Radiomics model–based algorithm for preoperative prediction of pancreatic ductal adenocarcinoma grade

Tikhonova, V. S.; Karmazanovsky, Grigory; Kondratyev, Evvgeny V.; Gruzdev, I. S.; Mikhaylyuk, K. A.; Sinelnikov, Mikhail Y.; Revishvili, A. Sh.

doi:10.1007/s00330-022-09046-1

Cited by 8 publications

(11 citation statements)

References 26 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This provides a principle for pathological grading of PDAC through imaging studies. For example, Tikhonova et al 25 used a machine learning algorithm to establish a diagnostic model for image-based PDAC grading based on preoperative CT, using data from 91 patients to establish a diagnostic model. The AUC for pathological grading ≥ 2 (or 3) was 0.75 (or 0.66).…”

Section: Discussionmentioning

confidence: 99%

18 F-FDG-PET/CT based deep learning model for fully automated prediction of pathological grading for pancreatic ductal adenocarcinoma before surgery

Zhang

Bao

Liu

et al. 2023

Preprint

View full text Add to dashboard Cite

Background ：The determination of pathological grading has a guiding significance for the treatment of pancreatic ductal adenocarcinoma（PDAC）patients. However, there is a lack of an accurate and safe method to obtain pathological grading before surgery. The aim of this study is to develop a deep learning（DL）model based on 18F-FDG-PET/CT for a fully automatic prediction of preoperative pathological grading of pancreatic cancer. Results ：A total of 370 PDAC patients from January 2016 to September 2021 were collected retrospectively. All patients underwent 18F-FDG-PET/CT examination before surgery and obtained pathological results after surgery. A DL model for pancreatic cancer lesion segmentation was first developed using 100 of these cases and applied to the remaining cases to obtain lesion regions. After that, all patients were divided into training set, validation set and test set according to the ratio of 5:1:1. A predictive model of pancreatic cancer pathological grade was developed using the features computed from the lesion regions obtained by the lesion segmentation model and key clinical characteristics of the patients. Finally, the stability of the model was verified by 7-fold cross-validation.The Dice score of the developed PET/CT based tumour segmentation model for PDAC was 0.89. The area under curve (AUC) of the PET/CT-based DL model developed on the basis of the segmentation model was 0.74, with an accuracy, sensitivity, and specificity of 0.72, 0.73, and 0.72, respectively. After integrating key clinical data, the AUC of the model improved to 0.77, with its accuracy, sensitivity, and specificity boosted to 0.75, 0.77, and 0.73, respectively. Conclusion ：To the best of our knowledge, this is the first deep learning model to end-to-end predict the pathological grading of PDAC in a fully automatic manner, which is expected to improve clinical decision making.

show abstract

Section: Discussionmentioning

confidence: 99%

18 F-FDG-PET/CT based deep learning model for fully automated prediction of pathological grading for pancreatic ductal adenocarcinoma before surgery

Zhang

Bao

Liu

et al. 2023

Preprint

View full text Add to dashboard Cite

show abstract

“…Four studies showed a CT slice thickness of ≤ 1 mm [16,17,19,20], five a thickness of > 1-3 mm [18,[21][22][23][24], two a thickness of > 3 to 5 mm [25,26], and one had no slice thickness specified [27] (Table 3). Nine of the 12 studies used singlecentre data [16][17][18][20][21][22][23][24][25], 2 studies used data from 2 different centres (one each as a training cohort, one each a validation cohort) [26,27], and one study used data from 5 centres [19].…”

Section: Cohort Size and Ct Techniquementioning

confidence: 99%

“…A CT pancreas protocol was utilised in 5 out of 12 studies [17,18,22,26,27], while no pancreas protocol was applied in 5 out of 12 studies [19-21, 24, 25], and 2 studies did not provide information on whether patients underwent a CT pancreas protocol [16,23]. CT contrast phase used for segmentation varied among studies, with five studies using the pv-phase [16,17,19,20,26], two studies using the late arterial phase [18,24], four studies using multiple phases [22,23,25,27], and one study using no contrast agent [21].…”

Section: Cohort Size and Ct Techniquementioning

confidence: 99%

“…RF selection and analysis methodologies were reproducible, albeit highly variable. In three studies, RFs were chosen a priori [18,20,23], while the remaining nine studies extracted features from the respective software libraries. Software used was PyRadiomics [19,20,26] and LIFEx in three [17,21,23], IBEX [18,24] and MATLAB in two [22,27], and AnalysisKit [16] and Artificial Intelligent Kit in one study [25].…”

Section: Radiomic Feature Extraction and Selectionmentioning

confidence: 99%

“…Analysed outcome variables applied either time-to-event endpoints modelled using Cox-regression techniques [17-22, 25, 26] or logistic regression models for assessment of PDAC grades [23,24], local response [18,21], or superior mesenteric-portal vein invasion [16].…”

Section: Statistical Analysis and Model Developmentmentioning

confidence: 99%

See 2 more Smart Citations

Current state of radiomic research in pancreatic cancer: focusing on study design and reproducibility of findings

et al. 2023

View full text Add to dashboard Cite

Objectives To critically appraise methodology and reproducibility of published studies on CT radiomics of pancreatic ductal adenocarcinoma (PDAC). Methods PRISMA literature search of MEDLINE, PubMed, and Scopus databases was conducted from June to August 2022 relating to CT radiomics human research articles pertaining to PDAC diagnosis, treatment, and/ or prognosis, utilising Image Biomarker Standardisation Initiative-compliant (IBSI) radiomic software. Keyword search included [pancreas OR pancreatic] AND [radiomic OR [quantitative AND imaging] OR [texture AND analysis]]. Analysis included cohort size, CT protocol used, radiomic feature (RF) extraction, segmentation, and selection, software used, outcome correlation, and statistical methodology, with focus on reproducibility. Results Initial search yielded 1112 articles; however, only 12 articles met all inclusion/exclusion criteria. Cohort sizes ranged from 37 to 352 (median = 106, mean = 155.8). CT slice thickness varied among studies (4 using ≤ 1 mm, 5 using > 1 to 3 mm, 2 using > 3 to 5 mm, 1 not specifying). CT protocol varied (5 using a single portal-venous (pv)-phase, 5 using a pancreas protocol, 1 study using a non-contrast protocol). RF extraction and segmentation were heterogeneous (RF extraction: 5 using pv-phase, 2 using late arterial, 4 using multi-phase, 1 using non-contrast phase; RF selection: 3 pre-selected, 9 software-selected). 2D/3D RF segmentation was diverse (2D in 6, 3D in 4, 2D and 3D in 2 studies). Six different radiomics software were used. Research questions and cohort characteristics varied, ultimately leading to non-comparable outcome results. Conclusion The current twelve published IBSI-compliant PDAC radiomic studies show high variability and often incomplete methodology resulting in low robustness and reproducibility. Clinical relevance statement Radiomics research requires IBSI compliance, data harmonisation, and reproducible feature extraction methods for non-invasive imaging biomarker discoveries to be valid. This will ensure a successful clinical implementation and ultimately an improvement of patient outcomes as part of precision and personalised medicine. Key Points • Current state of radiomics research in pancreatic cancer shows low software compliance to the Image Biomarker Standardisation Initiative (IBSI). • IBSI-compliant radiomics studies in pancreatic cancer are heterogeneous and not comparable, and the majority of study designs showed low reproducibility. • Improved methodology and standardisation of practice in the emerging field of radiomics has the potential of this non-invasive imaging biomarker in the management of pancreatic cancer.

show abstract

Obsolescence of nomograms in radiomics research

D’Antonoli

Mercaldo

2023

Eur Radiol

View full text Add to dashboard Cite

Radiomics model–based algorithm for preoperative prediction of pancreatic ductal adenocarcinoma grade

Cited by 8 publications

References 26 publications

18 F-FDG-PET/CT based deep learning model for fully automated prediction of pathological grading for pancreatic ductal adenocarcinoma before surgery

18 F-FDG-PET/CT based deep learning model for fully automated prediction of pathological grading for pancreatic ductal adenocarcinoma before surgery

Current state of radiomic research in pancreatic cancer: focusing on study design and reproducibility of findings

Obsolescence of nomograms in radiomics research

Contact Info

Product

Resources

About