and PRECISE criteria in men on active surveillance for prostate cancer: A multicentre preliminary experience of different ADC calculations

Giganti

Pecoraro

Fierro

et al. 2020

Magnetic Resonance Imaging

View full text Add to dashboard Cite

A B S T R A C TPurpose: The PRECISE score estimates the likelihood of radiological progression in patients on active surveillance (AS) for prostate cancer (PCa) with serial multiparametric magnetic resonance imaging (mpMRI). A PRECISE score of 1 or 2 denotes radiological regression, PRECISE 3 indicates stability and PRECISE 4 or 5 implies progression.We evaluated the inter-reader reproducibility of different apparent diffusion coefficient (ADC) calculations and their relationship to the PRECISE score. Material and methods: Baseline and follow-up scans (on the same MR systems) of 30 patients with visible lesions from two different institutions (University College London and Sapienza University of Rome) were analysed by two radiologists (one from each site). The PRECISE score was initially assessed in consensus. At least six weeks later, to reduce the likelihood of being influenced by the consensus PRECISE reading, each radiologist independently calculated ADC for the following: lesion, non-cancerous tissue and urine in the bladder. Normalised ADC ratios were calculated with respect to normal prostatic tissue (npADC) and urine. Spearman's correlation (ρ), intraclass correlation coefficients (ICC), differences in ADC and ROC curves were computed. Results: Interobserver reproducibility was very good (ρ > 0.8; ICC > 0.90). Lesion ADC (0.91 vs 0.73 × 10 −3 mm 2 /s; p=0.025) and npADC ratio (0.68 vs 0.53; p=0.012) at follow-up mpMRI were different between patients with radiological regression or stability vs progression. Cut-offs of 0.77 × 10 −3 mm 2 /s (lesion ADC) and 0.59 (npADC ratio) could differentiate the two groups (area under the curve: 0.74 and 0.77, respectively). Conclusion: The ADC, npADC ratio and the PRECISE score should be recorded for MRI-based AS.

show abstract

“…24,25 The evolution of nerve texture analysis is represented by the advent of radiomics, an advanced quantitative image analysis that extracts a large amount of data from medical images, with the final outcome of providing information that is not visible to the human eye. [38][39][40][41][42][43][44][45][46][47][48][49][50][51][52][53][54] The possibility of studying the phenotype of peripheral nerves on images acquired with standard proto-cols and analyzing these images with widely available radiomics software packages could open new possibilities to study peripheral nerve pathology far beyond simple CSA evaluation. 7,9,10,18 MRI of peripheral nerves is also challenging mostly due to the thin nature of the nerves, the difficulties in selecting appropriate nerve boundaries, the difficulties in image interpretation, and the complex anatomy.…”

Section: Nerve Echotexture Evaluationmentioning

confidence: 99%

Peripheral Nerves: Not Only Cross-sectional Area in the Era of Radiomics

Tagliafico

González

Rossi

et al. 2020

Semin Musculoskelet Radiol

View full text Add to dashboard Cite

The peripheral nervous system is increasingly being investigated using medical imaging as a complement or in association with electrodiagnostics tests. The application of imaging techniques, such as ultrasound (US) and magnetic resonance imaging (MRI), allows detailed visualization of the peripheral nervous system. According to the European Society of Musculoskeletal Radiology, the use of US for nerve evaluation is strongly encouraged. In addition, the role of US is further enhanced by the wide application of US-guided techniques to diagnose or to treat peripheral nerve disorders.Standard evaluation of peripheral nerves on US usually relies on cross-sectional area evaluation with different cutoff values in the osteofibrous tunnels and outside them. In several anatomical areas, side-to-side comparison is highly recommended because it helps distinguish subtle variations by using the unaffected limb as an internal control.US is widely used to perform US-guided interventional procedures on peripheral nerves. The recent development of radiomics and machine and deep learning applied to peripheral nerves may reveal new insights beyond the capabilities of the human eye. Radiomics may have a role in expanding the diagnostic capabilities of US and MRI in the study of peripheral nerve pathology, especially when the cross-sectional area is not markedly increased.

show abstract

Radiomic features on MRI enable risk categorization of prostate cancer patients on active surveillance: Preliminary findings

Abstract: 3 Technical Efficacy: Stage 2 J. Magn. Reson. Imaging 2018.

Cited by 94 publications

References 42 publications

Machine and deep learning methods for radiomics

Machine and deep learning methods for radiomics

DWI and PRECISE criteria in men on active surveillance for prostate cancer: A multicentre preliminary experience of different ADC calculations

Peripheral Nerves: Not Only Cross-sectional Area in the Era of Radiomics

Contact Info

Product

Resources

About