scholarly journals Multiple Survival Outcome Prediction of Glioblastoma Patients Based on Multiparametric MRI

2021 ◽  
Vol 11 ◽  
Author(s):  
Bin Wang ◽  
Shan Zhang ◽  
Xubin Wu ◽  
Ying Li ◽  
Yueming Yan ◽  
...  
Keyword(s):  
Clinical Decision Making ◽  
Predictive Accuracy ◽  
Clinical Decision ◽  
Multiparametric Mri ◽  
Cox Proportional Hazards Model ◽  
The Cancer Genome Atlas ◽  
Training Set ◽  
Term Survival ◽  
Mri Features ◽  
Validation Set

PurposeConstruction of radiomics models for the individualized estimation of multiple survival stratification in glioblastoma (GBM) patients using the multiregional information extracted from multiparametric MRI that could facilitate clinical decision-making for GBM patients.Materials and MethodsA total of 134 eligible GBM patients were selected from The Cancer Genome Atlas. These patients were separated into the long-term and short-term survival groups according to the median of individual survival indicators: overall survival (OS), progression-free survival (PFS), and disease-specific survival (DSS). Then, the patients were divided into a training set and a validation set in a ratio of 2:1. Radiomics features (n = 5,152) were extracted from multiple regions of the GBM using multiparametric MRI. Then, radiomics signatures that are related to the three survival indicators were respectively constructed using the analysis of variance (ANOVA) and the least absolute shrinkage and selection operator (LASSO) regression for each patient in the training set. Based on a Cox proportional hazards model, the radiomics model was further constructed by combining the signature and clinical risk factors.ResultsThe constructed radiomics model showed a promising discrimination ability to differentiate in the training set and validation set of GBM patients with survival indicators of OS, PFS, and DSS. Both the four MRI modalities and five tumor subregions have different effects on the three survival indicators of GBM. The favorable calibration and decision curve analysis indicated the clinical decision value of the radiomics model. The performance of models of the three survival indicators was different but excellent; the best model achieved C indexes of 0.725, 0.677, and 0.724, respectively, in the validation set.ConclusionOur results show that the proposed radiomics models have favorable predictive accuracy on three survival indicators and can provide individualized probabilities of survival stratification for GBM patients by using multiparametric and multiregional MRI features.

scholarly journals A CT-based radiomics nomogram for predicting prognosis of coronavirus disease 2019 (COVID-19) radiomics nomogram predicting COVID-19

2021 ◽  
Vol 94 (1117) ◽  
pp. 20200634
Author(s):  
Hang Chen ◽  
Ming Zeng ◽  
Xinglan Wang ◽  
Liping Su ◽  
Yuwei Xia ◽  
...  
Keyword(s):  
Clinical Decision Making ◽  
Predictive Performance ◽  
Clinical Decision ◽  
Ct Images ◽  
Prediction Performance ◽  
Clinical Factors ◽  
Training Set ◽  
Clinical Model ◽  
Potential Biomarker ◽  
Validation Set

Objectives: To identify the value of radiomics method derived from CT images to predict prognosis in patients with COVID-19. Methods: A total of 40 patients with COVID-19 were enrolled in the study. Baseline clinical data, CT images, and laboratory testing results were collected from all patients. We defined that ROIs in the absorption group decreased in the density and scope in GGO, and ROIs in the progress group progressed to consolidation. A total of 180 ROIs from absorption group (n = 118) and consolidation group (n = 62) were randomly divided into a training set (n = 145) and a validation set (n = 35) (8:2). Radiomics features were extracted from CT images, and the radiomics-based models were built with three classifiers. A radiomics score (Rad-score) was calculated by a linear combination of selected features. The Rad-score and clinical factors were incorporated into the radiomics nomogram construction. The prediction performance of the clinical factors model and the radiomics nomogram for prognosis was estimated. Results: A total of 15 radiomics features with respective coefficients were calculated. The AUC values of radiomics models (kNN, SVM, and LR) were 0.88, 0.88, and 0.84, respectively, showing a good performance. The C-index of the clinical factors model was 0.82 [95% CI (0.75–0.88)] in the training set and 0.77 [95% CI (0.59–0.90)] in the validation set. The radiomics nomogram showed optimal prediction performance. In the training set, the C-index was 0.91 [95% CI (0.85–0.95)], and in the validation set, the C-index was 0.85 [95% CI (0.69–0.95)]. For the training set, the C-index of the radiomics nomogram was significantly higher than the clinical factors model (p = 0.0021). Decision curve analysis showed that radiomics nomogram outperformed the clinical model in terms of clinical usefulness. Conclusions: The radiomics nomogram based on CT images showed favorable prediction performance in the prognosis of COVID-19. The radiomics nomogram could be used as a potential biomarker for more accurate categorization of patients into different stages for clinical decision-making process. Advances in knowledge: Radiomics features based on chest CT images help clinicians to categorize the patients of COVID-19 into different stages. Radiomics nomogram based on CT images has favorable predictive performance in the prognosis of COVID-19. Radiomics act as a potential modality to supplement conventional medical examinations.

scholarly journals MRI-Based Radiomics Nomogram To Predict Intraoperative Hemorrhage of Placenta Previa

2021 ◽  
Author(s):  
Ye Song ◽  
Liping Zhu ◽  
Dali Chen ◽  
Yongmei Li ◽  
Qi Xi ◽  
...  
Keyword(s):  
Predictive Accuracy ◽  
Placenta Previa ◽  
Preoperative Evaluation ◽  
Curve Analysis ◽  
Training Set ◽  
Decision Curve Analysis ◽  
Preoperative Prediction ◽  
Mri Features ◽  
Wide Range ◽  
Validation Set

Abstract Background: Placenta previa is associated with higher percentage of intraoperative and postpartum hemorrhage, increased obstetric hysterectomy, significant maternal morbidity and mortality. We aimed to develop and validate a magnetic resonance imaging (MRI)-based nomogram to preoperative prediction of intraoperative hemorrhage (IPH) for placenta previa, which might contribute to adequate assessment and preoperative preparation for the obstetricians.Methods: Between May 2015 and December 2019, a total of 125 placenta previa pregnant women were divided into a training set (n = 80) and a validation set (n = 45). Radiomics features were extracted from MRI images of each patient. A MRI-based model comprising seven features was built for the classification of patients into IPH and non-IPH groups in a training set and validation set. Multivariate nomograms based on logistic regression analyses were built according to radiomics features. Receiver operating characteristic (ROC) curve was used to assess the model. Predictive accuracy of nomogram were assessed by calibration plots and decision curve analysis. Results: In multivariate analysis, placenta position, placenta thickness, cervical blood sinus and placental signals in the cervix were signifcantly independent predictors for IPH (all p < 0.05). The MRI-based nomogram showed favorable discrimination between IPH and non-IPH groups. The calibration curve showed good agreement between the estimated and the actual probability of IPH. Decision curve analysis also showed a high clinical benefit across a wide range of probability thresholds. The AUC was 0.918 ( 95% CI, 0.857-0.979 ) in the training set and 0.866( 95% CI, 0.748-0.985 ) in the validation set by the combination of four MRI features.Conclusions: The MRI-based nomograms might be a useful tool for the preoperative prediction of IPH outcomes for placenta previa. Our study enables obstetricians to perform adequate preoperative evaluation to minimize blood loss and reduce the rate of caesarean hysterectomy.

scholarly journals An easy-to-operate web-based calculator for predicting the progression of chronic kidney disease

2021 ◽  
Vol 19 (1) ◽  
Author(s):  
Qian Xu ◽  
Yunyun Wang ◽  
Yiqun Fang ◽  
Shanshan Feng ◽  
Cuiyun Chen ◽  
...  
Keyword(s):  
Chronic Kidney Disease ◽  
Kidney Disease ◽  
Clinical Decision Making ◽  
Area Under The Curve ◽  
Progression Free Survival ◽  
Clinical Decision ◽  
Training Set ◽  
Web Based ◽  
Free Survival ◽  
Validation Set

Abstract Background This study aimed to establish and validate an easy-to-operate novel scoring system based on simple and readily available clinical indices for predicting the progression of chronic kidney disease (CKD). Methods We retrospectively evaluated 1045 eligible CKD patients from a publicly available database. Factors included in the model were determined by univariate and multiple Cox proportional hazard analyses based on the training set. Results Independent prognostic factors including etiology, hemoglobin level, creatinine level, proteinuria, and urinary protein/creatinine ratio were determined and contained in the model. The model showed good calibration and discrimination. The area under the curve (AUC) values generated to predict 1-, 2-, and 3-year progression-free survival in the training set were 0.947, 0.931, and 0.939, respectively. In the validation set, the model still revealed excellent calibration and discrimination, and the AUC values generated to predict 1-, 2-, and 3-year progression-free survival were 0.948, 0.933, and 0.915, respectively. In addition, decision curve analysis demonstrated that the model was clinically beneficial. Moreover, to visualize the prediction results, we established a web-based calculator (https://ncutool.shinyapps.io/CKDprogression/). Conclusion An easy-to-operate model based on five relevant factors was developed and validated as a conventional tool to assist doctors with clinical decision-making and personalized treatment.

scholarly journals Multimodal imaging patterns predict survival in recurrent glioblastoma patients treated with bevacizumab

2016 ◽  
Vol 18 (12) ◽  
pp. 1680-1687 ◽  
Author(s):  
Ken Chang ◽  
Biqi Zhang ◽  
Xiaotao Guo ◽  
Min Zong ◽  
Rifaquat Rahman ◽  
...  
Keyword(s):  
Clinical Decision Making ◽  
Progression Free Survival ◽  
Clinical Decision ◽  
Recurrent Glioblastoma ◽  
Machine Learning Techniques ◽  
Imaging Biomarkers ◽  
Imaging Features ◽  
Mri Features ◽  
Fluid Attenuated Inversion Recovery ◽  
Multimodal Mri

Abstract Background Bevacizumab is a humanized antibody against vascular endothelial growth factor approved for treatment of recurrent glioblastoma. There is a need to discover imaging biomarkers that can aid in the selection of patients who will likely derive the most survival benefit from bevacizumab. Methods The aim of the study was to examine if pre- and posttherapy multimodal MRI features could predict progression-free survival and overall survival (OS) for patients with recurrent glioblastoma treated with bevacizumab. The patient population included 84 patients in a training cohort and 42 patients in a testing cohort, separated based on pretherapy imaging date. Tumor volumes of interest were segmented from contrast-enhanced T1-weighted and fluid attenuated inversion recovery images and were used to derive volumetric, shape, texture, parametric, and histogram features. A total of 2293 pretherapy and 9811 posttherapy features were used to generate the model. Results Using standard radiographic assessment criteria, the hazard ratio for predicting OS was 3.38 (P < .001). The hazard ratios for pre- and posttherapy features predicting OS were 5.10 (P < .001) and 3.64 (P < .005) for the training and testing cohorts, respectively. Conclusion With the use of machine learning techniques to analyze imaging features derived from pre- and posttherapy multimodal MRI, we were able to develop a predictive model for patient OS that could potentially assist clinical decision making.

Abstract W P266: A Simple Score for Prediction of Outcome after Cerebral Venous Thrombosis

Stroke ◽  
2015 ◽  
Vol 46 (suppl_1) ◽  
Author(s):  
Miguel A Barboza ◽  
Erwin Chiquete ◽  
Antonio Arauz ◽  
Jonathan Colín ◽  
Alejandro Quiroz-Compean ◽  
...  
Keyword(s):  
Decision Making ◽  
Venous Thrombosis ◽  
Predictive Value ◽  
Cerebral Venous Thrombosis ◽  
Goodness Of Fit ◽  
Clinical Decision Making ◽  
Case Fatality ◽  
Clinical Decision ◽  
Classification Systems ◽  
Cox Proportional Hazards Model

Background and purpose: Cerebral venous thrombosis (CVT) not always implies a good prognosis. There is a need for robust and simple classification systems of severity after CVT that help in clinical decision-making. Methods: We studied 467 patients (81.6% women, median age: 29 years, interquartile range: 22-38 years) with CVT who were hospitalized from 1980 to 2014 in two third-level referral hospitals. Bivariate analyses were performed to select variables associated with 30-day mortality to integrate a further multivariate analysis. The resultant model was evaluated with the Hosmer-Lemeshow test for goodness of fit, and on Cox proportional hazards model for reliability of the effect size. After the scale was configured, security and validity were tested for 30-day mortality and modified Rankin scale (mRS) >2. The prognostic performance was compared with that of the CVT risk score (CVT-RS, 0-6 points) as the reference system. Results: The 30-day case fatality rate was 8.7%. The CVT grading scale (CVT-GS, 0-9 points) was integrated by stupor/coma (4 points), parenchymal lesion >6 cm (2 points), mixed (superficial and deep systems) CVT (1 point), meningeal syndrome (1 point) and seizures (1 point). CVT-GS was categorized into mild (0-3 points, 1.1% mortality), moderate (4-6 points, 19.6% mortality) and severe (7-9 points, 61.4% mortality). For 30-day mortality prediction, as compared with CVT-RS (cut-off 4 points), CVT-GS (cut-off 5 points) was globally better in sensitivity (85% vs 37%), specificity (90% vs 95%), positive predictive value (44% vs 40%), negative predictive value (98% vs 94%), and accuracy (94% vs 80%). For 30-day mRS >2 the performance of CVT-GS over CVT-RS was comparably improved. Conclusion: The CVT-GS is a simple and reliable score for predicting outcome that may help in clinical decision-making and that could be used to stratify patients recruited into clinical trials.

Development and validation of a multiparametric MRI-based radiomics signature for distinguishing between indolent and aggressive prostate cancer

2021 ◽  
pp. 20210191
Author(s):  
Liuhui Zhang ◽  
Donggen Jiang ◽  
Chujie Chen ◽  
Xiangwei Yang ◽  
Hanqi Lei ◽  
...  
Keyword(s):  
Prostate Cancer ◽  
Multiparametric Mri ◽  
Curve Analysis ◽  
Training Set ◽  
Threshold Probability ◽  
Roc Curve Analysis ◽  
Aggressive Prostate Cancer ◽  
Therapeutic Decisions ◽  
Validation Set ◽  
Radiomics Signature

Objectives: To develop and validate a noninvasive MRI-based radiomics signature for distinguishing between indolent and aggressive prostate cancer (PCa) prior to therapy. Methods: In all, 139 qualified and pathology-confirmed PCa patients were divided into a training set (n = 93) and a validation set (n = 46). A total of 1576 radiomics features were extracted from the T2WI (n = 788) and DWI (n = 788) for each patient. The Select K Best and the least absolute shrinkage and selection operator (LASSO) regression algorithm were used to construct a radiomics signature in the training set. The predictive performance of the radiomics signature was assessed in the training set and then validated in the validation set by receiver operating characteristic (ROC) curve analysis. We computed the calibration curve and the decision curve to evaluate the calibration and clinical usefulness of the signature. Results: nine radiomics features were identified to form the radiomics signature. The radiomics score (Rad-score) was significantly different between indolent and aggressive PCa (p < 0.001). The radiomics signature exhibited favorable discrimination between the indolent and aggressive PCa groups in the training set (AUC: 0.853, 95% CI: 0.766 to 0.941) and validation set (AUC: 0.901, 95% CI: 0.793 to 1.000). The decision curve analysis showed that a greater net benefit would be obtained when the threshold probability ranged from 20 to 90%. Conclusions: The multiparametric MRI-based radiomics signature can potentially serve as a noninvasive tool for distinguishing between indolent and aggressive PCa prior to therapy. Advances in knowledge: The multiparametric MRI-based radiomics signature has the potential to noninvasively distinguish between the indolent and aggressive PCa, which might aid clinicians in making personalized therapeutic decisions.

scholarly journals Genome-Wide Identification of a Novel Autophagy-Related Signature for Colorectal Cancer

Dose-Response ◽  
2019 ◽  
Vol 17 (4) ◽  
pp. 155932581989417 ◽  
Author(s):  
Zhi Huang ◽  
Jie Liu ◽  
Liang Luo ◽  
Pan Sheng ◽  
Biao Wang ◽  
...  
Keyword(s):  
Colorectal Cancer ◽  
Signaling Pathway ◽  
Risk Score ◽  
Low Risk ◽  
Training Data ◽  
The Cancer Genome Atlas ◽  
Training Set ◽  
Data Set ◽  
Validation Set ◽  
Cox Analysis

Background: Plenty of evidence has suggested that autophagy plays a crucial role in the biological processes of cancers. This study aimed to screen autophagy-related genes (ARGs) and establish a novel a scoring system for colorectal cancer (CRC). Methods: Autophagy-related genes sequencing data and the corresponding clinical data of CRC in The Cancer Genome Atlas were used as training data set. The GSE39582 data set from the Gene Expression Omnibus was used as validation set. An autophagy-related signature was developed in training set using univariate Cox analysis followed by stepwise multivariate Cox analysis and assessed in the validation set. Then we analyzed the function and pathways of ARGs using Gene Ontology and Kyoto Encyclopedia of Genes and Genomes (KEGG) database. Finally, a prognostic nomogram combining the autophagy-related risk score and clinicopathological characteristics was developed according to multivariate Cox analysis. Results: After univariate and multivariate analysis, 3 ARGs were used to construct autophagy-related signature. The KEGG pathway analyses showed several significantly enriched oncological signatures, such as p53 signaling pathway, apoptosis, human cytomegalovirus infection, platinum drug resistance, necroptosis, and ErbB signaling pathway. Patients were divided into high- and low-risk groups, and patients with high risk had significantly shorter overall survival (OS) than low-risk patients in both training set and validation set. Furthermore, the nomogram for predicting 3- and 5-year OS was established based on autophagy-based risk score and clinicopathologic factors. The area under the curve and calibration curves indicated that the nomogram showed well accuracy of prediction. Conclusions: Our proposed autophagy-based signature has important prognostic value and may provide a promising tool for the development of personalized therapy.

A Four Protein Plasma Fingerprint of Acute Graft Versus Host Disease (GVHD) Predicts Long Term Survival.

Blood ◽  
2007 ◽  
Vol 110 (11) ◽  
pp. 38-38 ◽  
Author(s):  
Sophie Paczesny ◽  
Sung Choi ◽  
Thomas Braun ◽  
Carrie Kitko ◽  
Krijanovski Oleg ◽  
...  
Keyword(s):  
Acute Gvhd ◽  
Strong Association ◽  
Conditioning Regimen ◽  
Target Organ ◽  
Training Set ◽  
Term Survival ◽  
Long Term Survival ◽  
Validation Set

Abstract There are no independent laboratory diagnostic tests for acute GVHD. We first identified 7 potential plasma biomarkers (IL-2R-α, CRP, IL-8, TIMP-1, TNFR1, HGF, CA-19.9) of acute GVHD using a combination of proteomic approaches and antibody microarrays. We next conducted a retrospective analysis using plasma samples from 424 patients at the University of Michigan under IRB approval. We obtained samples at the first clinical signs of acute GVHD prior to treatment and at equivalent time points in patients without GVHD (Table 1). The median duration of follow-up was 420 days with a minimum follow-up of 180 days. Patients with veno-occlusive disease, idiopathic pneumonia syndrome, or septic shock were not included. We measured plasma levels of the 7 proteins by sequential ELISA. Logistic regression models with and without leave-one-out-cross-validation (LOOCV) tested the correlation of the laboratory values with the diagnosis of acute GVHD using area under the receiver-operating-characteristic (ROC) curves (AUC). The training set consisted of 282 randomly selected patients; the validation set included the remaining 142 patients. The final optimal fingerprint of four proteins excluded CRP because of its association with non-specific inflammation and included IL-2R-α, TNFR1, IL-8 and HGF, with AUCs of 0.91 and 0.89 in the training set (without and with LOOCV, respectively) and 0.86 in the validation set. The fingerprint had a strong association with grade of GVHD (p&lt;0.001) and target organ (p=0.002) at onset; interestingly, HGF had the strongest association. Using a predicted probability of acute GVHD of at least 50%, the fingerprint had a 72% sensitivity and 89% specificity. When we categorized the predicted risk of acute GVHD into low (0.00–0.59), moderate (0.60–094) and high (0.95–1.00), the plasma fingerprint predicted long-term survival (Figure 1, p&lt;0.001). We conclude that this plasma protein fingerprint has good sensitivity, high specificity, strong association with initial grade and target organ of acute GVHD, and effectively stratifies patients into three risk groups for GVHD that correlate with long term survival. Figure Figure Table 1: Patients characteristics GVHD- (N=242) GVHD+ (N=182) Age-yr Median (range) 45 (1–69) 49 (1–71) Donor type (%) MRD: 169 (70%) MRD: 105 (58%) URD: 73 (30%) URD: 77 (42%) Conditioning regimen Intensity (%) Full: 182 (75%) Full: 114 (63%) Reduced: 60 (25%) Reduced: 68 (37%) Day after BMT of samples : median (range) 30 (7–104) 29 (5–119) Grade at GVHD Onset (%) Grade 0 Grade 1 Grade 2 Grade 3–4 242 (57%) 48 (12%) 100 (24%) 34 (7%) Organ Target at GVHD Onset (%) n/a Skin Gut Liver Combined 119 (65%) 38 (21%) 7 (4%) 18 (10%)

Hsa-miR-31-3p expression in FFPE tumor samples as a predictor of anti-EGFR response in patients with metastatic colorectal cancer.

2013 ◽  
Vol 31 (15_suppl) ◽  
pp. 3562-3562 ◽  
Author(s):  
Gilles Manceau ◽  
Jean-Baptiste Bachet ◽  
Benoist Chibaudel ◽  
Francois Liebaert ◽  
Raphaële Thiébaut ◽  
...  
Keyword(s):  
Colorectal Cancer ◽  
Metastatic Colorectal Cancer ◽  
Cox Proportional Hazards Model ◽  
Expression Level ◽  
Prognostic Impact ◽  
Wild Type ◽  
Training Set ◽  
Ffpe Samples ◽  
Validation Set ◽  
Third Line

3562 Background: In metastatic colorectal cancer (mCRC), KRAS mutations are associated with resistance to anti-EGFR antibodies. To identify, in wild-type KRAS mCRC patients, markers that predict response to anti-EGFR antibodies, we focused on miRNAs. Methods: Expression profile of 1145 miRNAs was done on 84 colorectal tumors and 5 normal colon mucosae. Correlations between miRNAs expression level and survival were based on frozen samples of a training set from a retrospective series of 33 patients treated by cetuximab and irinotecan and of two validation set from prospective collections of 38 patients treated by cetuximab or panitumumab based chemotherapy or by panitumumab and irinotecan as third-line, using an adjusted Cox proportional hazards model. Validation on FFPE samples was done on 39 patients treated with panitumumab and irinotecan as third-line. Results: A predictive signature of 11 miRNA linked to the PFS was identified (p<0.01) but only one, hsa-miR-31-3p, exhibited significant different expression level between tumor from bad prognosis and good prognosis from the training set. We tested expression of this miRNA on the training set, and found a HR of 1.9 CI95% [1.1-2.9]. In validation set, the prognostic impact of hsa-miR-31-3p the HR was estimated to 1.9 CI95% [1.1-3.1]. Using multivariate model obtained from the training set to the two validation set, we predict the PFS of the patients (accuracy of prediction: AUC = 0.77). We classified the validation set according to a free PFS score (P=0.005) with a specificity of 62% CI95% [38%-82%] and a sensitivity of 82% CI95% [56%-96%] for the prediction model. A nomogram, established taking into account hsa-miR-31-3p expression level, predicted the progression risk (P<0.0001). Confirmation of the predictive value of hsa-miR-31-3p expression on survival risk progression was done on FFPE sample (p=0.0006). Conclusions: This is the first tool to select individual patients with a wild-type KRAS tumor for anti-EGFR therapy from frozen or FFPE samples.

scholarly journals SGPNet: A Three-Dimensional Multitask Residual Framework for Segmentation and IDH Genotype Prediction of Gliomas

2021 ◽  
Vol 2021 ◽  
pp. 1-9
Author(s):  
Yao Wang ◽  
Yan Wang ◽  
Chunjie Guo ◽  
Shuangquan Zhang ◽  
Lili Yang
Keyword(s):  
Brain Tumor ◽  
Clinical Decision Making ◽  
Three Dimensional ◽  
Clinical Decision ◽  
Error Rates ◽  
The Cancer Genome Atlas ◽  
Classification Error ◽  
Tumor Segmentation ◽  
Brain Tumor Segmentation ◽  
The Status

Glioma is the main type of malignant brain tumor in adults, and the status of isocitrate dehydrogenase (IDH) mutation highly affects the diagnosis, treatment, and prognosis of gliomas. Radiographic medical imaging provides a noninvasive platform for sampling both inter and intralesion heterogeneity of gliomas, and previous research has shown that the IDH genotype can be predicted from the fusion of multimodality radiology images. The features of medical images and IDH genotype are vital for medical treatment; however, it still lacks a multitask framework for the segmentation of the lesion areas of gliomas and the prediction of IDH genotype. In this paper, we propose a novel three-dimensional (3D) multitask deep learning model for segmentation and genotype prediction (SGPNet). The residual units are also introduced into the SGPNet that allows the output blocks to extract hierarchical features for different tasks and facilitate the information propagation. Our model reduces 26.6% classification error rates comparing with previous models on the datasets of Multimodal Brain Tumor Segmentation Challenge (BRATS) 2020 and The Cancer Genome Atlas (TCGA) gliomas’ databases. Furthermore, we first practically investigate the influence of lesion areas on the performance of IDH genotype prediction by setting different groups of learning targets. The experimental results indicate that the information of lesion areas is more important for the IDH genotype prediction. Our framework is effective and generalizable, which can serve as a highly automated tool to be applied in clinical decision making.

Sign in / Sign up

Export Citation Format

Share Document