Radiomics of Multiparametric MRI for Pretreatment Prediction of Pathologic Complete Response to Neoadjuvant Chemotherapy in Breast Cancer: A Multicenter Study

Clinical Cancer Research, Journal Year: 2019, Volume and Issue: 25(12), P. 3538 - 3547

Published: March 6, 2019

We evaluated the performance of newly proposed radiomics multiparametric MRI (RMM), developed and validated based on a multicenter dataset adopting radiomic strategy, for pretreatment prediction pathologic complete response (pCR) to neoadjuvant chemotherapy (NAC) in breast cancer.A total 586 potentially eligible patients were retrospectively enrolled from four hospitals (primary cohort external validation 1-3). Quantitative imaging features extracted T2-weighted imaging, diffusion-weighted contrast-enhanced T1-weighted before NAC each patient. With selected using coarse fine feature selection signatures constructed three sequences their combination. RMM was best signature incorporating with independent clinicopathologic risk factors. The assessed respect its discrimination clinical usefulness, compared that information-based model.Radiomic combining achieved an AUC 0.79 (the highest among signatures). further good performances hormone receptor-positive HER2-negative group triple-negative group. yielded 0.86, which significantly higher than model two cohorts.The study suggested possibility provided potential tool develop predicting pCR cancer.

Language: Английский

---

Predicting EGFR mutation status in lung adenocarcinoma on computed tomography image using deep learning

European Respiratory Journal, Journal Year: 2019, Volume and Issue: 53(3), P. 1800986 - 1800986

Published: Jan. 11, 2019

Epidermal growth factor receptor (EGFR) genotyping is critical for treatment guidelines such as the use of tyrosine kinase inhibitors in lung adenocarcinoma. Conventional identification EGFR genotype requires biopsy and sequence testing which invasive may suffer from difficulty accessing tissue samples. Here, we propose a deep learning model to predict mutation status adenocarcinoma using non-invasive computed tomography (CT). We retrospectively collected data 844 patients with pre-operative CT images, clinical information two hospitals. An end-to-end was proposed by scanning. By training 14 926 achieved encouraging predictive performance both primary cohort (n=603; AUC 0.85, 95% CI 0.83–0.88) independent validation (n=241; 0.81, 0.79–0.83), showed significant improvement over previous studies hand-crafted features or characteristics (p<0.001). The score demonstrated differences EGFR-mutant EGFR-wild type tumours Since routinely used cancer diagnosis, provides easy-to-use method prediction.

Language: Английский

---

Development and validation of an individualized nomogram to identify occult peritoneal metastasis in patients with advanced gastric cancer

Annals of Oncology, Journal Year: 2019, Volume and Issue: 30(3), P. 431 - 438

Published: Jan. 19, 2019

BackgroundOccult peritoneal metastasis (PM) in advanced gastric cancer (AGC) patients is highly possible to be missed on computed tomography (CT) images. Patients with occult PMs are subject late detection or even improper surgical treatment. We therefore aimed develop a radiomic nomogram preoperatively identify AGC patients.Patients and methodsA total of 554 from 4 centers were divided into 1 training, internal validation, 2 external validation cohorts. All patients' PM status was firstly diagnosed as negative by CT, but later confirmed laparoscopy (PM-positive n = 122, PM-negative 432). Radiomic signatures reflecting phenotypes the primary tumor (RS1) peritoneum region (RS2) built predictors 266 quantitative image features. Individualized nomograms incorporating RS1, RS2, clinical factors developed evaluated regarding prediction ability.ResultsRS1, Lauren type significant (all P < 0.05). A these three demonstrated better diagnostic accuracy than model alone net reclassification improvement The area under curve yielded 0.958 [95% confidence interval (CI) 0.923–0.993], 0.941 (95% CI 0.904–0.977), 0.928 0.886–0.971), 0.920 0.862–0.978) for internal, two cohorts, respectively. Stratification analysis showed that this had potential generalization ability.ConclusionCT both nearby significantly associated status. CT has an excellent ability PM, may have implications early AGC.

Language: Английский

---

Deep learning radiomic nomogram can predict the number of lymph node metastasis in locally advanced gastric cancer: an international multicenter study

Annals of Oncology, Journal Year: 2020, Volume and Issue: 31(7), P. 912 - 920

Published: April 15, 2020

•Evaluation of the lymph node metastasis (LNM) is basis individual treatment locally advanced gastric cancer (LAGC).•Deep leaning radiomic nomogram (DLRN) based on CT images can preoperatively determine number LNM in LAGC.•DLRN significantly superior to routinely used clinical N stages, tumor size, and model.•DLRN associated with overall survival LAGC. BackgroundPreoperative evaluation (LAGC). However, preoperative determination method not accurate enough.Patients methodsWe enrolled 730 LAGC patients from five centers China one center Italy, divided them into primary cohort, three external validation cohorts, international cohort. A deep learning was built multiphase computed tomography (CT) for determining We comprehensively tested DLRN compared it state-of-the-art methods. Moreover, we investigated value analysis.ResultsThe showed good discrimination all cohorts [overall C-indexes (95% confidence interval): 0.821 (0.785–0.858) 0.797 (0.771–0.823) 0.822 (0.756–0.887) cohort]. The performed better than model (P < 0.05). Besides, (n = 271).ConclusionA learning-based had predictive In staging-oriented cancer, this could provide baseline information Preoperative enough. analysis. 271).

Language: Английский

---

Quantitative radiomics studies for tissue characterization: a review of technology and methodological procedures

British Journal of Radiology, Journal Year: 2016, Volume and Issue: 90(1070)

Published: Dec. 12, 2016

Quantitative analysis of tumour characteristics based on medical imaging is an emerging field research. In recent years, quantitative features derived from CT, positron emission tomography and MR scans were shown to be added value in the prediction outcome parameters oncology, what called radiomics field. However, results might difficult compare owing a lack standardized methodologies conduct image analyses. this review, we aim present overview current challenges, technical routines protocols that are involved studies. The first issue should overcome dependency several scan acquisition reconstruction parameters. Adopting consistent methods subsequent target segmentation step evenly crucial. To further establish robust analyses, standardization or at least calibration different feature extraction settings required, especially for texture- filter-based features. Several open-source commercial software packages perform currently available, all with slightly functionalities, which makes benchmarking quite challenging. number calculated typically larger than patients studied, emphasizes importance proper selection model-building prevent overfitting. Even though many these challenges still need addressed before can brought into daily clinical practice, expected critical component integration image-derived information personalize treatment future.

Language: Английский

---

Defining the biological basis of radiomic phenotypes in lung cancer

eLife, Journal Year: 2017, Volume and Issue: 6

Published: July 21, 2017

Medical imaging can visualize characteristics of human cancer noninvasively. Radiomics is an emerging field that translates these medical images into quantitative data to enable phenotypic profiling tumors. While radiomics has been associated with several clinical endpoints, the complex relationships radiomics, factors, and tumor biology are largely unknown. To this end, we analyzed two independent cohorts respectively 262 North American 89 European patients lung cancer, consistently identified previously undescribed associations between radiomic features, molecular pathways, factors. In particular, found a relationship immune response, inflammation, survival, which was further validated by immunohistochemical staining. Moreover, number features showed predictive value for specific pathways; example, intra-tumor heterogeneity predicted activity RNA polymerase transcription (AUC = 0.62, p=0.03) intensity dispersion autodegration pathway ubiquitin ligase 0.69, p<10-4). Finally, observed prognostic biomarkers performed highest when combining radiomic, genetic, information (CI 0.73, p<10-9) indicating complementary data. conclusion, demonstrate approaches permit noninvasive assessment both tumors, therefore have potential advance decision-making systematically analyzing standard-of-care images.

Language: Английский

---

Prognostic Value of Deep Learning PET/CT-Based Radiomics: Potential Role for Future Individual Induction Chemotherapy in Advanced Nasopharyngeal Carcinoma

Clinical Cancer Research, Journal Year: 2019, Volume and Issue: 25(14), P. 4271 - 4279

Published: April 11, 2019

We aimed to evaluate the value of deep learning on positron emission tomography with computed (PET/CT)-based radiomics for individual induction chemotherapy (IC) in advanced nasopharyngeal carcinoma (NPC).

Language: Английский

---

In-depth mining of clinical data: the construction of clinical prediction model with R

Annals of Translational Medicine, Journal Year: 2019, Volume and Issue: 7(23), P. 796 - 796

Published: Dec. 1, 2019

Abstract: This article is the series of methodology clinical prediction model construction (total 16 sections this series). The first section mainly introduces concept, current application status, methods and processes, classification models, necessary conditions for conducting such researches problems currently faced. second episode these concentrates on screening method in multivariate regression analysis. third models based Logistic Nomogram drawing. fourth Cox proportional hazards fifth Section calculation C-Statistics logistic model. sixth two common C-Index R. seventh focuses principle Net Reclassification Index (NRI) using eighth IDI (Integrated Discrimination Index) ninth continues to explore evaluation utility after predictive construction: Decision Curve Analysis. tenth a supplement previous Analysis survival outcome data. eleventh discusses external validation twelfth in-depth R, including calculating concordance index discrimination (C-index) data set drawing calibration curve. thirteenth how deal with competitive risk fourteenth draw nomogram fifteenth identification outliers interpolation missing values. sixteenth introduced advanced variable selection linear model, as Ridge LASSO regression.

Language: Английский

---

Quality of science and reporting of radiomics in oncologic studies: room for improvement according to radiomics quality score and TRIPOD statement

European Radiology, Journal Year: 2019, Volume and Issue: 30(1), P. 523 - 536

Published: July 26, 2019

Language: Английский

---

Preoperative prediction of sentinel lymph node metastasis in breast cancer based on radiomics of T2-weighted fat-suppression and diffusion-weighted MRI

European Radiology, Journal Year: 2017, Volume and Issue: 28(2), P. 582 - 591

Published: Aug. 21, 2017

Language: Английский

---