Dataset Information

Deep learning-assisted radiomics facilitates multimodal prognostication for personalized treatment strategies in low-grade glioma.

ABSTRACT: Determining the optimal course of treatment for low grade glioma (LGG) patients is challenging and frequently reliant on subjective judgment and limited scientific evidence. Our objective was to develop a comprehensive deep learning assisted radiomics model for assessing not only overall survival in LGG, but also the likelihood of future malignancy and glioma growth velocity. Thus, we retrospectively included 349 LGG patients to develop a prediction model using clinical, anatomical, and preoperative MRI data. Before performing radiomics analysis, a U2-model for glioma segmentation was utilized to prevent bias, yielding a mean whole tumor Dice score of 0.837. Overall survival and time to malignancy were estimated using Cox proportional hazard models. In a postoperative model, we derived a C-index of 0.82 (CI 0.79-0.86) for the training cohort over 10 years and 0.74 (Cl 0.64-0.84) for the test cohort. Preoperative models showed a C-index of 0.77 (Cl 0.73-0.82) for training and 0.67 (Cl 0.57-0.80) test sets. Our findings suggest that we can reliably predict the survival of a heterogeneous population of glioma patients in both preoperative and postoperative scenarios. Further, we demonstrate the utility of radiomics in predicting biological tumor activity, such as the time to malignancy and the LGG growth rate.

SUBMITTER: Rauch P

PROVIDER: S-EPMC10258197 | biostudies-literature | 2023 Jun

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Deep learning-assisted radiomics facilitates multimodal prognostication for personalized treatment strategies in low-grade glioma.

Rauch P P Stefanits H H Aichholzer M M Serra C C Vorhauer D D Wagner H H Böhm P P Hartl S S Manakov I I Sonnberger M M Buckwar E E Ruiz-Navarro F F Heil K K Glöckel M M Oberndorfer J J Spiegl-Kreinecker S S Aufschnaiter-Hiessböck K K Weis S S Leibetseder A A Thomae W W Hauser T T Auer C C Katletz S S Gruber A A Gmeiner M M

Scientific reports 20230611 1

Determining the optimal course of treatment for low grade glioma (LGG) patients is challenging and frequently reliant on subjective judgment and limited scientific evidence. Our objective was to develop a comprehensive deep learning assisted radiomics model for assessing not only overall survival in LGG, but also the likelihood of future malignancy and glioma growth velocity. Thus, we retrospectively included 349 LGG patients to develop a prediction model using clinical, anatomical, and preopera ...[more]

PMID: 37302994

Similar Datasets

Project description:BackgroundNon-small-cell lung cancer (NSCLC) patients often demonstrate varying clinical courses and outcomes, even within the same tumor stage. This study explores deep learning applications in medical imaging allowing for the automated quantification of radiographic characteristics and potentially improving patient stratification.Methods and findingsWe performed an integrative analysis on 7 independent datasets across 5 institutions totaling 1,194 NSCLC patients (age median = 68.3 years [range 32.5-93.3], survival median = 1.7 years [range 0.0-11.7]). Using external validation in computed tomography (CT) data, we identified prognostic signatures using a 3D convolutional neural network (CNN) for patients treated with radiotherapy (n = 771, age median = 68.0 years [range 32.5-93.3], survival median = 1.3 years [range 0.0-11.7]). We then employed a transfer learning approach to achieve the same for surgery patients (n = 391, age median = 69.1 years [range 37.2-88.0], survival median = 3.1 years [range 0.0-8.8]). We found that the CNN predictions were significantly associated with 2-year overall survival from the start of respective treatment for radiotherapy (area under the receiver operating characteristic curve [AUC] = 0.70 [95% CI 0.63-0.78], p < 0.001) and surgery (AUC = 0.71 [95% CI 0.60-0.82], p < 0.001) patients. The CNN was also able to significantly stratify patients into low and high mortality risk groups in both the radiotherapy (p < 0.001) and surgery (p = 0.03) datasets. Additionally, the CNN was found to significantly outperform random forest models built on clinical parameters-including age, sex, and tumor node metastasis stage-as well as demonstrate high robustness against test-retest (intraclass correlation coefficient = 0.91) and inter-reader (Spearman's rank-order correlation = 0.88) variations. To gain a better understanding of the characteristics captured by the CNN, we identified regions with the most contribution towards predictions and highlighted the importance of tumor-surrounding tissue in patient stratification. We also present preliminary findings on the biological basis of the captured phenotypes as being linked to cell cycle and transcriptional processes. Limitations include the retrospective nature of this study as well as the opaque black box nature of deep learning networks.ConclusionsOur results provide evidence that deep learning networks may be used for mortality risk stratification based on standard-of-care CT images from NSCLC patients. This evidence motivates future research into better deciphering the clinical and biological basis of deep learning networks as well as validation in prospective data.

Project description:BackgroundMultimodal techniques-assisted resection of glioma under general anesthesia (GA) has been shown to achieve similar clinical outcomes as awake craniotomy (AC) in some studies. In this study, we aim to validate the use of multimodal techniques can achieve the maximal safe resection of high-grade glioma involving language areas (HGILAs) under GA.MethodsHGILAs cases were reviewed and collected between January 2009 and December 2020 in our center. Patients were separated into multimodal group (using neuronavigation, intraoperative MRI combined with direct electrical stimulation [DES] and neuromonitoring [IONM]) and conventional group (neuronavigation alone) and clinical outcomes were compared between groups. Studies of HGILAs were reviewed systematically and the meta-analysis results of previous (GA or AC) studies were compared with our results.ResultsFinally, there were 263 patients in multimodal group and 137 patients in conventional group. Compared to the conventional group, the multimodal group achieved the higher median EOR (100% versus 94.32%, P < 0.001) and rate of gross total resection (GTR) (73.8% versus 36.5%, P < 0.001) and the lower incidence of permanent language deficit (PLD) (9.5% versus 19.7%, P = 0.004). The multimodal group achieved the longer median PFS (16.8 versus 10.3 months, P < 0.001) and OS (23.7 versus 15.7 months, P < 0.001) than the conventional group. The multimodal group achieved a higher rate of GTR than the cohorts in previous multimodal studies under GA and AC (73.8% versus 55.7% [95%CI 32.0-79.3%] versus 53.4% [35.5-71.2%]). The multimodal group had a lower incidence of PLD than the cohorts in previous multimodal studies under GA (9.5% versus 14.0% [5.8-22.1%]) and our incidence of PLD was a little higher than that of previous multimodal studies under AC (9.5% versus 7.5% [3.7-11.2%]). Our multimodal group also achieved a relative longer survival than previous studies.ConclusionsSurgery assisted by multimodal techniques can achieve maximal safe resection for HGILAs under GA. Further prospective studies are needed to compare GA with AC for HGILAs.

Dataset Information

Deep learning-assisted radiomics facilitates multimodal prognostication for personalized treatment strategies in low-grade glioma.

Publications

Deep learning-assisted radiomics facilitates multimodal prognostication for personalized treatment strategies in low-grade glioma.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets